Servicio de Información Comunitario sobre Investigación y Desarrollo - CORDIS

FP7

DESICOS Report Summary

Project reference: 282522
Funded under: FP7-SPACE

Final Report Summary - DESICOS (New Robust DESIgn Guideline for Imperfection Sensitive COmposite Launcher Structures)

Executive Summary:
European Space industry demand for lighter and cheaper launcher transport systems. The project DESICOS contributed to these aims by a new design approach for imperfection sensitive compo-site launcher structures, exploiting the worst imperfection idea efficiently: the Single Pertur-bation Load Approach. Currently, imperfection sensitive shell structures prone to buckling are designed according the NASA SP 8007 guideline using the conservative lower bound curve. The guideline dates from 1968, and the structural behaviour of composite material is not considered appropriately, in particular since the imperfection sensitivity of shells made from such materials depends on the lay-up design. The buckling loads of CFRP structures may vary by a factor of about 3 just by changing the lay-up. This is not considered in the NASA SP 8007, which allows designing only so called "black metal" structures. Here is a high need for a new precise and fast design approach for imperfection sensitive composite structures which allows significant reduction of structural weight and design cost. For most relevant architectures of cylindrical and conical launcher structures (monolithic, sandwich - without and with holes) DESICOS investigated a combined methodology from the Single Perturbation Load approach and a specific stochastic approach which guarantees an effective and robust design. Investigations demonstrated, that an axially loaded unstiffened cylinder, which is disturbed by a large enough single perturbation load, is leading directly to the design buckling load 45% higher compared with the respective NASA SP 8007 design. All results and experience is summarized in a handbook for the design of imperfection sensitive composite launcher structures. The potential is demonstrated within 2 industrially driven use cases.
By getting the objectives DESICOS contributes to reduce launcher weight, development time, design and manufacturing costs, and to increase launcher capacity.
The main DESICOS results can be summarised as following:
1) Benchmarking results:
a. Collection of all worldwide existing papers to buckling experiments
b. Imperfection data base with existing measurements
c. ABQUS plug-in for improved modelling and evaluation of cylindrical and conical structures with different loads, boundary conditions, cut-outs, imperfections, …
2) Experimental data base on:
a. Material properties of different materials used in the project
b. Manufacturing of structures
c. Buckling experiments
3) New design approaches:
a. Modelling and analyses
b. New design approaches
c. Validation and application of the design approaches
4) Design and analysis handbook
a. Design and analysis handbook
b. Industrial validation

The main results were published in 30 peer-reviewed papers (see www.desicos.eu). All the results generated are used by the industrial partners. One can summarize that the application of analysis based design methods, using different approaches to represent the imperfections, seems to lead to less conservative KDFs than those obtained by the NASA SP. However, additional studies are needed to collect imperfection data of the real structures, and how these imperfections should be represented in an efficient way.

Project Context and Objectives:
Objectives
Launcher design requires fast and precise prediction of structural weight as well its weight distribution already in the early design phase, because in that phase different concepts of the whole launcher system have to be evaluated in order to identify the optimal one. Prediction of weight and its distribution have to be fast, because a great number of different system configu-rations must be considered in combination with all the other disciplines contributing to the launcher system design (system design loop). The prediction has to be precise, because less re-liable ones might lead to basic changes, later in the detailed design phase, which might also influence the design of the whole system. Such changes in later design phases are extremely costly in terms of time and money, they definitely have to be avoided. This aspect is of particular importance for innovative concepts with not sufficient background from experience. If the system design loop has come to an optimum result for the system configuration, the structural part of this configuration has to be optimised without change of the relations to the whole system (structural design loop). The structural design loop is part of the detailed design phase, it is followed by a detailed proof of the structural performance and reliability.
The project DESICOS contributes to the aims of the Call in the main by a new approach for robust design of fibre composite launcher structures, to be applied within the system and the structural design loops. As circular cylindrical shapes are dominating launcher structures, the focus of DESICOS was on this type of structures, however, cones were also considered in order to check the applicability of the basic DESICOS results to them.
The dimensioning criterion with the design of launcher structures is buckling not before ultimate load, thus they do not have an exploitable postbuckling area, as in comparison with the majority of aircraft structures. The most critical aspect for numerical buckling prediction is the structure’s sensitivity to imperfections, in particular to imperfections of shape and loading (deviations from perfect shape and from perfect loading distribution), the so-called traditional ones. This is characteristic of unstiffened or skin-dominant structures which are typically used in launcher structures. Nevertheless, also the non-traditional imperfections (deviations from the nominal values of the structural parameters stiffness, thickness, etc.) must be taken into account, in particular for CFRP structures. Currently, imperfection sensitive shell structures prone to buckling are designed according the NASA SP 8007 guideline dating from 1968. This guideline can be applied in a fast way, however it is extremely conservative and not precise enough for an optimal structure. It summarizes the results of a multitude of buckling tests with different types of structures, different quality of manufacturing and different quality of testing. The lowest buckling loads are taken as reference for designs, independent of the type of structure and the quality of manufacturing. So this guideline penalizes optimal structural types and best quality manufacturing. Moreover and most importantly, it does not consider appropriately the real struc-tural behaviour of composite materials, in particular since buckling load and imperfection sensitivity of CFRP shells depend on the lay-up (stacking). It has been numerically and experimentally shown that the buckling loads of axially compressed cylindrical composite structures may vary by a factor of about 3 just by changing the lay-up, and that structures optimised against buckling reveal high sensitivity to imperfections. All this is not considered in the NASA SP 8007, which was developed for metallic structures and allows only designing of "black metal" structures. “Black metal” means in this context a copy of typical metal designs and disregarding the influence of stacking by homogenisation – ignoring its relevant impact on load carrying capacities and stiffness tailoring of composites. There are no effective design guidelines for imperfection sensitive composite launcher structures prone to buckling, and so there is a high need for a new, improved and fast one which allows significant reduction of structural weight and costs.
For that purpose, DESICOS developed for the relevant architectures of cylindrically shaped primary composite launcher structures (monolithic, sandwich) design guidelines based on the “Single Perturbation Load Approach”, which are combined with the “Specific Stochastic Ap-proach” in order to guarantee a robust and in any case safe design. Investigations demonstrated on scaled structured, that the combination of these two approaches applied to an axially loaded unstiffened cylinder is leading directly to the design buckling load 45% higher compared with the respective NASA SP 8007 design. This increased allowable in buckling load corresponds approximately to 20% weight reduction, if the load is kept constant, and to a reduction of costs by about 15%. For conically shaped structures DESICOS evaluates the applicability of this combined approach, too. DESICOS considered the aforementioned primary space structures. A further aspect, DESICOS contributes to reduce operational costs for inspection and revalidation by monitoring the structural health and performance of reusable launchers, an actual ESA objective. It uses the experiments performed to validate the design guidelines to define a reliable procedure how to apply the Vibration Correlation Technique (extended Southwell method) to launcher structures; a technique which can predict experimentally the buckling load of manufactured structures without buckling it to evaluate their, so called “as manufactured” perfor-mance. All results are summarized in a handbook for the design of imperfection sensitive composite launcher structures. The potential is demonstrated within 2 industrially driven use cas-es.
By getting the objectives DESICOS contributes
- to reduce launcher weight,
- to reduce development time
- to reduce design and manufacturing costs, and
- to increase launcher capacity.

One can summarize that the application of analysis based design methods, using different ap-proaches to represent the imperfections, seems to lead to less conservative KDFs than those obtained by the NASA SP. However, additional studies are needed to collect imperfection data of the real structures, and how these imperfections should be represented in an efficient way.

Results
The main DESICOS results can be summarised as following.
WP 1 Benchmarking on selected structures with existing methods
The buckling behaviour of different imperfection sensitive structures was be simulated, taking into account the current and the Single Perturbation Approach. In order to select appropriate existing benchmarks (structure tests) for assessment of approach evaluation purposes, as well as to compile in detail the abilities and deficiencies of the approaches, benchmarking exercises were performed. The work started in Task 1.1 with selecting of real structure tests used by the industri-al partners or from literature. In Task 1.2 the partners harmonise the modelling aspects and ap-proaches (e.g. type and number of finite elements, boundary conditions, sandwich modelling, etc.) in order allow comparable results. In Task 1.3, the partners involved performed analyses on the benchmarks selected. They applied the Single Perturbation Approach and stochastic analyses as well as - for the purpose of comparison – the NASA SP8007 and the hierarchical approach in order to find out the limits as well as abilities and deficiencies of these approaches. In addition, the partner PFH developed an imperfection data base, which collects all available test results from the partners.
WP 2 Material characterisation and design of structures for buckling tests
This workpackage is focused on the identification of material properties and the design of imper-fection sensitive structures that shall be manufactured and tested in workpackage 4. The Task 2.1 on manufacturing and testing of specimens was shifted to reporting period 2 to be closer to the manufacturing of the large test structures. In this task DLR, RTU and TECHNION characterised the material properties of the monolithic specimens manufactured of UD IM7/8552 UD. The part-ners used different standards. TECHNION characterised also the ROHACELL 200WF core mate-rial. The specimens were manufactured by DLR, RTU and GRIPHUS. Task 2.2 started with the collection of all design constraints of test structures under consideration of benchmark results and accounting for the limitations of existing concepts. The partners distinguished between two types of designs: typical industrial structures and validation structures. The validation structures were designed to test specific limiting aspects of application of the design to be validated, with examples including small or large imperfection sensitivity. Industrial structures were designed with regard to industrial applications, mainly by existing procedures and requirements used in day-to-day industrial design practice. For these structures there exist usually multi-objective requirements concerning weight, load-carrying capacity and costs. Design parameters such as overall dimensions or lay-ups (i.e. stacking sequence and ply orientations) were chosen according to the design processes. All industrial partners provided constraints and suggestions according to their daily manufacturing and design practice and communicate them to the testing partners. In this context also the requirements of VCT was defined and taken into account. Furthermore, the industrial partners were directly involved in trade-off and refinement processes.
WP 3 Development and application of improved design approaches
Within Task 3.1 Modelling and analyses the partners considered a quite diverse field of methods and structures to simulate reliably the buckling phenomena of monolithic, stiffened and un-stiffened cylindrical and conical structures, partly including holes and openings. Furthermore, the contributions include different numerical analysis concepts and experimental methods. The contributions of this task consider state-of-the-art methods and novel approaches which are part of active research. In particular, they document modeling and analysis aspects of classical design problems in space engineering and compare traditional concepts and numerically sophisticated modeling and analysis techniques.
In Task 3.2 Development of new design approaches the partners developed new approchaes based on the SPLA and the stochastic approach. Various contributions of the partners considered a quite diverse field of methods, structures and concepts to simulate reliably the buckling phenomena of monolithic, stiffened and un-stiffened cylinders and conical structures, partly including holes and openings. Furthermore, the contributions include different numerical analysis concepts and experimental methods. The coordination of the various contributions was organized such that a bilateral exchange of experience and results supported the evaluation of the proposed design approaches.
In Task 3.3 the design concepts were validated by experiments. The application and validation data was chosen to evaluate best the single contributions of the partners and went beyond the test data proposed within the DESICOS framework. It is worth to mention that for some numerically analyzed structures no test scenarios exist, e.g. stiffened cylindrical structures. For these structures a corresponding verification of the investigated design approach was ensured by results from literature.
WP 4 Manufacture, inspection and testing of structures designed in WP 2
Within Task 4.1, the various manufacturers manufactured different shell type specimens to be tested under axial compression. GRIPHUS manufactured six sandwich shells and two sandwich cones tested by POLIMI and the TECHNION. DLR manufactured two monolithic laminated composite shells and two monolithic cones, beside a metal stringer stiffened shell, later tested by DLR. RTU manufactured and tested two cylindrical shells, one monolithic laminated composite and one metallic one, beside a series of shells further manufactured to test various capacities of their equipment. All the manufactured specimens were inspected and their initial geometric imperfections and their thickness distributions were mapped for further analysis. Within Task 4.2 the partners, DLR, RTU, POLIMI and TECHNION tested the various specimens using various equipment and their testing experience.
WP 5 Design handbook and industrial validation
In Task 5.1 Design and analysis handbook the partners summarized all results and experience. This could appear common to say that good design for aeronautic and space structures should ovoid structures which are too much imperfection sensitive, because they are particularly difficult to size with the good margin for being safe without being not too conservative. Several means are known to let the structures less imperfection sensitive, as to add longitudinal stiffeners, circular frames, and the combination of both, to pressurize enough the structural tank.

In Task 5.2 Industrial validation two industrial use cases studied by ASTRIUM ST – F were con-sidered. The goal of this task is to compare on real structures classical methods commonly applied in space industry (modal defect, axisymmetric defect) and new methods proposed during the DESICOS project (SPLA). These analysis based design have showed some improvement of KDFs compared to the NASA SP 8007. However, there are still open questions which are the best way to apply for real space structures.

The main results were published in 30 peer-reviewed papers (see www.desicos.eu). All the results generated are used by the industrial partners. One can summarize that the application of analysis based design methods, using different approaches to represent the imperfections, seems to lead to less conservative KDFs than those obtained by the NASA SP. However, additional studies are needed to collect imperfection data of the real structures, and how these imperfections should be represented in an efficient way.

Project Results:
1. Overview of the project objectives
1.1 Overview of the project objectives
Launcher design requires fast and precise prediction of structural weight as well its weight distribution already in the early design phase, because in that phase different concepts of the whole launcher system have to be evaluated in order to identify the optimal one. Prediction of weight and its distribution have to be fast, because a great number of different system configu-rations must be considered in combination with all the other disciplines contributing to the launcher system design (system design loop). The prediction has to be precise, because less re-liable ones might lead to basic changes, later in the detailed design phase, which might also influence the design of the whole system. Such changes in later design phases are extremely costly in terms of time and money, they definitely have to be avoided. This aspect is of particular importance for innovative concepts with not sufficient background from experience. If the system design loop has come to an optimum result for the system configuration, the structural part of this configuration has to be optimised without change of the relations to the whole system (structural design loop). The structural design loop is part of the detailed design phase, it is followed by a detailed proof of the structural performance and reliability.
The project DESICOS contributes to the aims of the Call in the main by a new approach for robust design of fibre composite launcher structures, to be applied within the system and the structural design loops. As circular cylindrical shapes are dominating launcher structures, the focus of DESICOS was on this type of structures, however, cones were also considered in order to check the applicability of the basic DESICOS results to them.
The dimensioning criterion with the design of launcher structures is buckling not before ultimate load, thus they do not have an exploitable postbuckling area, as in comparison with the majority of aircraft structures. The most critical aspect for numerical buckling prediction is the structure’s sensitivity to imperfections, in particular to imperfections of shape and loading (deviations from perfect shape and from perfect loading distribution), the so-called traditional ones. This is characteristic of unstiffened or skin-dominant structures which are typically used in launcher structures. Nevertheless, also the non-traditional imperfections (deviations from the nominal values of the structural parameters stiffness, thickness, etc.) must be taken into account, in particular for CFRP structures. Currently, imperfection sensitive shell structures prone to buckling are designed according the NASA SP 8007 guideline dating from 1968. This guideline can be applied in a fast way, however it is extremely conservative and not precise enough for an optimal structure. It summarizes the results of a multitude of buckling tests with different types of structures, different quality of manufacturing and different quality of testing. The lowest buckling loads are taken as reference for designs, independent of the type of structure and the quality of manufacturing. So this guideline penalizes optimal structural types and best quality manufacturing. Moreover and most importantly, it does not consider appropriately the real struc-tural behaviour of composite materials, in particular since buckling load and imperfection sensitivity of CFRP shells depend on the lay-up (stacking). It has been numerically and experimentally shown that the buckling loads of axially compressed cylindrical composite structures may vary by a factor of about 3 just by changing the lay-up, and that structures optimised against buckling reveal high sensitivity to imperfections. All this is not considered in the NASA SP 8007, which was developed for metallic structures and allows only designing of "black metal" structures. “Black metal” means in this context a copy of typical metal designs and disregarding the influence of stacking by homogenisation – ignoring its relevant impact on load carrying capacities and stiffness tailoring of composites. There are no effective design guidelines for imperfection sensitive composite launcher structures prone to buckling, and so there is a high need for a new, improved and fast one which allows significant reduction of structural weight and costs.
For that purpose, DESICOS developed for the relevant architectures of cylindrically shaped primary composite launcher structures (monolithic, sandwich) design guidelines based on the “Single Perturbation Load Approach”, which are combined with the “Specific Stochastic Ap-proach” in order to guarantee a robust and in any case safe design. Investigations demonstrated on scaled structured, that the combination of these two approaches applied to an axially loaded unstiffened cylinder is leading directly to the design buckling load 45% higher compared with the respective NASA SP 8007 design. This increased allowable in buckling load corresponds approximately to 20% weight reduction, if the load is kept constant, and to a reduction of costs by about 15%. For conically shaped structures DESICOS evaluates the applicability of this combined approach, too. DESICOS considered the aforementioned primary space structures. A further aspect, DESICOS contributes to reduce operational costs for inspection and revalidation by monitoring the structural health and performance of reusable launchers, an actual ESA objective. It uses the experiments performed to validate the design guidelines to define a reliable procedure how to apply the Vibration Correlation Technique (extended Southwell method) to launcher structures; a technique which can predict experimentally the buckling load of manufactured structures without buckling it to evaluate their, so called “as manufactured” perfor-mance.
By getting the objectives DESICOS contributes to reduce launcher weight, development time, design and manufacturing costs, and to increase launcher capacity.
The main DESICOS results can be summarised as following:
1) Benchmarking results
2) Experimental data base
3) New design approaches
4) Design and analysis handbook

The main results were published in 30 peer-reviewed papers (see www.desicos.eu). All the results generated are used by the industrial partners. One can summarize that the application of analysis based design methods, using different approaches to represent the imperfections, seems to lead to less conservative KDFs than those obtained by the NASA SP. However, additional studies are needed to collect imperfection data of the real structures, and how these imperfections should be represented in an efficient way.

1.2 Overview of the project objectives task by task
According the DoW the objectives were all achieved and can be summarised as follows:
- WP 1: Benchmarking on selected structures with existing methods (Deliverable D1.1)
• Task 1.1: Choice of benchmarks
• Task 1.2: Harmonisation of modelling aspects
• Task 1.3: Benchmarking on structures
- WP 2: Material characterisation and design of structures for buckling tests (Deliverable D2.1)
• Task 2.1: Characterisation of material properties (shifted to reporting period 2)
• Task 2.2: Design and analysis of test structures (Deliverable D2.2)
- WP 3: Development and application of improved design approaches
• Task 3.1: Modelling and analyses (Deliverable D3.1, D3.2 and D3.3)
• Task 3.2: Development of new design approaches (Deliverable D3.4 and D3.5)
• Task 3.3: Validation and application of the design concepts (Deliverable D3.6 and D3.7)
- WP 4: Manufacture, inspection and testing of structures designed in WP 2
• Task 4.1: Manufacture of the test structures and non-destructive inspection (Deliv-erable D.4.1 and D4.2)
• Task 4.2: Buckling tests (Deliverable D4.3 and D4.4)
- WP 5: Design handbook and industrial validation
• Task 5.1: Design and analysis handbook (Deliverable D5.1 and D5.2)
• Task 5.2: Industrial validation (Deliverable D5.3)
- WP 8: Exploitation and dissemination
• Task 8.1: Exploitation and dissemination (Deliverable D8.1, D8.2 and D8.3)
2. Work progress and achievements during the period
2.1 Update of the work plan
In general there are no changes to the original description of work concerning project structure, milestones deliverables or partners. The only change was the extension of the project duration by 6 months. The reason is that the testing activities could not start on time to the delay of the test activities of a different EU project. In the first reporting period the test matrix of the test structures in WP 4 and the distribution of the activities related to the new methods in WP 3 was slightly modified.
Updated project structure and milestones
There are no changes of the structure and milestones, except the project extension by 6 months.
2.2 Summary of progress towards objectives and details for each task
In general there are no changes to the original description of work. The expected objectives and results were achieved. The project achieved in the workpackages the following final results.
WP 1 Benchmarking on selected structures with existing methods
The buckling behaviour of different imperfection sensitive structures was be simulated, taking into account the current and the Single Perturbation Approach. In order to select appropriate existing benchmarks (structure tests) for assessment of approach evaluation purposes, as well as to compile in detail the abilities and deficiencies of the approaches, benchmarking exercises were performed. The work started in Task 1.1 with selecting of real structure tests used by the industri-al partners or from literature. In Task 1.2 the partners harmonise the modelling aspects and ap-proaches (e.g. type and number of finite elements, boundary conditions, sandwich modelling, etc.) in order allow comparable results. In Task 1.3, the partners involved performed analyses on the benchmarks selected. They applied the Single Perturbation Approach and stochastic analyses as well as - for the purpose of comparison – the NASA SP8007 and the hierarchical approach in order to find out the limits as well as abilities and deficiencies of these approaches. In addition, the partner PFH developed an imperfection data base, which collects all available test results from the partners. All results are summarised in Deliverable D1.1.
WP 2 Material characterisation and design of structures for buckling tests
This workpackage is focused on the identification of material properties and the design of imper-fection sensitive structures that shall be manufactured and tested in workpackage 4. The Task 2.1 on manufacturing and testing of specimens was shifted to reporting period 2 to be closer to the manufacturing of the large test structures. In this task DLR, RTU and TECHNION characterised the material properties of the monolithic specimens manufactured of UD IM7/8552 UD. The part-ners used different standards. TECHNION characterised also the ROHACELL 200WF core mate-rial. The specimens were manufactured by DLR, RTU and GRIPHUS. All results are summarised in Deliverable D2.1. Task 2.2 started with the collection of all design constraints of test structures under consideration of benchmark results and accounting for the limitations of existing concepts. The partners distinguished between two types of designs: typical industrial structures and valida-tion structures. The validation structures were designed to test specific limiting aspects of application of the design to be validated, with examples including small or large imperfection sensitivity. Industrial structures were designed with regard to industrial applications, mainly by existing procedures and requirements used in day-to-day industrial design practice. For these structures there exist usually multi-objective requirements concerning weight, load-carrying capacity and costs. Design parameters such as overall dimensions or lay-ups (i.e. stacking sequence and ply orientations) were chosen according to the design processes. All industrial partners provided constraints and suggestions according to their daily manufacturing and design practice and communicate them to the testing partners. In this context also the requirements of VCT was defined and taken into account. Furthermore, the industrial partners were directly involved in trade-off and refinement processes. All results are summarised in Deliverable D2.2.
WP 3 Development and application of improved design approaches
Task 3.1: Modelling and analyses
The objective of this task was the development of modeling and analysis strategies for the computation of buckling loads. Starting point of this task were the results of the benchmark tests of workpackage 1 and the design of the test structures as specified in workpackage 2. Furthermore, there was a strong interrelation with the test results of workpackage 4 which provided the necessary reference test data for validation purposes. The results of this workpackage and related tasks are further documented in workpackage 5.
The contributions of the partners considered a quite diverse field of methods and structures to simulate reliably the buckling phenomena of monolithic, stiffened and un-stiffened cylindrical and conical structures, partly including holes and openings. Furthermore, the contributions include different numerical analysis concepts and experimental methods.
The findings of task 3.1 are considered as an extension to the handbook Buckling of structures in space engineering, ESA Requirements and Standards Division, Noordwijk, The Netherlands, 2010 which provides a standard guideline for stability and buckling analysis in space engineering.
The contributions of this workpackage consider state-of-the-art methods and novel approaches which are part of active research. In particular, they document modeling and analysis aspects of classical design problems in space engineering and compare traditional concepts and numerically sophisticated modeling and analysis techniques.
The partner contributions addressing modeling and analysis strategies include the following as-pects:
• selection and application of model specific parameters: this aspect includes the modeling of laminate composite shells including material specification, the modeling and influence behavior of essential and natural boundary conditions. Modeling examples were considered for stiffened and un-stiffened cylinders including the specific modeling aspects with regard to stiffeners and to smeared models (see e.g. contributions RWTH Aachen, TU Delft, RTU). Furthermore, convergence studies were considered to reveal an optimal mesh size in terms of a trade-off between sufficient accuracy and model size (see e.g. contributions DLR, TU Delft) and to provide a clear picture of the convergence properties of the state-of-the-art algorithms provided in Abaqus.
• deterministic modeling of imperfections: the single perturbation load approach (SPLA) as a focal point of innovation in imperfection modeling was applied and studied under the aspect of practical modeling issues. The characteristics of this modeling strategy for geometric imperfections were critically evaluated throughout workpackage 3. The application of the SPLA was partly documented in this subtask to prepare the utilization of the approach within the novel design strategies proposed in task 3.2 (see e.g. contribution LUH). Furthermore, linear mode imperfections, axis-symmetric Fourier series–based imperfection models and multiple load imperfections were considered and compared with the SPLA (see e.g. contributions DLR, RWTH). The considered structures included cylindrical and conical shells.
• probabilistic analyses utilizing stochastic methods for different imperfection types: this aspect includes a broad overview about available tools for probabilistic analyses and a de-tailed discussion of the performance of the utilized stochastic methods. Beside the classical modeling strategies for geometric imperfections other approaches e.g. evolutionary and spectral methods were considered (see e.g. contribution CRC-ACS). The probabilistic modeling aspects further included other imperfection types such as thickness imperfections, load imperfections, ply stacking and material imperfections (see e.g. contributions, CRC-ACS, TU Delft).
• application of a pre- and post-processing plug-in for Abaqus, developed by PFH and DLR: the developed plug-in provides a user-friendly graphical user interface for the specification and analysis of laminate composite shells considering imperfection modeling. It further supports the presentation of the analysis results and facilitates the overall modeling and analysis pipeline for numerical simulations of composite shell structures with Abaqus (contribution DLR, PFH). The application was documented with a tutorial-like hand-book.
• documentation of the modeling pipeline of a reduced order model for the reliable analysis of snap-back phenomena of thin shells: the Koiter-Newton method (contribution TU Delft) is a hybrid reduced order model which addresses the issue of analysis efficiency in terms of a reduced computational effort and a reliable and accurate analysis result. Beside an overview about the simulation pipeline of the Koiter-Newton model, test results and a convergence study document the principal behavior of the method.
Based on the documented analysis results of this sub-task knock-down factors were partly derived and compared to established analytic/empirical formulas to provide the complete modeling, analysis and design chain.

All detailed results are summarised in Deliverable D3.3.
Task 3.2: Development of new design approaches
Deliverable D3.6 marks the endpoint of Task 3.2 - Development of new design approaches. The objective of this task was the development of a new robust design approach and variants addressing the buckling of imperfection sensitive composite structures. The new design approaches are essentially a combination of the single perturbation load approach considering geometric imperfections and a stochastic approach which takes other imperfections into account.
Starting point of the development of this task were the results of the benchmark tests of workpackage 1 and the design of the test structures as specified in workpackage 2. Furthermore, there was a strong interrelation with the test results of workpackage 4 which provided the necessary reference test data for validation purposes. The results of this workpackage and related tasks were further documented in workpackage 5.
The various contributions of the partners considered a quite diverse field of methods, structures and concepts to simulate reliably the buckling phenomena of monolithic, stiffened and un-stiffened cylinders and conical structures, partly including holes and openings. Furthermore, the contributions include different numerical analysis concepts and experimental methods. The coordination of the various contributions was organized such that a bilateral exchange of experience and results supported the evaluation of the proposed design approaches according to Figure 1.

Figure 1: Organisation of the development of improved design approaches in WP 3

The principal ideas, concepts and design & analysis strategies considered by the partners can be summarized as follows:
• simplified consideration of geometric imperfections with a single perturbation load ap-proach (SPLA): the idea of the SPLA to model geometric imperfections reliably and inde-pendent of the knowledge of a true geometric imperfection pattern was at the heart of this workpackage. In particular, a separate consideration of this approach independent of a sto-chastic consideration of other imperfections was intended to provide an additional knock-down factor which can be combined with the knock-down factor retrieved for other imper-fections considered in the analysis model. The contributions of all partners provide a sys-tematic analysis for a large variety of different geometry configurations, including cylindrical and conical shells, partly with openings and holes, different structural aspects including stiffened and un-stiffened shells and different boundary conditions. A major issue discussed in the contributions and repeatedly evaluated with the many analysis results was the reliability of the approach for practical purposes in the buckling design of shells. This discussion involved accuracy aspects with regard to test results, aspects on the representation of the SPLA of the worst imperfection, aspects on the application with regard to the correct location of the perturbation load, in particular for conical structures and structures with holes and openings, and finally aspects on the conservativeness of the applied approach in comparison with measured imperfections and other, empirical design formulae (see e.g. contributions LUH, TU Delft, RWTH Aachen, DLR, PFH, RTU, POLIMI).
• a study on the mechanical response of cylindrical shells with regard to an increasing single perturbation load: with this study (contribution PFH) a physical interpretation of the perturbation load was found to reveal the principal buckling mechanism for this imperfection case. Furthermore, the influence of low perturbation loads, length variations of the modeled cylinder and the influence of changing boundary condition on the buckling response was tested (contribution LUH) and evaluated to identify further characteristics of the proposed approach.
• stochastic analysis strategies to consider various imperfections: mainly Monte-Carlo based methods were considered and variants were tested or combined to consider imperfections in the analysis model, including geometric imperfections, load imperfections, material and ply-stacking imperfections and thickness imperfections (see e.g. contributions POLIMI, CRC-ACS, TU Delft). Based on the simulation results sensitivity analyses revealed the most dominant imperfection type and its contribution to a reduction of the knock-down factor. For conical structures with and without holes and openings various semi-vertex angles specifying the surface inclination were considered and systematically evaluated with regard to the buckling load and the reduction of the KDF (contributions CRC-ACS, POLIMI).
• derivation and evaluation of different design strategies based on the deterministic and sto-chastic analysis results in terms of modified knock-down factors: the consideration of a separate knock-down factor for deterministically derived analysis results considering geo-metric imperfections and for stochastically derived analysis results, considering either geometric imperfections or other imperfections, was one of the major issues to be addressed in this workpackage and in particular in this subtask. Various combinations for a robust design rule of cylindrical structures were considered including e.g.
o two independently derived knock-down factors considering geometric imperfections on the basis of the SPLA or on the basis of measured imperfection models and a stochastically derived knock-down factor representing all other imperfections. A simple multiplicative concatenation of the knock-down factors was tested and compared with the conservative design of the NASA-SP guideline (contributions TU Delft, RWTH Aachen, DLR, PFH, RTU).
o a semi-analytical approach (contribution LUH) included the SPLA into a probabilistic analysis instead of a separate consideration.
• design rules for stringer stiffened shell models: for cylindrical stiffened shells a smeared model approach was tested to derive a suitable knock-down factor (see contribution RWTH, TU Delft) and compared with established analytical design rules. The suitability of the SPLA for stringer stiffened shells was studied with test cylinders and compared to models considering measured geometric imperfection.
• predictions on basis of the vibration correlation technique: in contrast to the numerical methods mainly considered in this workpackage the vibration correlation technique is an experimental approach considering the correlation of natural vibrations of the shell struc-ture and axial loading to determine the buckling load (contribution Technion). This tech-nique was revisited as an alternative non-destructive approach for the proposed test cylin-ders of workpackage 2. The potential, the drawbacks and limitations of the method were carefully documented with several test examples. The derived buckling loads provided a first indicator for the validation of the numerical models.
All detailed results are summarised in Deliverable D3.6.

Task 3.3: Validation and application of the design concepts

Deliverable D3.8 marks the endpoint of Task 3.3 – Validation and application of the design ap-proaches. The objective of this task was to unify the ideas and developments of task 3.2 and to evaluate the design approaches in terms of their reliability, accuracy and applicability. Task 3.3 considered the test structures, specified in workpackage 2 and tested within workpackage 4. Fur-thermore, a number of available test structures and benchmark problems from previous projects and literature were used to verify numerical and empirical developments and to validate the applied modeling strategies. The results of this workpackage and related tasks are further documented in workpackage 5.
The application and validation data within this subtask was chosen to evaluate best the single contributions of the partners and went beyond the test data proposed within the DESICOS framework. It is worth to mention that for some numerically analyzed structures no test scenarios exist, e.g. stiffened cylindrical structures. For these structures a corresponding verification of the investigated design approach was ensured by results from literature. The application of the new developed design approaches and used methods was partly integrated into the final report of Task 3.2, Deliverable D14 (Development of new design approaches).
The contributions of this subtask can be summarized as follows:
• stochastic results for unstiffened cylindrical shells were validated with test results. A split analysis of the different considered imperfections was performed to reveal the most severe imperfection type and to identify the single contributions in the reduction of the final design load (see e.g. contribution CRC-ACS). The stochastic analysis for all considered imperfection types showed an improved knock-down factor (0.55) compared to the knock-down factor of the NASA SP 8007 guideline (0.32). In the same study it was revealed that loading imperfections are the most relevant ones leading to a highest reduction of the design load.
• evaluation of stochastic results for truncated conical structures: the SPLA derived knock-down factor was found to be significantly lower than the corresponding result of the sto-chastic analysis for small semi-vertex angles (<30°) specifying the inclination angle of the conical surface. For larger semi-vertex angles the SPLA result was in the range of the stochastic results, independent of the chosen lay-up of the composite shell (contribution CRC-ACS).
• identification of the range of validity of the SPLA for different lengths of the conical structures and different vertex angles.
• conical structure with measured mid-surface, thickness and fiber fraction imperfections: two cones were tested (see contribution DLR) and the numerical prediction was validated: an improved knock-down factor by roughly a factor two was found in comparison with the knock-in factor proposed by the NASA-SP 8007 guideline. Furthermore, it was observed that the SPLA result matches well the experiment performed without perturbation load (as-built) and that the SPLA stayed at a slightly higher level in the range of the experimental results with perturbation load.
• validation of the SYLDA structure: the experimental test result were significantly lower than the predicted numerical results from a stochastic analysis. Several indicators for this discrepancy were identified and discussed (see contribution POLIMI).
• application of the probabilistic perturbation load approach: the knock-down factors derived from the probabilistic perturbation load approach were found to be in the range of the NASA-SP 8007 design guideline (see e.g. contributions TU Delft, LUH). The validation revealed a conservative design load compared to the test results. However, the probabilistic perturbation load approach led in general a slightly less conservative design than the NASA-SP 8007 design.
The high diversity of structures considered in DESICOS and the many different modeling & analysis strategies and design lines followed by the partners, correspondingly revealed diverse results in terms of appropriateness and suitability of the initially proposed design rule for a robust design. The conceptual ideas behind the proposed approach, including a deterministically derived factor which represents geometric imperfections on the basis of the single perturbation load approach and another, stochastically derived factor representing other imperfections, aimed at improving the design concept proposed in the NASA-SP 8007 design guideline for cylindrical launcher structures.
Based on the results and findings of the project partners some conclusions can be drawn as follows:
• the basic idea to replace a measured geometric imperfection model with a dimple imperfection from a single perturbation load is desirable and facilitates the modeling of geometric imperfections significantly.
• the single perturbation load leads to an essential and representative geometric imperfection, however, in general the generated imperfection cannot be considered as the worst imperfection which limits the generality of the approach. Basically, the SPLA generated imperfection can be considered as a relevant imperfection.
• the applicability of the SPLA seems reasonable for thin metallic and laminate composite structures. The applicability for sandwich structures and stiffened structures is of limited relevance since their structural properties destroy the principal effect of generating a dimple imperfection on a thin surface.
• for stiffened structures the SPLA derived knock-down factors matched well the factors proposed by Almroth (ρ90).
• geometric imperfections such as holes or openings may dominate the imperfection model with effect on the relevance of the SPLA for the buckling response. The effect and the need of the SPLA for such structures need to be further investigated.
• knock-down factors derived from the SPLA, in general, are significantly higher compared with the knock-down factors of the NASA-SP 8007 guideline. Other imperfections such as loading imperfections, thickness imperfections or lay-up imperfections, to mention a few, must be considered to obtain a representative knock-down factor for robust buckling de-sign.
• the combination of two knock-down factors representing geometric imperfections from the SPLA and other imperfections confirms in some cases the initially proposed concept leading to a less conservative design, but fails in other cases which prevents a general conclusion about the reliability and applicability of the approach. Future research should investigate more sophisticated strategies for the concatenation of independently derived knock-down factors. Furthermore, the number of numerically predicted and validated test samples for each structural type must be significantly increased to confirm the observations to date.
• stochastic models considering various types of imperfections, in general, lead to less con-servative knock-down factors compared to the NASA-SP 8007 guideline but show some sensitivity on the input imperfection data, the sample size and the specified probability level. Despite the robustness of stochastic methods in terms of a reasonable prediction, their applicability in industrial design scenarios is limited due to a high modeling effort and a high computational cost.
• both knock-down factors, derived from the SPLA and from stochastic analyses, are inde-pendent of the R/t-ratio used in the NASA-SP 8007 guideline, thus they are not subject to geometric restrictions.
• the vibration correlation technique seems to be a promising approach to obtain buckling loads in a non-destructive way from the experiment. A first empirical approach was devel-oped. It needs to be further investigated and validated in future activities.

The majority of simulation scenarios considered in workpackage 3 led to an improved design compared with the established, commonly accepted but overly conservative NASA-SP 8007 design guideline. The consideration of the most predominant imperfection types for the considered shell structures revealed a significant potential for a more efficient design in terms of stability and reliability, worth to be investigated further in future. The numerical modeling and analysis schemes developed and applied in DESICOS are capable to capture comprehensively and accurately the dominant model properties which influence buckling and stability. Further model characteristics, e.g. influences of adjacent structures and resulting boundary conditions, should be considered in future research for an improved and robust structural design.

All detailed results are summarised in Deliverable D3.8.

WP 4 Manufacture, inspection and testing of structures designed in WP 2

Task 4.1: Manufacture of the test structures and non-destructive inspection
Within Task 4.1, the various manufacturers manufactured different shell type specimens to be tested under axial compression. Griphus manufactured six sandwich shells and two sandwich cones tested by POLIMI and the Technion.
DLR manufactured two monolithic laminated composite shells and two monolithic cones, beside a metal stringer stiffened shell, later tested by DLR.
RTU manufactured and tested two cylindrical shells, one monolithic laminated composite and one metallic one, beside a series of shells further manufactured to test various capacities of their equipment.
All the manufactured specimens were inspected and their initial geometric imperfections and their thickness distributions were mapped for further analysis.
Task 4.2: Buckling tests
Within Task 4.2 the partners, DLR, RTU, POLIMI and TECHNION tested the various specimens using various equipment and their testing experience.
DLR tested their specimens using their loading test rig, applying both the ARAMIS system and strain gages to collect the strains distribution across the loaded shells and cones. End-shortening vs. the axial compression curves were generated for cones and shells and the buckling loads were determined. DLR applied experimentally the single perturbation load approach (SPLA) method by applying a lateral small load, and investigated its influence on the buckling loads of the various tested specimens. Consistent results were obtained for the shells and the cones tested, and in good correlation with the predicted buckling loads.
RTU designed and manufactured a special test rig to test the monolithic composite shell simulating the ISS+JAVE structure. They used both laser scanning and photogrammetry to scan the specimen both for geometric imperfection and thickness variation. RTU tested this specimen and other specimens from composite and metal materials, using their ARAMIS system and strain gages. They applied the single perturbation load approach method –the SPLA method, on their tested specimens, showing that the method is not always feasible. They showed that applying a lateral load of 200 N for the ISS+JAVE shell type structure did not influence its buckling behavior.
POLIMI tested their four sandwich shell type structures using their loading rig. Strain gages were bonded back-to-back on the shell surface to monitor the load distribution along the circumference and to define the buckling load. The end shortening of the shells were determined using LVDTs, and the buckling load was detected from the end-shortening vs. compressive applied load curve, yielding consistent results. The sandwich shell type structures buckling behavior was characterized by a total collapse of the shell with de-bonding between the skin and the core and breakage of the skin along the circumference of the tested specimen. The complete destruction of the specimen after the application of the compressive load, prevented the re-use of the shell after removing the applied load, which is normally characteristic to monolithic composite cylindrical shell. Therefore the intended application of the SPLA method was not applicable for these cases.
TECHNION tested their four sandwich type structures two shell and two cones, using their loading rig. Strain gages were bonded back-to-back on the shell surface to monitor the load distribution along the circumference and to define the buckling load. The end shortening of the shells were determined using LVDTs, and the buckling load was detected from the end-shortening vs. compressive applied load curve, yielding consistent results. The same behavior experienced by POLIMI for their tested shells, namely a total collapse of the shell or the cone with de-bonding between the skin and the core and breakage of the skin along the circumference of the tested specimen, was also found during the tests being performed at the TECHNION. The complete destruction of the specimen after the application of the compressive load, prevented the re-use of the shell after removing the applied load, which is normally characteristic to monolithic composite cylindrical shell. Therefore the intended application of the SPLA method was not applicable for these cases. The influence of the axial compression on the natural frequency of the tested specimens was measured using the hammer method. This was used to non-destructively predict the buckling loads of the specimens.
Another important activity which was conducted in the framework of WP4, was to apply the VCT (Vibration Correlation Technique) approach to non-destructively determine the buckling loads of thin walled monolithic composite shells. A close cooperation among DLR, RTU and TECHNION lead to a series of tests being performed on shell Z36, manufactured and tested by DLR. The tests were performed at the DLR Braunschweig premises, with RTU equipment and the participation of the RTU, TECHNION and DLR personnel, yielding very consistent and promising results.
Based on the work performed within WP4 the following conclusions can be drawn:
• Various types of thin walled structures (shells and cones) were manufactured and non-destructive inspections were performed.
• Geometrical initial imperfections were measured to be included in FE codes for better predictions of the buckling behavior of the specimens.
• The SPLA method for determining a new less conservative knock-down curve was applied for different types of shell type structures.
• Shell type structures, made of metal and laminated composite materials were tested under axial compression loading.
• Shell and cone type structures, made of sandwich and laminated composite materials were tested under axial compression loading, showing interested collapse modes.
• The VCT approach was applied on various types of shell type structures, showing promising results. More work should be done to yield an effective tool to non-destructively predict in-situ buckling of thin walled structures.

All results are summarised in the Deliverables D4.2 and D4.4.

WP 5 Design handbook and industrial validation

Task 5.1: Design and analysis handbook

I) About the “new” methods

This is mainly for the two “new” methods particularly deeply studied within the DESICOS project, the SPLA method, the stochastic method, or some combination of both.

SPLA:

• SPLA proved in the project to represent only geometrical imperfections.
• SPLA is representing the global buckling which is not the worst imperfection as for in-stance local buckling may occur first. It has to be find out in the future activities if repre-senting the global buckling by SPLA is sufficient for the design.
• Loading imperfections are not covered by SPLA.
• For sandwich structures:

o SPLA was studied on small test structures. The numerical studies led to clear re-sults. But it could not be confirmed experimentally on cylindrical sandwich speci-mens, because the structures failed too early by material damage after the first buckling test without applying SPLA.
o Numerical study performed for industrial Use case is not so clear.
o Thus, the application of SPLA for real sandwich structures is therefore not clear enough.

Stochastic method:

• This method seems more powerful to represent the physics.
• This method usually requires sufficient computational effort, higher than other methods.
• Also, the difficult issue is to know (or to choose a priori) the stochastic values for the dif-ferent parameters.
• To simplify this issue, as the DESICOS study has shown that the geometric imperfections are mostly the dominant parameter for imperfection sensitive structures, even when composite structures, there could be a proposal to limit to treat as stochastic the geometric parameter, and to apply an additional Knock Down Factor for taking account all other parameters. For that, DESICOS gives results to help to choose some adequate additional KDF for the application in the projects.
• However, despite such possible simplification, the resulting KDF would still depend on the choice of the ratio a / t (where “a” is the amplitude of the imperfections, and “t” the thickness). That means the same difficulty than for the methods using modal or axisymmetric imperfections. So the advantage to use complex stochastic method, compared to more simple ones method is not obvious. Current more simple methods are nonlinear analysis with modal shape pattern for geometric imperfections, or axisymmetric pattern when more appropriate (in particular if well-defined thickness transition or singularity along the meridian).
• In fact, to establish a reliable design of a sensitive structure without too much con-servatism, it should be known the pattern and amplitude of the most significant geometric imperfections induced by the manufacturing process of this particular structure. This has to be measured on a batch including an enough number of real manufactured specimens.

II) About the possible improvement of the NASA KDF, for less conservative design

• The application of analysis based design methods, using different approaches to represent the imperfections, seems to lead to less conservative KDFs than those obtained by the NASA SP.
• Additional studies are needed to collect imperfection data of the real structures, and how these imperfections should be represented in an efficient way.

III) Design recommendations

This could appear common to say that, and not always possible to apply in the projects, but good design for aeronautic and space structures should ovoid structures which are too much imperfection sensitive, because they are particularly difficult to size with the good margin for being safe without being not too conservative.

Several means are known to let the structures less imperfection sensitive, as to add longitudinal stiffeners, circular frames, and the combination of both, to pressurize enough the structural tank.

All results are summarised in Deliverable D5.2.

Task 5.2: Industrial validation

This task summarizes the industrial use cases studied by ASTRIUM ST – F in the frame of the DESICOS project. The goal of this task is to compare on real structures classical methods com-monly applied in space industry (modal defect, axisymmetric defect) and new methods proposed during the DESICOS project (SPLA). Two use cases are studied: the SYLDA5 E/CA and the Bare Tank A5ME. These analysis based design have showed some improvement of KDFs compared to the NASA SP 8007. However, there are still open questions which are the best way to apply for real space structures.

Conclusion of the buckling analysis of the SYLDA5

This section has presented the different classical ways to analyse the buckling behaviour of the SYLDA5 cylinder. The most critical numerical KDF is 0.66, obtained by taking into account a modal defect. The gain compared to NASA KDF is only 8%. It is too low to remove one ply in the design of the composite shell and to expect a mass gain. About SPLA, this method gives a KDF of 0.51 if considering the first transition point. The convergence seems to be obtained by about 0.30, for unrealistic perturbation load. The application of SPLA to this example of the cylinder of the SYLDA5 is not clear.

Conclusion of the buckling analysis of the bare tank A5ME

This section has presented the different classical ways to analyse the buckling behaviour of the cylinder of the Bare Tank A5ME. The numerical KDF the most critical is 0.47. This is obtained by a non-linear analysis, taking into account a modal defect, with a chosen realistic amplitude a/t equal to 1.4. The gain compared to NASA KDF (0.34) is 38%. This result of the present study is consistent with previous AST-F studies which have shown some conservatism in the NASA KDF for unstiffened tank cylinders at zero or low pressure (here the studied case). About SPLA, at low level of the perturbation load, this method gives KDF 0.68, this point being defined by a clear transition. But the confidence in this result may be questionable. But there is no clear convergence when continuing to increase sufficiently the perturbation load. The application of SPLA to this example of the cylinder is also not clear.

All results are summarised in Deliverable D5.3.
2.3 Deviations from the project workprogramme and corrective actions
Except the extension of the project duration the objectives were all achieved according the DoW.
2.4 List of deliverables and milestones
All deliverables were finished by the end of the project.

D1.1 Benchmarking on selected structures with existing methods
D2.1 Data base of material properties
D2.2 Design and analysis of test structures
D3.1 Modelling and analyses (First preliminary report)
D3.2 Modelling and analyses (Second preliminary report)
D3.3 Modelling and analyses
D3.4 New design approaches (First preliminary report)
D3.5 New design approaches (Second preliminary report)
D3.6 New design approaches
D3.7 Validation and application of the design approaches (Preliminary report)
D3.8 Validation and application of the design approaches
D4.1 Manufacture of the test structures and non-destructive inspection (Preliminary report)
D4.2 Manufacture of the test structures and non-destructive inspection
D4.3 Buckling tests (Preliminary report)
D4.4 Buckling tests
D5.1 Design and analysis handbook (Preliminary report)
D5.2 Design and analysis handbook
D5.3 Industrial validation
D6.1 Final report
D7.1 Final cost statement
D8.1 Project presentation
D8.2 Plan for using and disseminating knowledge
D8.3 participation and awareness (users group, internet and conference)

The following table summarizes the list of all milestones which are directly linked to deliverables. All activities linked to the milestones were achieved.

Milestone number Milestone name WP number Lead benefi-ciary number Delivery date Comments
MS1 MS1 WP1, WP2 1 12 D1.1, D2.1 and D2.2 are finished
MS2 MS2 WP3, WP4 1 18 D3.1, D3.3 and D4.1 are finished.
MS3 MS3 WP3, WP4 1 36 D3.2, D3.4, D3.7, D4.2
and D4.3 are finished.
MS4 MS4 WP3, WP4, WP5 1 42 D3.3, D3.5, D3.8, D4.4 and D5.1 are finished.
MS5 MS5 WP5,WP6, WP7 1 42 All deliverables are finished

3. Project management during the period
3.1 Consortium management tasks and achievements
All coordination tasks due within the were performed according to the planning as laid down in the Description of Work. The workpackages were co-ordinated by the WP leaders as planned. There were no major problems and the management tasks were in general achieved as planned. The project was controlled by the regular project meetings as well as by the internal reports and deliverables. Details to all meetings and co-ordination activities are mentioned in Section 3.3.
3.2 List of project meetings, dates and venues
During the reporting period the following meetings were organised or attended. The table shows also all meetings.

3.3 Project planning and status
There are no deviations from the planned milestones and deliverables. There are also no changes to the legal status of any of the beneficiaries.
3.4 Project website
The project website www.inicop.org is updated regularly. It uses improved using web 2.0 technol-ogies which allows the partners and other interested users as NASA to upload new information by their own.
3.5 Comments and information on co-ordination activities
The project started on 1 February 2012. The Kick-off Meeting was held on 28-29 March 2012 in Braunschweig (Germany) with DLR as host. Every 6 month there was a project meeting at which all partners were presented. One additional workpackage meeting on WP 3 are organised at TU-Delft. Between the meetings numerous working contacts among the partners happened on bilateral basis. Communication among the partners and with the EC is predominantly performed by e-mail. In order to alleviate communication, a ‘Partner Communication Table’ with partner, representative(s), e-mail, phone and fax numbers, and address as entries was prepared and distributed. It is updated periodically. In order to alleviate common data keeping and exchange of big data sets like reports or test results, DLR established a team site with very specific rights of access. The server is confined to the DESICOS partners only.
DLR co-operates in parallel with NASA. It regulated an exchange of experience and test results related to the DESICOS project. The agreement was signed in 2014 and will end 2016, however, it is planned to extend it by 2020.
Prof. Degenhardt spent a research semester at NASA in Langley (USA) from December 2012 until March 2013. He co-ordinated the project via e-mail from that place.
On 23-25 March 2015 DLR organised the 3rd Int. Conf. on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures with a DESICOS workshop as final event of the DESICOS project.
During the reporting period many papers on DESICOS results were presented by partners at inter-national conferences and full papers appeared in the international journals. Links to all papers are provided at the DESICOS homepage www.desicos.eu.

4. List of publications
All publications are listed in chapter 6 of this report and on the Website Website www.desicos.eu.

5. Partner communication table

6. DESICOS - List of publications

I. Full papers in scientific journals (reviewed)

1. Degenhardt R., Kling A., Zimmermann R., Odermann F., F. C. de Araújo, “Dealing with imperfection sensitivity of composite structures prone to buckling”, Book “Advances in Computational Stability Analysis”, http://dx.doi.org/10.5772/45810)
2. Degenhardt R., Castro S., Arbelo M., Zimmerman R., Kling A., Khakimova R., “Future structural stability design for composite space and airframe structures“,Int. Journal of Thin-Walled Structures, Vol. 81, (2014), pp.29-38
3. Castro S., A. Arbelo M., Zimmermann R., Khahimova R., Degenhardt R., Hilburger M., “Geometric imperfections and lower-bound methods used to calculate knock-down factors for composite cylindrical shells “, Int. Journal of Thin-Walled Structures, Vol. 74, (2014), pp. 118–132
4. Castro S., A. Arbelo M., Zimmermann R., Khakimova R., Degenhardt R., “Exploring the constancy of the global buckling load after a critical geometric imperfection level in thin-walled cylindrical shells for less conservative knock-down factors“,Int. Journal of Thin-Walled Structures, Vol. 72, (2012), pp.76–87
5. Arbelo M., Degenhardt R., Castro S., Zimmermann R., “Numerical characterization of imperfection sensitive composite structures”, Int. Journal of Composites Structures, Vol. 108, (2014), pp. 295-303
6. Arbelo M., Zimmermann R., Castro S., Degenhardt R., “Comparison of new Design Guidelines for Composite Cylindrical Shells prone to Buckling”, ”, Int. Journal of Composites Structures, (accepted)
7. Khakimova R., Zimmermann R., Castro S., Arbelo M., Degenhardt R., “The single perturbation load approach applied to imperfection sensitive conical composite structures”, Int. Journal of Thin-Walled Structures, Vol. 84, (2014), pp. 369–377
8. Kepple J., Herath M., Pearce G., Prusty G., Thomson R., Degenhardt R., “Stochastic Analysis of Imperfection Sensitive Unstiffened Composite Cylinders using Realistic Imperfection Models”. Int. Journal of Composite Structures Vol. 126, (2015), pp. 159-173
9. Arbelo M., Almeida S., Danadon M., Rett S., Degenhardt R., Castro S., Kalnins K., Ozolins O., “Vibration correlation technique for the estimation of real boundary conditions and buckling load of unstiffened plates and cylindrical shells“, Int. Journal of Thin-Walled Structures, Vol. 79, (2014), pp.119-128
10. Castro S., Mittelstedt C., Monteiro F., A. Arbelo M., Ziegmann G., Degenhardt R., Linear buckling predictions of unstiffened laminated composite cylinders and cones under various loading and boundary conditions using semi-analytical models, Int. Journal of Composite Structures, Vol. 118 (2014), pp. 303-315
11. Castro S., Mittelstedt C., Monteiro F., A. Arbelo M., Degenhardt R., “A semi-analytical approach for the linear and non-linear buckling analysis of imperfect unstiffened laminated composite cylinders and cones under axial, torsion and pressure loads “, Int. Journal of Thin-Walled Structures, (submitted)
12. Arbelo M., Kalnins K., Almeida S., Ozolins O., Skukis E., Castro S., Degenhardt R., “Experimental and numerical estimation of buckling load on unstiffened cylindrical shells using vibration correlation technique (accepted)
13. Di Pasqua M., Khakimova R., Castro S., Arbelo M., Riccio A., Degenhardt R., “The influence of Geometrical parameters on the buckling behavior of conical shell by the Single Perturbation Load Approach”, Int. Journal of Applied Composite Materials, DOI 10.1007/s10443-014-9414-3
14. Arbelo, M., Castro, S., Herrmann, A., Khakimova, R., Degenhardt, R., Zimmermann R., Investigation of buckling behavior of composite shell structures with cutouts", Int. Journal of Applied Composite Materials, DOI ACMA-D-14-00710
15. Castro S., Mittelstedt C., Monteiro F., Ziegmann G., Degenhardt R., Evaluation of non-linear buckling loads of geometrically imperfect composite cylinders and cones with the Ritz method, Int. Journal of Composite Structures, 10.1016/j.compstruct.2014.11.050
16. Kepple J., Herath M., Pearce G., Prusty G., Thomson R., Degenhardt R., “Improved stochastic methods for modelling imperfections for buckling analysis of composite cylindrical shells”, Engineering Structures Vol. 100, (2015), pp.385-398
17. Kepple J., Herath M., Khakimova R., Pearce G., Prusty G., Thomson R., Degenhardt R., “Comparing the Accuracy of the Single Perturbation Load Approach Applied to Unstiffened Composite Truncated Cones with Realistic Stochastic Methods”. Int. Journal of Thin-Walled Structures (submitted)
18. Khakimova R., Burau F., Zimmermann R., Degenhardt R., Siebert M., “Design and manufacture of conical shell structures using pregreg laminates“, Int. Journal of Composites Part A (submitted)
19. Kalnins K., Arbelo M., Ozolins O., Skukis E., Castro S., Degenhardt R., “Experimental non-destructive test for estimation of buckling load on unstiffened cylindrical shells using vibration correlation technique”, Thin-Walled Structures. Volume 94, Pages 273-279
20. Khakimova R., Porceluzzi G., Zimmermann R., Degenhardt R., Castro S., “Empirical formulae for the design load of Single Perturbation Load Approach: derivation, verification and validation“, Int. Journal of Thin-Walled Structures, (submitted)
21. A. Orifici, C. Bisagni, "Perturbation-based imperfection analysis for composite cylindrical shells buckling in compression", Composite Structures, 106 (2013): 520-528.
22. M. Alfano, C. Bisagni, "Probability-based methodology for buckling investigation of sandwich composite shells with and without cut-outs", International Journal of Structural Stability and Dynamics (submitted).
23. Kalnins, K., Arbelo, M.A., Ozolins, O., Skukis, E., Castro, S.G.P., Degenhardt, R. Experimental non-destructive test for estimation of buckling load on unstiffened cylindrical shells using vibration correlation technique, Shock and Vibration Journal, submitted
24. Linus Friedrich, Theodor Andres Schmid Fuertes, Kai-Uwe Schröder, Comparison of theoretical approaches to account for geometrical imperfections of unstiffened isotropic thin walled cylindrical shell structures under axial compression. Thin-Walled Structures, 92(0), pp. 1 – 9
25. Kepple J, Herath M, Khakimova R, Pearce G, Prusty G, Thomson R, Degenhardt R., Comparing the single perturbation load approach with stochastic methods for the buckling of composite truncated cones, Int. Journal of Composite Structures 2015 (submitted)

II. Conference papers

1. Degenhardt R., Zimmermann R., Kling A., Wilckens D., “Future Structural stability Design for Composite space and Airframe Structures”, 6th Int. Space World, Frankfurt, Germany, 2-4 November, 2011
2. Degenhardt R., “Future Structural stability Design for Composite space and Airframe Structures”, 32th Iberian Latin American Congress on Computational Methods in Engineering (CILAMCE), Ouro Preto, Brazil, 13-16th November 2011, (invited keynote)
3. Degenhardt R., Zimmermann R., Kling A., Wilckens D., “Challenges to design imperfection sensitive composite launcher structures”, 4th Int. Conference on Structural Stability and Dynamics, Jaipur, India, 4-6 January 2012, (invited keynote)
4. Degenhardt R., “Challenges in the Validation of Stability Sensitive Multiaxialy Loaded CFRP Structures”, DFG Workshop “Inauguration of the new Multiaxial Test Rigs at TU Braunschweig and Hamburg University of Technology”, Hamburg, Germany, 3 May 2012
5. Degenhardt R., “Future design of composite launcher structures”, 20th International Annual Conference on Composites Engineering (ICCE 20), Beijing China, 22-28 July, 2012
6. De Groof V., Oberguggenberger M., Haller H., Degenhardt R., Kling A., „Quantitative assessment of random field models in finite element buckling analyses of composite cylinders “, ECCOMAS, Vienna, Austria, 10 – 14 September, 2012
7. Oberguggenberger M., De Groof V., Haller H., Degenhardt R., Kling A., „A quantitative assessment of random field models in finite element buckling analyses of composite cylinders“, DGLR Congress, Berlin, Germany, 10 – 14 September, 2012
8. Castro S., Degenhardt R., “Stability behaviour of imperfection sensitive composite structures”, 7th Int. Conf. Supply on the Wings, Frankfurt, Germany, 6-8 November, 2012
9. Arbelo M., Degenhardt R., Castro S., Zimmermann R.,, “Investigations related to most critical imperfections of composite structures”, 7th Int. Conf. Supply on the Wings, Frankfurt, Germany, 6-8 November, 2012
10. Degenhardt R.,, “IFAR - International Forum for Aviation Research”, 7th Int. Conf. Supply on the Wings, Frankfurt, Germany, 6-8 November, 2012
11. Degenhardt R., Castro S., “Future structural stability design for composite space and airframe structures”, 6th International Conference on Coupled Instabilities in Metal Structures, Strathclyde University, Scotland, UK, 3-5 December 2012, (invited keynote)
12. Kepple J., Prusty G., Pearce G., Kelly D., Thomson R., “A New Multi-Objective Robust Optimisation Methodology.” Proc. of 7th Australasian Congress on Applied Mechanics, The University of Adelaide, Adelaide, Australia. Adelaide: ACAM 7, 2012. ISBN 978-1-9221076-1-9.
13. Linus Friedrich, Hans.-G. Reimerdes, Imperfection sensitivity of circular cylindrical shells of varying length subjected to axial compression, 54th Structural Dynamics and Materials Conference 2013 : Boston, Massachusetts, USA, 8 - 11 April 2013
14. Linus Friedrich, Theodor Andres Schmid Fuertes, Hans-G. Reimerdes, Comparison of Imperfection Modelling Approaches for Unstiffened Thin-Walled Cylindrical Shell Structures, DLRK 2013 Stuttgart
15. Arbelo M., Zimmermann R., Castro S., Degenhardt R., “Comparison of new Design Guidelines for Composite Cylindrical Shells prone to Buckling”, ICCST-9 Conference in Sorrento, Italy, 24-26 April, 2013
16. Degenhardt R., “Challenges and opportunities future structures made of CFRP”, Global Composites panel at the Int. Conference SAMPE 2013 - Education & Green Sky – Materials Technology for a Better World, Long Beach, CA (USA), May 6-9, 2013
17. Skukis, E., Kalnins, Ozolins, O. Ozolins, Assesment of the Effect of Boundary Conditions on Cylindrical Shell Modal Responses, Civil Engineering ’13: 4th International Scientific Conference: Proceedings, Latvia, Jelgava, 16-17 May, 2013., pp.41-48.
18. De Groof V., Oberguggenberger M., Haller H., Degenhardt R., Kling A., „A case study of random field models applied to thin-walled composite cylinders in finite element analysis”, ICOSSAR 2013, New York, USA, June 16-20, 2013
19. Arbelo M., Degenhardt R., Castro S., Zimmermann R., “Numerical characterization of imperfection sensitive composite structures”, 17th International Conference on Composite Structures (ICCS17), Porto, Portugal, June 17-21, 2013
20. Castro S., Arbelo M., Zimmermann R., Degenhardt R., “On the buckling mechanism of imperfection sensitive monolithic thin-walled unstiffened composite cylinders – physical observations to support less conservative knock-down factors”, 17th International Conference on Composite Structures (ICCS17), Porto, Portugal, June 17-21, 2013
21. Degenhardt R., “Challenges and Opportunities for Future Aircrafts made of CFRP”, 21st International Annual Conference on Composites Engineering (ICCE 21) in Tenerife, Spain, 21-27 July, 2013
22. Khakimova R., Castro S., Arbelo M., Degenhardt R. Rohwer K., Zimmermann R., Quappen G., Hinsch S., “Studies of Impefection Sensitive Conical Composite Structures”, 21st International Annual Conference on Composites Engineering (ICCE 21) in Tenerife, Spain, 21-27 July, 2013
23. Xu H., Hui D., Castro S., Arbelo M., Degenhardt R., “Imperfection Sensitivity of Antisymmetric Cross-Ply Cylindrical Shell Under Axial Compression Using Hui's Postbuckling Theory”, 21st International Annual Conference on Composites Engineering (ICCE 21) in Tenerife, Spain, 21-27 July, 2013
24. Keple J., Prusty G., Pearce G., Kelly D., Thomson R., Degenhardt R., “Influence of Imperfections on Axial Buckling Load of Composite Cylindrical Shells”, 19th Int. Conf. on Composite Materials, Montreal, Canada, 28 July -3 August, 2013
25. Skukis, E., Kalnins, K., Chate, A. Preliminary assessment of correlation between vibrations and buckling load of stainless steel cylinders, Shell Structures: Theory and Applications - Proceedings of the 10th SSTA 2013 Conference. Volume 3, 2014, Pages 325-328
26. Kalnins, K., Ozoliņš, O., Arbelo M. Castro S., Degenhardt, R., “Verification Study on Buckling Behaviour of Composite Cylinder with Eccentric Supports”, 51th Israel Annual Conference on Aerospace Sciences, Tel-Aviv, Israel, 19-20 February, 2014
27. Arbelo M., Khakimova R., Castro S., Degenhardt R. Rohwer K., Zimmermann R., “Improving the correlation of finite element models using vibration correlation technique on composite cylindrical shells”, 51th Israel Annual Conference on Aerospace Sciences, Tel-Aviv, Israel, 19-20 February, 2014
28. Khakimova R., Castro S., Arbelo M., Degenhardt R. Rohwer K., Zimmermann R., “Investigating the buckling behaviour of imperfection sensitive conical composite structures subjected to Single Perturbation Load Ap-proach”, 51th Israel Annual Conference on Aerospace Sciences, Tel-Aviv, Israel, 19-20 February, 2014
29. Degenhardt R. “Future design for imperfection sensitive composite launcher structures”, 51th Israel Annual Conference on Aerospace Sciences, Tel-Aviv, Israel, 19-20 February, 2014
30. Castro S., Arbelo M., Khakimova R., Degenhardt R. Rohwer K., Zimmermann R., “Ritz Method for the Analysis of Unstiffened Laminated Composite Cylinders and Cones under Axial Compression”, 51th Israel Annual Conference on Aerospace Sciences, Tel-Aviv, Israel, 19-20 February, 2014
31. Degenhardt R., “New robust design guideline for imperfection sensitive composite launcher structures – The DESICOS project”, European conference on spacecraft structures, materials & environmental testing”, Braunschweig , Germany, 1-4 April, 2014
32. Khakimova R., Zimmermann R., Castro S., Arbelo M., Degenhardt R.,”Optimization of the manufacturing process of conical shell structures using prepreg laminates”, European conference on spacecraft structures, materials & environmental testing”, Braunschweig , Germany, 1-4 April, 2014
33. Arbelo, M., Castro, S., Kalnins, K., Ozoliņš, O., Khakimova, R., Degenhardt, R. “Experimental characterization of buckling load on imperfect cylindrical shells using the multiple perturbation load approach”, European conference on spacecraft structures, materials & environmental testing”, Braunschweig , Germany, 1-4 April, 2014
34. Kalnins, K., Ozoliņš, O., Arbelo, M., Degenhardt, R, Castro, P. Experimental characterization of buckling on composite cylindrical shells with eccentric supports European conference on spacecraft structures, materials & environmental testing”, Braunschweig , Germany, 1-4 April, 2014
35. Kalnins, K., Arbelo, M., Castro, P., Ozoliņš, O., Khakimova, R., Degenhardt,. R. Experimental characterization of Buckling load on imperfect cylindrical shells using the multiple perturbation load approach (2014) XVIII International Conference Mechanics of Composite Materials MCM2014, June 2-6, Jurmala, Latvia, pp 116
36. Leite E., Skukis E., Kalnins K, Auzins J., Arbelo M.A, Identification of mechanical properties for PANEX (c) unidirectional carbon fiber composite (2014) XVIII International Conference Mechanics of Composite Materials MCM2014, June 2-6, 2014, Jurmala, Latvia, pp 116
37. Degenhardt R., “Challenges and opportunities for the stability design of future composite space and airframe structures”, 1st International Conference on Mechanics of Composites in Long Island, USA, 8-12 June, 2014 (invited keynote)
38. Khakimova R., Zimmermann R., Castro S., Arbelo M., Degenhardt R.,” An empirical formula of the design load for conical isotropic shell structures obtained by use of Single Perturbation Load Approach”, 1st International Conference on Mechanics of Composites, USA, 8-12 June, 2014
39. Castro S.; Mittelstedt C.; Arbelo M.; Degenhardt R., “Semi-analytical tools for the Single Perturbation Load Approach using the Ritz Method, 1st International Conference on Mechanics of Composites, in Long Island, USA, 8-12 June, 2014
40. Arbelo, M., Castro, S., Herrmann, A., Khakimova, R., Degenhardt, R., Investigation of buckling behavior of carbon fiber-reinforced composite shell structures with openings. Proceedings of the 1st International Conference on Mechanics of Composites, USA, 8-12 June, 2014
41. Kalnins, K., Arbelo, M.A., Ozolins, O., Defenhardt, R. Evaluation of Multiple Perturbation Load approach for experimental characterisation of Design buckling load on imperfect cylindrical shells, The 34th International Seminar of The Students' Associations, Warsaw, Poland, 2 September 2014
42. Kepple J, Herath M, Pearce G, Prusty G, Thomson R, Degenhardt R., “Improved methods for modelling imperfections for buckling analysis of composite cylindrical shells”, 29th Congress of the Int. Council of the Aeronautical Sciences (ICAS), St. Petersburg, Russia, 7-12 September, 2014
43. Degenhardt R., “Challenges and opportunities for future structures made of CFRP”, 2nd Brazilian Conference on Composite Materials (BCCM2), São José dos Campos-SP/Brazil, 16-19 September 2014 (invited keynote)
44. Khakimova R., Castro S., Degenhardt R., Wilckens D., Kepke M., Hildebrandt B., Odermann F., „Buckling experiments on imperfection sensitive thin-walled structures using additional perturbation loads “,3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
45. Khakimova R., Degenhardt R., „Assessment of the Single Perturbation Load Approach on composite conical shells“,3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
46. Degenhardt D., Hilburger M., Castro S., Khakimova R., Degenhardt R., „Buckling studies under non-uniform loading“,3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
47. Degenhardt R., „New Robust Design Guideline for Imperfection Sensitive Composite Launcher Structures - The DESICOS project“, 3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
48. Castro S., Khakimova R., Ziegmann G., Degenhardt D., Degenhardt R., „Simulation of geometric imperfections and uneven edges in thin-walled cylinders“, 3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Lam-inated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
49. Castro S., Mittelstedt C., Monteiro F. A.C., Arbelo M., Degenhardt R., Ziegmann G.,, „A semi-analytical approach for linear and non-linear analysis of unstiffened laminated composite cylinders and cones under axial, torsion and pressure loads“, 3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
50. Castro S., Mittelstedt C., Arbelo M., Degenhardt R., Khakimova R., Hilburger M., Ziegmann G., ,,Non-linear buckling response of unstiffened laminated composite cylinders using different geometric imperfections“, 3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
51. Castro S., Arbelo M., Degenhardt R., Ziegmann G.,, „Single perturbation load approach: new definition for P1 and explaining the constancy of the buckling load“, 3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
52. Arbelo M., Kalnins K., Ozoliņš O., Castro S., Degenhardt R., „Buckling of imperfection sensitive shell structures: experi-mental characterization of the knock-down factor using the Multiple Perturbation Load Approach “, 3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
53. K. Kalnins, E. Skukis, O. Ozolins and M. A. Arbelo . Experimental determination of the Buckling load of composite cylindrical shells using Vibration Correlation Technique, 3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
54. Linus Friedrich, Hans-G. Reimerdes, Kai-Uwe Schröder, Advanced sizing strategies for preliminary design of orthotropic grid stiffened shell structures, 3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
55. Linus Friedrich, Athanasios Dafnis, Hans-G. Reimerdes, Kai-Uwe Schröder, Influence of load application on the collapse load of imperfection sensitive shell structures, 3rd Int. Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, Braunschweig, Germany, 25-27 March, 2015
56. Arbelo M., Kalnins K., Ozolins O., Castro S., Degenhardt R., “Numerical characterization of the knock-down factor on unstiffened cylindrical shells with initial geometric imperfections”, 20th Int. Conf. on Composite Materials (ICCM20), Copenhagen, Denmark, 19-24 July, 2015
57. M. Alfano, C. Bisagni, “Probabilistic buckling analysis of composite and sandwich cylindrical shells”, 55th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, 13-17 January 2014, National Harbor, Maryland, USA.
58. M. Alfano, C. Bisagni, “Reliability assessment of buckling response of axially compressed sandwich composite shells with and without cut-outs,” 3rd International Conference on Buckling and Postbuckling Behaviour of Composite Laminated Shell Structures, 25-27 March 2015, Braunschweig, Germany.
59. M. Alfano, C. Bisagni, “A probabilistic approach for buckling analysis of sandwich composite cylindrical shells,” to presented at AIDAA 2015 Conference on Aeronautics and Astronautics, 17-19 November 2015, Torino, Italy.

Potential Impact:
1. Potential impact
1.1 Impacts listed in the work programme
DESICOS responds to the Activity ‘Space transportation technologies’. It contributes to increase the innovation capacity of future developments by proposing and evaluating new concepts and disruptive technologies, offering a wider range of choice for the next developments. To get this, DESICOS covers activities in the domain of innovative materials and simulation methodologies, resulting in increased capacities, reduced weight and cost of launcher structures. It developed an innovative simulation methodology for the design of future launcher structures made from fibre composite material (CFRP), which deals with their dominant design criterion, i.e. to avoid buck-ling. With the extension of the Single Perturbation Load Approach (SPLA) to composite structures and by combining it with the Specific Stochastic Approach a methodology was created that is as fast as needed in the early system design phase and at the same time as precise as needed later in the structural design loop. So it substantially contributes to reduce weight and cost. These savings can be used to increase capacities. The methodology comprises experimentally validated design approaches, design guidelines and a handbook containing the entire outcome for the purpose of proper application of the results and of standardisation.
Contribution to independent European access to space
Worldwide, customers of space transportation decide within a really existing transportation market. Independent European access to space requires competitive European launchers, which are ready to earn money compensating at least a substantial part of their development cost, otherwise it means extreme cost for Europe, if European launchers offered to space transportation cannot compete within the market. With the innovative design methodology for composite material launcher structures – leading to reduction of weight by about 20% and reduction of cost by about 15% concerning the buckling critical structural launcher components – DESICOS substantially contributes to the competitive power of European launcher structures.
Contribution to enhance capacities for space exploration
Reduction of weight due to the DESICOS results will be used for increased loading capacity or partly for additional propellant in order to increase launcher range. With covered budgets for space activities the reduction of cost can be employed to at least partly finance increased loading for space exploration.
Consolidation of the long term sustainability and improvement of the economical aspects of a domain known to be demanding in terms of reliability
DESICOS increases the competitive power of the European launchers in the worldwide space market, thus it contributes to consolidate the long term sustainability and to improve the economical aspect by its new fast and precise design methodology for composite launcher structures. It responds to the demand for high reliability by its precision and the inclusion of a stochastic procedure with pre-assigned requested reliability.
Increase of the innovation capacity for future developments
DESICOS elaborates a new design approach for composite cylindrical launcher structures. In addition, with first steps, it also investigates the applicability of the new approach for the design of non-cylindrical structures like cones or domes endangered by buckling.
Economic impact on SMEs in the space sector
The DESICOS consortium contains one industrial SME partner, the main contribution of which to the project is the manufacture of the structures to be tested by the partners TECHNION and POLIMI. This includes all the aspects of manufacturing like identification of material properties, non-destructive inspection of the manufactured parts, etc. DESICOS install a Users Group with members interested in being informed about the DESICOS findings already in course of the pro-ject. Particular effort is devoted to get SME members. Cost for participation as a member is con-strained as far as possible, e.g. by coupling DESICOS presentations with ESA’s presentation activities in order to reduce travel cost. Information also will be submitted safely to the members by internet.
1.2 Steps to bring about the impact
The basic technical steps needed are reflected by the workpackage structure of the project:
- Benchmarking on selected structures with existing methods
- Material characterisation and design of structures for buckling tests
- Development and application of improved design approaches
- Manufacture, inspection and testing of structures
- Design guidelines, handbook and industrial validation
In addition, exploitation and dissemination activities are performed in the course of the project, at its end, and afterwards. For example, a DESICOS Users Group, with end users included, is estab-lished. The members were informed regularly about the progress of the DESICOS project and they were invited to attend any workshop or conference dedicated to DESICOS. On 23-25 March 2015 an international conference with partner presentations on the technical results and a DESICOS workshop in order to demonstrate the tools to the public was organised.
1.3 Requirement for European approach, and benefit for European society
The complexity in the project tasks and the transnational relations in the space industry require an approach on the European level, in which the knowledge and experience in
• developing of design concepts in the field of stability,
• manufacturing and,
• testing
by highly reputed partners of research and industry from different countries in Europe are effec-tively combined. They developed advanced knowledge and approaches and finally exploit the outcome. No single country or institution has the ability with respect to research personnel and facilities to investigate the subject to full extent. The DESICOS consortium therefore includes leading space industry in Europe and the most competent researchers and institutions in the field of structural stability.
The application of the improved design approaches promotes an enhanced cooperation between European industrial enterprises. A standardization of the data base and design guidelines leads to a basis for future common European standards, the acceptance of which only can be achieved by European approaches. Researchers use the output for future designs on buckling of composite space structures.
One of the major contributions to the competitive power of the European space industry is the continuous availability and quality of a highly skilled labour force. In order to further secure high employment by a vital civil space sector Europe must intensify its research into extensive utilisation of advanced lightweight structures. The design teams for new spacecraft represent employment at the highest qualification level. The state of the art in launcher design is now rapidly changing due to the increasing role that the use of composite materials for primary structures and numerical simulations play. The new design approaches lead to enhanced job satisfaction of the personnel in charge. As a long-term effect, application of the new design procedures as a possible part of concurrent engineering will be a substantial contribution to secure highly skilled employment.
Cohesion between the EU within its present borders on the one hand and closer connections with its neighbours and with the USA and Australia (an important market) on the other hand is promoted in DESICOS. Moreover, by bringing ten partners from five EU-countries (ESA included, and one partner from the new EU member state Latvia) as well as two partners from one associated country (Israel) together, their cohesion within an enlarged EU is strengthened (cf. Figure 2.2).
DESICOS contributes to European technological and scientific leadership in the area of light-weight design, and therefore reinforces competitive power of European research and industry. The developed design approaches are used directly for a lighter launcher and a more efficient design process. They reduce elaborate test procedures, and allow for an efficient concurrent engineering process of improved space structure, due to their potential to reduce analysis time substantially. The data bases and the design handbook contribute to an advanced standardisation on European level, which strengthen the competitive power of European industry and act as a focus to rapidly improve utilisation of advanced composite technologies.
Application of the developed design procedure
• reduce the operating costs of new launchers
• weight reduction and therefore increase of payload to orbit
• reduce cost and time of the development due to an efficient concurrent engineering process.
Thus, DESICOS contributes to an increased productivity and to higher profit within the space market.
DESICOS developed bridges between society and technology by increasing the social resources and by answering in an effective manner the societal needs of global information. The application of the new tools leads to more efficient and safer launchers. Thus, space based offers to the public become more affordable for a wider range of society and will bring people of the world into closer connection by information exchange. In addition, reduction of weight directly reduces raw material and propellant consumption and thus contributes to quality of life.
DESICOS
• offers an excellent technological basis for lighter launcher structures,
• promotes European cooperation of research and industry,
• improves the knowledge of engineers, the skill of which is an indispensable resource.
1.4 Related national and international research activities
Several partners of DESICOS are involved in complementary international projects. The strong research activity of the partners together with the existing research relations lead to an efficient cooperation with an additional potential for synergies across concurrently running projects. The main related projects with participation of DESICOS partners are

• DEVILS “Design and Validation of Imperfection-Tolerant Laminated Shell Structures” (European FP 4): DLR, POLIMI
• GARTEUR AG 25 “Postbuckling and Collapse Analysis”: DLR
• HICAS “High Velocity Impact of Composite Aircraft Structures” (BRITE III, DG XII In-dustrial and Materials Technologies Programme): DLR
• POSICOSS “Improved Postbuckling Simulation for Design of Fibre Composite Stiffened Fuselage Structures” (European FP 5): DLR, POLIMI, RTU, RWTH, TECHNION
• ALCAS ”Advanced low cost aircraft structures” (European 6th FP): DLR, RTU
• MOJO “Modular Joints for Aircraft Composite Structures” (European FP 6): DLR, CRC-ACS
• COCOMAT “Improved MATerial Exploitation at Safe Design of Composite Airframe Structures by Accurate Simulation of Collapse” (European FP 6): DLR, CRC-ACS, POLIMI, RTU, RWTH and TECHNION.
• VIVACE “Value Improvement through a virtual Aeronautical Collaborative Enterprise” (European FP 6): DLR
• Buckling handbook (ESA, ECSS-E-30-24): DLR, LUH, POLIMI, TECHNION, RWTH
• Probabilistic aspects of buckling knock down factors – Test and analysis (DLR-ESA study), DLR, ESA
• ROSETTA Lander “Philae” (ESA mission, 1994-2004), DLR, ESA
• Mars Netlander (1998-2003), DLR
• Solar Sail I (DLR project, 1997-2001), DLR
• Methodology for nonlinear analysis of large launcher and spacecraft structures (ESA study, 2006-2007), DLR, ESA
• Study on Carbon Fibre Tube Inserts, Part 1 + Part 2 (ESA studies), DLR, ESA
• DAEDALOS, “Dynamics in Aircraft Engineering Design and Analysis for Light Optimized Structures, (European FP 7): POLIMI, DLR, LUH, RWTH, TECHNION
• MAAXIMUS, “More Affordable Aircraft Structure Lifecycle through Extended, Integrated and Mature Numerical Sizing” (European FP 7): DLR, POLIMI; LUH, RWTH
• JTI Clean Sky, (European FP 7, Level 3 project)
• glFEM, “generic linking of Finite Element based Models”, (European FP 7): DLR, LUH
• Solar Sail II (DLR project, since 2008), DLR
• AISat (DLR project, since 2010), DLR
• Asteroid Finder (DLR project, since 2009), DLR
• Verbundprojekt “Kryogene Oberstufe (DLR-Astrium, 2008-2012), DLR, ASTRIUM-D
• DeOrbit Sail (European FP7)
• MAPICC 3D, (European FP7, Level 2), “One-shot Manufacturing on large scale of 3D up graded panels and stiffeners for lightweight thermoplastic textile composite structures”, RTU, POLIMI
• COSMOS, (European FP7, SPACE, CA/SSA, 2008-2011), “Cooperation of Space NCPs as a Means to Optimise Services”, RTU, DLR
Complementarity and relation of DESICOS to other projects
EU projects:
• finished project COCOMAT and the running project MAAXIMUS. The main difference is on the structures considered as well as the focus of research. COCOMAT and MAAXIMUS consider imperfection tolerant structures used in aircraft applications. DESICOS considers imperfection sensitive structures. Typical applications are sandwich structures, skin dominant structures or isogrid structures. Such structures are not only used in space (cf. the use cases of DESICOS), but also in aeronautics.
• DAEDALOS project. DAEDALOS is investigating the dynamic damping behaviour of typical aerospace structures which are mainly not imperfection sensitive. DESICOS is developing design approaches for imperfection sensitive structures under static loading. So there is no overlapping within these two projects.
• JTI Clean Sky Project. No work of the type to be performed in DESICOS will be per-formed in the technical platforms of the FP7 JTI Clean Sky Project.
• All projects related to space, as they do not cover the topic stability (cf. ESA statement below).

ESA programmes:
ESA statement: It is confirmed that the DESICOS project is complementary to ESA activities in this domain. Indeed, ESA has previously performed related work, including cylinder buckling test programmes as well as the development of analysis methodologies, and the output from such studies are foreseen to form a valuable input for this project. There were no ESA technology developments projects running or planned which are overlapping with the tasks and objectives of the DESICOS project. Finally, it should be noted that the results of this project constitute a valua-ble contribution to the Buckling Handbook [74] completed under ESA leadership.
NASA programmes:
There is a synergy between NASA programs for space structures research and development and the DESICOS proposal. In particular, NASA provided its expertise and resources, to the best of its ability, in the following areas of the project: development and validation of new analysis-based shell design technologies, use of random fields in modelling parameter variability in design, reliability-based design methods. In addition, NASA to develop and experimentally validate lower-bound buckling load estimates that would complement the research and development activities outlined in the DESICOS project.

2. Dissemination and exploitation of project results,
2.1 Exploitation and dissemination
Exploitation of the project results took place on several levels. The industrial partners used the new design approaches for superior products, thus increasing competitive power of European in-dustry. The research partners exploited the results through offering more advanced solutions to the industry seeking their expertise. They also used the project results to develop improved so-lutions in the future. The exploitation manager was responsible for the Plan of Using and Disseminating Knowledge and its updates during the course of the project. The exploitation manager also co-ordinated the exploitation activities among the project partners and lead the exploitation task within Workpackage 6 on management and exploitation. In order to exploit the results further, the consortium established a Users Group with partners from industry, in particular from SMEs. This instrument proved to be successful within the COCOMAT project. It informs other industrial partners during the running project about the progress and outcome in more detail. This Group provided an excellent platform for dissemination of project results and their applications. In general, anybody can join this group in order to get access to detailed results. However, it allows the consortium to prevent that certain industry gets access to the project results earlier than the European industry. Detailed planning and execution of these activities were a part of the exploitation task.
Project results were disseminated at two levels: detailed and general. At the detailed level tech-nical reports were produced and distributed between all the project partners. Those reports de-scribe the technical details and the conclusions of the research. At the general level project results were provided to the Users Group, papers were published in international journals, presented at appropriate international conferences and made available to Internet users by the DESICOS homepage. At the end of the project DLR organised a scientific conference on the DESICOS topic, and a DESICOS workshop. This ensured that the increased technological capability of Europe and its partners in taking full advantage of composite material superior characteristics becomes evident to the world-wide aerospace community. Any dissemination were made in accordance with the consortium agreement.
The project brought together spacecraft design engineers from the industrial future-project offices and research engineers. The educational benefit from this goes along two ways. On the one hand, the spacecraft designers experience the state of the art in performing effective analysis of imperfection sensitive structures, and this improves their practise and judgement for future design projects. On the other hand the researchers learn to understand the need for efficient design cycles including the delicate balance between turnaround times, cost, and quality.
The advanced technical and scientific know-how obtained during the project improves the profes-sional level of the engineers and students of the participating partners, and the dissemination activities promote in addition engineers and researchers outside of the consortium.
In detail, the exploitation and dissemination was raised by the DESICOS consortium in the following way:
- Establishment of a Users Group with partners from industry and authorities.
- Information via the internet homepage www.desicos.eu
- 30 Scientific publications in international journals
- 59 Presentations at international conferences or workshops
- On 25-27 March 2015 a scientific conference on buckling behaviour of fibre composite structures and a DESICOS workshop to demonstrate the capabilities of the new approaches.
- Flyer and poster about the DESICOS project.
The most effective exploitation of the new approaches and knowledge is given by the experienced partners in a well-balanced consortium with strong participation of the space industry. Complete control of the structural simulation enabled the industrial partners to fully exploit the potential of their composite structures.
2.2 Management of intellectual property
The policy for securing Intellectual Property Rights and for licensing were determined in the Consortium Agreement. Some principles of the intellectual properties were the following:
- Any new approach, software or invention generated by a partner is the property of that partner, that partner will patent any patentable results.
- If any joint invention results from the co-operation under the consortium agreement, i.e. joint inventions made by employees of more than one partner, and if the features of that joint invention cannot patented separately, the parties concerned can jointly apply for patent protection.
- A continuous and substantial interaction between the personnel of the partners took place during the project. The partners agree that the partner generating foreground information shall own any foreground information but all partners shall be entitled to use such infor-mation for DESICOS purposes without any financial compensation to or the consent of the owing partner.
- A considerable amount of background information was being brought into the project, which remain the property of the specific partner and may, in some cases, be commer-cially confidential. This specific know-how is owned by the specific partner and was not shared within this project. The exception is the transfer of know-how in order to enable a cooperating partner to perform tasks vital to achieve the project goals. The legal modalities are usually sorted out on bilateral level.
- In order to gain maximum synergy all results and developments obtained by the partners within the project aremade available among the partners in the consortium. This includes all test results and improved software tools. Some results, which are obtained after having applied new methods on company specific elements or sub-components, might be confidential due to competitive power reasons. However, the dissemination level of the associated deliverables and its confidentiality ranking are regulated in the list of deliverables.

It is worthwhile to note that such technological development as it was performed in the DESICOS project involves a significant collaborative effort, with each partner engaged on a particular ele-ment of the task, which is assembled to address the overall objectives. A project of the DESICOS’ nature can only succeed if data are exchanged openly between the partners but limiting the disclosure of the information to the required dissemination to the public.

List of Websites:
www.desicos.eu

Related information

Contact

Kutz, Felicitas (Contract Manager)
Tel.: +49 531 295 2119
Fax: +49 531 295 2875
E-mail
Record Number: 182546 / Last updated on: 2016-05-12
Information source: SESAM