Final Report Summary - ADD CONTROL (Advanced Solutions for Waste Water Treatment) Executive summary:Tougher effluent quality standards, increasing wastewater loads, and energy consumption and operation costs reduction requirements are nowadays forcing wastewater treatment plants (WWTP) to work under more and more stringent conditions. Many automatic control strategies have shown their usefulness at simulation scale, or even pilot plant scale for improving WWTP performance without structural changes in the plant. However, they are experiencing limited full-scale plant applicability mainly due to the still low reliability of the data provided by online sensors and analysers and the ideal behaviour of sensors and actuators commonly assumed by existing WWTP-specific simulation platforms.The ADD CONTROL project (see http://www.addcontrol-fp7.eu for further details) has addressed this problem and proposes an extension of the WWTP models and modelling architectures used in traditional WWTP simulation platforms considering next to the classical mass transformations (transport, physico-chemical phenomena, and biological reactions) all the instrumentation, actuation and automation and control components, where their real (not ideal) behaviour is modelled, including actuators power consumption. In this way, all aspects considerably influencing the real controllers' performance and strategy are accounted for.Regarding mathematical modelling, a three-layer modelling architecture separating the mathematical models of the mass-flows derived from the water treatment process (tanks, reactors, settlers, hydraulic connections, hydraulic flows, etc.) from the virtualization of the data-flows associated with instrumentation, actuation and control devices has been implemented: (1) the treatment process or 'mass' layer describing the biochemical, physico-chemical, equilibrium, liquid-gas transfer, hydraulic and other processes that take place in the different reactors of the plant, based on the plant-wide modelling specification; (2) the 'instrumentation and actuation' layer, describing sensors' and actuators' real behaviour, including signal delays, noise and failures and power consumption; (3) the 'automation and control layer', describing automatic control devices (on/off controllers, PID controllers, fuzzy controllers,...) and some signal processing algorithms (sampling, filtering, A/D and D/A conversions,...). This architecture allows readily integrating the different mathematical models and replicating the classical control architecture of real plants.Regarding SW (SW) development, a tool based on two well known and widely used modelling and simulation platforms has been developed: (a) the WESTand#174;/Tornado platform (mikebydhi.com) a water quality systems modelling and simulation SW platform; and (b) the Matlaband#174;/Simulinkand#174; platform (see http://www.mathworks.com for details), a general purpose modelling and simulation SW tool. The (1) 'mass' layer is implemented on WESTand#174;/Tornado using the MSL modelling language, whereas the (2) 'instrumentation and actuation' layer and the (3) 'automation and control' layer are implemented on WESTand#174;/Tornado using the Modelica (see http://www.modelica.org for details) modelling language on the one hand, and on Matlaband#174;/Simulinkand#174; on the other.Regarding testing and validation, the developed SW tool has been used to design and test automatic controllers for: (a) a full-scale municipal WWTP; and (b) a pilot-scale industrial WWTP, the controllers having been successfully implemented and validated. Thus, the final objective of this project, the development of a simulation tool to allow a rapid and reliable transition from the simulation of the controllers to their implementation in real plant, has been advantageously reached.Project context and objectives:IntroductionIn Europe, significant progress has been made in the last decade with respect to population connected to wastewater treatment facilities. In 2007, about 70 % of the population was connected to any kind of urban wastewater treatment (Eurostat, 2011). Not only the quantity, but also the quality of wastewater treatment improved as the percentage of population connected to wastewater treatment facilities with tertiary treatment increased from 29 % in 1997 to 48 % in 2007 (Eurostat, 2011).Nonetheless, many existing wastewater treatment plants (WWTPs) are being (or still need to be) upgraded with advanced treatment technologies to meet more stringent effluent criteria and standards for sludge disposal. In addition, many existing urban treatment plants are facing increasing wastewater loads that need to be treated. Similar tendencies are being observed for industrial wastewater treatment. Current regulations are forcing industries to install more and more wastewater treatment systems and to improve the effluent quality of their wastewaters before discharging into sewers or receiving waters.Instrumentation, control, and automation (ICA) is a promising technology for improving treatment plant performance, and a valuable alternative to structural modifications of WWTPs, such as increasing reactor volumes. It has been shown, for example, that ICA may increase the capacity of biological nutrient removing WWTPs by 10 - 30 % (Olsson et al., 2005). Furthermore, ICA can help to improve treatment plant performance in terms of process robustness, energy consumption, and operating costs while still ensuring the required effluent quality. Two examples from Olsson et al. (2005) are given below that demonstrate possible reductions in energy consumption and operating costs when implementing ICA:1. The dosage of chemicals for phosphorus precipitation can be reduced by 18 % when using load proportional dosage of chemicals instead of constant dosage, and by 40 % when implementing feedback control based on in-situ ortho-phosphate measurements.2. Aeration energy can be lowered by about 30 % when implementing dissolved oxygen (DO) control with constant setpoint instead of applying a constant air flow. Implementation of DO control with variable setpoint can decrease the energy demand for aeration even further.In the future, the issue of decreasing the energy consumption of a WWTP will gain importance, especially with rising energy prices. Designing control strategies for reducing operating costs and energy consumption of a treatment plant involves moving from the individual process level control to plant-wide control, considering different individual process levels together.Nowadays, mathematical modelling and simulation have become essential tools for supporting not only the design and operation of wastewater treatment plants (WWTPs) but also the analysis and synthesis of control algorithms. The original benchmark simulation model No. 1 (BSM1) protocol and its sequels, the long-term BSM1 (BSM1_LT) and the benchmark simulation model No. 2 (BSM2), are three important simulation benchmarks aimed at providing researchers with standard methodologies for the objective comparison of control strategies (Copp et al., 2002). Since their publication, multiple automatic control approaches have been tested and verified by using these protocols: conventional PID controllers (Yong et al., 2006a); fuzzy controllers (Yong et al., 2006b); model-based predictive controllers (Zarrad et al, 2004; Holenda et al., 2007); and feedback / feedforward controllers (Stare et al., 2007). However, the extensive work undertaken in the simulation domain has not been accompanied by the practical implementation of advanced controllers in full-scale WWTPs. One reason for this is the fact that the design of good operational control systems requires working on more practical aspects than just the design of the control algorithms, i.e. the behaviour of sensors and actuators (Rosen et al., 2008), communication systems, databases and other technical aspects. Moreover, next to process stability and reliability, also final effluent quality and energy consumption gain importance in the design of a control strategy. Another reason is that existing WWTP-specific simulation frameworks have traditionally been intended for design and operation studies rather than for control law definition purposes. In practice, these SW tools fail to allow a rapid transition from the simulation of the control strategy to its implementation in full-scale plants.These existing WWTP-specific simulation packages (e.g. BioWinand#8482;, GPS-Xand#8482;, SIMBAand#174;, WESTand#174;) typically provide some controller components to enable the implementation of simple control strategies in a WWTP model. The realised control strategies are, however, based on ideal behavior of sensors and actuators as signal noise and potential sensor and actuator failures are not considered. In order to design 'practical control' products, real behaviour of sensors and actuators, communications between different hardware components, and data treatment need to be taken into account to ensure robust controller behaviour and adequate plant performance. Tools for control design and data treatment needed for designing practical control solutions are not available in WWTP-specific simulation packages, but in general-purpose simulation SW packages (e.g. Matlaband#174;/Simulinkand#174;, Scilab). General-purpose simulation SW packages, on the other hand, lack model libraries of processes specific for wastewater treatment. Thus, neither existing WWTP-specific simulation packages nor general-purpose simulation SW packages are suitable for designing, testing, and validating practical control solutions.Therefore, this work proposes an extension of the traditional WWTP models and modelling architectures used in current WWTP-specific simulation tools, focusing on current needs for practical control. The proposed approach's principle is to simulate all the plant's components determining the final control solution: from biological tanks to instrumentation, actuation and automation systems. That is, the classical concept of WWTP simulation, in which the focus, hitherto, has been on the modelling of mass transformations (transport, physico-chemical phenomena, and biological reactions), is extended by a WWTP virtualization, where modelling covers all the components involved in the operation of a plant.Objectives:The main objective of the ADD CONTROL project is to design, implement, and validate a new simulation tool for practical control in WWTPs. In order for the general objective to be reached, the following partial goals have to be fulfilled:1. Mathematical modelling of all those components that, in general, make up a WWTP, considering components close to the treatment process (tanks, biological reactors, settlers) as well as components closer to the automation and control (sensors, actuators, PLCs, SCADAs .2. Interfacing components according to a hierarchical architecture that emulates the real architectures for automation and control in full-scale WWTPs3. Open environment to support the virtualisation of different WWTP configurations and treatment technologies (conventional activated sludge systems, membrane bioreactors, biofilm systems, anaerobic digesters, etc.).4. Validation in real case-studies. The performance of the SW tool is assessed through two full-scale plant scenarios: (1) urban wastewater scenario; and (2) industrial wastewater scenario. In both cases, advanced controllers are designed and tested using the virtualization SW tool. Then, these controllers have been implemented and validated in their respective full-scale plant.StructureThe ADD CONTROL project is structured into six technical work packages (WPs). An additional WP deals with project management (WP1). WP2 - WP7 are dedicated to the following:WP2: Definition of technical specifications for the ADD CONTROL tool, including the SW architecture and the SW libraries to implement WP3, WP4, WP5: Design, development and integration of the three SW layers according to the technical specifications WP6: Practical validation of ADD CONTROL in two full-scale plant scenarios (an urban and an industrial WWTP), in order to demonstrate the usefulness of the tool WP7: Create the path for an effective exploitation of the project results and an effective dissemination, during and after the project life Project results:Thank to the success of the project, the gathered knowledge will put Europe in pole position with respect to tackling optimisation of wastewater treatment processes in a sustainable way. In this sense, exploitable results can be commercialized together in a joined offer or as separate results. These are the exploitable results developed during the project:1. SW library models2. WWTP emulator SW3. advanced controllers4. enhanced treatment technologies. In overall, a new simulation SW platform, based on new process and system models, aimed at the design and implementation of new advanced controllers and already offering enhanced treatment technologies has been developed. Next, a description of these results is given:1) SW library modelsIn this project, an extension of the WWTP models and modelling architectures used in traditional WWTP simulation platforms has been implemented. Next to the classical mass transformations (transport, physico-chemical phenomena, and biological reactions), models for all the instrumentation, actuation and automation and control components have been developed, including their real behaviour (signal noise, signal delay, faults, drift, offset) and actuators' power consumption. That is, all aspects considerably influencing the real controllers' performance and strategy have been modelled.Regarding modelling architecture, a three-layer modelling structure separating the mathematical models of the mass-flows derived from the water treatment process (tanks, reactors, settlers, hydraulic connections, hydraulic flows, etc.) from the virtualisation of the data-flows associated with instrumentation, actuation and control devices has been implemented:1) the treatment process or 'mass layer' describing the biochemical, physico-chemical, equilibrium, liquid-gas transfer, hydraulic and other processes that take place in the different reactors of the plant, based on the plant-wide modelling specification; 2) the 'instrumentation and actuation layer', describing sensors' and actuators' real behaviour, including signal delays, noise and failures and power consumption; 3) the 'automation and control layer', describing automatic control devices (on/off controllers, PID controllers, fuzzy controllers,...) and some signal processing algorithms (sampling, filtering, A/D and D/A conversions,...). This architecture allows readily integrating the different mathematical models and replicating the classical control architecture of real plants.Next, a description of each of these three layers is given: The mass layerAs mentioned above, the 'mass' layer relates to the physical processes of a WWTP. This layer is represented and modelled in terms of unit processes (UP), where each UP represents a physical element of the plant. The UPs are modeled according to the plant-wide modelling (PWM) specification (Grau et al., 2007), based on the following two main features:- a common state-vector is shared by all UPs;- conservation of mass and charge is considered in all the transformations.Model specification language (MSL) is the language chosen for modelling the UPs. Taking this into account, two PWM categories have been implemented:- CN_AnD: Carbon, nitrogen and anaerobic digestion. This category gathers all the components and transformations which dynamically describe aerobic, anoxic and anaerobic COD biodegradation and N removal.- C2NP_AnD: Carbon, nitrogen, phosphorous and anaerobic digestion. This category gathers all the components and transformations which dynamically describe aerobic, anoxic and anaerobic COD biodegradation, P biological removal, N removal in two steps (NO3-NO2-N2) and anaerobic ammonia oxidation (the Anammox process).Regarding the UPs of the 'mass' layer, the main UPs by this work implemented are next listed:- buffer tank (completely stirred tank (CST))- completely stirred open tank reactor (O-CSTR)- completely stirred closed tank reactor (C-CSTR)- intermittently stirred open tank reactor (O-ISTR)- settling tank,where the description of the internal transformations of those tanks considering biological reactions will depend on the PWM category selected. The instrumentation and actuation layerThe 'instrumentation and actuation' layer deals with the mathematical modeling of sensors and actuators with 'real behaviour'. For sensors, a 'generic' sensor model has been built, no matter pH-probes, DO-sensors, flow rate sensors, temperature probes ... are to be modelled. The idea is not to take an actual sample from a process unit such as an activated sludge tank, but to distort a perfect signal into a more realistic signal.A virtual sensor gets its input (i.e. the ideal signal) from the 'mass' layer and it outputs a mimicked real signal taking into account: 1) sensor dynamics (response time, delay and temporal resolution (sample-and-hold)); 2) signal noise; and 3) sensor faults. The virtual sensor model takes a signal input from the 'mass' layer and produces a sensor signal and a sensor fault signal which are communicated to the 'automation and control' layer.The implementation of the above 'generic' virtual sensor, some of the associated sub-models and the fault model is partly based on the work of Alex et al. (2009), Corominas et al. (2010) and Rosen et al. (2008).Concerning actuators, the modelling of pumps and aeration blowers has been addressed, since those are the most important energy consumers in wastewater treatment plants. An extensive literature review revealed that much knowledge is available on the theoretical functioning of pumps and blowers from a design point of view, but that no dynamic energy consumption models are readily available for implementation. Even though some commercial SW packages for calculating and optimising features for energy consumption of (mostly) pumps already exist, often these are static instead of dynamic. This research work proposes an approach to create dynamic models for the calculation of pumping and (bubble) aeration power, based on steady-state models and considering an ideal behaviour, and with the possibility to test different pump and blower types and different control mechanisms often used in practice. Regarding pumps, centrifugal pumps and positive displacement pumps have been addressed. For these, generic models based on generic pump curves, generic systems curves and generic approaches to efficiency calculation have been developed. The general approach deals with delivering the required flow rate from a single pump. However, this would in most cases not be the situation in a real plant where the total flow rate between unit operations is the result of using two or more pumps in parallel. Thus, power consumption calculation for pump groups of different layouts (individual discharge pipes, one common discharge pipe) has also been implemented. The Markov chain concept followed by the fault modelling strategy used for sensors (Rosen et al., 2008) has also been applied to actuators by adapting the associated states and transition probabilities, both for the single pump model and the pump group model.Regarding blowers, centrifugal blowers and positive displacement blowers have been addressed. Energy consumption for blowers, similar to that of pumps, is a function of air flow rate, efficiencies and discharge pressure. Thus, generic models based on generic blower curves, generic systems curves and generic approaches to efficiency calculation have also been developed here. However, significant differences and complexities are introduced to aeration blower applications due to the compressibility of air. Characteristics such as air density, relative humidity, altitude and temperature influence the required airflow to the system and therefore also the energy requirement of the blower. These aspects, together with the selected process control strategy, are the key issues that have been considered when evaluating the energy consumption of aeration systems.All sensor and actuator models have been implemented in Modelica and in the Matlab/Simulink platform.The automation and control layerThe 'automation and control' layer deals with modelling approaches for controllers and signal processing. Controller models range from simple on/off controllers and common PID controllers to advanced controllers, such as fuzzy logic controllers and model based or adaptive controllers. The described modelling approaches for signal processing cover signal converters, signal analysis, digital filters, fault detection, data input and output as well as pseudo real-time simulation and OPC communication. Regarding fault detection, it is noted that a fault detection strategy based on receiving and processing fault signals sent from sensors and actuators and therefore, i.e. on the integration with the 'instrumentation and actuation layer' has been implemented, where the fault signal sent by the sensors and actuators depends on the fault state of sensors / actuators.All the models have been implemented in the Matlab/Simulink platform. Parts of the models have also been implemented in Modelica.2) WWTP Emulator SWThe developed simulation tool is based on the Tornado framework for modeling and virtual experimentation (Claeys et al., 2006) and the WESTand#174; product suite (Vanhooren et al., 2003). TORNADO supports simulation of complex systems and provides SW interfaces to standard SW platforms, such as Matlab/Simulink, .NET, and C++. WESTand#174; is an environment for dynamic modeling and simulation of water quality systems, such as wastewater treatment plants, sewers, and urban catchments, and is built on top of Tornado.Typically, the wastewater treatment processes and, the instrumentation, actuation and control systems are modelled using domain specific SW tools. WEST/Tornado (Claeys et al., 2006) and Matlab/Simulink correspondingly two well known and widely used modelling and simulation platforms. This work considers two implementation architectures: (a) one platform-based implementation and; (b) two platforms-based hybrid implementation. Thus, the (1) 'mass' layer is built within the WEST/Tornado platform, whereas the (2) 'instrumentation and actuation' layer and the (3) 'automation and control' layer are built in both the WEST/Tornado and the M/S platforms. Having the entire plant modelled in one platform makes the interfacing between different models easier, whereas using a hybrid solution allows exploiting the power of every specialized SW platform at the expense of making the interfacing between different models more complicated.The 'subsystem' functionality of M/S has been employed to embed the 'instrumentation and actuation' and 'automation and control' layers separately. A specific M/S block has been used to communicate these two M/S-based layers with the WEST/Tornado-based 'mass' layer. Since both M/S and WEST/Tornado have their own numerical solvers to simulate their models, a synchronization mechanism has been required to ensure a unique simulation time in both platforms. Typically, models in the 'mass' layer have to be simulated using 'stiff' solvers; in contrast, fixed-step solvers seem to be appropriate for simulating the two other layers. This feature and the fact that the 'mass' layer has been embedded into a specific M/S block leads to a hierarchical architecture where the M/S solver is on top of the WEST/Tornado solver. 3) Advanced controllersOnce the WWTP emulator SW and the SW library of models were available, the usefulness of the platform for the development of practical control solutions has been evaluated. Two different case studies have been analyzed by means of the proposed simulation framework:(a) full scale secondary treatment (water line) of the Mekolalde WWTP (Bergara, Spain);(b) pilot scale anaerobic digester located at INRA and fed by the industrial distillery wastewater of the distillery in Ornaisons, France (SCAD).Regarding the secondary treatment (water line) of the Mekolalde WWTP, once proven that the simulator for Mekolalde was fully operative for use, the next step was to explore its appropriateness to test and validate practical controllers before their implementation in a real control device. Taking into account that this plant must cope with nitrogen removal requirements, a decentralised control schema based on three non-interacting feedback controllers has been implemented, each of them devoted to control respectively:(1) the concentration of suspended solids in the mixed liquor (MLSS); (2) the ammonia concentration in the last aerobic reactor; and (3) the nitrates in the first anoxic tank. Simulations carried out for testing the controllers showed their correct performance. Regarding the pilot scale anaerobic digester, four different control strategies have been designed and tested. Two of the controllers used PI controllers from the controller library of the 'automation and control' layer. In addition, two model-based controllers, for total VFA control and methane gas flow rate control, have been developed and implemented in Matlab/Simulink. In all cases, aiming to mimic realistic conditions, signal noise has been activated at the 'instrumentation and actuation layer'. Simulations carried out for testing the controllers show that both conventional PI controllers and model-based controllers are able to keep the defined set-points despite variations in the influent COD concentration to the digester, and despite signal noise in the measurements.These controllers have been validated in the real case-studies, obtaining positive results and validating the developed simulation SW platform, models and advanced controllers.4) Enhanced treatment technologiesAs previously mentioned, an urban WWTP and an industrial WWTP have been used as case-studies during the project. Thank to the advanced controllers develop using the ADD CONTROL simulation SW, the performance of the treatment technologies of these plants have been improved.During the validation task of the project, it has been also validated that the developed 'mass' layer is able to predict the behaviour of the real plants. This means that the ADD CONTROL simulation SW is appropriate for the design of both treatment technologies and advanced controllers.Regarding the secondary treatment (water line) of the Mekolalde WWTP, it corresponds to a conventional pre-denitrifying activated sludge process made up of four tanks arranged in series. The first tank operates under non aerated conditions, necessary to perform the denitrification of nitrates. On the other hand, a blower injects air into the other three reactors for supplying the oxygen required to keep the effluent ammonia concentrations below upper limits. After the implementation of the advanced controllers, the performance of the conventional activated sludge (without controllers) process was improved in terms of effluent quality and energy consumption. The enhanced activated sludge process is able to: (1) obtain the similar ammonia concentration in the effluent while reducing the nitrates concentration in the effluent; (2) reduce the power consumption of the biologic reactor by operating at lower oxygen concentrations and internal recirculation flow rates. Thanks to the positive results obtained in Mekolalde, it is convenient to analyse the implementation of the enhanced activated sludge process in existing or new urban WWTPs plants.Regarding the industrial anaerobic digester treating distillery wastewater, although four advanced controllers were developed and analysed at simulation level, the following two control strategies were implemented and tested in the industrial anaerobic digester plant: 1) pH control using an incremental PI controller; and 2) linearising control of the methane gas flow rate. Experimental results from the implementation of the pH control with an incremental PI controller show good results: the controller is able to keep the desired setpoint despite changes in COD influent concentration and load. Results obtained with the linearising control of the methane flow rate are also promising. Thank to this enhanced industrial anaerobic digester, the operation of the digester is more stable (pH controller) and the biogas production is increased. In consequence the efficiency of the digester is improved.Therefore, the real validations show that the simulation tool is useful to analyse the performance of the treatment technologies and improve their performance.Dissemination activities:There have been done different dissemination activities (publications in specialised and non specialised international journals, congresses, workshops and exhibitions and training activities to small and medium-sized enterprises (SMEs)). All presentations and posters related with ADD CONTROL have showed the flag of the EU, the logo of the Seventh Framework Programme (FP7) and the ADD CONTROL logo.Exploitation of results:Nowadays, there are many planned to-do tasks but there are also some achievements. Most interest came from environmental engineering companies related to advanced controllers and enhanced treatment technologies. At the moment exploitation of results could be summarized in the following way:- Many visits and contacts have been done.- Industrial associations and their members were contacted through email and physically, taking advantage of meetings and congresses. - Regarding end-users each SME is at the moment in contact with different plants. Not only European but also North-Africa and South America, where the implementation of enhanced treatment technologies and advanced controllers is being analysed.Thus, the consortium started with the first exploitation activities during the project, which are still in process, and will focus in the exploitation of results after the end of the project.Use and dissemination of foregroundA plan for use and dissemination of foreground (including socio-economic impact and target groups for the results of the research) shall be established at the end of the project. It should, where appropriate, be an update of the initial plan in Annex I for use and dissemination of foreground and be consistent with the report on societal implications on the use and dissemination of foreground.The plan should consist of:Section A This section should describe the dissemination measures, including any scientific publications relating to foreground. Its content will be made available in the public domain thus demonstrating the added-value and positive impact of the project on the European Union (EU).Section BThis section should specify the exploitable foreground and provide the plans for exploitation. All these data can be public or confidential; the report must clearly mark non-publishable (confidential) parts that will be treated as such by the European Commission (EC). Information under Section B that is not marked as confidential will be made available in the public domain thus demonstrating the added-value and positive impact of the project on the EU.Potential impact:As partners are completely complementary, there is need of distribution of market only between Naskeo and AQC. This is solved dividing European market, each in its local European area. AQC will exploit results from Germany (not including) to the East (including Norway, Sweden, Finland, Denmark, Easter countries and reaching the Mediterranean sea) and Naskeo the Western and South Europe (Belgium, Germany, Switzerland, Italy and the Western Europe)In order to maximise the success of the commercial exploitation, the target group for dissemination and exploitation have been identified:- Control engineering companies: Not only MSI will accede to WWTP emulator but also every control engineering company is a potential customer of this result. Thus, M4W will show via workshops and meetings the advantages of this solution. Firstly, to actual clients and then to potential users. - Environmental engineering companies. MSI will commercialize advanced controllers to NASKEO and AQC covering both Industrial and Urban WWTP. Moreover, regarding urban WW MSI will analyse environmental engineering companies with which solid commercial alliances will be established. MSI will contact to both end users and environmental companies in order to demonstrate the benefits of the project results.- End users: Industry and urban WWTP. Apart from actions to be done by MSI, Naskeo and AQC will exploit their current commercial routes in order to find potential end users of the developed solution. Initially, Naskeo will focus its commercial activities in Industry WWTP. However, it is planned that in a near future Naskeo may accede to Urban WWTP. AQC will focus its commercial activity in industrial and urban WWTP.Project website: http://www.addcontrol-fp7.eu/Mondragon Sistemas de Informacion, S.COOPMember of the management board: Mr Jose Maria Sagarna, R&D ManagerE-mail: jmsagarna@msigrupo.com20140 Andoain, SpainTel.: +34-943-594400Fax: +34-943-590536Mostforwater NVMember of the management board: Mr Dirk Van der Stede, CEOE-mail: fc@mostforwater.com8500 Kortrijk, BelgiumTel. : +32-563-54390Fax: +32-563-60230Naskeo EnvironmentMember of the management board: Mr Frederic Sylvain, R&D ManagerE-mail: Sylvain.frederic@naskeo.com92240 Malakoff, FranceTel.: +33-(0)15-7210216Fax: +33-(0)15-7213471Societe Cooperative Agricole D'OrnaisonsMember of the management board: Dr Axel Tapissier, General ManagerE-mail: distillerie.ornaisons@wanadoo.fr11200 Ornaisons, FranceTel.: +33-468-272811Fax: +33-468-272417Aguas De Gipuzkoa S.A.Member of the management board: Mr Oscar Fernández, Quality ManagerE-mail: oscarf@gipuzkoakour.com20018 Donostia, SpainTel. : +34-943-311801Fax: +34-943-211959Centro De Estudios e Investigaciones Tecnicas de GipuzkoaMember of the management board: Dr Mikel Maiza, Project ManagerE-mail: mmaiza@ceit.es20018 Donostia, SpainTel.: +34-943-212800Fax : +34-943-213076Universiteit GentMember of the management board: Dr Ingmar Nopens, Postdoctoral ResearcherE-mail: Ingmar.nopens@ugent.beB-9000 Gent, BelgiumTel.: +32-926-45935Fax: +32-926-46220Institut National De La Recherche AgronomiqueMember of the management board: Mr Jean-Philippe Steyer, Staff DirectorE-mail: steyer@supagro.inra.fr11100 Narbonne, FranceTel.: +33-468-425163Fax: +33-468-425160Aqua-Contact Praha V.O.S.Member of the management board: Dr Libor Novak, Consulting EngineerE-mail: Libor.novak@aqua-contact.cz55101 Jaromer, Czech RepublicTel.: +42-022-4311424Fax: +42-022-4311424 Related documents Final Report - ADD CONTROL (ADVANCED CONTROL SOLUTIONS FOR WASTE WATER TREATMENT)
