Objective
Objectives of the project
HVAC systems are equipments usually implemented for maintaining satisfactory comfort conditions in buildings. However, their operation requires at least 50% of the overall energy consumption in office buildings. Moreover, improper tuning and commissioning of HVAC control systems often result in excessive ventilation rates, excessive (or insufficient) temperature levels, poor network balance,... and thus in increased energy consumptions and discomfort for occupants.
Many advanced control strategies have already been developed in the last fifteen years, specially predictive and optimal control strategies. However, they have not been produced at industrial level, since they were extremely difficult to implement and no tuning strategy had been developed for assisting building control systems installers in this task.
The aim of this project is thus to develop optimum control strategies for HVAC systems, based on multicriteria analysis and to develop systematic tuning strategies for these controllers. Two main innovations will be implemented:
- The use of rule-based controllers (fuzzy logic controllers) . This will enable the implementation of real multicriteria control strategies incorporating expert knowledge. In the frame of this project, criteria of particular importance will be: energy consumptions, occupants, thermal comfort, indoor air quality, while criteria such as peak load electrical demand, network management and balance, equipment stability and maintenance will also be considered.
- The development and comparison of smart setting and tuning techniques for these controllers: It will enable a rational operation and improved performance of fuzzy controllers and is a necessary condition for implementing complex control techniques. Moreover this development will reduce commissioning periods since control strategies will be developed following a systematic protocol. These techniques will be implemented using optimisation and learning procedures (genetic algorithms and neural networks) on the basis of evaluations performed through so called "fitness functions" incorporating the criteria cited above (i.e. energy consumption, comfort,...) and whose structure will be developed within this project.
Technical approach
The project will be performed with seven tasks. Task 1 will consist in selecting the most relevant equipments to work on with regard to energy saving and marketing potential, to control complexity, control parameters will also be selected in this task. Task 2 will consist in developing modelization tools required for developing smart tuning strategies. Task 3 will consist in constructing fuzzy controllers and selecting appropriate smart tuning techniques. These smart tuning strategies will be developed and simulated in task 4. Resulting from these developments hardware and software prototypes will be developed in task 5. Fuzzy control and smart tuning strategies will then be experimentally tested in task 6.All the results will finally be analysed in the final task . The marketing potential will be particularly studied as well as possible extensions to other equipments and buildings.
Expected achievements
The final outcomes will be:
- A set of fuzzy controller prototypes and their application software for implementation of smart tuning control strategies. The systems will then be ready for commercial development provided that additional real scale testing is performed and software interface is improved.
- A handbook on fuzzy control applied to HVAC systems.
- A set of reports for each task developed in this project
- Databases on equipments and building typology and experimental data.
A WWW server (http://www.entpe.fr/Genesys) will be maintained in order to present the project, its objectives and its main results.
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringcontrol systems
- engineering and technologymechanical engineeringthermodynamic engineering
- engineering and technologyenvironmental engineeringair pollution engineering
- natural sciencescomputer and information sciencessoftwaresoftware applications
- natural sciencescomputer and information sciencesartificial intelligencecomputational intelligence
You need to log in or register to use this function
Call for proposal
Data not availableFunding Scheme
CSC - Cost-sharing contractsCoordinator
69518 Vaulx-en-Velin
France