Partenaires et organisations internationales
(Anglais)
|
TNO-BOUW (NL), COORDINATEUR, ALCOA (NL), CONPHOEBUS (I), LESO-PB/EPFL (CH), METSAPUU OY (FIN), MUOVIITALA OY (FIN), TIIVITUOTE OY (FI)N, VTT (FIN)
|
Résumé des résultats (Abstract)
(Anglais)
|
The project Smart Window was aimed at the elaboration of a window component, including all the control circuitry and algorithms to control 'smartly' the integrated blind and ventilation opening, in order to increase the user's comfort (thermal, lighting, air quality) while saving the maximum possible energy used by the remaining building services not integrated in the window (heating, cooling, central ventilation, artificial lighting). The control algorithms are characterized by:
· the use of fuzzy logic rule bases for describing the operation of partial control systems (blinds, ventilation, possibly artificial lighting when connected);
· the use of self-adaptive techniques, allowing a continuous adaptation of the system to changeable conditions (weather, building characteristics, user's behaviour), and using essentially artificial neural networks;
· the use of a global 'cost function' allowing the long term adaptation of the integrated control system, through the minimization of that function by genetic algorithms.
Several research projects can be considered as the basis for the Smart Window project. In Switzerland, the projects Predictive Stochastic Control (a stochastic predictive heating controller using dynamic programming for finding an optimal command sequence), DELTA (a blind controller using fuzzy logic), NEUROBAT (bio-mimetic heating controller using artificial neural networks) have been carried on during the last years by LESO-PB in collaboration with other institutions and companies; in Europe, the project EDIFICIO (an integrated adaptive and predictive controller using 'soft computing' algorithms for heating, cooling, ventilation, blinds and artificial lighting) has been progressing in cooperation with some of the partners of the Smart Window project. At LESO-PB/EPFL, the algorithm development has been done in common with the corresponding development work for the EDIFICIO project.
The Smart Window project included several tasks. After a preliminary study, the three most important tasks were the algorithm development and simulation, the window design and manufacturing, and the experimental tests. The LESO was the task leader for the algorithm development and simulation, with a strong cooperation with Conphoebus. The window manufacturing was done by the industrial partners (four companies). The experimental tests were carried out in four inhabited buildings covering very different climates (from Italy to Finland), among them the LESO building.
Several versions of the Smart Window component were supposed to be developed; in particular, a stand-alone version (without any connexion to the other technical equipment like heating or artificial lighting), a version connected only to the artificial lighting system (which has a strong logical relationship with the window controller behaviour), and another version connected to an existing building bus (therefore able to take into account completely the other technical equipment). Finally, a limited number of these versions have been tested by measurements in real inhabited buildings.
|
