Data-based Integration Modeling of Building Groups and Distributed Energy Systems and Self-systems and Self-learning Optimal Control
Abstract: With the rapid increase of construction scale, energy consumption of buildings in China has become a very serious problem nowadays. This project will establish a unified modeling method for cluster of buildings and distributed energy sources based on the vast amount of data. It will also establish self-learning optimal coordination control schemes for such unified systems. First, data-based feature analysis and modeling for the unified systems of cluster of buildings with distributed energy sources will be conducted using cluster analysis and neural networks. Second, considering the requirements of comfort and structural properties of buildings, self-learning optimal control methods based on adaptive dynamic programming will be established for three different cases of the unified systems of single-building with distributed energy sources, i.e., systems without distributed energy sources, with energy storage devices, and with both energy storage devices and clean renewable energy sources. Third, distributed iterative adaptive dynamic programming methods and multiagent-based adaptive dynamic programming approach will be developed for optimal coordination control of unified systems of cluster of buildings with distributed energy systems. Fourth, optimal coordination control will be established for the unified systems of cluster of buildings with distributed energy sources with uncertainties and time-delays based on multiagent systems theory. Finally, a platform will be established for energy consumption analysis and optimization of smart community. Results from this project will be applied to real systems with notable economic and social impacts. This project will push forward the frontiers of coordination control of cluster of buildings with distributed energy sources. On a larger scale, this project will help to improve energy efficiency, to ease the energy shortage, and to promote the healthy development of the national economy. Therefore, this project has a lasting strategic importance.
Keywords: adaptive dynamic programming; optimal control systems; data-based control; adaptive learning systems; data-based control applications
Contact:
LIU Derong
E-mail: derong.liu@ia.ac.cn
The State Key Laboratory of Management and Control for Complex Systems