Agent systems for intelligent and robust control of complex energy systems in non-residential buildings as part of the higher-level energy system
Duration: 3 years (05/2019 – 04/2022)
- RWTH Aachen, E.ON Energy Research Center, Institute for Energy Efficient Buildings and Indoor Climate
- Robert Bosch GmbH, Future Systems for Building Technology, Management & Automation
- Friederich-Alexander-University Erlangen-Nuremberg, Control Engineering (LRT)
The high complexity of energy systems in non-residential buildings leads to difficulties realizing an efficient operation in practice while taking into account comfort requirements. State of the art is a central control of the building energy system. As a result, the individual subsystems are not sufficiently coordinated with other system components to enable robust and efficient operation due to the complexity of the system.
The pursued approach is to reduce the complexity by distributing the tasks to individual software agents. In this context, an agent is a autonomous unit with the aim to achieve defined goals independently. This requires communication with other agents and the environment. In an agent system, different agents solve one or more tasks together.
The goal of the project is the development of an agent system for non-residential buildings to increase the reliability, durability and efficiency of the components used. The first step is to define evaluation criteria and develop user models for a demand-oriented energy supply. Additionally, it will be evaluated, how a flexible operation of the building can be used to support the electrical distribution grid. Subsequently, we implement the agents as a part of a test platform. The agent system should configure itself automatically and control the various components in a self-learning and predictive manner. After simulating the developed agents, a demonstration of the designed agent system in an office building is realized. Finally, we will compare the obtained results with the state of the art regarding to the defined evaluation parameters.
We gratefully acknowledge the financial support provided by the BMWi (Federal Ministry of Economic Affairs and Energy), promotional reference 03ET1495A.