Predictive and flexible system controller for hybrid power supply systems

Abstract

Regarding the overall efficiency of hybrid power supply systems, a system controller as an addition to device controllers was developed to contribute to its increase. An essential premise in the development of such a controller was the modular structure and the implementation as an open source solution. By using the MQTT protocol, the controller can operate within an IoT-network and hence be used universally, e.g. in a Software- (SiL) or in the Hardware-in-the-loop (HiL) environment as well as in field test. In order to make optimal use of the flexibilities provided by thermal or electrical storages, MPC (Modell Predictive Control) functionalities were implemented in the controller, which, however, are supported by rule-based algorithms on a “fallback levelâ€. The target values of the control can be adjusted variably and include economical as well as ecological aspects in a weighted way. The open structure of the controller makes it possible to easily include other system components in the control concept and also to operate the controller either locally, in an edge device or in a cloud environment. The performance of the controller was demonstrated by SiL- and HiL-tests. Here, annual characteristic values were determined on the basis of representative days. The representative days were selected using a cluster procedure. Overall, energy and cost savings can be demonstrated by using the controller, which are particularly effective in the optimized use of electricity from photovoltaic systems in combination with heat pumps.