Proposal of an Intelligent, Predictive Fuzzy Controller for Off-Grid Devices

Off-grid electrical devices, such as autonomous lamps, can operate without being connected to the power grid thanks to integrated renewable-source based energy generators and storage facilities. However, they may not be able to fulfill all of the user's requirements in unfavourable conditions. Controllers currently used in off-grid devices use simplistic algorithms and strategies. This paper presents a proposal of an intelligent, fuzzy logic-based controller for off-grid devices with prediction and planning capabilities. Artificial neural networks of various types are used for prediction of generator performance, monitoring of component performance and detection of events such as partial malfunction. Fuzzy logic is used for to facilitate the definition of rules reflecting the user's preferences, and to define the internal strategies of the controller. The predicted behaviour of the power-related components is used to select the operating mode which fulfils the prioritised user's requirements to the maximum extent, while minimizing the chance of running out of power.