The paper describes a novel adaptive fuzzy logic controller based energy management concept (A-FLC-EM) for a stand-alone photovoltaic (PV) hybrid system with battery and hydrogen storage path. The reference application is a single family home. The basic idea is to switch and optimally adjust the energy management parameters according to identified changes of distinct longterm energy supply and/or demand situations. Key elements of the offline learning phase are the analysis of the energy time series and the automatic determination of distinct energy situations on the basis of a segmentation algorithm and a vector of suitable statistical features calculated for a short-term, sliding observation window. A bottom-up approach is used, ranking and selecting statistical features that are particularly good at distinguishing certain long-term energy situations. The selected features form the basis for a clustering algorithm to detect and describe distinct energy situations. For each energy situation, the calculation of optimal energy management parameters is performed for a training data set employing particle swarm optimization (PSO). The performance of the novel A-FLC-EM is demonstrated compared to a conventional fuzzy logic controller based energy management (FLC-EM) with an all-year fixed parameter setting. Qualitative and quantitative improvements as well as further challenges are discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.