Experimental investigation of a novel smart energy management system for performance enhancement of conventional solar photovoltaic microgrids

Abstract

Solar photovoltaic microgrids are reliable and efficient systems without the need for energy storage. However, during power outages, the generated solar power cannot be used by consumers, which is one of the major limitations of conventional solar microgrids. This results in power disruption, developing hotspots in PV modules, and significant loss of generated power, thus affecting the efficiency of the system. These issues can be resolved by implementing a smart energy management system for such microgrids. In this study, a smart energy management system is proposed for conventional microgrids, which consists of two stages. First power production forecasting is done using an artificial neural network technique and then using a smart load demand management controller system which uses Grey Wolf optimiser to optimize the load consumption. To demonstrate the proposed system, an experimental microgrid setup is established to simulate and evaluate its performance under real outdoor conditions. The results show a promising system performance by reducing the conventional solar microgrids losses by 100% during clear sunny conditions and 42.6% under cloudy conditions. The study results are of relevance to further develop a smart energy management system for conventional microgrid Industry and to achieve the targets of sustainable development goals.