Optimization of photovoltaic battery swapping station based on weather/traffic forecasts and speed variable charging

• Published in: Applied energy 2020-04, Vol.264, p.114708, Article 114708
• Main Authors: Feng, Jiawei, Hou, Shengya, Yu, Lijun, Dimov, Nikolay, Zheng, Pei, Wang, Chunping
• Format: Article
• Language: English
• Subjects: Battery aging; Capacity optimization; Operation method optimization; Peak-shifting; PV-BSS
• ISSN: 0306-2619; 1872-9118
• DOI: 10.1016/j.apenergy.2020.114708

•Photovoltaic-battery swapping station economic model is developed.•Battery degradation as a function of the charging speed is implemented to the model.•Weather and traffic forecasts are accounted.•Optimized speed variable charging method is proposed.•Superiority of the speed-variable charging is verified by a case study. An economic model of integrated Photovoltaic - Battery Swapping Station (PV-BSS) is developed in this work. Speed-variable charging taking into account battery degradation models of modern lithium-ion batteries is combined with weather and road traffic forecasts for the first time to maximize the economic and environmental impacts of this emerging technology. Annual net revenue is used as an objective function, and the power balance, charging rate limit, and consumer demand are set as model constraints. Particle swarm optimization (PSO) algorithm complemented with non-linear programming features is applied to optimize the operation of the integrated PV-BSS. The extensive model developed in this work let us propose an optimized speed variable charging method with clear economic and environmental advantages over the other methods considered in the literature so far. A case study with specific values of input parameters fed into the model is performed to verify the applicability of the model and the reliability of its predictions. The case study confirms that the proposed speed-variable charging method dispatches the PV-generated energy in an intelligent manner and shifts the energy of the grid from the peak to the valley. The model is easily extendable, allows implementation of additional metrics of interest and could be an important enabler of the upcoming smart grid revolution.