A capacity configuration study of energy storage units for  power plants based on a novel load prediction technology

Abstract

As the electricity demand for human activities increases, the power grid operation faces constantly rising pressure. The capacity configuration of energy storage systems has recently been a widespread research issue, especially regarding renewable energy system research. An unqualified capacity configuration leads to an insufficient power supply to the power grid. Therefore, this paper proposes a capacity allocation of energy storage power plant based on power load predicting technology, combining the grey predicting model, improved BP neural network predicting model and multiple linear Regression predicting models to establish a grey regression neural network predicting model to predict the future power load; constructing a photovoltaic (PV) energy storage power plant capacity configuration model based on the predicting results, and solving the model using multi-objective snake optimization algorithm. The optimal value of PV energy storage plant capacity configuration is obtained using a multi-objective snake optimization algorithm. The results show that the grey regression neural network prediction model proposed in this paper has reliable prediction results and a high accuracy rate, and the correlation coefficient is as high as 99.7% in the regression analysis; the energy storage power plant can share about 150 kW power loads in the grid during the peak electricity consumption period, which effectively reduces the pressure on the grid operation.

Keywords: Photovoltaic, algorithm, capacity allocation, peak shaving and valley filling