Energy Storage for Hourly Dispatching Utility-Scale Solar PV Power using HOMER Pro - A Cost Investigation

Abstract

The objective of this study is to investigate the expenditure of different kinds of energy storage systems (ESSs) for the economical dispatching of solar power at one-hour increments for an entire day for megawatt-scale grid-connected photovoltaic (PV) arrays. Accurate forecasting of PV power is vital for generation scheduling and cost-effective operation. A multilayer perceptron Artificial Neural Network (ANN) is utilized to predict PV irradiance one hour ahead of time, which performs well with good convergence mapping between input and target output data. Moreover, this research proposes a state of charge (SOC) control algorithm based on an adaptive neuro-fuzzy inference system (ANFIS) that can accurately estimate the grid reference power for each one-hour dispatching period, which is necessary for ensuring the ESS completes each dispatching period with its starting SOC and has sufficient capacity for next-day operation. Finally, an economic comparison is presented utilizing the Hybrid Optimization of Multiple Energy Resources (HOMER Pro) software to develop a cost-effective ESS for an hourly PV power dispatching scenario.