Model Predictive Control based Energy Collaborative Optimization Management for Energy Storage System of Virtual Power Plant

摘要

This paper presents an energy collaborative optimization management for an energy storage system (ESS) of virtual power plant (VPP) based on model predictive control (MPC). This method uses long-short term memory (LSTM) neural network to obtain the one hour-ahead forecasting information for the load, the generation of wind and photovoltaic within the jurisdiction of VPP. With the minimum economic cost of VPP as the optimization goal, the optimal scheduling is solved by an improved particle swarm optimization (PSO) algorithm in the concept of the MPC framework. Through the comparison with the conventional VPP optimization solution, the numerical results clearly demonstrated that the proposed method improves the utilization of distributed generators (DGs) and reduces the impact of prediction errors on the optimization results.