Distributed MPC of the Standalone Hybrid Wind and Solar Generation System Based on Neural Network Modeling

摘要

In this paper, we proposed a distributed model predictive control (DMPC) strategy for optimal management and operation of a wind and solar hybrid generation system which consists of wind and photovoltaic generation subsystems, a battery bank and an ac load. Due to the nonlinearity of the system, data-driven modeling with back propagation (BP) neural network is used for each power generation subsystem. And it is linearized online to calculate the control policy from a quadratic programming problem. The performance of the distributed model predictive control is shown through computer simulation, which illustrates the efficacy and feasibility of the proposed method.