VRB/EDLC Hybrid Storage System with PID Controller Based on BP Neural Network

Abstract

In order to stabilize fluctuations caused by intermittent energy and complement different storage elements advantages, we designed a hybrid energy storage combining VRB pack with EDLC bank. This hybrid energy storage device has two DC/DC convertors; DC/DC(A) controls VRB pack's power and DC/DC(B) stabilizes the voltage on DC bus. By doing this, it is easy to make power schedule for VRB according to renewable energy forecast and SOC of battery, while EDCL bank stabilizes the short-time power fluctuation, most of which result from forecasting error. To achieve the above goals, self-adaptive PID controller based on BP neuron network is applied in DCDC(B), for switching-mode convertor is strongly non-linear system and traditional PID controller is not suitable to the case of large changed load. Simulations in Matlab Simulink are presented to testify the novel control method and show this hybrid energy storage device can work efficiently.