Artificial Neural Network-based PI control of battery energy storage system connected with MMC-HVDC system

Offshore wind farm (OWF) has received considerable attention as it is growing exponentially worldwide. Nowadays, a modular multilevel converter-based high voltage direct current (MMC-HVDC) system is a promising topology for connecting the offshore wind farm to the grid because of its modular characteristics and low switching frequency operation. Moreover, it does not require an expensive filter as it can generate almost pure sine waves. The energy storage system is used to improve the transient performance of MMC-HVDC and to manage the surplus energy in the HVDC system. Moreover, it is used to offer auxiliary services to the electrical grid. The lithium-ion battery energy storage system is used widely because of its superior performance in terms of high energy density and stability. A DC-DC boost converter is used as an interface between BESS and MMC-HVDC. This paper proposed a control strategy for a DC-DC boost converter based on artificial neural network-based proportional integral control (ANN-based PI). Firstly, the proportional-integral (PI) control method is used as an expert system; then, the dataset is utilized for training the ANN. To prove the effectiveness of the proposed control method, the system undergoes testing in the presence of a three-phase to ground fault near the grid, which is then compared to the performance of the PI controller. Furthermore, the performance of ANN is measured using MATLAB/Simulink
^®
.