Droop Control Design of MTDC Systems with Large-Scale Renewable Integration based on ANFIS

Abstract

Large-scale integration of renewable energy in highvoltage direct-current (HVDC) systems has resulted in significant technical and economic challenges. Maintaining DC voltage stability and power balance is one of the critical issues to achieve the stable operation of a multi-terminal HVDC (MTDC) system. This paper presents a control strategy based on adaptive network-based fuzzy inference system (ANFIS) for the MTDC system with large-scale renewable integration. The droop controller based on ANFIS is designed by the DC voltage deviation and active power margin of the converter stations. To improve the performance in the training process of ANFIS, particle swarm optimization (PSO) is utilized to optimize the prerequisite and conclusion parameters of ANFIS. Compared with the traditional droop control approach, the ANFIS-based controller can track the active power and DC voltage deviation of the converter stations in real time, allowing it to reduce the DC voltage deviation and reasonably distribute unbalanced power in the case of fluctuations in renewable energy power output. Simulations are conducted to compare with the traditional control method under different working conditions. The comparative simulation results show the effectiveness and robustness of the proposed control strategy.