基于递归径向基神经网络滑模的多功能柔性多状态开关控制方法OA

In recent years,the increasing penetration of new energy and electric vehicles poses significant challenges to the current optimization and power quality management of distribution networks.In response to the problems of the stochastic and intermittent nature of distributed generation,a control method for a flexible multi-state switch(FMS)based on recursive radial basis function neural network(RRBFNN)sliding mode is proposed in this paper.The objective is to achieve power interaction,flexible arc suppression for multi-terminal single-phase ground faults,and enhance the disturbance rejection capabilities of the FMS.Beginning with the consideration of parameter perturbations,an improved RRBFNN sliding mode control method is introduced to overcome the inherent chattering in traditional sliding mode control,reduce reliance on the precise mathematical model of the system,and mitigate transient impacts during grid connection.A calculus-based sliding mode surface is employed for flexible arc suppression control,and the control law for the zero-sequence voltage is theoretically derived,enhancing fault current suppression rate.The stability and convergence of the proposed method are further demonstrated through Lyapunov's theorem.Finally,a simulation model of a three-port FMS with its control system is developed in Matlab/Simulink.The feasibility and effectiveness of the proposed strategy are verified through simulation comparisons.