基于DE-RBF神经网络的短电弧脉冲电源恒压控制

Constant Voltage Control of Short Arc Pulse Power Supply Based on DE-RBF Neural Network

为获得较好的短电弧脉冲电源稳压效果,实现短电弧加工高效率、低损耗的目的,提出了一种基于差分进化算法(DE)优化RBF神经网络的PID控制策略。建立了MATLAB/Simulink电源系统仿真模型,仿真对比分析了传统PID及RBFPID控制策略,并进行了短电弧铣削加工对比实验。结果表明:新型控制策略下的脉冲电源输出稳定、抗干扰性强、稳压效果好,加工效率提升59.7%,相对电极损耗减少3.07%。

In order to obtain better voltage stabilizing effect of short arc pulse power supply and to achieve the purpose of high efficiency and low loss for short arc processing,a Proportion Integral Derivative(PID)control strategy based on differential evolution algorithm(DE)to optimize Radial Basis Function(RBF)neural network is proposed.The simulation model of MATLAB/Simulink power system was established.The traditional PID and RBF PID control strategies were compared and analyzed,and the comparison experiment of short arc milling was carried out.The results show that the output of the pulse power supply under the new control strategy has stable output,strong anti-interference and good voltage regulation effect.The processing efficiency is increased by 59 7%and the relative electrode loss is reduced by 3 07%.