陷波滤波器谐振抑制参数优化方法

Optimization Method of Resonant Suppression Parameters of Notch Filters

针对伺服系统谐振抑制中,陷波滤波器的深度和宽度不合理而使谐振抑制不理想的问题,提出陷波滤波器谐振抑制参数优化方法。利用速度偏差的FFT变换得到谐振频谱图,以最大幅值处的频率确定陷波频率;在频谱图上以曲线任意点的幅值与最大幅值的比值确定深度参数,并且以所对应的频率与谐振频率的宽度确定宽度参数;为了保证参数的合理性,以滤波后谐振频率处的幅值为适应度,利用改进粒子群优化算法对陷波深度和宽度参数进行优化,避免因噪声干扰和负载变化而使陷波参数陷入局部最优。实验结果表明:该方法能在线分析机械谐振并且匹配最优参数,可以大幅度降低机械谐振。

In the resonant suppression of the servo system, unreasonable depth and width of the notch filter may lead to poor resonance suppression effect. In this paper, an optimization method of resonant suppression parameters of notch filters is proposed. The resonant spectrum is obtained by FFT transformation of velocity deviation, and the notch frequency is determined by the frequency at the maximum amplitude. The depth parameter is determined by the ratio of the amplitude at any point of the spectrum curve to the maximum amplitude, and the width parameter is determined by the width between the corresponding frequency and the resonant frequency. In order to ensure the rationality of the parameters, the amplitude at the resonant frequency after filtering is used as the fitness, and the notch depth and width parameters are optimized by the improved particle swarm optimization（PSO） algorithm. This method can avoid the parameters trapping into local optimum due to noise interference and load change. The experimental results show that this method can analyze the mechanical resonance and match the optimal parameters on-line and significantly reduce the mechanical resonance.