改进粒子群优化小波阈值的矿用钢丝绳损伤信号处理方法研究

为有效提取矿用钢丝绳损伤信号的特征值,采取小波分析对损伤信号去噪。针对损伤信号中存在小奇异点的特性,对小波分析中的阈值获取和阈值函数选取两方面改进。首先利用粒子群算法优化经验值,并基于Birge-Massart策略获取阈值。提出一种改进的小波阈值函数算法。该函数加入了可调变量,改善了已有软、硬阈值函数去噪中的不足点,通过仿真实验的信号结果和信噪比(SNR)对比几种阈值函数去噪算法,最终得出,采用优化经验值并改进小波域值函数的去噪算法相比于其他方法,更能完整保留原始信号,去噪效果好。

Mine wire rope damage signal processing method based on improved particle swarm optimization wavelet threshold

In order to extract the characteristic value of the damage signal of the mining wire rope,wavelet analysis is used to denoise the damage signal.For the characteristics of small singular points in the damage signal,the threshold acquisition and threshold function in wavelet analysis are improved.First,the particle swarm optimization algorithm is used to optimize the empirical value,and the threshold is obtained based on the Birge-Massart strategy.An improved wavelet threshold function algorithm is proposed.This function adds adjustable variables,which improves the shortcomings of existing soft and hard threshold function denoising.The signal results of simulation experiments are compared with several threshold function denoising algorithms,and the signal-to-noise ratio(SNR)data evaluation denoising effect.Finally,the denoising algorithm which optimizes the empirical value and improves the wavelet domain value function is more able to completely retain the original signal than the other methods,and the denoising effect is good.