面向复杂旋转机械故障诊断的混沌粒子群算法研究  

Research on Chaos Particle Swarm Optimization Algorithm for Fault Diagnosis of Complex Rotating Machinery

作　　者：黄惇晶 HUANG Dun-jing(Minxi Vocational&Technical College,Longyan 364000,Fujian)

机构地区：[1]闽西职业技术学院,福建龙岩364000

出　　处：《电脑与电信》2023年第5期95-99,109,共6页Computer & Telecommunication

摘　　要：针对传统旋转机械故障诊断算法存在的诊断精度低、分类性能差的不足,提出一种面向复杂机械故障诊断的混沌粒子群算法。基于S变换提取非平稳故障信号的特征,并利用高斯窗口函数抑制噪声干扰,提升故障信号的局部适应度;利用混沌理论中的混沌振子,提取微弱故障信号的振幅和频率,并基于粒子群算法在全局范围内寻找最优的模型参数组合;依据混沌优化理论改善经典粒子群算法的局部寻优性能,在全局范围内获取最优解。实验结果表明,在3种不同的电机载荷条件下,提出算法均能够准确识别出故障样本,且针对不同故障类型的平均分类准确率在98%以上。Aiming at the shortcomings of traditional fault diagnosis algorithms for rotating machinery,such as low fault diagnosis rate and poor classification performance,a chaotic particle swarm optimization algorithm for fault diagnosis of complex mechanical structure is proposed.The feature of non-stationary fault signal is extracted based on S transform,and the Gaussian window function is used to suppress the noise interference and improve the local fitness of the signal.The amplitude and frequency of weak fault signal are extracted by chaotic oscillator in chaos theory,and the optimal parameter combination is found in the global range based on particle swarm optimization algorithm.According to chaos optimization theory,the local optimization performance of PSO is improved and the global optimal solution is obtained.The experimental results show that the proposed algorithm can accurately diagnose the fault samples under three different motor loads,and the fault classification accuracy is more than 98%.

关 键 词：旋转机械 混沌粒子群 S变换 混沌振子 全局寻优 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]