基于假设检验和支持向量机的旋转机械故障诊断方法

发布时间：2018-01-12 17:38

本文关键词：基于假设检验和支持向量机的旋转机械故障诊断方法　出处：《中国机械工程》2017年07期 　论文类型：期刊论文

【摘要】：针对旋转机械故障诊断中存在的早期非平稳微弱故障信号特征提取困难、故障诊断不准确等问题,提出了一种基于自适应假设检验滤波和支持向量机(SVM)的故障诊断方法。该方法采用统计学假设检验原理来评估参考信号(噪声信号)和原始信号(故障信号)在频域上的相似性,删除具有高相似性的频域成分;通过粒子群优化算法获得最佳的显著性水平α;定义评估因子Ipq来评价假设检验滤波的效果。最后通过SVM来逐次诊断轴系构造异常。验证结果表明该方法能够有效地诊断出传动轴不对中和不平衡的故障类型。
[Abstract]:In order to solve the problems in fault diagnosis of rotating machinery, it is difficult to extract the feature of early non-stationary weak fault signal, and the fault diagnosis is not accurate. A fault diagnosis method based on adaptive hypothesis test filtering and support vector machine (SVM) is proposed. The statistical hypothesis test principle is used to evaluate the reference signal (noise signal) and the original signal (. Fault signal) similarity in frequency domain. Deleting the frequency-domain components with high similarity; The best significant level 伪 is obtained by particle swarm optimization. The evaluation factor Ipq is defined to evaluate the filtering effect of hypothesis testing. Finally, the shafting structure anomaly is diagnosed by SVM step by step. The results show that this method can effectively diagnose the misalignment and imbalance of the drive shaft. Type of barrier.
【作者单位】：江南大学江苏省食品先进制造装备技术重点实验室;三重大学;
【基金】：国家科技支撑计划资助项目(2015BAF16B02)
【分类号】：TH17
【正文快照】： 0引言振动信号故障特征的提取直接关系到故障诊断的准确性和早期故障预报的可靠性,是设备状态监测和故障诊断的关键[1-2]。大多数旋转机械在实际运行状态下,其故障信号是非平稳、非线性的,同时故障信号中包含过多的干扰信号,故障信号能量微弱,故而故障特征的提取十分困难。在

【相似文献】