基于多元高斯分布的零件尺寸异常检测  

Part Size Anomaly Detection Based on Multivariate Gaussian Distribution

作　　者：吴林慧 何毅斌[1,2] 陈宇晨 杜伟 汪强 WU Lin-hui;HE Yi-bin;CHEN Yu-chen;DU Wei;WANG Qiang(School of Mechanical and Electrical Engineering,Wuhan Institute of Technology,Wuhan 430205,China;Hubei Key Laboratory of Chemical Equipment Strengthening and Intrinsic Safety,Wuhan Institute of Technology,Wuhan 430205,China)

机构地区：[1]武汉工程大学机电工程学院,武汉430205 [2]武汉工程大学化工装备强化与本质安全湖北省重点实验室,武汉430205

出　　处：《组合机床与自动化加工技术》2022年第4期92-95,共4页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：湖北省科技厅重大专项(2016AAA056);化工装备强化与本质安全湖北省重点实验室开放基金项(2018KA01);武汉工程大学第十二届研究生教育创新基金(CX2020043)。

摘　　要：针对零件尺寸异常检测中废品样本难以获取,并且特征维度不高的特点,提出一种建立基于数据分布特性的多元高斯分布模型对零件尺寸是否超差进行异常检测的方法。首先,通过利用极大似然估计计算模型的参数;其次,为了简化协方差矩阵求逆时的复杂性,使用平方根分解优化模型;最后,利用问题转化后的改进粒子群优化算法求解最优阈值。实验中将零件的长、宽、圆的大小作为模型的输入,通过真实数据验证该方法的准确率达到97.5%,耗时缩短到2.433 s,性能显著提升,表明此方法可以有效应用于尺寸超差故障检测。Aiming at waste in anomaly detection component size sample is difficult to obtain, and the characteristics of the feature dimension is not high, establish a multivariate gaussian distribution model based on data distribution characteristics of whether parts size out-of-tolerance anomaly detection, calculation model by using maximum likelihood estimation of parameters, in order to simplify the complexity of the covariance matrix inversion, using square root decomposition optimization model, finally using problem after the improved particle swarm optimization algorithm to solve the optimal threshold.In the experiment, the length, width and circle size of the parts are taken as the input of the model.The real data verify that the accuracy of the method reaches 97.5%,the time consumption is reduced to 2.433 s, and the performance is significantly improved.It shows that this method can be effectively applied to the size out of tolerance fault detection.

关 键 词：异常检测 多元高斯分布 极大似然估计 平方根分解 粒子群优化算法 

分 类 号：TH16[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]