激光增材制造零部件故障自动分类方法

Automatic classification method of laser additive manufacturing component faults

不同状态下的零部件故障特征不同,为了准确分辨零部件故障,为解决故障提供依据,提出了激光增材制造零部件故障自动分类方法。基于阶次分析原理提取了激光增材制造零部件的故障特征,并获取了零部件故障特征向量;利用支持向量机中的一对一分类方法,对获取的零部件故障特征进行分类,基于分类结果对其进行校正,达到提高故障分类精度的目的,最终实现激光增材制造零部件故障自动分类。通过对该方法进行内圈故障和外圈故障两种不同状态下进行分类识别效果测试,验证了该方法具有较高的激光增材制造零部件故障分类准确性。

The fault characteristics of parts in different states are different.In order to accurately distinguish the fault of parts and provide a basis for solving the fault,an automatic fault classification method of parts in laser additive manufacturing is proposed.Based on the principle of order analysis,the fault features of laser additive manufacturing parts are extracted,and the fault feature vector of parts is obtained;The one-to-one classification method in support vector machine is used to classify the obtained fault features of parts,and correct them based on the classification results,so as to improve the accuracy of fault classification,and finally realize the automatic fault classification of parts in laser additive manufacturing.By testing the classification and recognition effect of this method in two different states of inner ring fault and outer ring fault,it is verified that this method has high fault classification accuracy of laser additive manufacturing parts.