电弧焊接凝固开裂的模糊分类

Fuzzy Classification of Solidification Cracking in Arc Welds

针对低合金钢电弧焊工艺焊缝热裂纹形成条件判别问题，提出一种新的模糊分类器，其每条规则能以不同的概率表达一类以上的类别。首先，将分类器扩展为用混合模型估计类别条件密度的二次Bayes分类器，使其具有Takagi-Sugeno型模糊推理规则。其次，将Gath-Geva聚类算法改进为一种监督式聚类算法，用于辨识。最后，通过Fisher类间可分性实现输入参数的选择，简化分类器结构。该方法用于焊缝凝固开裂的分类识别。

For the solidification cracking during the arc welding of low alloy steels, a new fuzzy classifier structure is proposed which each rule can represent more than one class with different probabilities. The obtained classifier can be considered as an extension of the quadratic Bayes classifier that utilizes mixture of models for estimating the class conditional densities. A supervised clustering algorithm based on improved Gath-Geva algorithms has been worked out for the identification of this fuzzy classifier with Takagi-Sugeno type inference rules. The relevant input variables of the fuzzy classifier with an interpretable rule base have been selected based on the analysis of the clusters by Fisher＇s interclass separable criteria. This new approach is applied to the weld hot cracking process conditions classification problem.