基于特征优化与鲸鱼算法的刀具磨损状态识别模型

Tool Wear State Recognition Model Based on Feature Optimization and Whale Optimization Algorithm

摘要: 为了提高对刀具磨损状态识别的精度，提升识别效率，针对铣刀的磨损状态提出一种基于鲸鱼算法(Whale Optimization Algorithm, WOA)与支持向量机(Support Vector Machine, SVM)的刀具磨损状态识别模型。本文首先对采集到的刀具磨损信号进行预处理，并进行多域信号分析，进行特征提取；其次，利用主成分分析(PCA)对特征向量进行优化选择，得到冗余度低的特征向量；然后利用WOA优化SVM的参数，惩罚参数 与核参数 ；最后利用优化好的WOA-SVM分类器实现刀具磨损状态的识别。通过实验对比分析，相比于SVM、PSO-SVM模型，WOA-SVM模型准确率最高，达到97.89%，且参数优化时间也比PSO-SVM模型缩短了47.35%，从两个方面验证了WOA-SVM模型的优越性。

Abstract: To improve the accuracy and efficiency of tool wear state recognition, a tool wear state recognition model based on the Whale Optimization Algorithm (WOA) and Support Vector Machine (SVM) is proposed for milling cutter wear states. In this paper, the collected tool wear signals are prepro-cessed and analyzed in multiple domains for feature extraction. Principal Component Analysis (PCA) is then utilized to optimize and select feature vectors with low redundancy. WOA is used to optimize the SVM parameters, including the penalty parameter and kernel parameter. Finally, the optimized WOA-SVM classifier is employed to achieve tool wear state recognition. Through experi-mental comparative analysis, the WOA-SVM model achieved the highest accuracy of 97.89%, which is superior to the SVM and PSO-SVM models. Additionally, the parameter optimization time was re-duced by 47.35% compared to the PSO-SVM model, which further demonstrates the superiority of the WOA-SVM model.