Helical Gear Machining Quality Estimation Based on Processing Parameters Importance Analysis

Abstract

It is a difficult problem to predict the machining error of helical gears, among which high-dimensional processing parameter variables are an obstacle. To solve this problem, a helical gear machining quality prediction method based on processing parameters importance analysis (MQP-PPIA) is proposed. First, Kernel Principle Component Analysis (KPCA) algorithm is used to preliminary reduce the dimension of the high-dimensional quality inspection data of helical gear. Then, an improved Birch clustering algorithm with initial parameters auto-generator is proposed. The one-dimensional quality grade label obtained by clustering algorithm is used as the decision attribute of rough set algorithm for parameter importance analysis and attribute reduction. Finally, the Voting Regression (VR) algorithm is adopted to establish the mapping relationship between the reduced processing parameters and the quality inspection indicators to predict the helical machining error. According to experimental verification, the proposed Auto-Birch method has a good clustering effect, and the adopted VR algorithm also has a good prediction performance for helical gear machining error prediction.