基于铣削力与神经网络刀具磨损状态识别研究

Research on Tool Wear State Recognition Based on Milling Force and Wavelet Neural Network

以铣削难加工材料———高锰钢加工过程为研究对象,建立了以铣削力作为监测信号的铣刀磨损监测实验系统。应用小波包理论对铣削力信号进行分析和消噪处理,并提取了信号的能量特征作为神经网络的输入向量。基于神经网络极强的非线性映射能力及分类能力,选用小波包分析与BP网络结合的方式对刀具磨损状态进行识别。建立了模式识别BP网络结构,构造了网络训练样本及测试样本,对网络进行了训练、仿真及验证测试,结果表明该网络能够对刀具磨损状态进行准确的识别,对刀具的在线监测具有良好的现实意义。

The dissertation takes the course of milling the hard -to -cut material, High Manganese Steel as the research object, establishes milling cutter wear state monitoring experimental system with the milling force as the monitoring signal. The dissertation adopts wavelet package theory to analyze and de - noise the milling force signal, and extracts energy feature of the signal as import vectors of Neural Network. Because Neural Network has strong non -linear mapping capability and classified capability, the dissertation adopts the combine of wavelet package analysis and BP （ Back Propagation ） Neural Network to recognize the tool wear state, establishes the configuration of BP network, structures training swatch and testing swatch. The results of training, simulating and validating network indicate it can exactly recognize the tool wear state and has significant realistic meaning to tool online monitoring.