基于BP神经网络的不同转速下切削声音识别

Identifying Cutting Sounds of Different Spindle Speed With a BP Neural Network

机械加工发出的声音可用于制造过程智能化监控。加工时参数不同,发出的声音也不相同。对不同转速下切削声音加以识别,可为进一步综合考虑其它切削加工参数和条件下的声音识别打下基础。为识别不同转速下的切削加工声音,设计了一个三层BP神经网络,且先对采集声音进行小波包分解,求出分解后各频率段成分的能量,归一化处理后构成特征向量,再将处理后的信号分为训练样本集和测试样本集,对网络进行训练和测试。测试结果表明该网络具有训练速度快和识别准确率高的特点。

Sounds of machining can be used to monitor machining process intelligently. They are not the same un- der different cutting conditions. Those sounds of different spindle speeds are recognized first. This work can be further extended to the investigation of other contributing conditions. Here, A three level BP neural network is designed to identify those sounds of different spindle speed. And these sound signals are decomposed by wavelet packet first, the energies of every frequency band are calculated, they form into character vector. Then, the sounds disposed are divided into two parts： training samples and test samples. They are used to train and test the BP network. The test result shows that the network has the advantage of fast training speed and high recognition ratio.