论文部分内容阅读
针对强力输送带钢丝芯漏磁检测信号成分复杂、噪声干扰严重、钢丝芯断丝损伤程度难以识别等问题,采用小波-支持向量机算法对钢丝芯断丝信号进行提取及识别。首先用小波分析技术滤除检测信号的噪声,并根据钢丝缺陷先验知提取断丝特征,形成断丝特征样本集;采用支持向量机对断丝损伤信号进行识别分级;最后使用实际数据进行实验验证。结果表明,该算法识别结果与实际情况基本一致,为钢丝芯损伤监测研究提供了一种新的方法。
Aiming at the problems of complex signal detection of magnetic flux leakage in wire core of power transmission belt, serious interference of noise and difficulty in recognizing the damage degree of broken wire core, wavelet packet support vector machine (SVM) algorithm is used to extract and identify the broken signal of wire core. Firstly, the noise of the detection signal is filtered by wavelet analysis, the broken wire feature is extracted according to the prior knowledge of the steel wire defect to form the broken wire feature sample set, and the support vector machine is used to identify and classify the broken wire damage signal. At last, the experimental data verification. The results show that the recognition result of the algorithm is basically consistent with the actual situation, and provides a new method for the damage monitoring of steel core.