论文部分内容阅读
矿用车辆变速箱的运行状态判断主要依靠人工拆卸、观察的方式,检测诊断的效率低下,甚至隐含着较大的安全隐患。以矿用车辆变速箱圆柱滚子轴承为研究对象,利用小波包对轴承故障振动信号进行分解,区分比较不同状态信号的总能量熵,利用支持向量机原理对矿用车辆变速箱圆柱滚子轴承故障进行诊断。试验表明,应用SVM和小波包能量熵可以对轴承故障进行有效诊断,且准确率较高,研究对矿用车辆变速箱圆柱滚子轴承的故障诊断有着现实的指导意义。
Mining vehicles to determine the operating status of the gearbox mainly rely on manual disassembly, observation, detection and diagnosis of inefficient, and even implies a greater security risk. Taking the gearbox cylindrical roller bearing of mining vehicle as the research object, wavelet packet is used to decompose the vibration signal of bearing fault to distinguish and contrast the total energy entropy of different state signals. The support vector machine theory is applied to the cylindrical roller bearing of mining vehicle gearbox Fault diagnosis. Experiments show that the application of SVM and wavelet packet energy entropy can effectively diagnose the bearing faults with high accuracy. The research has a realistic guiding significance to the fault diagnosis of the cylindrical roller bearing of gearbox for mining vehicles.