基于机器学习的齿轮箱故障诊断
The Fault Diagnosis of Gearbox Based on Machine Learning

齿轮箱是用于增加输出扭矩或改变电机速度的机械装置，被广泛应用于如汽车、输送机、风机等机械设备中。在齿轮箱运行过程中，可以通过加速度传感器采集振动信号来判断齿轮箱是否出现异常，传统对数据进行分析的方式耗时大，取而代之的机器学习诊断方式能够有效的诊断其中的问题数据判断机械故障情况。基于此问题，本文提出了基于支持向量机(SVM)、决策树分类和BP神经网络分类的三种分类模型，建立故障数据分类模型，分别进行故障数据的分类和对比，最终得出最好的分类模型。
The gear box is a mechanical device used to increase the output torque or change the speed of the motor, which is widely used in mechanical equipment such as automobiles, conveyors and fans. During the operation of the gearbox, the vibration signal collected by the acceleration sensor can be used to judge whether the gearbox is abnormal. The traditional method of analyzing the data takes a lot of time, and instead, the machine learning diagnosis method can effectively diagnose and pre-dict the problem data to judge the mechanical failure. To solve this problem, this paper puts for-ward three classification models based on support vector machine (SVM), decision tree classification and BP neural network classification, establishes the fault data classification model, classifies and compares the fault data respectively, and finally obtains the best classification model.