摘要
以采煤机虚拟样机实验为基础,以小波包分解和神经网络为辅助,设计一种多参数复杂机械故障诊断方法,使设备在投入使用之前就具有一套适应自身的故障诊断系统:即根据采煤机设计图样进行齿轮故障形式下的刚柔耦合虚拟样机实验,以虚拟样机实验中的惰轮轴为测试点,采集受力信号,经过小波分解,以小波分解子带能量值组成神经网络输入向量,对神经网络进行训练,得出可以根据惰轮轴受力信号诊断齿轮故障的神经网络系统。将故障虚拟样机应用到整体执行机构上,建立多个采样点作为故障信号的输出,采集受力信息,经反复实验分析得出一个收敛的神经网络故障诊断系统;重复进行虚拟样机实验,提取信号,验证系统的可靠性。
Presenting a fault diagnosis method in complex machine based on the shearer virtual prototyp experiment. The method is building a series of fault virtual prototyp experiment for its weight of testpoint, and then acquiring the fault-expression-vector corresponds to weight information by wavelet decomposition and neural network analysis. Through the analysis of neural network, building a classification system to diagnose the fault of machine. The fault virtual prototyp was applied to whole machine, and various testpoint was set. At length, the method was verified by simulation signals and engineering examples of mechanical fault diagnosis effectively.
作者
赵丽娟
付东波
李明昊
Zhao Lijuan;Fu Dongbo;Li Minghao(College of Mechanical Engineering,Liaoning Technical University,Fuxin 123000,Liaoning,China)
出处
《现代制造工程》
CSCD
北大核心
2018年第11期142-148,共7页
Modern Manufacturing Engineering
基金
国家自然科学基金项目(51674134)
关键词
故障诊断
柔性体
虚拟样机
神经网络
fault diagnosis
flexible bodies
virtual prototype
neural network
