摘要
针对某口径高速自动机转膛体衬套温度升高,发生膨胀,摩擦阻力增大,导致击发不响的非平稳性故障,利用现场采集的数据,在分析自动机工作原理的基础上,将小波变换与神经网络进行结合,对实测信号进行分解重构,提取隐藏的潜在故障信息,评估各部件健康状态,实现对高速自动机的故障诊断和预测。最后根据实例数据分析和MATLAB仿真,从预测输出曲线基本拟合,以及预测输出值与实际值之间的误差百分比控制在5%以内表明小波神经网络方法非常适合应用于设备故障预测,是一种效果显著的方法。
In view of the non-stationary fault of a certain caliber high-speed automatic machine,which is caused by the increase of temperature,expansion and friction resistance,this paper uses the data collected in the field,analyzes the working principle of the automaton,combines wavelet transform with neural network,decomposes and reconstructs the measured signal,extracts hidden potential fault information,and evaluates each part The fault diagnosis and prediction of high-speed automata are realized.Finally,according to the case data analysis and MATLAB simulation,the prediction output curve is basically fitted,and the error percentage between the predicted output value and the actual value is controlled within 5%.It shows that the wavelet neural network method is very suitable for equipment fault prediction,and it is an effective method.
作者
张昕昕
敬伟
王鹏
张宁超
Zhang Xinxin;Jing Wei;Wang Peng;Zhang Ningchao(School of Electronic Information Engineering,Xi'an Technological University,Xi'an 710021,China)
出处
《国外电子测量技术》
2020年第8期11-16,共6页
Foreign Electronic Measurement Technology
基金
陕西省科技厅重点研发计划(2019GY-075)
西安市科技局科技创新引导项目(201805031YD9CG15(1))
西安市科技局高校人才服务企业项目(2019217214GXRC008CG009-GXYD8.1)资助。
关键词
自动机
故障预测
小波分析
神经网络
automaton
fault prediction
wavelet analysis
neural network
