基于SC-BP的设备故障预警研究

Research on Fault Alarm of Equipment Based on SC-BP

为实现对火炮系统健康状态监测和故障预警,满足从“故障修”向“状态修”转变的发展需求,采用拟合模型算法,建立设备标准化电流优化BP神经网络预测模型(SC-BP),分析运行数据的内在逻辑和相关性,设计设备正常运行状态下的数据参数库,并与实时采集的设备状态数据进行比较,根据拟合优度进行辨识,通过比较设备实时数据与历史正常数据之间的相似关系来判断设备状态,并显示故障关联信息,为工作人员分析故障原因提供了参考,并对算例进行了分析验证。研究结果表明,该故障预警方法能够辨识设备性能异常状况,使维护人员能够在设备故障形成早期发现征兆,提前采取措施以避免设备故障的发生。

In order to monitor and warn the health status of the artillery system and to meet the development requirement of changing from“failure repair”to“status repair”,a prediction model of equipment standardized current optimization BP network(SC-BP)is established with the algorithm of fitting model,the internal logic and correlation of the operation data are analyzed,the database of parameters of the normal operation state of the equipment is designed,and the data of the database are compared with those of the equipment status collected in real-time.Identification is carried out according to the goodness of fitting,the status of the equipment is judged by comparing the similar relationship between the real-time data of the equipment and the normal historical data,and the fault related information is displayed,the reference is provided for the operators to analyze the cause of the failure.Finally,an example is analyzed and verified.The research results show that the fault alarm method can identify the abnormal condition of the device performance and can make the maintenance personnel to discover the symptoms before the device failure occurrence and take measures in advance to avoid the occurrence of the device failure.