摘要
目的研制一种新型的故障诊断系统,解决高速旋转机械复合型故障诊断的问题.方法将神经网络理论应用于故障诊断中,把信号采集、状态监测、信号分析和智能诊断组合在一起,并以H1C型涡轮增压器为诊断研究对象,进行了该系统的试验验证.结果得出的该系统的诊断结果与理论上分析得出的结果基本一致.结论证实了所研究的该系统是可信的,它能够准确、快速地诊断出设备存在的故障.基于神经网络理论的机械故障智能诊断系统在解决复合型故障诊断方面更具有优越性与高效性,并为其现场应用于实时监测和实时诊断奠定了试验基础和技术支持.
A new-style fault intelligent diagnosis of high speed rotating machinery is studied, which solves the problem of its compound fault diagnosis. Neural network is used to diagnose the fault. The system combines signal sampling, condition monitoring, signal analysis and intelligent diagnosis into one. Then H1C supercharger is studied and the verifying test of the system is done, the result of which is consistent with the theoretical one. The reliability of the system is verified, which can diagnose the fault of machinery accurately and quickly. The intelligent diagnosis system is employed to resolve compound fault diagnosis of high speed rotating machinery. The superiority in diagnosing compound fault is proved and the foundation for practical utilization and real-time fault diagnosis is established based on neural network.
出处
《沈阳建筑大学学报（自然科学版）》
EI
CAS
2005年第6期770-773,共4页
Journal of Shenyang Jianzhu University：Natural Science
基金
国家自然科学基金项目(50275025)
关键词
故障诊断
旋转机械
状态监测
神经网络
fault diagnosis, rotating machinery, condition monitoring, neural network
