摘要
本文针对压缩机性能试验系统,提出了基于神经网络的故障检测及可靠性分析方法。首先,通过分析实际运行数据,筛选稳态工况下数据。其次,利用稳态无故障数据训练神经网络模型,获得压缩机性能测试系统的无故障神经网络模型。然后,通过实验分别引入压缩机缺油故障、冷凝器缺水故障、传感器固定偏差和漂移故障。最后,采用训练好的神经网络模型对故障系统进行可靠性分析,检测到了不同故障的发生。实验验证的结果表明,本文提出的基于神经网络的可靠性分析模型,对于压缩机性能测试系统具有较好的检测能力。
A fault detection and reliability analysis method based on neural network are proposed for compressor performance test system in this paper.Firstly,by analyzing the actual operating data,the data under steady-state conditions are screened.Secondly,the fault-free neural network model of compressor performance test system is obtained by using the steady-state fault-free data training neural network model.Thereafter,the compressor oil shortage fault,condenser water shortage fault and sensor fixed bias and drift fault are introduced through the experiment.Finally,the trained neural network model is used to analyze the reliability of the fault system,and the occurrence of different faults is detected.The experimental results show that the reliability analysis model based on neural network proposed in this paper has better fault detection ability for compressor performance test system.
作者
王崇亮
张帅
李前舸
杜志敏
晋欣桥
WANG Chongliang;ZHANG Shuai;LI Qiange;DU Zhimin;JIN Xinqiao(Institute of Refrigeration and Cryogenics,Shanghai Jiao Tong University,Shanghai 200240,China)
出处
《制冷技术》
2019年第3期22-29,共8页
Chinese Journal of Refrigeration Technology
基金
国家自然基金(No.51376125)
关键词
压缩机性能测试
故障检测
可靠性研究
神经网络
稳态判别
Compressor performance test
Fault detection
Reliability research
Neural networks
Steady-statediscrimination
