基于人工神经网络的制冷机组泄漏故障诊断

Leakage Fault Diagnosis in Refrigerating Units Based on Neural Networks

研究了冷库制冷机组制冷剂泄漏的特征 ,分析了泄漏引发的流量和压力参数变化情况 ,并提取了反映泄漏故障的特征指标 ,建立了能对制冷机组运行状态进行分类的神经网络模型 。

The characteristics of refrigerant leakage fault in refrigerating units are analyzed and a series of characteristic indexes of pressure signals are described to extract the information of leakage fault . A backpropagation neural network model with the input matrix composed of the characteristic indexes of pressure is proposed to classify the running state of refrigerating units in order to diagnose and forecast the leakage fault.\;