技术中心

【中文摘要】 针对制冷系统渐变故障的特性和监测参数维数过大、变量间相关严重、易受干扰而信噪比低造成系统故障检测困难的难题,建立了基于主元分析的制冷系统渐变故障检测模型。应用实验数据对模型进行了验证和评估。结果表明,基于主元分析的制冷系统渐变故障检测模型具有良好的检测性能,对早期故障具有非常好的敏感性和及时捕捉的能力,并能够清晰地呈现出渐变故障发展的过程和趋势。 【英文摘要】 According to the characteristics of gradual faults in refrigeration systems,a fault detection model based on Principal Component Analysis (PCA) was presented to solve the problems such as large-scale monitored parameters,multicollinearity,liability to interference and low signal-to-noise ratio. The present fault detection (FD) model was validated and evaluated by the experiment. The results show that the PCA-FD model can meet the detection performance requirement and has higher sensitivity to incipient faults. Moreover,it can clearly display the development process and trend of the gradual faults. 【中文关键词】 热工学; 检测模型; 主元分析; 渐变故障 【英文关键词】 Pyrology; Detection model; Principal component analysis (PCA); Gradual fault