一次回风空调系统的故障诊断研究

The Fault Diagnosis of Modular Air Handling Units

空调系统的故障诊断研究对于节约空调系统能耗、减少设备之间的磨损、延长设备的使用寿命、提高空调系统的运行效率均具有重要的意义。由于空调系统的复杂性、空调设备的多样性,空调系统在实际运行过程中会存在多种多样的问题,包括硬件设备的老化、阀门的堵塞、传感器失灵、换热器换热效率降低等,都会造成能源消耗增加、能源浪费等问题。因此针对空调系统的特点研究一套适用于该系统的故障诊断方法,实现对系统运行过程中常见故障的诊断,主要针对组合式空气处理机组。通过实现模拟再现常见的7种故障,利用一阶自回归方程建立特征参数残差图,由残差分析得到不同故障条件下特征参数的残差变化,进一步针对残差变化趋势以送分温度残差为基准建立故障诊断分类器,并细化为7条故障诊断规则集;同时,通过Matlab Script节点实现Matlab与LabVIEW的混合编程,并调用训练好的模糊神经网络模型实现故障诊断结果的可视化显示。

The fault diagnosis of HVAC systems is important for saving energy,equipment maintenance,prolonging service life of equipment and improving the operating efficiency of air-conditioning system.Because of the complexity of the air-conditioning system and the diversity of air-conditioning equipment,there will be a variety of problems in the operation process including aging of hardware equipment,valve blockage,sensor failure,reduced heat exchange efficiency of the heat exchanger,etc.,which cause problems such as increased energy consumption and energy waste.Therefore,according to the characteristics of the HVAC system,this paper studies a set of fault diagnosis methods suitable for the system to realize the diagnosis of common faults in the process of system operation.This paper mainly focuses on the modular air handling units.Seven common faults are simulated,and the residual plots of characteristic parameters are established through regressive equation.The residual changes of characteristic parameters under different fault conditions are obtained by residual analysis.Fault diagnosis classifier is established on the basis of the residual temperature of the residual change trend,and it is refined into seven fault diagnosis rule sets;The mixed programming of Matlab and LabVIEW is realized by Matlab Script node realizes,and the visualization of fault diagnosis results can be achieved by the fuzzy neural network model.