基于混合仿真平台的空调系统PCA故障检测方法研究及改进  

The Study and Improvement of PCA Fault Detection Method for HVAC System Based on Hardware-in-loop Simulation Platform

作　　者：郭轶波[1] 晋欣桥[1] 杜志敏[1] 

机构地区：[1]上海交通大学机械与动力工程学院制冷与低温工程研究所,上海200240

出　　处：《科技通报》2014年第3期139-143,共5页Bulletin of Science and Technology

基　　金：国家自然科学基金(50976066)

摘　　要：为了实现建筑节能和保证空调系统正常运行,当前的建筑空调控制系统已逐渐融入了各种故障检测策略。与此同时,研究人员也提出了越来越多的空调系统故障检测策略,这些故障检测策略的验证,多通过纯仿真平台进行,但是由于纯仿真平台与实际空调系统的差距,经纯仿真平台验证的故障检测策略在实际应用时,会面临着效率下降的问题。因此,本文提出了利用混合仿真平台来对故障检测策略进行验证。以空调系统PCA故障检测方法为例,通过对比实验发现,纯仿真平台和混合仿真平台验证结果的确有较大差别。在此基础上,利用小波变换对PCA故障检测方法做出改进,实验结果表明,基于小波PCA的故障检测方法,能够提高故障检测效果,具有更好的实用性。For the purpose of saving energy and ensuring healthy operation of heating, ventilation and air conditioning （HVAC） systems, various of fault detection strategies have been embedded into the Building Automation System. At the same time, more and more fault detection strategies are proposed by researchers. The validations of these strategies are mainly conducted with digital simulation platform. Due to the difference of digital simulation platform and real HVAC system, the fault detection strategies validated with digital simulation platform may be inefficient in real application. In this study, hardware-in-loop simulation platform is used for the validation of fault detection strategies. Take fault detection strategy based on principal component analysis （PCA） for example, it can be found that the validation result with digital simulation platform differs from the validation result with hardware-in-loop simulation platform. Thus, wavelet transform is used to improve the efficiency of fault detection strategy. Experiment results show that fault detection strategy based on wavelet transform and PCA is more effective and practical.

关 键 词：混合仿真 故障检测 主成分分析 小波变换 

分 类 号：TU83[建筑科学—供热、供燃气、通风及空调工程]