基于主成分分析法的空调系统传感器自动故障诊断

Abstract

由于各种各样的故障,空调系统在整个建筑生命周期内其性能很少能够达到设计目标。为了改善空调系统的性能,在整个建筑生命周期内对其进行定期或连续检测是很必要的。随着空调系统的日益复杂化和对快速可靠检测的需求,人工检测已远远不能满足要求,这使自动检测成为了必须。目前国际上研究建筑空调系统自动检测的技术很热门,但较少考虑空调系统传感器的准确性。传感器读数准确性是空调系统可靠控制和检测的先决条件。本文提出了一种基于主成分分析法的鲁棒故障诊断策略,该策略可以有效地对空气处理单元中的传感器进行自动检测。结合建筑自动化系统,该诊断策略可以实现空调系统传感器故障的在线检测。 HVAC systems seldom perform as intended under the original design concept.As a quality-oriented process,commissioning should be conducted regularly or continuously throughout the building life cycle.Considering labor,time and cost consumed by manual commissioning,automatic commissioning is more competent for life cycle commissioning.This paper investigates existing techniques for implementing automatic commissioning.It is found that,although many efforts have been put into developing automatic commissioning techniques,sensor validation was always overlooked.Accurate sensor measurements are prerequisite to reliable commissioning.This paper presents a diagnostic tool for automatic commissioning of sensors in typical air-handling units(AHU).The diagnostic tool implements a robust sensor fault detection and diagnosis(FDD) strategy based on the principal component analysis(PCA) method.The diagnostic tool is integrated in a real building automation system(BAS) for automatic commissioning of AHU sensors in a simulated building.The application of automatic online commissioning of AHU sensors is demonstrated.