基于主成分法的VAV系统传感器故障检测与诊断

众所周知，VAV系统对控制的要求很高，作为控制系统中非常关键的元件——传感器，一旦出现故障将直接影响控制系统的决策，从而使VAV系统的运行偏离设计要求。因此，VAV系统传感器的故障检测与诊断研究是很有必要的。本文采用主成分分析法(PCA，Principal Component Analysis)对传感器故障的检测、确认与重构进行分析，以期获得一种可行的方案。

Sensor Fault Detection and Diagnosis Research in VAV System Based on Principal Component Analysis

It is well-known that control system is crucial to VAV system. Sensors as a key component of control system make a direct impact on decision-making of control system. Their fault makes the operation of VAV system differ from the designer's objective. As a result, research of sensor fault detection and diagnosis is very necessary in VAV system. An approach for sensor fault detection, identification and reconstruction via Principal Component Analysis (PCA) was presented in this paper. The multivariate statistical method of PCA is a very useful tool for reducing the number of variables in a data set and for obtaining useful two-dimensional views of a multi-dimensional data set. The PCA model partitions the measurement space into a Principal Component Subspace (PCS) where normal variation occurs, and a Residual Subspace (RS) that faults may occupy. When the actual fault is assumed, the maximum reduction in the Squared Prediction Error (SPE) is achieved. A fault-identification index was defined in terms of SPE. Some examples were provided to verify the method in the paper.