Fault Diagnosis for Sensors in HVAC Systems Using Wavelet Neural Network

Various sensor faults in heating, ventilation and air conditioning (HVAC) systems usually result in more energy consumption or poor indoor air quality. Generally, two fault detection and diagnosis methods have been developed and widely applied in HVAC systems. One is the model-based, and the other is the data-driven. Each of these two methods has its own characteristic and application condition. The model-based method deeply relies on the accuracy of the mathematic model built. While the data-driven approach requires excellent training data. In this paper, wavelet neural network, which combines the wavelet analysis and neural network, is presented to diagnose the fixed and drifting biases of sensors in the HVAC systems. Wavelet analysis is employed to process measurement. data to obtain the eigenvector matrix representing the operation characteristic information of the system through decomposing the original data. Neural network, which is well trained to learn various conditions using the eigenvectors, is used to diagnose the sensor faults. Simulation tests in this paper illustrate that wavelet neural network can successfully diagnose fixed and drifting biases of sensors in HVAC systems.