Law-based sensor fault diagnosis and validation for building air-conditioning systems

This paper presents an automatic strategy that can be used in a building energy management and control system to detect, diagnose, and evaluate soft sensor faults in building air-conditioning systems. The strategy is based on fundamental conservation (mass and energy conservation) relations and accommodates changes of plant performance and working conditions. The existence and magnitude of non-abrupt biases in chilled water flow meters and temperature sensors due to miscalibration and drift are detected. Sensor biases are estimated by minimizing the sum of the squares of the mass and energy balance residuals of selected control volumes. Validation of the strategy on a central chilling system is presented.