Nonlinear joint state-parameter observer for VAV damper position estimation

Variable Air Volume (VAV) based Heating Ventilation and Air Conditioning (HVAC) systems are common in large non-residential buildings. The dynamic model of a VAV system along with the Air Handling Unit (AHU) and the zones has a nonlinear characteristic. In this paper, a nonlinear model based joint state and parameter observer is proposed to estimate the VAV damper position in such systems. First, a Takagi-Sugeno (T-S) equivalent model for the AHU-VAV-Zone model is obtained using sector nonlinearity approach. The damper position estimation problem is then posed as a time varying parameter estimation problem. A procedure based on existing literature results on T-S joint state and parameter estimation is implemented. Simulation results show the effectiveness of this approach. A bank of observers based approach is then described that can help in detecting and isolating VAV damper faults in the system using the state and parameter estimates.