Wavelet based residual evaluation for fault detection and isolation

Fault detection and isolation (FDI) is an important issue for safe operation in industrial processes. To avoid false alarms, the FDI scheme must be robust enough to handle all unknown input that might confuse the fault detection system. The aim objective of this work is to use wavelets to increase the robustness of residuals to measurement noise. Our approach is tested in simulation on an alcoholic fermentation process. The faults are modelled as changes in the system parameters and residuals are generated using a set of adaptive observers.