Abrupt and incipient fault detection and compensation for a 4-tank system benchmark

Fault detection and compensation play a key role in enhancing the high demand for performance and security in technological systems. This paper proposes an active fault-tolerant control scheme that detects and compensates for actuator faults in a 4-tank system benchmark. The faults are modeled as a drastic gain loss in actuators (pumps), which could lead to a large loss in the nominal performance. The model-based approach uses a recursive least squares parameter estimation algorithm to form a fault detection and diagnosis subsystem and utilizes a parametric eigenstructure assignment method to reconfigure the state feedback controller. The designed controller is simulated for the nonlinear system, and the results demonstrate promising performance increases in faulty cases in comparison with the nonfault-tolerant controller.