Nonlinear model predictive control of a boiler unit: A fault tolerant control study

Abstract

This paper deals with a nonlinear model predictive control designed for a boiler unit. The predictive controller is realized by means of a recurrent neural network which acts as a one-step ahead predictor. Then, based on the neural predictor, the control law is derived solving an optimization problem. Fault tolerant properties of the proposed control system are also investigated. A set of eight faulty scenarios is prepared to verify the quality of the fault tolerant control. Based of different faulty situations, a fault compensation problem is also investigated. As the automatic control system can hide faults from being observed, the control system is equipped with a fault detection block. The fault detection module designed using the one-step ahead predictor and constant thresholds informs the user about any abnormal behaviour of the system even in the cases when faults are quickly and reliably compensated by the predictive controller.

A simple scheme for semi-recursive identification of Hammerstein system nonlinearity by Haar wavelets

“…Furthermore, the limit properties are robust to the correlation structure of the external noise and the structure of the system dynamics. Combination of these asymptotic properties with implementation-relevant A simple scheme for semi-recursive identification of Hammerstein system nonlinearity by Haar wavelets 515 numerical stability and computational simplicity makes the presented algorithm an interesting offer in the system identification area, e.g., fault detection (Chen et al, 2011;Patan and Korbicz, 2012). …”

Data-driven models for fault detection using kernel PCA: A water distribution system case study

“…The FTC mechanism can then properly modify the controller (e.g., by adapting the performance index and the constraints in the MPC approach) or replace it with another one, better fit to the actual plant operating condition. The examples of such approach can be found, e.g., in the works of Patan and Korbicz (2012) or Brdyś et al (2008).…”

Comparison of speaker dependent and speaker independent emotion recognition

“…ANNs are usually trained in supervised mode, using the back-propagation algorithm. Another type of ANN architecture is represented by recurrent networks, used, e.g., in automatic control systems described by Patan and Korbicz (2012).…”