Fault Tolerant Control of Nonlinear Processes with Adaptive Diagonal Recurrent Neural Network Model
Fault tolerant control (FTC) using an adaptive recurrent neural network model is developed in this paper. The model adaptation is achieved with the extended Kalman filter (EKF). A novel recursive algorithm is proposed to calculate the Jacobian matrix in the model adaptation so that the algorithm is simple and converges fast. A model inversion control with the developed adaptive model is applied to nonlinear processes and fault tolerant control is achieved. The developed control scheme is evaluated by a simulated continuous stirred tank reactor (CSTR) and effectiveness is demonstrated.
KeywordsHide Layer Extended Kalman Filter Model Predictive Control Recurrent Neural Network Nonlinear Process
Unable to display preview. Download preview PDF.
- 1.Almeida, L.B.: Backpropagation in Non-Feedforward Networks in Neural Computing Architectures, pp. 75–91. MIT Press, Cambridge (1989)Google Scholar
- 3.Chang, T.K.: Fault Tolerant Control for Nonlinear Processes Using Adaptive Neural Networks. PhD Thesis, Scholl of Engineering, Liverpool John Moores University, U.K. (2002)Google Scholar