Computer-Aided Controller Design for a Nonlinear Process Using a Lagrangian-Based State Transition Algorithm

Abstract

Model parameter estimation and controller tuning of the nonlinear clarification process are primary issues in the sugar industry. This paper proposes a Lagrangian-based state transition algorithm (LSTA) for controlling the clarification process and identifying the model parameters. First, the standard system identification procedure is used to estimate the model parameters for the clarifier. Then, the LSTA is used to enhance the accuracy of the identified system parameters. Once a suitable model is obtained, the proposed LSTA is used to fine-tune the optimal controller parameters off-line. We choose two types of evolutionary algorithms for comparison to show the efficacy of the proposed design. The results of the simulations indicate that the proposed controller achieves optimum transient and tracking performance in all cases. For fast convergence and the ability to provide the global optimum when estimating the modeling and controller parameters, the LSTA performance was found to be superior to that of the other algorithms.