A COMPARATIVE STUDY OF DIFFERENT OPTIMIZATION TECHNIQUES IN MODELLING AND PREDICTIVE CONTROLS

This study summarizes different optimization techniques in modeling and predictive controls. Optimization plays an important role in Model Predictive Control (MPC). The optimization techniques help to evaluate the best solution to a mathematical model by comparing it with visual data. The different optimization techniques include the Polak method, Cauchy method, Newton method, Modified Newton method, Steep Decent method, David-Fletcher-Powell (DFP) method, and Momentum method. These optimization techniques can fit linear and non-linear models with n-number input variables. This can help to improve the performance of the MPC controller by addressing the challenges of the data being controlled. The solution to this problem is addressed in this study, including the mathematical modeling of optimization techniques. Simulation is conducted to verify the comparison test and model a nonlinear dynamic process. The output of the MPC controller shows a linear agreement with the open-loop test, for all the optimization techniques, with some deviation in the overshoot, rise times and settling times.