In CRESED PERFORMAn CEOFA HYBRIDOPTImIZERFOR SImuLATIOn BASED COnTROLLERPARAETERIZATIOn

Abstract

 In an integrated automated design, the controller parameterization is often accomplished by using simple empirical
 formulae. Therefore, the resultant system often fails to adequately verify the specified values because The optimizer
 determines a potential superior solution based on the simulator's results. A representation of the whole system under
 examination forms the simulator's central component. Consequently, it is necessary to solve an optimization problem (eq. 1)
 [1]. actions. This study presents a way for parameterizing the controller using simulation-based optimization techniques.
 This lets the user specify F (ⁱ) min(F (ⁱ) (1)certain limitations, such as the complementary sensitivity function
 (CSF), can affect the control loop's dynamic behavior. Additionally, other optimization criteria may be used. The
 execution time is a major determinant of effective offline and controller internal optimization techniques, and it may be
 decreased by using a hybrid optimization approach. Therefore, the study compares the performance of the global Particle
Swarm-Optimization (PSO) algorithm in its straight form and the global PSO algorithm combined with the The evaluation
 of the actual solution is represented by the real-valued function F(), often known as the fitness function [2]. Generally
 speaking, punishment values are used to impose limitations. A punishment value is applied to the assessment of the actual
 solution in the event that a constraint is broken. The optimizer avoids and depreciates it as a result. Equation 2 is used to
 compute the assessment of a solution. Main_Criterion (xn) F (xn) Nelder-Mead (NM) local optimization technique to a
 hybrid optimizer (HO) using examples