Fitness dependent optimizer based computational technique for solving optimal control problems of nonlinear dynamical systems
Document Type
Article
Publication Date
1-1-2023
Abstract
This paper presents a pragmatic approach established on the hybridization of nature-inspired optimization algorithms and Bernstein Polynomials (BPs), achieving the optimum numeric solution for Nonlinear Optimal Control Problems (NOCPs) of dynamical systems. The approximated solution for NOCPs is obtained by the linear combination of BPs with unknown parameters. The unknown parameters are evaluated by the conversion of NOCP to an error minimization problem and the formulation of an objective function. The Fitness Dependent Optimizer (FDO) and Genetic Algorithm (GA) are used to solve the objective function, and subsequently the optimal values of unknown parameters and the optimum solution of NOCP are attained. The efficacy of the proposed technique is assessed on three real-world NOCPs, including Van der Pol (VDP) oscillator problem, Chemical Reactor Problem (CRP), and Continuous Stirred-Tank Chemical Reactor Problem (CSTCRP). The final results and statistical outcomes suggest that the proposed technique generates a better solution and surpasses the recently represented methods in the literature, which eventually verifies the efficiency and credibility of the recommended approach.
Keywords
Optimization, Convergence, Approximation algorithms, Optimal control, Linear programming, Artificial neural networks, Search problems, Bernstein polynomials, Dynamical systems, Fitness dependent optimizer, Genetic algorithm, Nonlinear optimal control problems, Optimization problem, Optimization techniques
Divisions
sch_ecs
Funders
King Saud University [Grant No. RSPD2023R585]
Publication Title
IEEE Access
Volume
11
Publisher
Institute of Electrical and Electronics Engineers
Publisher Location
445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA