This paper demonstrates the development of a novel metaheuristic algorithm which has an enhanced diversification and intensification features and aims to construct an efficient mechanism via the developed algorithm to control a magnetic object suspension. Manta ray foraging optimization (MRFO) algorithm together with generalized opposition-based learning (GOBL) technique and Nelder-Mead (NM) simplex search method is used to define the general frame of the developed algorithm. The developed novel algorithm (Ob-MRFONM) employs the NM method for better intensification whereas integrates the GOBL technique for better diversification. The constructed Ob-MRFONM algorithm is confirmed to have enhanced capabilities via evaluations against well-known unimodal and multimodal benchmark functions. The developed algorithm is also utilized to reach optimum values of a real PID plus second-order derivative (PIDD 2 ) controller employed in a magnetic object suspension system to demonstrate the capability of the Ob-MRFONM algorithm for such a complex real-world engineering problem. It is worth noting that this paper is the first report in the literature that demonstrates the application of a real PIDD 2 controller in a magnetic object suspension system. To reach better capability, a new objective function is used for minimization. Then, the proposed approach is comparatively evaluated in terms of statistical and nonparametric statistical analysis, convergence, transient response, disturbance rejection, controller effort and effects of the noise signal. The demonstrated results also confirm the highly competitive capability of the proposed algorithm for the complex magnetic object suspension system. The nonlinear model of the system is also used in this study in order to validate the linearized model.
KeywordsManta ray foraging optimization • Nelder-Mead method • Opposition-based learning • Real PIDD 2 controller • Magnetic ball suspension system • Nonlinear plant B Davut Izci