No abstract

No abstract

No abstract

In the present study, we suggest a modified version of heterogeneous multi-swarm particle swarm optimization (MSPSO) algorithm, that allows the amelioration of its performance by introducing an adaptive inertia weight approach. In order to bring about a balance between the exploration and exploitation characteristics of MSPSO allowing to promote information exchange amongst the subswarms. However, the classical MSPSO algorithm search behavior has not always been optimal in finding the optimal solution to certain problems, which results in falling into local optimum leading to premature convergence. The most advantages of the MSPSO there are easy to implement and there are few parameters to adjust. The inertia weight (w) is one of the most Particle Swarm Optimization’s (PSO) parameters. Controlling this parameter could facilitate the convergence and prevent an explosion of the swarm. To overcome the above limitations, this paper proposes a heterogeneous multi swarm PSO algorithm based on PSO number selection approach centred on the idea of particle swarm referred to as Multi-Swarm Particle Swarm Optimization algorithm with Factor selection strategy (FMSPSO). In the various process implementations of the particle swarm search, different parameter selection strategies are adopted to ameliorate the global search ability. The proposed FMSPSO is able to improve the population’s diversity and better explore the entire feature space. The statistical test and indicators that are reported in the specialized literature demonstrate that the suggested approach is superior in terms of efficiency to nine other popular PSO algorithms in solving the optimization problem of complex problems. The approach suggests that FMSPSO reaches a very promising performance for solving different types of optimization problems, leading eventually to higher solution accuracy.

This paper presents the vector control of Induction Motor (IM) supplied by a photovoltaic generatorwhich is controlled by an adaptive Proportional-Integral (PI) speed controller. The proposed solution is used toovercome the induction motor rotor resistance variation problem, which can affect negatively the performanceof the speed control. To overcome the rotor resistance variation, an adaptive Proportional-Integral controller isdeveloped with gains adaptation based on Adaptive Neuro-Fuzzy Inference System (ANFIS) in order to guaranteea high performances of electric drive systems against the parametric variations. The proposed control algorithmis tested by Matlab-Simulink. Analysis of the obtained results shows the characteristic robustness to disturbancesof the load torque and to rotor resistance variation compared to the classical PI control and Model ReferenceAdaptive System (MRAS) rotor resistance observers.