Applications of Metaheuristics in Power Electronics

Abstract

The impact of power electronics in modern power system is profound. The grid integration of distributed generations (DGs) including renewable energy systems (RESs) employs the power electronic converters. For the economic operation of power system, the optimization is required to reduce its number of components, complexity, installation cost, running cost, electrical losses, and harmonic contents etc. Several conventional iterative methods were applied in these optimization problems. However, they suffer to a large extent from various drawbacks such as convergence to local minima, complexity in programming, large computational time, requirement of proper initial guess and intuition etc. Although there are various methods proposed to solve the optimization problem, the metaheuristics on the other hand have proven their capabilities in solving the problems related to optimization in many engineering fields. In power electronics, the optimization is required in circuit design, filter design, intelligent controllers design, parameters computation, modeling of new topologies, harmonic mitigation, losses evaluation, finding of safe operating areas of power electronic components etc. The evolutionary algorithms (EAs) and metaheuristics are very beneficial in solving these problems as these do not require intuition or past experience instead they work on the law of evolution and social behavior of groups. The advantages of metaheuristics over conventional optimization methods are saving in computation time, cost, ease of programming, lesser mathematical complexity. In grid-connected applications, metaheuristics have also shown their excellence, as they enhance the quality of power along with optimization of cost, size, and efficiency of power system network. In this chapter, EAs and metaheuristics proposed for the various power conversion applications such as FACTs controllers and devices, power filters, multilevel inverters, dc-dc converters, and PWM converters have been discussed. The merits and demerits of metaheuristics over conventional optimization methods are discussed. The detailed comparative analysis of various metaheuristics in power electronics systems is presented. The future perspectives of metaheuristics and EAs in power electronics is also discussed in this study.