Application of Evolutionary Algorithms for Harmonic Profile Optimization in Symmetric Multilevel Inverter used in Medical Electronic Equipments

摘要

Background: Inverters are finding applications in electrical field where conversionfrom Direct Current (DC) to Alternate Current (AC) has become an inevitable option. The need ofMulti Level Inverter (MLI) is to provide a high output power from medium voltage source. Highpower quality is the very basic important requirement for MLI used in medical electronic equipment.Harmonic elimination in MLI is a challenging one and is solved by optimum switching ofpower electronic switches present in the MLI topology.Methods: Several control strategies have been proposed for harmonic elimination in multilevelinverters. Newton-Raphson method is the conventional and iterative based method to obtain optimumswitching angle to minimize the Total Harmonic Distortion (THD). But it requires an initialguess of switching angles which is very close to the exact solution. To overcome this, harmonicelimination is converted into an optimization task and is solved by using evolutionary algorithmssuch as Genetic Algorithms (GA) and Particle Swarm Optimization (PSO). GA and PSO offer optimumswitching angles to minimize THD and used to increase the robustness of the system.Results: The performance of proposed symmetric 21 level multilevel inverter with GA and PSOtechniques are analysed and the objective of minimum THD is obtained by using MATLABSIMULINKsoftware.In experimental setup, FPGA controller has been used to generate the PulseWidth Modulation (PWM) control signals according to the proposed soft computing based switchingstrategy. The experimental result is used to verify the ability of the proposed system for thegeneration of desired output voltage with minimum THD.Conclusion: The proposed symmetric MLI with evolutionary algorithm based switching is a goodchoice for medical electronic equipment used in hospital to obtain quality power with minimumTHD.