Electrical Performance Analysis of Thermoelectric Generator HZ-20 with particle swarm optimization Algorithm

Abstract

Introduction: The Thermoelectric Generator (TEG) represents a recent development renewable energy technologies, harnessing heat energy to generate electricity from various waste energy sources. the maximum power point (MPPT) is an essential approach to maximizing energy extraction from thermoelectric generators. the position of this point is not fixed but it moves according to the difference temperature and load. Many classic methods and controllers have been widely developed and implemented to track the maximum power point. The Perturb and Observe algorithm was  chosen, due to  its  simplicity and convergent  capacities. However, the operating point oscillates around the MPP that increase the loss of energy in the power of the TEM.

Objectives: of the paper is a comparative analysis of an optimization technique Particle Swarm Optimization (PSO) along with Perturb & Observe (P&O) for the extraction of maximum power from the thermoelectric generator.

Methods: Particle Swarm Optimization PSO is based on two fundamental principles: the retention of information on past performance, and communication between the agents forming the swarm. The chosen DC-DC converter approach, a one directional boost converter, is used in combination with the MPPT approach to construct the power conditioning unit. The TEG modeling was integrated with the Perturb and Observe (P&O) MPPT technique and the Particle Swarm Optimization (PSO) MPPT method.

Results: In this paper we will evaluate the performance of PSO and P&O algorithms for maximum power point tracking with thermoelectric generator  connected to load via Boost  converter is verified on MATLAB/Simulink environment. The simulation results shows that the two techniques defeat the maximum power problems and the Particle Swarm Optimization (PSO) method has advantages compare to Perturb and Observe (P&O) method. From this comparison it is observed that faster convergence is achieved in PSO algorithm when compared to P&O algorithm.