Experimental investigation of thermoelectric generator system

Abstract

This was accomplished by mathematical models for the dependent Pi terms in the quantitative data-based modelling. This aims to use dimensionless analysis to figure out what factors influence the efficiency of a thermoelectric generator device (TEG). Simple mathematical models were developed based on actual experiments in this research work to predict the pattern of voltage, current, and power generated in TEG modules. These mathematical models are one of the most effective ways to explain experimental findings and gain a better understanding of the experimental method under study. We often find ourselves in the situation of having to verify if data matches an equation while testing mathematical models against data. Experimental studies rely on experimental principles to explain which components are the most relevant, and they are based on observation to draw conclusions on how an experimental system performs well or not. Making the necessary observations without disrupting the experimental setup, however, can be difficult. The aim of this study is to see how the independent variables affect the dependent or response variable. An artificial neural network is used to model the mathematical formulation of the device that includes the TEG module (ANN). By intentionally making local changes in their thermoelectric generator experimental set-up, this experimental modelling and ANN simulation approach allows them to obtain a system-wide view. The development of logarithmic best fit mathematical models is used to evaluate the effects of the experiments. Different scatter graphs were plotted in this study to show the output of the dependent variable current ΠD2 (experimental, model, and ANN) vs. the independent variable relative to the heat source Π1.