Identification of the most influential variables on the power generation of an automotive thermoelectric generator through a global sensitivity analysis

Abstract

A methodology to determine the most influential independent input variables on the maximum power generation of an automotive thermoelectric generator is developed. A validated numerical model is used to predict the maximum power generation of a thermoelectric generator composed of several thermoelectric modules installed in the exhaust pipe of an internal combustion engine. The Morris sensitivity analysis and the least absolute shrinkage and selection operator feature selection approach are employed to identify the most influential variables. The amount of independent variables selected to carry out the analysis are 18 and they are embedded in different fields such as hydraulic, thermal, electrical, chemical, geometrical and design. Results show that the most influential variables are the inlet temperature of the hot fluid and the Seebeck coefficient and electric resistance of the thermoelectric modules. The thickness of the thermoelectric modules has the least influence on the maximum power generation. These findings could be useful to other researchers to develop simpler mathematical models without compromising the accuracy

Albert Massaguer is a post-doctoral researcher at the GREFEMA research group of the University of Girona. This work has been partially funded by the Spanish Government (Ministerio de Ciencia e Inovación) under contracts TED2021-130536A-I00 and the European Union by “NextGenerationEU”/PRTR. This research has also received funding from the European Union’s Horizon 2020 research and innovation program under the project UncorrelaTEd, grant agreement No. 863222

9