Abstract
The performance of thermoelectric generators (TEGs) besides depending on material’s figure of merit, is also a function of the module design like the heat exchanger design, thermoelectric leg size, configuration, and contact resistances. Optimizing these parameters using conventional experimental methods is quite time-consuming and inconvenient. Taguchi optimization method could be used an efficient technique to predict the performance of a TEG module with less number of experimental runs as compared to the customary techniques. The most commonly used materials in thermoelectric applications in the medium temperature range (400–900 K) are the PbTe alloys and Mg 2 (Si-Sn) based alloys. N-type Mg 2 (Si-Sn) alloys often being integrated with higher manganese silicide (HMS) for fabricating TEGs, pose problems in maintaining the long-term stability of such generators owing to the imperfectly matched thermomechanical properties between these materials. Therefore, it is required to evaluate a TEG system that is completely made out of magnesium based thermoelectric alloys (both p- and n-type). The current study examines the scope of Mg 2 (Si-Sn) thermoelectric generators based on the numerical modeling technique along with employing Taguchi and ANOVA techniques for improving the overall system output. A TEG system comprising of 36 modules is built using numerical modeling which is followed by optimization of pivotal design parameters through the Taguchi and ANOVA methods. Optimal control factor settings obtained from the Taguchi method result in output power and conversion efficiency values of 90.04 W and 9.08% respectively. The predicted optimal output Y predicted is also calculated, the respective values obtained for output power and conversion efficiency are 59.9 W and 9.12%. From ANOVA analysis, it is observed that the most significant variable for power output is the cross-sectional area with a percentage contribution of 35.22% and that for the conversion efficiency is the temperature difference between the two ends of the TEG having a percentage contribution of 97.73%.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have