Transient model and parameters matching of a segmented solar thermoelectric system
Solar thermoelectric generators (STEGs) with segmented thermoelectric (TE) materials have higher energy conversion efficiency than traditional STEGs with uniform TE materials. This paper constructs a transient model to investigate the dynamic heat conduction characters of the segmented STEGs. The so...
Saved in:
| Published in | International journal of green energy Vol. 21; no. 3; pp. 521 - 534 |
|---|---|
| Main Authors | , , , |
| Format | Journal Article |
| Language | English |
| Published |
Taylor & Francis
19.02.2024
|
| Subjects | |
| Online Access | Get full text |
| ISSN | 1543-5075 1543-5083 1543-5083 |
| DOI | 10.1080/15435075.2023.2200438 |
Cover
| Summary: | Solar thermoelectric generators (STEGs) with segmented thermoelectric (TE) materials have higher energy conversion efficiency than traditional STEGs with uniform TE materials. This paper constructs a transient model to investigate the dynamic heat conduction characters of the segmented STEGs. The solutions of temperature field and energy conversion efficiency are obtained. The parameters matching problem of segmented TE materials is also studied based on the developed transient model. Numerical results show that there exists an optimum parameter matching at which the efficiency could attain its local maximum value. In addition, tuning and matching segmented TE material parameters could enhance the efficiency and might increase thermal stress simultaneously. Therefore, in this paper, an optimization purpose function is defined to analyze the effect of parameters matching on both performance and structural safety. Neural network (NN) is also used to find segmented STEGs with high efficiency and low thermal stress. This work may offer useful information for the optimal design of real segmented STEGs and promote their practical applications. |
|---|---|
| Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| ISSN: | 1543-5075 1543-5083 1543-5083 |
| DOI: | 10.1080/15435075.2023.2200438 |