Metaheuristic based single and multiobjective optimization of thermoelectric generator

Irene Jacob; Ravita Lamba; Rajesh Kumar; Francisco J. Montero

doi:10.1016/j.applthermaleng.2023.121790

Metaheuristic based single and multiobjective optimization of thermoelectric generator

Irene Jacob, Ravita Lamba, Rajesh Kumar, Francisco J. Montero

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

Abstract

Analytical optimization methods are not applicable for some complex practical problems that lead to explore the numerical/metaheuristic optimization methods. Thermoelectric devices are also complex systems and need metaheuristic methods for their performance optimization. This paper analyses various metaheuristic techniques for performance optimization of thermoelectric generator (TEG). Single and multiobjective optimizations of TEG have been performed to optimize its power output and efficiency. The hot and cold side temperatures, load resistance, length of semiconductor legs, and fill factor of the TEG are used as performance parameters. Single-objective optimization is performed using 13-metaheuristic algorithms and their results are compared in terms of solution optimality and computational time acceptability. It has been observed that the cuckoo search (CS) algorithm is the most suitable in terms of accuracy and speed. CS produced the optimum power output and efficiency of 19.20837 W and 2.4427 % respectively in minimum time. The multiobjective optimization is also performed for simultaneous optimization of the power output and efficiency. Shannon entropy approach is found to be the best technique to determine optimal point on the Pareto front for multiobjective optimization. The optimal values of decision variables, T_h T_c, R_L, L and FF for multiobjective optimization are found to be 321.195 K, 281.195 K, 0.14250 Ω, 0.5 mm and 0.9 respectively and it produced optimum power output and efficiency of 19.20837 W and 2.25932 % respectively. The statistical and box plot analysis have also been carried out for all the algorithms. The determination of the most suitable algorithm for the faster performance optimization of the TEG, hereby provides the pathways to ascertain the most accurate solution for the power output and efficiency obtainable from the TEG system and hence, delivering the ability to decide the right parameters for the construction of the TEG system with the best performance.

Original language	English
Article number	121790
Journal	Applied Thermal Engineering
Volume	236
DOIs	https://doi.org/10.1016/j.applthermaleng.2023.121790
Publication status	Published - 10 Jan 2024
Externally published	Yes

Keywords

Efficiency
Metaheuristic algorithms
Power output
Single and multi-objective optimization
Thermoelectric generator

ASJC Scopus subject areas

Energy Engineering and Power Technology
Mechanical Engineering
Fluid Flow and Transfer Processes
Industrial and Manufacturing Engineering

Access to Document

10.1016/j.applthermaleng.2023.121790

Cite this

@article{a66665d550994b22998bcfe8f664b2f5,

title = "Metaheuristic based single and multiobjective optimization of thermoelectric generator",

abstract = "Analytical optimization methods are not applicable for some complex practical problems that lead to explore the numerical/metaheuristic optimization methods. Thermoelectric devices are also complex systems and need metaheuristic methods for their performance optimization. This paper analyses various metaheuristic techniques for performance optimization of thermoelectric generator (TEG). Single and multiobjective optimizations of TEG have been performed to optimize its power output and efficiency. The hot and cold side temperatures, load resistance, length of semiconductor legs, and fill factor of the TEG are used as performance parameters. Single-objective optimization is performed using 13-metaheuristic algorithms and their results are compared in terms of solution optimality and computational time acceptability. It has been observed that the cuckoo search (CS) algorithm is the most suitable in terms of accuracy and speed. CS produced the optimum power output and efficiency of 19.20837 W and 2.4427 % respectively in minimum time. The multiobjective optimization is also performed for simultaneous optimization of the power output and efficiency. Shannon entropy approach is found to be the best technique to determine optimal point on the Pareto front for multiobjective optimization. The optimal values of decision variables, Th Tc, RL, L and FF for multiobjective optimization are found to be 321.195 K, 281.195 K, 0.14250 Ω, 0.5 mm and 0.9 respectively and it produced optimum power output and efficiency of 19.20837 W and 2.25932 % respectively. The statistical and box plot analysis have also been carried out for all the algorithms. The determination of the most suitable algorithm for the faster performance optimization of the TEG, hereby provides the pathways to ascertain the most accurate solution for the power output and efficiency obtainable from the TEG system and hence, delivering the ability to decide the right parameters for the construction of the TEG system with the best performance.",

keywords = "Efficiency, Metaheuristic algorithms, Power output, Single and multi-objective optimization, Thermoelectric generator",

author = "Irene Jacob and Ravita Lamba and Rajesh Kumar and Montero, {Francisco J.}",

note = "Publisher Copyright: {\textcopyright} 2023 Elsevier Ltd",

year = "2024",

month = jan,

day = "10",

doi = "10.1016/j.applthermaleng.2023.121790",

language = "English",

volume = "236",

journal = "Applied Thermal Engineering",

issn = "1359-4311",

publisher = "Elsevier Ltd.",

}

TY - JOUR

T1 - Metaheuristic based single and multiobjective optimization of thermoelectric generator

AU - Jacob, Irene

AU - Lamba, Ravita

AU - Kumar, Rajesh

AU - Montero, Francisco J.

PY - 2024/1/10

Y1 - 2024/1/10

N2 - Analytical optimization methods are not applicable for some complex practical problems that lead to explore the numerical/metaheuristic optimization methods. Thermoelectric devices are also complex systems and need metaheuristic methods for their performance optimization. This paper analyses various metaheuristic techniques for performance optimization of thermoelectric generator (TEG). Single and multiobjective optimizations of TEG have been performed to optimize its power output and efficiency. The hot and cold side temperatures, load resistance, length of semiconductor legs, and fill factor of the TEG are used as performance parameters. Single-objective optimization is performed using 13-metaheuristic algorithms and their results are compared in terms of solution optimality and computational time acceptability. It has been observed that the cuckoo search (CS) algorithm is the most suitable in terms of accuracy and speed. CS produced the optimum power output and efficiency of 19.20837 W and 2.4427 % respectively in minimum time. The multiobjective optimization is also performed for simultaneous optimization of the power output and efficiency. Shannon entropy approach is found to be the best technique to determine optimal point on the Pareto front for multiobjective optimization. The optimal values of decision variables, Th Tc, RL, L and FF for multiobjective optimization are found to be 321.195 K, 281.195 K, 0.14250 Ω, 0.5 mm and 0.9 respectively and it produced optimum power output and efficiency of 19.20837 W and 2.25932 % respectively. The statistical and box plot analysis have also been carried out for all the algorithms. The determination of the most suitable algorithm for the faster performance optimization of the TEG, hereby provides the pathways to ascertain the most accurate solution for the power output and efficiency obtainable from the TEG system and hence, delivering the ability to decide the right parameters for the construction of the TEG system with the best performance.

AB - Analytical optimization methods are not applicable for some complex practical problems that lead to explore the numerical/metaheuristic optimization methods. Thermoelectric devices are also complex systems and need metaheuristic methods for their performance optimization. This paper analyses various metaheuristic techniques for performance optimization of thermoelectric generator (TEG). Single and multiobjective optimizations of TEG have been performed to optimize its power output and efficiency. The hot and cold side temperatures, load resistance, length of semiconductor legs, and fill factor of the TEG are used as performance parameters. Single-objective optimization is performed using 13-metaheuristic algorithms and their results are compared in terms of solution optimality and computational time acceptability. It has been observed that the cuckoo search (CS) algorithm is the most suitable in terms of accuracy and speed. CS produced the optimum power output and efficiency of 19.20837 W and 2.4427 % respectively in minimum time. The multiobjective optimization is also performed for simultaneous optimization of the power output and efficiency. Shannon entropy approach is found to be the best technique to determine optimal point on the Pareto front for multiobjective optimization. The optimal values of decision variables, Th Tc, RL, L and FF for multiobjective optimization are found to be 321.195 K, 281.195 K, 0.14250 Ω, 0.5 mm and 0.9 respectively and it produced optimum power output and efficiency of 19.20837 W and 2.25932 % respectively. The statistical and box plot analysis have also been carried out for all the algorithms. The determination of the most suitable algorithm for the faster performance optimization of the TEG, hereby provides the pathways to ascertain the most accurate solution for the power output and efficiency obtainable from the TEG system and hence, delivering the ability to decide the right parameters for the construction of the TEG system with the best performance.

KW - Efficiency

KW - Metaheuristic algorithms

KW - Power output

KW - Single and multi-objective optimization

KW - Thermoelectric generator

UR - http://www.scopus.com/inward/record.url?scp=85175199068&partnerID=8YFLogxK

U2 - 10.1016/j.applthermaleng.2023.121790

DO - 10.1016/j.applthermaleng.2023.121790

M3 - Article

AN - SCOPUS:85175199068

SN - 1359-4311

VL - 236

JO - Applied Thermal Engineering

JF - Applied Thermal Engineering

M1 - 121790

ER -

Metaheuristic based single and multiobjective optimization of thermoelectric generator

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this