Abstract:With the scientific exploration in deep space and deep sea, traditional fossil energy and solar energy cannot supply energy for deep space aircrafts and deep sea vehicles continuously, new energy power systems with the features of long endurance, high reliability and high energy density are imperative. Small nuclear reactor has become one of the best solutions for deep space energy demand in the future due to its strong environmental adaptability and long-life time. To improve the reliability of the static heat pipe reactor, the thermoelectric generators (TEGs) are utilized to convert fission heat into electrical energy. Focusing on the demands of improving the conversion efficiency, a new three segmented TEG is proposed based on the Bi 2 Te 3 , CoSb 3 and Half-Heusler thermoelectric materials in this paper. Combined with the application requirements of heat pipe reactor, the geometric configuration such as different heights and the cross-sectional area of the three different materials in the PN legs is designed and optimized under different heat flux boundary conditions. The influences of contact electrical resistance and contact thermal resistance on the conversion efficiency have been investigated. When the contact electrical resistivity increases from 0.01 μΩ cm 2 to 1000 μΩ cm 2 , the conversion efficiency loss increases from 2.6% to 80.1%. As the surface roughness rises from 0.1 μm to 100 μm, the conversion efficiency loss increases from 0.9% to 42.7%. Based on the heat flux boundary conditions provided by the heat pipes, the thermoelectric conversion performances of three segmented TEG under different external loads are investigated. Through the geometric optimization of the three segmented TEG with total height 20 mm, the thermoelectric conversion efficiency improves from 6.9% to 15.34%. For the simulation heat flux density conditions of 70 kW m −2 , 75 kW m −2 , 80 kW m −2 , 85 kW m −2 and 90 kW m −2 , the maximum output power of a thermoelectric generator can reach 8.56W, 9.71W, 10.85W, 12.00W, and 13.11W, respectively. The output characteristics of the three segmented TEGs meet the design demands of the static conversion system.

Artificial Neural Network for Geometric Design and Optimization of Three-Stage Segmented Thermoelectric Generators

Predicting the Optimal Performance of a Concentrated Solar Segmented Variable Leg Thermoelectric Generator Using Neural Networks

Geometrical Optimization of Segmented Thermoelectric Generators (TEGs) Based on Neural Network and Multi-Objective Genetic Algorithm

Optimization of a unileg thermoelectric generator by the combination of Taguchi method and evolutionary neural network for green power generation

Thermo-mechanical optimization of thermoelectric generators using deep learning artificial intelligence algorithms fed with verified finite element simulation data

A synergistic approach to optimizing the performance of a concentrating solar segmented variable area leg thermoelectric generator using numerical methods and neural networks

Design and optimization of three segmented thermoelectric generator for nuclear reactor application

Experiment-Validated multiphysics modeling, generalizable deep learning and interpretable global sensitivity analyses for thermoelectric generators

Optimization analysis of a segmented thermoelectric generator based on genetic algorithm

Performance prediction and optimization of annular thermoelectric generators based on a comprehensive surrogate model

Optimal Design and Multi-Parameter Sensitivity Analysis of a Segmented Thermoelectric Generator

Advantage of a Thermoelectric Generator with Hybridization of Segmented Materials and Irregularly Variable Cross-Section Design

Maximum Efficient Power Performance Analysis and Multi-Objective Optimization of Two-Stage Thermoelectric Generators

Geometric optimization of thermoelectric generator using genetic algorithm considering contact resistance and Thomson effect

Geometry design for maximizing output power of segmented skutterudite thermoelectric generator by evolutionary computation

Enhancing the performance of thermoelectric generators using novel segmental arrangement of multi-functional gradient materials

3E investigation and artificial neural network optimization of a new triple-flash geothermally-powered configuration

Machine Learning Assisted Microchannel Geometric Optimization—A Case Study of Channel Designs

Transient real-weather 4E optimization of two-stage segmented thermoelectric generators for enhanced solar energy conversion

MODELING AND OPTIMAL DESIGN FOR MICRO-THERMOELECTRIC GENERATOR

Geometry and stage number optimization of a concentrating solar multistage segmented thermoelectric generator by exploiting different optimization schemes