Multi-objective Optimization And Performance Analysis Of Natural Convection Cooling System For Thermoelectric Cooler

Thermoelectric refrigeration technology has attracted much attention due to its nonpolluting and noise-free characteristics.At present,the thermoelectric cooling capacity and cooling efficiency are still at a low level.As an efficient heat transfer unit,the heat pipe radiator can solve the problem of small area of thermoelectric cooler and large heat flow density.However,under the condition of natural convection,the total thermal resistance of radiator is large which limits the further improvement of thermoelectric refrigeration performance.Therefore,a numerical research on radiator is carried out in order to reduce the natural convection thermal resistance and improve the refrigeration performance of the thermoelectric chip.In this paper,the numerical model of the heat pipe radiator is established firstly,and the heat transfer characteristics of the heat sink under natural convection conditions are obtained and analyzed by computational fluid dynamics.The results show that the temperature distribution in fin array of the radiator is not uniform due to the interspersed position of the heat pipe.In the process of heat transfer between fins and air,there is a small air flow velocity and a high resistance along the path,so that the heat transfer coefficient of the fin array is low,and the heat dissipation effect is poor.In order to improve the heat transfer performance of the heat pipe radiator,a non-equal fin array design with continuously varying heights was proposed,and the influence of the maximum height difference p and the spacing s on the heat transfer characteristics of the radiator is investigated.Compared with the equal high fin arrays under the same conditions,the non-equal high fin arrays have advantages in reducing the flow resistance,and with the increase of the maximum height difference p,the advantage is obvious.Further study on non-equal high fin arrays shows that the heat transfer performance and weight of the system are significantly affected by the fin spacing s and the fin height difference p.Under the combined effect of heat dissipation area and heat dissipation,the heat transfer performance of the heat sink is nonmonotonic.Combining the advantages of multi-objective genetic algorithm and neural network model,a complete optimization process of heat dissipation system is built in this paper.The purpose is to be able to comprehensively consider the thermal resistance and quality requirements of the heat sink,and obtain the relatively optimal design parameters under different weight values.The results show that the heat transfer performance of the radiator is improved by 7.8%,the cost is reduced by 50.72%.Optimization parameters are distributed within the Ribbon range.The design variables are predicted by the neural network model and the error can be kept within 5%.The optimized heat sink structure is used for heat dissipation of the thermoelectric cooler,and the thermoelectric refrigeration performance is improved.Based on the optimization results,the theoretical model is established in this paper,and the effects of cooling temperature Tc ' and which thermal resistance Rcv on the performance of thermoelectric refrigeration system are discussed respectively.It is found that the maximum cooling capacity of thermoelectric refrigeration system,Qcmax,changes linearly with Tc' and Rcv.Compared with the thermal resistance on the cold side of the thermoelectric cooler,the cooling temperature has a greater effect on the performance of TEC.