The Design And Optimization Of Thermoelectric Generator With Great Temperature Difference

Thermoelectric device can convert thermal energy to electrical power directly via the Seebeck effect. It has been of great interest in low-temperature thermal energy utilizing applications and space power systems because of its well-known advantages such as no moving parts, good stability, high reliability, environmental friendliness and long operation life. The design theory of thermoelectric devices has been a hot study issue for domestic and international researchers.Aiming at the problem that the constant properties model (CPM) just can calculate the performances of a thermoelectric module with small temperature difference, a variable properties model (VPM) considering the temperature dependence of the thermoelectric material properties was established. Based on the basic thermoelectric coupling differential equation, VPM builded a set of thermoelectric balance equations by dividing the thermoelectric legs into many elements, and then a computer procedure including iterative algorithm was developed to solve the equations and calculate the performances of thermoelectric device. The theoretical results agree well with the test results, which show that VPM can evaluate the performances of the thermoelectric moudle accurately, no matter how large the temperature difference is. Therefore, VPM can be used to guide the design of thermoelectric devices with great temperature difference.Due to the fact that every thermoelectric material just has a high ZT value in a narrow temperature range, the leg of the thermoelectric device with great temperature difference always consists of two or more thermoelectric materials with different applicable temperature range, which makes to the segmented thermoelectric device (STD). The design of the STD is a complicated problem. Based on the VPM, a kind of multi-parameter and nonlinear optimization method, Powell, was applied to solve the optimization problem that searching the optimal value for geometric parameters of each thermoelectric material and making the thermoelectric device gain the largest output power per mass. Comparing to the optimization results obtained according to the ZT curve and the CP curve of thermoelectric materials, the output power per mass of the STD optimized by the Powell optimization algorithm, has been improved by 8.0% and 11.5%, respectively.Finally, the application of the thermoelectric device with great temperature difference optimized by the Powell optimization algorithm in the concentration solar thermoelectric system (CSTS) was studied. Based on the best available properties of thermoelectric materials reported in the literature, the best possible performances of the CSTS with the single material thermoelectric device and the segmented thermoelectric device made of the above materials were predicted. The theory-calculated efficiency of the CSTS using the BiSbTe—In0.2Ce0.15Co4Sb12 STD optimized by Powell reaches 11.8%, comparing to the CSTS adopting the BiSbTe single material thermoelectric device and the In0.2Ce0.15Co4Sb12 single material thermoelectric device, this efficiency has been improved 61.6% and 12.4%, respectively. The optimization design of the thermoelectric device with great temperature difference makes important sense for improving the generating efficency of CSTS.