Impacts of using helical coils (HCs) on the performance improvements of thermoelectric device mounted channels and modeling by using soft computing techniques

Abstract

Impacts of using of helical coil (HC) array in the thermoelectric generator mounted a channel on the power generation characteristics during forced convection are analyzed. The number HCs in the array (between 1 and 5), number of turns (between 1 and 5), axial pitch (between ap and 4ap), hot and cold stream Reynolds number (Reh and Rec between 250 and 1000) on the power generation are explored. Higher powers are obtained with the installation of the HCs in both channels. The power generation rises with increase in the number of HCs in the array, number of turns and fluid stream Re. However, an optimum value of axial pitch is observed to achieve the highest power generation. The optimum value is found the three times of the major radius of the HC. The power rises by 18.4% at Rec = 100, when the number of coils is increased from 1 to 5. However, the power rises by about 31.4% for five HCs in the array as compared to bare channel. At Reh = 1000, the power generation increments of 37.7% and 31.96% are achieved when cases with the highest and lowest cold stream Re are compared. The HC is an effective tool for controlling the power generation characteristics. An efficient method for reducing the computational cost of the fully coupled system with HCs is offered by using feed-forward neural networks while the computational cost is reduced from 8 hours-18 min to 12 min.