Abstract A wide range of technical applications, including solar power, waste heat recovery, electronics thermal management, and heat exchangers, employ thermoelectric generators. They can be mounted in between channels / cavities where hot and cold fluid streams exist. In this study, two novel methods of enhancing the power generation from thermoelectric generator device mounted in between vented cavities are proposed by combined utilization of active heater/cooler rectangular blocks and blade-shaped nanoparticles in base fluid. Finite element method investigation is conducted numerically for a range of hot and cold stream Reynolds numbers (250–1000), non-dimensional hot and cold block sizes (0.01 $$-$$ - 0.4), and heating/cooling increments (0–10), with nanoparticle loading limited to 0.03. Higher values of Reynolds number results in a rise in thermoelectric generator power. When comparing the cases of lowest and highest Reynolds number combinations, a 219 $$\%$$ % increase in power is achieved. The thermoelectric generator power will rise by around 27.5 $$\%$$ % when the object size reaches its maximum. However, for moderate object sizes, up to 31.6 $$\%$$ % reduction in power generation can be realized. Greater temperature differences result in a linearly rising power generation, with an achievable power increase of up to 22 $$\%$$ % . When nanoparticle loading in the base fluid for both cavities is raised to its maximum value, the resultant power increases by around $$30\%$$ 30 % . Thermoelectric generator power rises by 67.8 $$\%$$ % when an active heater/cooler with nanofluid is used in vented cavities, as opposed to the reference scenario of employing no object and only water. The thermoelectric generator device’s hot and cold interface temperatures are accurately estimated using the artificial neural network based method. The estimated temperature can be used as boundary condition for the solution of the governing equations in the thermoelectric generator device domain which will decrease the computational cost when dealing with very complex channel configurations.
Read full abstract