Modelling and experimental study of thermoelectric generator

Abstract

Thermoelectric generators can convert thermal energy into electrical energy. The working principle is based on the Seebeck effect, the first thermoelectric effect discovered by Estonian physicist Thomas Seebeck in 1821. This led to the invention of the thermoelectric generator and discovery of other thermoelectric effects, namely the Peltier effect and Thomson effect which are related to the change of thermal energy within the thermoelectric module. Due to having a low electric power output and efficiency, thermoelectric generators are not commercially viable. In recent years, with the increasing demand for clean and sustainable energy, the thermoelectric generator has gained research interest as its intended purpose is to recycle waste heat into electric power directly. In this project, the steady-state heating and pulsed heating performance of the thermoelectric generator will be studied with the aim to increase the efficiency enhancement capability of pulsed heating over steady-state heating. As experimentation substitutes the waste heat supplied with an electrically powered heating cartridge, it is costly and time consuming. Thereby, modelling of the thermoelectric generator will be carried out for both steady-state and transient conditions. The experiments conducted examines variables which affect the overall performance of the thermoelectric generator. These variables include the high-low state voltage input ratio, high-state heating time, low-state heating time and cycle period. It was observed that the heating time and temperature were dependant on each other, thereby the optimal heat cycle was obtained by controlling the minimum and maximum temperature of a heat cycle. The modelling of the power output is based on a one-dimensional quasi-equilibrium steady-state condition where the effects of radiation heat loss was considered as negligible. Power output was quantified by an energy balance equation taking the semiconductor legs as a control volume considering the effects of fourier heat conduction and the three thermoelectric effects. The temperature profile simulation utilises a finite element analysis approach through the ANSYS software. Both models are validated against manufacturer’s date and experimental data collected in this project. The motivation of this project is to improve the employability of the thermoelectric generator by increasing its efficiency which is its limiting factor. Being able to operate the thermoelectric generator at higher efficiency can possibly see higher adoption rates as a step towards environmental sustainability. The development of the model is also motivated by the awareness of sustainability, by reducing the amount experimentation required through the usage of a model can allow further research and development to be done without the need of powering multiple equipment, especially the heating cartridges which has a high-power consumption rate. The results of the experiments showed a pulsed heating efficiency enhancement of 31.07% over steady-state heating. The mathematical model for predicting power output performed with an average error of 16.85% for steady-state heating and 8.19% for transient heating compared to experimental results. The learning outcome of the experiments showed that controlling the maximum and minimum temperature of the heat cycle had a significant impact on the efficiency enhancement capability. Optimising the heating phase and energy recovery phase will result in higher efficiency enhancement. The experimental results also ascertained the research findings that a larger high-low state voltage input ratio will produce better performance. This is because the heating phase, which is majority of the energy input, will be significantly shortened.