Inverse prediction and optimization analysis of a solar pond powering a thermoelectric generator

A given temperature difference across the upper and the lower convective zone of a solar pond is commonly sought in thermoelectric power generation. Based on this consideration, this work is aimed at predicting the lengths of various zones of a solar pond to ensure a minimum temperature potential throughout the year between its upper and lower convective zones. For predicting the critical lengths of various zones of the solar pond, at first, the heat energy conservation-based model available in the literature is modified by accounting the effect of salinity and temperature on various thermal parameters. The model is satisfactorily-validated with similar model and experimental data reported in the literature. Thereafter, considering the requirement of a thermoelectric power generator (TEG), an inverse problem is solved with the aid of a genetic algorithm-based optimization method to predict feasible lengths of various zones satisfying a minimum temperature potential across TEG considering suitable thermal resistances. The present results reveal improved pond dimensions achieving a better temperature profile at a lower total height than that available in the literature. Further, case studies of diverse meteorological conditions of India are carried out and it becomes apparent that, around the year, multiple combinations of convective and non-convective regions of the solar pond can ensure the required minimum (or more) temperature difference across relevant zones of the solar pond. Finally, the present study also reveals that the temperature of the upper convective zone is largely governed by the thickness of this zone, whereas, the thickness of the non-convective zone is largely responsible for the temperature within the storage zone. The present study provides a novel inverse methodology to predict and optimize the suitable dimensions of various regions of a salt-gradient solar pond to ensure a minimum temperature potential across the year for thermoelectric power generation.