Analysis of portable solar concrete ponds by using coal cinder to trap thermal energy of sustainable building using artificial intelligence

•This study proposes a new method of enhancing thermal energy storage in portable solar concrete ponds by adding coal cinder.•The study uses machine learning techniques, specifically Particle Swarm Optimization (PSO), to analyze the energy efficiency of the solar pond.•The results show that the energy performance of the solar pond decreases as the size and thickness of the non-convective zone decreases.•The study offers thermal analysis of a salt gradient solar pond with a constant temperature differential of 30°C and 20°C.•The annual average efficiency of the solar pond is found to be 23% and 16% for ponds with depths of 1.5 m and 1.0 m, respectively, making it a promising solution for low-temperature energy needs, such as space heating. Cinder is a waste product made from the coal residue left over from power plant blast furnaces. Solar energy is a promising renewable energy source, but because of its erratic availability, energy storage technologies must be used. Our research suggests employing coal cinder (CC) to boost the thermal energy stored in transportable solar concrete ponds, combining rock mechanics with renewable energy. The concepts of renewable energy and rock mechanics are combined in this invention. Additionally, because it doesn't release any pollutants into the sky, it is great for low heat applications like building cooling and heating. However, due to its sporadic nature, solar energy requires effective energy storage to be used at night and in cloudy weather. By adding coal cinders to the bottom layer of a solar concrete pond in order to increase its heat energy, this work integrates rock mechanics with sustainable energy. A variety of small-scale research use the standard micro solar concrete ponds. Also offered is a numerical calculation for predicting energy development over a sizable area of a solar pond with a salt gradient. The coal cinders may significantly raise the temperature at the pond's bottom. With a constant temperature differential of 30 °C and 20 °C throughout the gradient layer for two different pond sizes with depths of 1.5 m and 1.0 m, this study provides thermal analysis of a salt gradient solar pond using machine learning as Particle Swarm Optimization (PSO) and extracting heat from the Lower Convective Zone (LCZ). Going through the regression indices (RMSE, R2, r), the energy efficiency (heat) of the first 1.5-meter-deep concrete pond varies between 22.1 and 11.3% throughout the winter. The energy performance of