Abstract
The cost-effectiveness and energy efficiency of evaporative cooling systems have made them a popular solution for thermal management in agriculture, livestock, electronics, biotechnology, and food processing. As the thermal performance of various evaporative cooling systems such as direct, indirect and Maisotsenko systems strongly relies on certain meteorological and system parameters, so herein, we develop a method to assess the impact of these parameters on the thermal performance of various evaporative cooling systems, including direct, indirect, and Maisotsenko systems. A surrogate model by coupling Gaussian process regression algorithm for hyper-parameters optimization with a deep neural network is developed and assessed to visualize and highlight the impact of each considered meteorological and system parameter on the thermal performance of the considered evaporative cooling systems. The findings of this study reveal that these systems are mainly dependent on the meteorological parameters (dry and wet bulb temperature, dew point temperature, relative humidity, and enthalpy). On the contrary, the system's parameters such as area and inlet velocity have no impact for the considered systems and data range. Finally, the simplified, efficient, and highly accurate model is pro-posed to assess the thermal performance of the investigated evaporative cooling systems. The final proposed model is able to predict the thermal performance of the considered evaporative cooling systems with an accuracy of R2 = 0.999 within the tested data range.