Abstract: Natural disaster statistics indicate that the damage caused by wind and hail to photovoltaic structures has shown an increasing trend over the years. There is a lack of research on the wind-hail coupling effect both domestically and internationally. Therefore, it is necessary to study the impact resistance of photovoltaic structures under the wind-hail coupling effect. Such research holds a significant practical significance in accurately predicting the impact resistance of photovoltaic structures against wind and hail. In this study, an integrated device developed by the research team for simulating hail impact was used to conduct experiments on the wind-hail coupling mechanism. Wind speed and turbulence were taken as variables to systematically investigate the peak impact force of hail particles with different diameters on photovoltaic structures. The experimental results were used to validate and guide the establishment of a BP neural network structure for predicting the impact force of single hail particles under wind and hail conditions. Furthermore, a genetic algorithm was employed to optimize the BP neural network, resulting in the development of a GA-BP neural network. The results indicate that: the peak impact force of hail increases with the diameter and ejection velocity of hail particles, as well as the wind speed, while it decreases with increasing turbulence. Additionally, under the same ejection velocity, larger hail particles are more significantly affected by the wind speed and turbulence. Compared to the traditional BP neural network, the GA-BP neural network demonstrates higher prediction accuracy and generalization ability, enabling more precise prediction of the peak impact force of single hail particles under the wind-hail coupling effect.