Influence of compressed air dust removal on trough solar system in summer based on alpine areas: heat collection performance analysis and neural network prediction research

Based on the high irradiance in alpine areas, this study thoroughly investigates the impact of compressed air on the photothermal performance of trough solar systems before and after dust removal through experiments. The proportional relationship between energy generation and consumption was explored through dust removal airflow with different pressures to improve energy utilization. Furthermore, the Artificial Bee Colony algorithm was established to optimize Long Short-Term Memory for predicting the heat collection performance of the system. The research findings reveal that for every 0.1 MPa increase in inlet pressure, the specular reflectivity of the concentrator increases by 2.76 % on average, and the clearance rate increases by 3.55 %. When the dust density decreases by 3.61 g/m2, the reflectivity increases by 10 % on average, and the heat collection capacity per unit time increases by 17.11 kJ. The heat collection efficiency also improves by 10.43 %. The heat collection performance is observed to change with the dust level, and an airflow pressure of 0.5 MPa can provide the highest energy production-consumption ratio for the system The simulation results indicate that Long Short-Term Memory can significantly improve prediction accuracy after optimization using Artificial Bee Colony algorithm, which provides a scientific basis for dust removal management and performance optimization of the trough solar energy system in alpine areas.