Assessment of using atmospheric aerosol Transport models for predicting the soiling rate in CSP plants

Abstract

Particle deposition on solar mirrors deteriorates the performance of concentrating solar power (CSP) plants considerably. To save water for cleaning activities and to increase the plant performance, the empirical soiling model was developed by DLR to calculate the daily decline in cleanliness (soiling rate) of CSP collectors based on meteorological and aerosol particle input parameters. Until now, the model can only be applied to site locations where measurements of the necessary Input parameters are available. In this work, a new approach is investigated to extend the soiling model application to various locations: aerosol transport models simulate particle concentrations and many other parameters covering large areas, so they can provide the parameters that are needed as input for the soiling model. To implement this idea while following an integral stepwise process, the soiling model’s sensitivity for input data in a different resolution of time and aerosol particle size bins is investigated by adapting ground measurement data. The meteorological and particulate matter parameters provided by two transport models, the NMMB/BSC Dust model and the CAMS model by ECMWF, are compared to the ground measurement data. Finally, the data is adapted to the soiling model format, its performance with the new input data is validated and the best available input data source and configuration is determined. The validation of the soiling model for modelling PV soiling shows that the soiling model is universally applicable technology-wise too.