Solar thermal output forecasting that is derived from the solar irradiance forecasting, is very much exposed to high forecasting error. This is because of the heavy dependence on the solar irradiance prediction accuracy that could be very low in certain situations owing to the high uncertainty in weather conditions. By considering this fact, this paper proposes to develop a solar thermal output forecasting model using measured solar irradiance data, instead of the predicted data. The proposed model applies K-Nearest Neighbors (K-NN) algorithm to generate 24-hour ahead forecasting data on solar thermal output from a solar parabolic trough system For the purpose of illustrating the forecasting model performance, Kuala Lumpur, Malaysia is used as a case study, with PolyTrough 1800 model is selected as the solar parabolic trough collector under investigation. Simulation has been carried out using Matlab software to verify the effectiveness of the proposed K-NN-based forecasting model. The results show that the model is able to produce acceptable results in certain conditions.
|Journal||IOP Conference Series: Earth and Environmental Science|
|Publication status||Published - 2019 Oct 29|
|Event||2019 3rd International Conference on Sustainable Energy Engineering, ICSEE 2019 - Shanghai, China|
Duration: 2019 May 24 → 2019 May 26
ASJC Scopus subject areas
- Environmental Science(all)
- Earth and Planetary Sciences(all)