Research on Photovoltaic Prediction Model Based on Hybrid Model Algorithm

Photovoltaic power generation, or solar power generation, is a clean power generation method that is also a highly concerned power generation method under the macro issue of carbon reduction and environmental protection. Monitoring its work efficiency is very important. Therefore, this article believed that establishing a photovoltaic prediction model to predict the work efficiency of photovoltaic power generation equipment can effectively solve this problem. Photovoltaic power generation relies on sunlight, so meteorological factors have a direct impact on its work efficiency. Therefore, this article hoped to consider the impact of meteorological factors on photovoltaic power generation through WRF (Weather Research and Forecasting Model), and design a photovoltaic prediction model that can effectively predict. Moreover, this article also conducted comparative statistics on the relevant indicators of the successfully designed model at the end. The average absolute error of the improved model was only 0.47°C, and the original model was as high as 1.87°C. The improved model achieved a prediction accuracy of 89%, while the original model was only 67%. These fully demonstrated that the improved photovoltaic prediction model based on the hybrid model algorithm in this article was competent for prediction work.