研究目的
To propose a hybrid probabilistic estimation method for predicting the probability of PV power generation for every hour in one day, improving the accuracy and practicability of the prediction model.
研究成果
The proposed hybrid model, combining SPGP and IMGWO, demonstrates higher prediction accuracy and practicability compared to other models, effectively reflecting the fluctuation characteristics of PV power generation and supporting grid stability and safety.
研究不足
The study does not explicitly mention limitations, but potential areas for optimization could include the handling of extremely variable weather conditions and further reduction of computational complexity.
1：Experimental Design and Method Selection:
The study employs a hybrid model combining RF for variable selection, FCM for data division based on weather patterns, and SPGP improved by IMGWO for forecasting.
2：Sample Selection and Data Sources:
Historical power data from a solar power station from 2012 to 2014, focusing on data from 6 to 18 o'clock.
3：List of Experimental Equipment and Materials:
Not explicitly mentioned.
4：Experimental Procedures and Operational Workflow:
RF reduces input variables; FCM divides data into sunny and cloudy types; SPGP and IMGWO are used for forecasting.
5：Data Analysis Methods:
Evaluation based on RMSE, interval value, and coverage value for point and interval predictions.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
