研究目的
To improve the accuracy of photovoltaic (PV) power forecast by proposing a short-term PV output forecast method that considers error calibration under typical climate categories.
研究成果
The proposed method effectively improves the accuracy of short-term PV power forecasts by considering error calibration under typical climate categories. The NKDE simulation fits better than normal distribution simulation, and the fluctuation analysis combined with Latin hypercube sampling provides boundaries to avoid overcompensation and divergence of error.
研究不足
The study is limited by the specific geographical and climatic conditions of Hebei Province, China, and the year 2013 data. The methodology's effectiveness in other regions or under different climatic conditions is not verified.
1：Experimental Design and Method Selection:
The study defines typical climate categories to classify historical PV power data and employs nonparametric kernel density estimation (NKDE) to simulate probability density function (PDF) curves for each category. Latin hypercube sampling and a multi-scenario technique are used for sampling the RE. Fluctuation analysis is conducted to obtain compensation values of the RE.
2：Sample Selection and Data Sources:
Historical PV power data from Hebei Province, China in 2013 are used, classified into 16 typical climate categories based on weather conditions and seasons.
3：List of Experimental Equipment and Materials:
MATLAB is used for simulation. The study involves a Polycrystalline Silicon Photovoltaic Array (10 kW) in the State Key Laboratory of New Energy Power System at the North China Electric Power University.
4：Experimental Procedures and Operational Workflow:
The methodology includes defining climate categories, classifying PV power data, generating PDF curves for RE, performing PV power forecast using WNN with ADP, sampling RE values, conducting fluctuation analysis, and correcting forecast values with fitted RE values.
5：Data Analysis Methods:
The study uses Mean absolute error (MAE), root mean square error (RMSE), and Mean absolute percent error (MAPE) to measure forecast errors.
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
