Time-phased optical power prediction method and system based on meteorological classification

Through the time-divided optical power prediction method based on meteorological classification, using meteorological characteristics and binary classifiers to predict meteorological types, selecting appropriate prediction models for optical power prediction, solving the problem that the variability of meteorological factors affects the prediction accuracy and improving the accuracy of prediction Sex and robustness.

CN119005745APending Publication Date: 2024-11-22TOGEEK
0 Cites 1 Cited by

Patent Information

Application Number
CN202411063447.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-05
Publication Date
2024-11-22

Smart Images

  • Figure CN119005745A_ABST
    Figure CN119005745A_ABST
Patent Text Reader

Abstract

The invention provides a weather classification-based time-phased optical power prediction method and system, a storage medium and electronic equipment, and relates to the field of optical power prediction. According to the method, the future photovoltaic output time is divided into a plurality of time periods, and the statistical value of the downward short wave radiation flux in each time period is used as the input of three binary classifiers in sequence until the meteorological type in each time period is predicted; if the weather type is sunny, cloudy or rainy and snowy, respectively selecting prediction models in one-to-one correspondence with the binary classifiers, and if the weather type is unknown, selecting a general prediction model; and taking the statistical value of the meteorological characteristics in each time period as the input of the selected model, predicting the optical power of each time period, and obtaining a complete optical power prediction result after statistics. Meteorological classification and time-phased optical power prediction are carried out, and the influence of variable meteorological factors on optical power prediction is fully considered. In addition, a general prediction model is added to deal with unknown weather, and the robustness and generalization of the model are greatly improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Cited By

  • Weather type classification strategy and photovoltaic power generation power prediction method thereof

    CN119742755A