Restricted Boltzmann machine-based photovoltaic power generation short-term power probability prediction method

Abstract

The present invention relates to a restricted Boltzmann machine-based photovoltaic power generation short-term power probability prediction method. According to the method, an improved depth-restricted Boltzmann algorithm is adopted. The method includes the following steps that: a genetic algorithm optimized depth-restricted Boltzmann machine-based learning prediction model is built; the similar dates of a date to be predicted are searched through a gray correlation coefficient method; and with the historical electricity generation power data of the similar dates and the weather factor information of the predicted date from data weather forecasts adopted as the input quantity of the model, the output power information of a photovoltaic power generation system in the predicted date adopted as the output quantity of the prediction model, and actually measured information of one month adopted as the sample data of the prediction model, the photovoltaic power generation output power in 1 to 24 hours is predicted, and the probability density function of a predicted point is obtained; and a genetic algorithm is adopted as an optimization model, and the crossover rate of the genetic algorithm is 0.6, the value of the variation rate of the genetic algorithm is 0.2; and the number of the structural layers of the depth-restricted Boltzmann machine is set to be four layers. The method of the present invention is advantageous in high prediction accuracy.

Description

technical field [0001] The invention relates to a method for forecasting photovoltaic power generation. Background technique [0002] Photovoltaic power generation has the advantages of less pollution and flexible scale, and has been widely used. However, since the photovoltaic power generation system is significantly affected by environmental factors, it has characteristics such as uncertainty, volatility, and intermittency, which is not conducive to the safe dispatch and energy management of the power grid, and increases the operation risk of the power grid. The greater the volatility of the photovoltaic power generation system, the lower the prediction accuracy of the deterministic prediction of its power generation. Therefore, the probabilistic prediction of the short-term power of photovoltaic power generation can reflect the information of photovoltaic power generation more comprehensively. Security scheduling and energy management are of great significance. [0003]...

AI Technical Summary

Problems solved by technology

[0004] For photovoltaic power generation systems, since their power generation is significantly affected by natural environmental factors, it is difficult to accurately predict the short-term power of photovoltaic power generation.

Images