Photovoltaic power station power generation sub-band prediction method based on improved EMD-LSTM combination model

Abstract

The invention provides a photovoltaic power station power generation sub-band prediction method based on an improved EMDLSTM combination model. The method comprises the steps: preprocessing a non-stable and non-linear photovoltaic power station output data sequence through an improved data decomposition method to effectively improve the prediction precision; carrying out continuation and windowingon the photovoltaic power station output sequence through a neural network to effectively separate components with different fluctuation characteristics in output data; grouping power components withsimilar fluctuation by adopting a run length judgment method to divide into three frequency bands of high frequency, medium frequency and low frequency, so that the characteristics are more centralized, the problem that the number of components after adaptive decomposition under different power generation working conditions is uncertain is solved, and the prediction speed is increased; and combining a data decomposition method with a long-term and short-term memory network, so that long-term memory of power generation data can be achieved, the problem of long dependence of a traditional neural network in prediction is avoided, and the combined model is more suitable for solving the problem of long-period, strong-fluctuation and nonlinear photovoltaic power station output prediction and has a good time sequence prediction effect.

Description

technical field [0001] The invention mainly relates to the technical field of power generation capacity evaluation of photovoltaic power plants and power grid dispatching in power systems, and specifically relates to a frequency division prediction method for power generation of photovoltaic power plants based on an improved EMD-LSTM combination model. Background technique [0002] Since photovoltaic power plants need to use sunlight resources for power generation, affected by many uncertain factors such as meteorological factors, atmospheric conditions, and natural environments, the solar irradiance reaching the ground is usually characterized by strong randomness and volatility, which leads to Fluctuation and intermittency of power station output. When the grid-connected power generation of photovoltaic power plants exceeds a certain proportion, the output fluctuations connected to the grid will pose a great threat to the stable and safe operation of the power system and t...

AI Technical Summary

Problems solved by technology

However, due to the characteristics of nonlinearity, non-stationarity and timing of the output of photovoltaic power plants, and the complexity of environmental influencing factors, the accuracy of the results is poor if the prediction is directly based on the original data and a single model

Images