Photovoltaic Power Neural Network Prediction Method Based on Chaotic Phase Space Optimization and Reconstruction

Abstract

The invention relates to a neural network prediction method for photovoltaic power generation power optimized and reconstructed by chaotic phase space. The improved C-C method is used to solve the embedding dimension m and delay τ of the time series of photovoltaic power generation power, and the chaotic phase space of photovoltaic power generation power is calculated. Reconstruction; use ensemble empirical mode decomposition and peak frequency band division to decompose and reconstruct the data of each dimension in phase space, optimize and reconstruct chaotic attractors, and reduce the impact of random power fluctuations on prediction accuracy; use GA to optimize the BP neural network Initial weights and thresholds, building a GA‑BP neural network prediction model to learn the evolution law of chaotic attractors, realizing photovoltaic power generation prediction, and obtaining power prediction values. The invention realizes the prediction of photovoltaic power generation, improves the convergence speed of the algorithm and the accuracy and stability of the prediction. Moreover, the prediction method of the present invention does not need weather and power station system data, and the prediction accuracy and stability are high.

Description

technical field [0001] The invention relates to a photovoltaic technology, in particular to a neural network prediction method for photovoltaic power generation power based on chaotic phase space optimization and reconstruction. Background technique [0002] Photovoltaic power generation has been continuously promoted and applied around the world in recent years due to its advantages such as cleanness, no pollution, and easy distribution and promotion. However, affected by meteorological factors such as solar radiation and cloud coverage, photovoltaic power generation has strong fluctuations and randomness, and its grid connection will affect the safety, stability and economic operation of the grid. Photovoltaic power forecasting can know the future output of photovoltaic power plants in advance, which can not only provide important decision support for power dispatching, but also cooperate with energy storage systems to smooth power fluctuations and reduce the adverse impac...

AI Technical Summary

Problems solved by technology

[0005] The present invention is aimed at the existing problems of photovoltaic power prediction, and proposes a neural network prediction method for photovoltaic power generation power based on chaos phase space optimization and reconstruction, using chaos theory to analyze the historical evolution law of photovoltaic power generation sequence, and using ensemble experience mode Decomposition and peak frequency band division are used to optimize the reconstructed chaotic attractor and extract the implicit fluctuation information of the data to solve the traditional forecasting method's dependence on meteorological data and the problem of deviation accumulation caused by inaccurate numerical weather forecast data; then use GA (Genetic Algorithm) Optimize the initial weight and threshold of the BP neural network, build a GA-BP neural network prediction model, learn the evolution law of the photovoltaic power reconstruction chaotic attractor, realize the prediction of photovoltaic power generation, and improve the algorithm convergence speed and prediction accuracy and stability

Images