Quantile Regression Based Genetic Support Vector Machine Photovoltaic Power Interval Prediction Method

Abstract

The invention discloses a quantile regression-based genetic support vector machine photovoltaic power interval prediction method, which obtains data samples by extracting the solar radiation value, temperature value and photovoltaic power of historical data, performs normalization preprocessing, and then passes The genetic algorithm optimizes the parameters of the support vector machine to overcome the volatility and randomness of photovoltaic power generation, builds a prediction model, and obtains high-precision photovoltaic deterministic prediction power; then, by analyzing the photovoltaic power prediction error of the prediction model, the quantile regression variable is determined, and Construct the corresponding quantile regression model according to uncertain weather factors to realize photovoltaic power interval prediction; the present invention does not need to assume the error distribution of photovoltaic prediction power, and obtains accurate photovoltaic power interval prediction ranges under different confidence conditions, which can be used for power system scheduling decision-making and operation. Risk assessment provides richer information and solves the technical problems of unsatisfactory PV power interval prediction results.

Description

technical field [0001] The invention belongs to the technical field of photovoltaic power generation prediction, and in particular relates to a quantile regression-based genetic support vector machine photovoltaic power interval prediction method. Background technique [0002] As the global fossil energy shortage and environmental pollution become increasingly serious, photovoltaics, as a renewable energy source, have rapidly increased their grid-connected capacity. Photovoltaic power generation has the advantages of simple structure, cleanness, safety, no noise, and high reliability. However, since photovoltaic power generation is affected by solar radiation intensity, battery components, temperature, weather clouds and some random factors, the system operation process is an unbalanced random process, and its power generation and output power fluctuate greatly and are uncontrollable. The performance is particularly prominent. After this power generation method is connected...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the above problems, the present invention proposes a genetic support vector machine photovoltaic power interval prediction method based on quantile regression, which realizes the improvement of photovoltaic power prediction accuracy and obtains accurate interval prediction range, and solves the problem of unsatisfactory photovoltaic power interval prediction results. technical problem

Images