Fisher information processing-based single-weather factor short-term load prediction method

Abstract

The invention relates to a fisher information processing-based single-weather factor short-term load prediction method. According to the method, historical sample data are analyzed, so that single-weather factor fisher information window data corresponding to load are obtained; the fisher information value of the single-weather factor fisher information window data is calculated according to the single-weather factor fisher information window data; weighting processing is performed on the fisher information value, so that a fisher information weighted input variable is obtained; and the inputvariable and historical load data are inputted into an artificial intelligence prediction model. Fisher information weighting processing is performed on single meteorological factors, so that the cumulative effect of the meteorological factors on the load can be reflected more comprehensive and reasonable, so that weather-sensitive load can be accurately predicted.

Description

technical field [0001] The invention relates to the technical field of electric power grid load forecasting, in particular to a single meteorological factor short-term load forecasting method based on fisher information processing. Background technique [0002] Short-term load forecasting is the basis for safe and economical operation of power systems. Its prediction accuracy is affected by many interdependent factors, among which meteorological factors have a great influence on short-term load forecasting, and the cumulative effect of meteorological factors will cause large errors in the load forecasting results. Therefore, the influence of cumulative effects should be considered in load forecasting. Among all meteorological factors, temperature has the most serious influence on load. During high temperature weather in summer, the power grid load increases sharply with the increase of temperature, especially the continuous high temperature weather will also have a cumulati...

AI Technical Summary

Problems solved by technology

The literature "Analysis of the Influence of Real-time Meteorological Factors in Short-term Load Forecasting and Its Processing Strategy" "Power Grid Technology, 2006" gives a method for processing the cumulative effect of temperature and humidity in the neural network forecasting model, but it needs to add a large number of input layer variables , which undoubtedly increases the training time and prediction time of the network

Images