Method for forecasting short-term load of distribution network based on human body comfort index

Abstract

The invention relates to a method for forecasting the short-term load of a distribution network based on a human body comfort index. The method is characterized in that a concept of the human body comfort index in meteorology is introduced in combination with the characteristics of the distribution network, a calculation method of the daily human body comfort index classified on the basis of temperatures is provided, a load-human body comfort index change rate is calculated by taking the load curve similarity of the daily human body comfort index similarity as a reference according to the load shape similarity and the daily human body comfort index distance, and a daily load value to be forecasted is calculated by the change rate and the reference curve. The holiday load is calculated by an averaging method of normal-day calculation and holiday calculation so as to prevent the few holiday samples and the abrupt temperature change from affecting the load. The method is based on the human body comfort index change, the regional and seasonal differences are prevented from affecting the forecasting value, the algorithm is concise and convenient, a small sample space is needed, the large-scale calculation is facilitated, and the actual demand for forecasting the short-term load of the distribution network can be met.

Description

technical field [0001] The invention belongs to the field of power grid load forecasting, and in particular relates to a short-term load forecasting method of a power distribution network. Background technique [0002] Distribution network short-term load forecasting is an important part of distribution network load forecasting. It is the main basis for formulating distribution network operation mode and realizing optimal operation, and is also an important basis for checking distribution network security. At present, most of the distribution network load forecasting focuses on medium and long-term load forecasting, and the research on short-term load forecasting mostly focuses on the transmission network field. The main methods are divided into traditional methods and intelligent methods. Traditional methods include time series forecasting, regression analysis, least squares, exponential smoothing, etc.; intelligent methods include gray forecasting, expert system, wavelet a...

AI Technical Summary

Problems solved by technology

At present, the impact of meteorological factors is mostly changed in a fixed way to change each influencing factor into a dimensionless unified value by selecting the influencing factor. This method is greatly affected by human selection factors.

The intelligent method requires a large amount of sample space, and the training time is long, and it is not adaptable to the characteristics of a large number of data prediction points and a small amount of historical values ​​in the distribution network.

Images