Principal component analysis and Verhulst model-based saturated load prediction method for urban power grid

Abstract

The invention discloses a principal component analysis and Verhulst model-based saturated load prediction method for an urban power grid. The method is characterized by comprising the steps of screening factors influencing load change through correlation analysis according to historical urban network total load data and statistical data of various factors possibly influencing the load change; performing feature extraction by adopting principal component analysis to obtain principal components of the factors influencing the load change; using a multivariate model to describe an association relationship between a load and the factors influencing the load change; using a Verhulst model to analyze a development law of the factors influencing the load change, thereby obtaining a future-year development trend of the factors influencing the load change; substituting future-year development trend data of the factors influencing the load change into an association model, thereby obtaining a change trend of a future-year urban network total load; and according to a future-year load growth condition, determining saturation time, a saturation stage starting point value and a saturation extremevalue of the urban network total load.

Description

technical field [0001] The invention relates to the field of saturated load prediction in urban distribution network planning, and relates to a method for predicting urban power grid saturated load based on principal component analysis and Verhulst model. Background technique [0002] Urban power grid saturation load forecasting is the basis of long-term urban distribution network planning, and it has great guiding significance for economical and reasonable power grid construction. According to the relevant regulations on the determination of saturated load in the "Technical Guidelines for Distribution Network Planning and Design" (DL / T 5729-2016), that is, according to the load saturation with a load growth rate of less than 2% or an electricity growth rate of less than 1% for five consecutive years When the regional economic and social level develops to a certain stage, the growth of power consumption slows down and remains relatively stable on the whole, and the load is s...

AI Technical Summary

Problems solved by technology

[0003] In order to make the constructed model more reasonable, consider the factors more comprehensively, and predict the results more accurately, today’s urban power grid saturated load forecasting methods are often carried out by proposing an improved model or constructing a combined forecasting model, but with the process of saturated load forecasting With the continuous increase of factors considered, while the forecasting effect is continuously improved, the following two problems also appear. One is the multicollinearity problem among factors that occurs due to the increase in factors considered when constructing the load forecasting model; the other is When determining the saturated load, since the factors affecting the load change cannot reach the saturation value at the same time, it does not mean that the factors affecting the load change also enter the saturation stage when the load enters the saturation stage. In the past, the load entered the saturation stage according to the factors affecting the load change. stage analysis ideas, most of the forecast results lag behind

Images