Unlock instant, AI-driven research and patent intelligence for your innovation.

Distribution network electric heating load prediction method based on improved BP neural network algorithm

A BP neural network and load forecasting technology, applied in neural learning methods, biological neural network models, forecasting, etc., can solve the problems of small weight changes, prone to flat areas, slow convergence process, etc., to increase the range of weight changes, Avoid output flat areas and improve economic benefits

Active Publication Date: 2018-03-02
STATE GRID XINJIANG ELECTRIC POWER CO ECONOMIC TECH RES INST +2
View PDF10 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] However, after this technology is applied to electric heating load forecasting, there are problems: the weight is given randomly, and a large number of repeated experiments lead to a slow convergence process; the correction amount for the given weight is very small, so the optimization is through local improvement If the direction is gradually adjusted, it is difficult to achieve satisfactory prediction results; when the output is close to the boundary value 0 or 1, it is easy to appear a flat area, and the weight change is very small, which is easy to stop the training process

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Distribution network electric heating load prediction method based on improved BP neural network algorithm
  • Distribution network electric heating load prediction method based on improved BP neural network algorithm
  • Distribution network electric heating load prediction method based on improved BP neural network algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0069] Obtain historical data and parameter setting step S1: Obtain the construction area of ​​new buildings planned in the power supply area over the years, the utilization rate of wind power generation over the years, and the electric heating load over the years; set m=3 layers of neural network, that is, an input layer, One hidden layer, one output layer, the total number of neurons is n=5, so the input layer neurons are x 1 , x 2 , the hidden layer neuron is x 3 , x 4 , the output layer neuron is x 5 , the expected output of the electric heating load is y, and the number of data sets calculated by the neural network is V max =year-1, year is the number of years of historical data, let the number of iterations it=1, the number of training groups variable v=1, given the maximum number of iterations it max , given the output error limit ε y , the error ε of a given neuron close to the upper limit 上 and the error ε close to the lower limit 下 , given the learning rate η ...

Embodiment 2

[0085] Step S1: Obtaining step

[0086] Taking a city in Xinjiang as an example, the following table shows the planned new building area, wind power utilization rate, and electric heating load value of the city from 2013 to 2016;

[0087]

[0088]

[0089] Step S2: Calculate the correlation coefficient

[0090] The normalized values ​​are shown in the table below:

[0091] years

[0092] The first set of data is the planned new building area in 2012 and the utilization rate of wind power generation to predict the electric heating load value in 2013, so that the number of iterations it=1, it max =500, v=1, output error limit ε y = 0.01, ε 1 = 0.01, ε 0 =0.01, θ 3 =0.1756, θ 4 =0.0472, θ 5 = 0.0953, η = 0.542;

[0093] The calculated correlation coefficient between the new building area and the electric heating load is ξ 1 = 0.92, the correlation coefficient between wind power utilization rate and electric heating load is ξ 2 =0.95;

[0094] Step S3: s...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a distribution network electric heating load prediction method based on an improved BP neural network algorithm. The method comprises the following steps of S1, acquiring historical data and carrying out parameter setting; S2, calculating a correlation coefficient; S3, carrying out weight coefficient assignment; S4, calculating and outputting each layer of neurons; S5, outputting and determining a result; S6, determining top and bottom limitations; S7, calculating a neuronal learning error; S8, carrying out weight coefficient correction based on a learning error; S9, randomly correcting a weight coefficient; S10, randomly correcting a bias and a learning rate; S11, determining whether the last set of data is calculated; and S12 carrying out final prediction. By using the distribution network electric heating load prediction method, a convergence speed is increased, a flat region is prevented from outputting, a weight change amplitude is increased, influences ofwind abandoning heating and a newly-constructed green building on electric heating promotion are considered and an electric heating load prediction result approaches an actual value.

Description

technical field [0001] The invention relates to a load forecasting method, in particular to a distribution network electric heating load forecasting method based on an improved BP neural network algorithm. Background technique [0002] Distribution network load forecasting is the basis of distribution network planning. However, with the access of a large number of electric heating loads, the load of the distribution network has been greatly increased in the heating system. The planning of the distribution network needs to be adjusted in time to adapt to large-scale electric heating loads. Therefore, accurate prediction of electric heating load is the premise and basis of current distribution network planning. [0003] The relationship between electric heating load forecast and historical data is relatively small, which is mainly affected by the following two factors: 1) One of the main purposes of promoting electric heating in my country is to abandon wind for heating, consu...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06Q10/04G06Q10/06G06Q50/06G06N3/08
CPCG06N3/084G06Q10/04G06Q10/06315G06Q50/06Y04S10/50
Inventor 周红莲李娟薛静杰华东张三春周会宾王燕李忠政郑伟东任知猷陈露锋孙家文李娴李清李光应孔锦绣罗攀刘自发王泽黎
Owner STATE GRID XINJIANG ELECTRIC POWER CO ECONOMIC TECH RES INST