Salt density interval prediction method based on phase space reconstruction and quantile regression

A technology of phase space reconstruction and quantile regression, applied in forecasting, data processing applications, instruments, etc., to improve forecasting accuracy, facilitate scientific decision-making, and improve forecasting accuracy

CN108564201APending Publication Date: 2018-09-21UNIV OF ELECTRONIC SCI & TECH OF CHINA
7 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
Publication Date
2018-09-21

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a salt density interval prediction method based on phase space reconstruction and neural-network quantile regression. A salt density interval is predicted through a manner of combination of phase space reconstruction theory of chaos time sequences and neural-network quantile regression. Specifically, phase space reconstruction of salt density time sequences is carried out first, then the reconstructed salt density sequences are used to construct a neural-network quantile regression model, and then the salt density interval is predicted. Therefore, the advantages of three aspects of chaos theory, neural networks and quantile regression are combined, a change law of salt density can be accurately characterized, and scientific decision-making is facilitated.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the technical field of insulator pollution prediction, and more specifically relates to a salt density interval prediction method based on phase space reconstruction and neural network quantile regression. Background technique

[0002] With the rapid development of the national economy, the scale of my country's power grid continues to increase, and the rated voltage level of the power grid continues to increase. Therefore, the pollution flashover accidents of the external insulation of the power transmission and transformation equipment in the power system are also becoming more and more serious. It is difficult to grasp the occurrence rules of pollution flashover accidents, and there are no current measures to effectively prevent such accidents.

[0003] Under normal circumstances, the inspection and maintenance personnel of the power system operation and management department usually take measures such as increasing the cree...

Examples

Embodiment

[0051] figure 1 It is a flow chart of the salt density interval prediction method based on phase space reconstruction and neural network quantile regression of the present invention.

[0052] In this example, if figure 1 As shown, the present invention is a salt density interval prediction method based on phase space reconstruction and neural network quantile regression. The method of using neural network quantile regression to obtain different salt density prediction values ​​at different quantiles to predict the interval of salt density changes includes the following steps:

[0053] S1, phase space reconstruction

[0054] S1.1, using the autocorrelation function method to determine the delay time t

[0055] Based on the salt-dense time series data from 2011 to 2014 provided by a power department, 300 sets of data are selected as test data, that is, the number of points in the salt-dense time series is N=300; when the delay time of the salt-dense time series is t The auto...