A two-layer spectral clustering method for power load curves considering wavelet entropy dimensionality reduction

A power load and spectral clustering algorithm technology, applied in computing, computer components, instruments, etc., can solve problems such as excessive data dimension and the inability of algorithms to adapt to different demand responses, so as to improve operating speed, improve stability, reduce The effect of data volume
CN108805213BActive Publication Date: 2021-11-23SHANDONG UNIV
0 Cites 1 Cited by

Patent Information

Authority / Receiving Office
CN ยท China
Patent Type
Patents(China)
Current Assignee / Owner
SHANDONG UNIV
Publication Date
2021-11-23

Smart Images

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

Abstract

The invention discloses a double-layer spectrum clustering method for electric load curves considering wavelet entropy dimensionality reduction, comprising: obtaining daily load data of electric loads and forming a data set; segmenting the electric load data in the data set into q intervals And calculate the wavelet entropy S of the original data set in the interval q q , according to the calculated wavelet entropy value and the threshold value of wavelet entropy to measure the degree of fluctuation of the data, the degree of fluctuation is greater than the specified threshold; on the contrary, the degree of fluctuation is relatively small; the number of loads whose wavelet entropy value is greater than the threshold value of wavelet entropy in the statistical interval q And calculate the proportion of the load data in the total load of the electric load; divide the interval with a proportion greater than the threshold into two sections, calculate the wavelet entropy value in the interval again and compare the fluctuation degree of the measurement data until the load with a large fluctuation degree in the interval accounts for The proportion of all loads is less than the threshold or the number of points in the interval cannot be equally divided, and the load curve data with variable time resolution is obtained; the double-layer spectral clustering obtains similar and refined load clusters.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical field of power systems, in particular to a double-layer spectrum clustering method for power load curves considering wavelet entropy dimension reduction. Background technique

[0002] The construction of the Energy Internet has promoted the development of big data on power distribution and consumption. The continuous accumulation of these energy-consuming data has brought certain difficulties to the implementation of demand response. The massive data of the power grid makes simple statistical analysis of load characteristics meaningless. , it is more unrealistic to adopt different control strategies for each load. It is necessary to classify the loads participating in demand response. The traditional classification method is to divide according to typical industries. However, the load curves of power users in the same industry may vary greatly , the load curves of power loads in different industries may also be ve...

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