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

Power load clustering method and device

A technology of power load and clustering algorithm, applied in the field of cluster analysis, can solve the problems of incompatibility between accuracy and speed, and achieve the effects of avoiding randomness and blindness, improving customer relationship, and reducing peak and valley loads

Inactive Publication Date: 2017-03-22
STATE GRID CORP OF CHINA +2
View PDF4 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0023] The embodiment of the present invention proposes a clustering method and device for power load data to solve the problem that the accuracy and speed of the existing clustering methods for power load data clustering cannot be balanced

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
  • Power load clustering method and device
  • Power load clustering method and device
  • Power load clustering method and device

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0086] In this embodiment, 96 points of data (such as Figure 7 Shown) for clustering, in the hadoop environment, use Spark for clustering operations.

[0087] Step 1. Dataset preparation

[0088] The sample database contains 96 point load data of 110,000 enterprise users in a certain city on a certain day. Canopy and Kmeans support parallel operations, so the model runs on the hadoop platform and uploads data to HDFS.

[0089] Step 2. Data vectorization

[0090] Divide the 96-point load data collected in the database by the contract capacity, standardize the data into standardized data between 0 and 1, and use the InputDriver class to convert the txt file into the file format (VectorWritable) required by the Canopy algorithm.

[0091] Step 3, Canopy clustering

[0092] 1. Vectorize the data set to get a List and put it into the memory, and use the Manhattan distance to calculate the distance between all samples to get: T 1 = 40.027.

[0093] 2. Eight categories are cal...

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 provides a power load clustering method and device. The method comprises the following steps: acquiring power load data; implementing Canopy clustering on the power load data to generate a plurality of canopy classes and canopy centers; and generating a power load clustering result by taking the canopy centers as K values and using a K-Means clustering algorithm. According to the power load clustering method and device provided by the invention, customers are clustered by using a method of combining the Canopy clustering with the K-Means clustering, so that the clustering speed and accuracy can be greatly increased, the randomness and blindness of K value selection can be avoided; and in addition, power companies can be targeted to provide active services to batch customers through the step of clustering and grouping electricity customers with different attributes and behavior characteristics and the step of analyzing the power load trends of the customers in the same group, and thus the aims of improving the customer relationships, increasing the customer satisfaction degree, preventing the electric charge risk, reducing the peak and valley load, achieving the high-quality service, lowering the cost and improving the efficiency can be achieved.

Description

technical field [0001] The invention relates to the field of cluster analysis, in particular to a clustering method and device applied in the field of electric load. Background technique [0002] This section is intended to provide a background or context for implementations of the invention that are recited in the claims. The descriptions herein are not admitted to be prior art by inclusion in this section. [0003] Cluster analysis, referred to as clustering, is a process of dividing data objects (or observations) into subsets. Each subset is a cluster such that objects in a cluster have high similarity but are not similar to objects in other clusters. The collection of clusters produced by cluster analysis is called a cluster. On the same dataset, different clustering methods may produce different clusters. The division is not done by people, but by clustering algorithms. An important use of cluster analysis is to divide target groups into groups with multiple indica...

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): G06Q50/06G06K9/62
CPCG06Q50/06G06F18/23213
Inventor 许鑫傅军朱天博周辛南王畅介志毅孙志杰汤佩霖边海叶
Owner STATE GRID CORP OF CHINA