Data mining-based method for recognizing electricity stealing behaviors of users

Abstract

The invention provides a data mining-based method for recognizing electricity stealing behaviors of users. The method comprises the following steps of: collecting data; analyzing the data; extractingfeature data; constructing an electricity stealing behavior recognition model; and obtaining an electricity stealing pre-warning degree. According to the method, electricity stealing-related data is collected and model training is continuously optimized to finally achieve the optimum electricity stealing behavior recognition model, so that the model is capable of carrying out correct electricity stealing behavior recognition according to basic attribute features of users, electricity utilization laws, industrial features, electricity utilizing time features and electric power industry operation features, the fighting efficiency, for electricity stealing behaviors, of electric power departments is improved through efficient and accurate electricity stealing behavior recognition, and significance is provided for ensuring the economic, reasonable and safe operation of electric power enterprises, improving the safe using of electric power and decreasing bad social moods.

Description

technical field [0001] The invention relates to a method for identifying a user's electricity stealing behavior based on data mining, and belongs to the technical field of electricity theft prevention analysis. Background technique [0002] With the continuous improvement of people's living standards and economic growth, the demand for electricity is also increasing. In the process of power supply in the power sector, illegal electricity theft has always been one of the major problems in the development of power grids. With the continuous change and development of electricity theft methods , so that the scope of influence of stealing electricity continues to expand, bringing huge uncertainty to the healthy and stable development of power companies, causing economic losses to power companies, and there are also major security risks. In recent years, due to illegal electricity theft The fire caused by it has become one of the main causes of fire. [0003] At present, the prev...

AI Technical Summary

Problems solved by technology

[0004] However, in the face of a huge user group, it takes a lot of manpower and material resources to check one by one, and the effect is not satisfactory. Therefore, an efficient and accurate anti-stealing method is needed to identify and lock the user's stealing behavior

[0005] In the prior art, a variety of methods are provided for anti-stealing electricity, such as a Chinese patent application number 201310148105.7, which discloses an anti-stealing method based on data mining of the electricity load management system. Using formulas to judge electricity stealing behavior is complicated, and it is easy to cause large error in the results, which takes a long time and is low in efficiency.

[0006] Another example is a Chinese patent application number 201711260280, which discloses a method for discovering electricity stealing behavior based on random forests, and a Chinese patent application number 201710842008, which discloses a method and application for determining the probability of electricity theft based on big data analysis of electricity consumption behavior. The Chinese patent No. 201710006620 discloses an anti-stealing method based on user behavior analysis, which uses more complex calculation formulas to judge the behavior of stealing electricity. Longer, less efficient question

[0007] Another example is a Chinese patent application number 201611157830, which discloses a single anomaly analysis anti-stealing early warning analysis method. This method only establishes one model, and uses a single model to analyze all data without adjusting its input and output results. The obtained The result is not optimal, and errors are prone to occur, making the results inaccurate, wrong power outages, and power failures, causing unnecessary troubles to users

Images