Electricity theft prevention method based on user behavior analysis

Abstract

The invention relates to an electricity theft prevention method based on user behavior analysis. The electricity theft prevention method comprises the steps of establishing an artificial neural network through using electricity theft parameters of a peak active power, an ordinary active power, a valley active power, a total reactive power, a total active power, a power factor and the like as input vectors and using an electricity theft coefficient as an output vector; performing normalization on the generated training sample, inputting the normalized training sample into the artificial neural network, and training the artificial neural network; inputting the acquired electricity theft parameters into the trained artificial neural network, and outputting the electricity theft coefficient; and determining whether a user has an electricity theft suspicion according to magnitude of the electricity theft coefficient. Compared with a traditional artificial case selection and checking method, the electricity theft prevention method has advantages of effectively realizing big data throughput, improving data analysis processing efficiency, and more accurately determining an electricity theft suspect.

Description

technical field [0001] The invention relates to the technical field of electricity theft analysis, in particular to an electricity theft prevention method based on user behavior analysis. Background technique [0002] For a long time, the phenomenon of illegal power theft has been seriously perplexing electric power companies. Some electricity units or individuals evade electric energy measurement by changing the wiring of electric energy meters, the influence of external strong magnetic fields, and connecting lines with branch circuits to achieve the purpose of stealing electricity, and use electricity stealing as a means of benefiting or reducing costs. Illegal electricity theft not only seriously violates the legal rights and economic benefits of power supply enterprises, but also causes a large loss of resources to the country, and also poses a huge threat to the safe use of electricity. According to statistics, nearly 25% of the electric shock and fire accidents that o...

AI Technical Summary

Problems solved by technology

[0004] However, enterprises and residents are user groups that frequently steal electricity. On the one hand, due to the huge number of users, frequent and periodic surveys cannot be used to check users' electricity theft behavior; on the other hand, there are insufficient methods for analyzing user data Efficient and accurate, so when a large-scale survey of electricity consumption behavior is carried out, for example, when a high-frequency and high-penetration surprise inspection is carried out in areas where electricity theft is particularly serious, the detected electricity consumption data is huge and cannot be analyzed quickly and accurately

Images