Abnormal user identification method based on GBDT algorithm and factor fusion

Abstract

The invention discloses an abnormal user identification method based on a GBDT algorithm and factor fusion. The abnormal user identification method comprises the following steps: step 1, data preprocessing: collecting user power consumption data and transformer area power supply data and preprocessing the user power consumption data and transformer area power supply data; step 2, feature extraction: screening out features related to electricity stealing from historical electricity consumption and electricity consumption transaction related features by adopting a support vector machine algorithm according to the relationship between each feature and the electricity stealing behavior; step 3, model construction: using the features related to electricity stealing screened in the step 2 as input data of an abnormal user identification model, and using GBDT as a modeling algorithm to construct the abnormal user identification model; and calculating a suspected abnormal electricity user through the abnormal user identification model, and calculating the electricity stealing probability P of the abnormal user according to a suspected electricity stealing probability formula. According tothe invention, accurate identification of electricity consumption behavior abnormity and calculation of the electricity stealing probability of the user can be effectively realized, so that the troubleshooting range is reduced, the troubleshooting key point is realized, the workload of troubleshooting personnel is reduced, and the submission accuracy is improved.

Description

technical field [0001] The invention belongs to the technical field of electric power information, and in particular relates to a method for constructing a topology of a unified distribution network grid based on graph fusion technology. Background technique [0002] With the acceleration of my country's modernization, the country's energy consumption is also increasing, especially the demand for electricity is increasing year by year. In this context, some criminals steal power resources by various means, and even some areas are very rampant. Stealing electricity not only seriously affects the normal order of power supply and consumption, but also brings serious economic losses to power grid enterprises, and also causes damage to power supply and transmission equipment, and even endangers the security of the power grid. [0003] At present, most of them adopt the method of daily inspection to detect anti-stealing behaviors, but this method of investigation is inefficient, ...

AI Technical Summary

Problems solved by technology

[0004] However, the prediction accuracy of the existing abnormal power consumption model and anti-stealing model is low, and there is no output user electricity theft probability, so that the anti-stealing work can focus on the investigation while reducing the scope of investigation.

Images