Electricity stealing user identification method based on support vector machine and related equipment

The embodiment of the invention discloses an electricity stealing user identification method based on a support vector machine and related equipment, and the method comprises the steps: firstly obtaining a set of electricity consumption data of each user recorded according to a metering period in a transformer area in a preset period and a set of lost electricity data recorded according to the metering period in the transformer area, and then cleaning the data from three aspects, comprising the steps of deleting abnormal records from data, cleaning and removing data influencing the modeling judgment accuracy of the support vector machine from the aspect of power consumption, and cleaning and removing noise data from the aspect of power loss. When the data is cleaned from the aspect of power loss, regression fitting operation is performed on the power consumption according to the power loss, and then the power consumption is eliminated according to set conditions, so that excessive cleaning can be avoided. And