User-transformer relation identification method based on electric quantity and voltage information fusion

Abstract

A user-transformer relation identification method based on electric quantity and voltage information fusion is based on hot restart stochastic gradient descent and one-class support vector machine information fusion, and comprises the following steps: (1) constructing a transformer area electric quantity parameterized model; (2) solving a globally optimal solution by using stochastic gradient iteration and a learning rate self-adjusting method, and carrying out preliminary identification; (3) obtaining a primary user transformer identification result through multiple times of judgment by using a sliding time window; (4) based on the preliminary identification result, forming a training sample by the voltage data of the users with the normal user-transformer relation; and (5) learning voltage characteristics of normal users in the transformer area user-transformer relation by using a first-class support vector machine, and constructing a user-transformer relation identification model to realize correct identification of the low-voltage transformer area user-transformer relation. According to the method, wrong users in the user-transformer relation file can be effectively identified, the recall ratio and the precision ratio are greatly improved, the troubleshooting cost is saved, a comprehensive judgment method based on fusion of different feature information of the transformer area is realized, and compared with a single feature information judgment method, the method has the advantages that the identification result is more stable and reliable.

Description

technical field [0001] The invention relates to the field of electric power distribution, and relates to household transformer relationship identification, in particular to a household transformer relationship identification method based on the fusion of electric quantity and voltage information. Background technique [0002] With the vigorous development of social economy and the improvement of people's living standards, electricity consumption has increased, and the line loss management of low-voltage station areas has become increasingly complex. In order to speed up the construction of the smart grid, fully tap the data resources on the power distribution side, and improve the quality of power supply services and customer satisfaction, the traditional extensive management has been unable to meet the needs of the times, and the lean management of the station area has become a new management trend . However, for a long time, distribution transformers have often been added...

AI Technical Summary

Problems solved by technology

However, this method needs to be tested for electricity, which affects people's normal electricity life, electricity experience and power supply reliability.

The carrier communication verification method is currently widely used. This method uses the carrier communication terminal on the side of the transformer and the handheld receiver on the side of the user's electric energy meter to perform carrier communication. The station area to which the user belongs is determined by analyzing the characteristics of the communication message. This method is limited to the user The electric meter supports the station area of ​​the carrier communication protocol, and its robustness is poor, it is susceptible to interference, and the recognition accuracy is only 70%-80%, and it cannot be carried out in the station area with large load fluctuations.

Images