Power distribution network fault diagnosis method based on deep feature clustering and LSTM

Abstract

The invention relates to the technical field of intelligent power grids, in particular to a power distribution network fault diagnosis method based on deep feature clustering and LSTM. The method comprises the following steps: extracting high-level features of time series data by utilizing a feature extractor built by a convolutional neural network, and performing semi-supervised clustering on allthe features extracted from the data, so as to obtain corresponding labels for label-free samples. Therefore, the fault type of the label-free sample can be determined and utilized. After samples ofdifferent types of faults are subjected to an oversampling algorithm, a classifier built by a recurrent neural network is used for classification and recognition, and then fault diagnosis is achieved.According to the invention, the label-free data can be utilized through the feature extractor and the semi-supervised clustering built through the convolutional network, the time sequence signals ofvarious monitoring quantities in the sample are successfully utilized through combination with the recurrent neural network, loss of time sequence information contained in original data is avoided, and the accuracy of fault diagnosis is effectively improved.

Description

technical field [0001] The invention relates to the technical field of smart grids, and more specifically, to a distribution network fault diagnosis method based on deep feature clustering and LSTM. Background technique [0002] With the development of science and technology, artificial neural network technology is widely used in many fields including the electric power industry. Based on artificial neural network technology and making full use of the rich fault data accumulated in real time in the distribution network, the identification and diagnosis of various faults can be realized, and more specific and detailed information can be provided for the staff to deal with faults, thereby improving fault line inspection, positioning, The efficiency of exclusion is of great significance to improve the reliability of electricity consumption by users. [0003] However, in the process of realizing distribution network with artificial neural network technology, there are the follo...

AI Technical Summary

Problems solved by technology

Insufficient collection and monitoring coverage of distribution network terminals leads to a large number of perception gaps in the distribution network, and the acquired fault information is incomplete, and the fault data of the power grid involves data security issues, and it is difficult to obtain a large amount of fault data

In addition, the current relevant researches all start from the waveform characteristics of the fault process. For example, Chinese patent CN110045227A, with a publication date of 2019.07.23, discloses a distribution network fault diagnosis method based on random matrix and deep learning. Through the fault diagnosis model, the Effective fault diagnosis information is obtained from the real-time data of the distribution network. However, manually extracting features to classify the faults is difficult, time-consuming and cumbersome, resulting in a small number of label fault data samples and a lack of typical data for the identification of each fault type. Unlabeled data is also underutilized

[0005] (2) Unbalanced fault data samples

The data is unbalanced, making it difficult to train adequately

[0006] (3) Timing data is not effectively utilized

Power grid fault characteristics are often closely related to time series information, but the traditional neural network separates the time correlation of sequence data, loses a lot of important information, and is difficult to achieve time-related classification and prediction

Images