Wuhan University

China University of Geosciences (wuhan)

State Grid Tianjin Electric Power Company

Zihao Li

Yigang He

Bin Hu

Shiqian Ma

Xiaoyu Liu

Yitong Hao

Changyang Liu

Zilong Zhang

Jianfeng Wang

Zhongxu Wen

Shiyu Yuan

Qingwu Gong

Zishuo Zhao

Yunfeng Ge

A method for diagnosing transformer fault based on a deep coupled dense convolutional neural network, includes: obtaining datasets of dissolved gas in oil of a transformer in normal and fault states; expanding the datasets by using an adaptive synthetic oversampling method; performing, in a form of a two-dimensional matrix, feature reconstruction on characteristic gas dissolved in the oil; building a transformer fault diagnosis model based on a deep coupled dense convolutional neural network; and dividing an expanded dataset into a training set and a test set, and taking the two-dimensional matrix as an input of the deep coupled dense convolutional neural network and a set label as an output to train the network to obtain a fault diagnosis model. The present disclosure can resolve a problem that a fault diagnosis accuracy rate of the transformer is low due to insufficient and unbalanced fault samples in the dissolved gas in the oil.

Method for diagnosing transformer fault based on deep coupled dense convolutional neural network

A method for determining a reasonable sampling interval with three-dimensional (3D) laser scanning includes: collecting rock joints; acquiring point cloud data of the rock joints with 3D laser scanning; performing preprocessing on acquired point cloud data; conducting indoor mechanical tests to determine rock mechanics parameters; calculating, with preprocessed point cloud data of the rock joints, joint roughness coefficients (JRCs) under different intervals obtained from statistical roughness parameters; fitting different relational curves according to a scatter plot between the different intervals and the JRCs to obtain a fitting equation; reversely calculating a JRC of the rock joint sample with the rock mechanics parameters to obtain a reversely calculated JRC; and substituting, according to a relational equation between the different intervals and the JRCs, the reversely calculated JRC into the fitting equation to determine a reasonable sampling interval for roughness evaluation.

Method and device for determining reasonable sampling interval with three-dimensional (3D) laser scanning, and storage medium

Growing community of inventors

Wuhan, China

Zihao Li