Diagnosis method of hydropower alarm events based on data augmentation and deep learning

被引：0

作者：

Sun G. ^{[1
]}

Zhang Y. ^{[1
]}

Tang J. ^{[2
]}

Tang F. ^{[2
]}

Wei Z. ^{[1
]}

Zang H. ^{[1
]}

Yang D. ^{[2
]}

机构：

[1] College of Energy and Electrical Engineering, Hohai University, Nanjing

[2] Yalong River Hydropower Development Co.，Ltd., Chengdu

来源：

Dianli Zidonghua Shebei/Electric Power Automation Equipment | 2023年 / 43卷 / 08期

关键词：

attention mechanism; deep learning; hydropower station alarm events; prior knowledge; text data augmentation;

D O I：

10.16081/j.epae.202302001

中图分类号：

学科分类号：

摘要：

Aiming at the shortcomings of traditional diagnosis methods of hydropower alarm events，such as low efficiency and low accuracy，a data augmentation method combining prior knowledge and a hierarchical attention deep learning framework based on bidirectional simple recurrent units++（Bi-SRU++） are designed. Aiming at the problem of imperfect hydropower alarm rules，the latent Dirichlet allocation-enhanced sequential inference model （LDA-ESIM） is used to construct the mapping mechanism between warning signals and warning features. Then，combined with the prior knowledge of hydropower alarm rules，an improved LDA method is proposed to augment the sample data. The hierarchical attention model learns the sample features and outputs the diagnosis results. The test example is actual alarm data of a hydropower centralized control center. The test results show that the proposed method can realize rapid diagnosis of hydropower alarm events with high accuracy in low resource training environment. © 2023 Electric Power Automation Equipment Press. All rights reserved.

引用

页码：88 / 95

页数：7

共 20 条

[1]

SHAN Xin, DAI Zemei, ZHANG Zhe, Et al., Research on and application of integrated smart alarm based on smart grid dispatching and control systems［J］, Automation of Electric Power Systems, 39, 1, pp. 65-72, (2015)

[2]

WANG Cuiyang, JIANG Quanyuan, TANG Yajie, Et al., Fault diagnosis of power dispatching based on alarm signal text mining［J］, Electric Power Automation Equipment, 39, 4, pp. 126-132, (2019)

[3]

LIU Ziquan, WANG Huifang, CAO Jing, Et al., A classification model of power equipment defect texts based on convolutional neural network［J］, Power System Technology, 42, 2, pp. 644-651, (2018)

[4]

FENG Bin, ZHANG Youwen, TANG Xin, Et al., Power equipment defect record text mining based on BiLSTM-Attention neural network［J］, Proceedings of the CSEE, 40, pp. 1-10, (2020)

[5]

Ziyu BAI, Guoqiang SUN, Haixiang ZANG, Et al., Identification technology of grid monitoring alarm event based on natural language processing and deep learning in China［J］, Energies, 12, 17, pp. 1-19, (2019)

[6]

SUN Guoqiang, SHEN Peifeng, ZHAO Yang, Et al., Intelligent recognition of power grid monitoring alarm event combining knowledge base and deep learning［J］, Electric Power Automation Equipment, 40, 4, pp. 40-47, (2020)

[7]

WEI Zhinong, SHI Dongming, ZHANG Ming, Et al., Intelligent identification method of power grid fault events considering sample classification imbalance［J］, Electric Power Automation Equipment, 41, 11, pp. 93-99, (2021)

[8]

YAN Peng, HUANG Xiaoxu, HUANG Yuhui, Et al., Online alarm recognition of power grid dispatching based on BERT-DSA-CNN and a knowledge base［J］, Power System Protection and Control, 50, 4, pp. 129-136, (2022)

[9]

Lantao YU, Weinan ZHANG, Jun WANG, Et al., SeqGAN：sequence generative adversarial nets with policy gradient, Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence, pp. 2852-2858, (2017)

[10]

Kai ZOU, EDA：easy data augmentation techniques for boosting performance on text classification tasks, 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing（EMNLP-IJCNLP）, pp. 6382-6388, (2019)

← 1 2 →