基于双向LSTM的电网调度日志分类

详细信息    查看全文 | 推荐本文 |

英文篇名：CLASSIFICATION OF DISPATCHING LOG OF POWER GRID BASED ON BIDIRECTIONAL LSTM 作者：徐家慧 ; 张明 ; 白静洁 ; 何慧 ; 赵扬 ; 白盛楠 英文作者：Xu Jiahui;Zhang Ming;Bai Jingjie;He Hui;Zhao Yang;Bai Shengnan;Nari Group Corporation/State Grid Electric Power Research Institute;BeiJing Ke Dong Electric Power Control System Co.,Ltd.;State Grid Jiangsu Electric Power Company;School of Control and Computer Engineer,North China Electric Power University; 关键词：电网调度日志 ; Skip-gram ; 词向量 ; 循环神经网络 ; 双向LSTM 英文关键词：Power grid dispatching log;;Skip-gram;;Word embedding;;Recurrent neural network;;Bidirectional LSTM 中文刊名：JYRJ 英文刊名：Computer Applications and Software 机构：南瑞集团(国网电力科学研究院)有限公司;北京科东电力控制系统有限责任公司;国网江苏省电力有限公司;华北电力大学控制与计算机工程学院; 出版日期：2019-01-12 出版单位：计算机应用与软件 年：2019 期：v.36 基金：国网江苏省电力有限公司科技项目—面向大电网智能监视与立体控制的调控关键技术研究与应用(J2018065) 语种：中文; 页：JYRJ201901027 页数：5 CN：01 ISSN：31-1260/TP 分类号：148-152

摘要

电网调度日志记录电网运行的各类信息,是分析调度过程、电网运行情况的重要数据来源。电网调度日志管理逐步智能化,调度日志分类任务也由人工操作转变为系统自动分类。为实现智能化分类,提出一种基于深度神经网络的电网调度日志分类方法。该方法基于电网调度日志训练出词向量,将词向量作为LSTM(Long Short-Term Memory)模型的输入。使用双向LSTM对电网调度日志进行分类。实验结果表明,该方法可以有效地对长度差别巨大的日志进行分类,并获得比传统分类方法更优的性能。
        Power grid dispatching log records all kinds of information of power grid operation,and is an important data source for analyzing dispatching process and power grid operation. The management of dispatching log is becoming more and more intelligent,and the task of classification is transformed from manual operation to automatic system classification. In order to realize intelligent classification,this paper presented a classification method of power grid dispatching log based on deep neural network. The method trained word embedding based on power grid dispatching log,then took word embedding as input of LSTM( Long Short-Term Memory) model,and used bidirectional LSTM to classify power grid dispatching logs. The experimental results show that this method can effectively classify logs with huge differences in length,and can achieve better performance than traditional classification methods.

引文

[1]邱剑,王慧芳,应高亮,等.文本信息挖掘技术及其在断路器全寿命状态评价中的应用[J].电力系统自动化,2016,40(6):107-112,118.
    [2] Wang S,Christopher D. Manning. baselines and bigrams:simple good sentiment and topic classification[C]//Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics:Short Papers Volume 2,2012.
    [3]张玉芳,彭时名,吕佳.基于文本分类TFIDF方法的改进与应用[J].计算机工程,2006,32(19):76-78.
    [4] Hinton G E,Salakhutdinov R R. Reducing the Dimensionality of Data with Neural Networks[J]. Science,2006,313(5786):504-507.
    [5] Turian J P,Ratinov L A,Bengio Y. Word Representations:A Simple and General Method for Semi-Supervised Learning[C]//Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics. Association for Computational Linguistics,2010:384-394.
    [6] Joachims T. Text categorization with Support Vector Machines:Learning with many relevant features[C]//European Conference on Machine Learning. Springer,Berlin,Heidelberg,1998.
    [7] Yang Y. An Evaluation of Statistical Approaches to Text Categorization[J]. Proc Amia Annu Fall Symp,1999,1(1-2):358-362.
    [8] Tang D Y,Qin B,Liu T. Document modeling with gated recurrent neural network for sentiment classification[C]//EMNLP,2015.
    [9] Zhang X,Lecun Y. Text Understanding from Scratch[EB].ar Xiv:1502. 01710,2015.
    [10] Kim Y. Convolutional neural networks for sentence classification[EB]. ar Xiv preprint ar Xiv:1408. 5882,2014.
    [11] Bengio Y,Ducharme R,Vincent P,et al. A neural probabilistic language model[J]. Journal of machine learning research,2003,3(3):1137-1155.
    [12] Collobert R,Weston J,Bottou L,et al. Natural Language Processing(almost)from Scratch[J]. Journal of Machine Learning Research,2011,12(2):2493-2537.
    [13] Mikolov T,Chen K,Corrado G,et al. Efficient estimation of word representations in vector space[EB]. ar Xiv preprint ar Xiv:1301. 3781,2013.
    [14] Hochreiter S,Jürgen Schmidhuber. Long Short-term Memory[J]. Neural Computation,1997,9(8):1735-1780.