摘要
为解决电力缺陷描述专业词汇较多分词准确率不佳以及单一神经网络模型自身存在不足的问题,提出了基于注意力机制优化组合神经网络的电力缺陷等级确定方法。该方法使用分布式字粒度向量对电力缺陷描述进行表示,使用由卷积神经网络和双向长短时记忆网络组成的卷积循环神经网络对电力缺陷描述的局部特征和序列特征进行特征提取,采用注意力机制对组合神经网络得到的语义特征进行权重分配,减少关键特征的丢失,进一步增强关键信息对分类结果的影响。以云南电网公司2014年—2019年间11万条缺陷描述数据作为实验对象,文中所提方法的Acc、MF_(1)值和WF_(1)值分别为0.9275、0.9112和0.9275,验证了该方法在电力缺陷等级确定中的有效性和可行性,为电网的智能化运行提供帮助。
In order to solve the problem that the accuracy of word segmentation in power defect descriptions is not good and the single neural network model has its own shortcomings,a determination method of defect grades in electrical equipment based on combination neural network optimized by attention mechanism is proposed in this paper.The distributed character granularity vector is used for representation of power defect descriptions.The local features and sequence features of power defect descriptions are extracted by using the convolutional recurrent neural network which is composed by convolutional neural network and bidirectional long short-term memory network.The attention mechanism is used to assign weights of the semantic features obtained by the combination neural network,so as to reduce the loss of key features and further enhance the influence of key information on the classification results.Taking 110000 defect description data of Yunnan Power Grid Company from 2014 to 2019 as experimental objects,the Acc,MF_(1) and WF_(1) values of the method proposed in this paper are 0.9275,0.9112 and 0.9275,which illustrates that the proposed method is effective and feasible in the determination of the power defect grades,and provides help for intelligent operation of power grid.
作者
程宏伟
高莲
于虹
李鹏
CHENG Hongwei;GAO Lian;YU Hong;LI Peng(School of Information,Yunnan University,Kunming 650500,China;Electric Power Research Institute of Yunnan Power Grid Co.,Ltd.,Kunming 650500,China)
出处
《电测与仪表》
北大核心
2024年第1期83-90,98,共9页
Electrical Measurement & Instrumentation
基金
国家自然科学基金资助项目(61763049)
云南省应用基础研究计划重点项目(2018FA032)。
关键词
卷积循环神经网络
字粒度
注意力机制
电力缺陷描述
状态评价
convolutional recurrent neural network
character granularity
attention mechanism
power defect descriptions
condition assessment
