摘要
配网业扩工程对供电企业开展电力供应工作具有重要意义。针对配电网业扩工程项目流程运转效率低下的问题,提出了一种基于数据增强和数据降维技术的配网业扩工程分类方法。该方法对原始数据进行数据增强,通过深度自编码器降低了数据维度,并进行特征提取及聚类分析。以某供电局配电网业扩工程项目数据为基础进行了仿真,实验结果表明:所用算法的分类准确性优于其他算法,所提方法能够合理分配业扩工程工期时长,实现对配网业扩工程的差异化管理,提高流程运转效率和客户满意度。
Distribution network expansion project is of great significance for power supply enterprises to carry out power supply work.Aiming at the low efficiency of the process operation of distribution network business expansion project,a classification method of distribution network business expansion project based on data enhancement and data dimension reduction technology is proposed.This method enhances the original data,reduces the data dimension through depth self encoder,and performs feature extraction and clustering analysis.Based on the data of a distribution network expansion project of a power supply bureau,the simulation results show that the classification accuracy of the algorithm used in this paper is better than other algorithms.The proposed method can reasonably allocate the duration of the industrial expansion project,realize the differential management of the distribution network industrial expansion project,and improve the process operation efficiency and customer satisfaction.
作者
周鑫
林镜星
谢志炜
张铮
梁濡铎
欧祖宏
ZHOU Xin;LIN Jingxing;XIE Zhiwei;ZHANG Zheng;LIANG Ruduo;OU Zuhong(Guangzhou Power Supply Bureau,Guangdong Power Grid Co.,Ltd.,Guangzhou 510013,China;College of Automation,Guangdong University of Technology,Guangzhou 510006,China)
出处
《中国电力》
CSCD
北大核心
2022年第12期91-97,共7页
Electric Power
基金
国家自然科学基金资助项目(61876040)
中国南方电网有限责任公司科技项目(080008KK52200010)。
关键词
配网业扩工程
生成对抗网络
深度自编码器
鲸鱼优化算法
distribution network expansion project
generative adversarial network
deep auto-encoder
whale optimization algorithm
