摘要
针对精密特种电网设备的运输监测问题,开展了面向电网设备供应链的多源数据融合感知算法研究。构建了电网设备供应链的多源数据融合感知系统架构,提出了基于深度置信网络(DBN)-Dempster Shafer(DS)证据理论的电网设备运输状态感知算法。同时利用该算法分别对来源于加速度、倾角及温湿度等多传感器的数据进行特征提取,通过DS证据理论对不同特征信息进行融合分析,以得到电网设备运输状态融合感知结果。算例分析结果表明,所提算法结合深层神经网络与决策层数据融合,能够深入挖掘多源数据之间的内在关联,且在电网设备运输状态感知上具有更高的准确性。在实际电网设备供应链监控中,其可准确监测电网设备的倾斜状态,并提高电网设备供应链的运输质量与安全水平。
Aiming at the transportation monitoring of precision special power grid equipment,this paper studies the multi⁃source data fusion sensing algorithm for power grid equipment supply chain.The multi⁃source data fusion sensing system architecture of power grid equipment supply chain is constructed,and a power grid equipment transportation state sensing algorithm based on Deep Belief Network(DBN)⁃Dempster Shafer(DS)evidence theory is proposed.The DBN algorithm is used to extract the features of the data from multiple sensors such as acceleration,inclination,temperature and humidity,and the fusion analysis of different feature information is carried out through DS evidence theory to obtain the fusion perception results of the transportation state of power grid equipment.The results of example analysis show that the proposed algorithm combined with deep neural network and decision⁃making level data fusion can deeply mine the internal relationship between multi⁃source data,and has higher accuracy in the perception of power grid equipment transportation state.At the same time,in the actual power grid equipment supply chain monitoring,it can accurately monitor the tilt state of power grid equipment and improve the transportation quality and safety level of power grid equipment supply chain.
作者
李岩
LI Yan(Department of Materials,State Grid Beijing Electric Power Company,Beijing 100031,China)
出处
《电子设计工程》
2023年第12期169-173,共5页
Electronic Design Engineering
基金
国网公司科技项目(JL71-15-042)。
关键词
电网设备
多源数据融合
深度学习
DS证据理论
power grid equipment
multi⁃source data fusion
deep learning
DS evidence theory
