摘要
工业数字孪生中物理设备与孪生体间精准建模需要同步海量感知数据,导致虚实映射实时性难以保障。针对该问题,提出一种适用于数字孪生的4层双向闭环架构,包含设备层-联邦层-云层-应用层。基于该架构,采用一种基于直方图增益的分布式联邦桶化决策树算法实现全局聚合。此外,设计自适应重分配梯度权重的剪枝算法加速模型收敛。多组实验结果表明,联邦聚合模型相较于基线方法准确率提升10%,11%,22%;收敛性提高15%,20%,27%;并且在最差信道环境下的额外通信开销可控制在300 mJ以内。
Accurate modeling between physical devices and digital twins in industrial digital twins requires synchronizing massive amounts of sensing data.This results in difficulty ensuring real-time virtual-physical mapping.To address this issue,this paper proposes a four-layer bidirectional closed-loop architecture suitable for digital twins,consisting of a device,federated,cloud,and application layers.We develop a distributed federated bucketized decision tree algorithm utilizing histogram-based gain for global aggregation based on this architecture.Additionally,we design a pruning algorithm with adaptive gradient-weight redistribution to accelerate model convergence.Experimental results on the dataset demonstrate that the proposed federated aggregation model achieves accuracy improvements of 10%,11%,and 22%compared to baseline methods.It also achieves convergence improvements of 15%,20%,and 27%.Moreover,the proposed model maintains additional communication overhead within 300 mJ,even under the worst channel conditions.
作者
孙士保
赵一凡
赵鹏程
刘剑锋
李新
SUN Shibao;ZHAO Yifan;ZHAO Pengcheng;LIU Jianfeng;LI Xin(Longmen Laboratory,Luoyang 471000,China;College of Information Engineering,Henan University of Science and Technology,Luoyang 471023,China;School of Software,Henan University of Science and Technology,Luoyang 471023,China;School of Vehicle and Mobility,Tsinghua University,Beijing 100084,China)
出处
《河南科技大学学报(自然科学版)》
北大核心
2025年第4期53-64,M0005,M0006,共14页
Journal of Henan University of Science And Technology(Natural Science)
基金
国家自然科学基金项目(51474095)
龙门实验室前沿探索项目(LMQYTSKT034)
河南省重点研发与推广专项(科技)项目(252102210158)
河南省高等学校重点科研项目(24B520010)。
关键词
工业物联网
数字孪生
联邦学习
桶化决策树
industrial internet of things
digital twin
federated learning
bucketized decision tree
作者简介
孙士保(1970-),男,河南固始人,教授,硕士生导师,主要从事工业物联网、机器学习方面的研究;通信作者:赵鹏程(1993-),男,河南平顶山人,讲师,主要从事工业物联网、多模态数据融合方面的研究.
