基于CS-ECA-BILSTM与KubeEdge的自适应智慧路灯边缘计算模型  

Adaptive intelligent street light edge computing model based on CS-ECA-BILSTM and KubeEdge

作　　者：郭泽鑫 林培杰 程树英 陈志聪 吴丽君 GUO Zexin;LIN Peijie;CHENG Shuying;CHEN Zhicong;WU Lijun(Institute of Micro-Nano Devices and Solar Cells,College of Physics and Information Engineering,Fuzhou University,Fuzhou,Fujian 350108,China)

机构地区：[1]福州大学物理与信息工程学院,微纳器件与太阳能电池研究所,福建福州350108

出　　处：《福州大学学报（自然科学版）》2024年第6期667-673,共7页Journal of Fuzhou University(Natural Science Edition)

基　　金：福建省科技厅引导性基金资助项目(2022H0008);福建省级科技创新重点资助项目(2022G02011)。

摘　　要：提出一种基于高效通道注意(ECA)网络和双向长短期记忆神经网络(BILSTM)的自适应智慧路灯边缘计算模型.首先,在BILSTM的基础上,融合布谷鸟算法、通道注意力机制,构建CS-ECA-BILSTM能见度预测模型,实现道路能见度预测;其次,为普通路灯控制因子单一的问题引入照度和色温因子,提高控制效率并降低路灯能耗;最后,在边缘端引入云原生理念,使用KubeEdge框架与容器技术部署路灯控制模型到边缘端,从而实现多路灯控制.实验结果表明,所提出CS-ECA-BILSTM模型的性能优于其他对比模型,可有效提高路灯能源利用率,且能实现在边缘端的运行.An adaptive intelligent street light edge computing model based on efficient channel atten-tion network(ECA)and bidirectional long short-term memory neural network(BILSTM)is proposed.Firstly,on the basis of BILSTM,CS-ECA-BILSTM visibility prediction model is constructed by combining Cuckoo Search algorithm and channel attention mechanism to realize road visibility predic-tion.Secondly,illuminance and color temperature factors are introduced to the problem of single control factors of ordinary street lights to improve control efficiency and reduce energy consumption of street lights.Finally,the cloud native concept is introduced at the edge end,and the street light control model is deployed to the edge end using KubeEdge framework and container technology,so as to achieve multi-street light control.The experimental results show that the performance of the CS-ECA-BILSTM model is superior to other comparison models,which can effectively improve the energy efficiency of street lamps,and can realize the operation at the edge.

关 键 词：智慧路灯 边缘计算 双向长短期记忆神经网络 注意力机制 容器技术 

分 类 号：TM923[电气工程—电力电子与电力传动] TP312[自动化与计算机技术—计算机软件与理论]