面向边缘端工业微服务部署的多自标白鲨优化算法

打开文本图片集

中图分类号：TP277 文献标志码：A

DOI： 10.7652/xjtuxb202508001 文章编号：0253-987X（2025）08-0001-10

Multi-Objective White Shark Optimizer for Edge-Side Industrial Microservice Deployment

LI Xiaobin，GOU Kunyao，YIN Chao （Green Intelligent Manufacturing Research Institute，Chongqing University， Chongqing 40o044，China）

Abstract： To address the challenge of balancing resource utilization and load balancing in edge nodes with limited resources during industrial microservice deployment，an improved multi-objective white shark optimizer （IMOWSO） is proposed. Considering the co-optimization of resource utilization and load conditions in edge nodes during industrial microservice deployment，a multi-objective optimization model is established with four key metrics： computational resource margin of edge nodes，service communication energy consumption， load balancing status，and storage resource margin. To solve this model，the improved algorithm enhances the original multi-objective white shark optimizer （MOWSO） in four aspects： initial population quality， convergence speed，ability to escape local optima， and preservation of high-quality solutions. Specifically，chaotic mapping initialization，adaptive weight factors， differential evolution operators， and an elite retention strategy are introduced. Experimental results demonstrate that compared to the original MOWSO， the proposed IMOWSO achieves optimizations of 13.8% ， 37.1% ， 63.9% ，and 47.4% （204号 in the four metrics，respectively. Furthermore，when compared to the second-generation genetic algorithm，the improvements reach 63.2% ， 53.2% ， 39.1% ，and 63.6% ，respectively，while also exhibiting faster convergence speed.

Keywords： industrial microservices deployment; edge nodes； multi-objective optimization; white shark optimization algorithm

在工业互联网环境下，伴随着工业设备的大量接入以及高并发的用户需求，微服务架构作为一种轻量级、模块化的软件架构风格，因其灵活性、可扩展性以及高可用性的特点[1-3]，越来越受到制造企业的青睐。（剩余14181字）