摘要
在传统的工业现场级网络中,存在大量具有不同时延需求的业务,如何满足不同业务的时延需求存在挑战。针对在工业场景中现场级网络应如何保障系统反馈控制实时性的问题,提出了一种基于模糊推理模型的优先级适配机制。该机制通过动态调整网络中数据流的优先级,以满足各类异构业务的传输需求,保证系统反馈控制的实时性;同时,为了使其能够与时间敏感网络的优先级结合,设计了离散量化输出模型,为实现确定性网络调度提供依据。最后通过仿真,验证了该机制的实时性与有效性。
In traditional industrial field-level networks,there are numerous services with diverse latency requirements,posing challenges to meet the delay requirements of different services.To address the issue of ensuring real-time feedback control in industrial scenarios,this paper proposes a priority adaptation mechanism based on fuzzy inference modeling.This mechanism dynamically adjusts the priority of data flows in the network to meet the transmission needs of various heterogeneous services,ensuring the real-time performance of system feedback control.Meanwhile,in order to integrate with the priority of time-sensitive networks,a discrete quantization output model is designed to facilitate deterministic network scheduling.Through simulation,the real-time capability and effectiveness of the proposed mechanism are validated.
作者
张子腾
王文烨
郑卓琳
袁亚洲
ZHANG Ziteng;WANG Wenye;ZHENG Zhuolin;YUAN Yazhou(School of Electrical Engineering,Yanshan University,Qinhuangdao 066004,China)
出处
《移动通信》
2023年第8期39-45,共7页
Mobile Communications
基金
国家重点研发计划“工业现场无线控制系统架构和设计方法”(2020YFB1708700)
国家自然科学基金项目“分布式工业网络中适配业务传输需求的多域资源协同优化研究”(62273295)
中央引导地方项目“业务需求驱动的多层级工业网络可靠接入与信息传输协同优化研究”(226Z0304G)。
关键词
优先级适配机制
模糊神经网络
动态优先级
时间敏感网络
priority adaptation mechanism
fuzzy neural network
dynamic priority
time-sensitive networks
