Efficient decentralized optimization for edge-enabled smart manufacturing: A federated learning-based framework

Huan Liu

doi:10.1016/j.future.2024.03.043

2024-08-01

PRE

DOI：10.1016/j.future.2024.03.043

原文链接

原文求助

摘要

En 中文

The volume of industrial data of smart manufacturing is growing rapidly. Edge computing has emerged as an advanced technique that provides scalable resources for Industrial Internet of Things (IIoT) devices to analyze industrial data and enhance productivity. However, real-time and privacy-protecting data services are crucial for comprehensive decision-making, particularly utilizing machine learning in edge-enabled smart manufacturing. Therefore, how to optimize efficiently the data processing has become a critical concern for manufacturers seeking to upgrade the overall manufacturing operations. In this paper, we propose a decentralized federated learning-based framework to ensure the processing of industrial data in a low-latency and secure manner. Our solution designs the combination between edge nodes and IIoT devices as a global consensus problem with equilibrium constraints. Moreover, we adopt the distributed Alternating Direction Method of Multipliers (ADMM) algorithm to optimize the proposed data processing model, considering its advantages of decomposability and parallelism. To flexibly handle various types of industrial data, such as low-rank and high-dimensional data, we present two types of inexact ADMM algorithms to provide efficient model training services, respectively. Furthermore, an integrated optimization flow is designed for data processing in edge-enabled smart manufacturing. Finally, using industrial datasets from a thermal power plant for steam prediction case, we show that the presented inexact algorithms can decrease the response time by up to 17.2% and 58% respectively compared to other existing algorithms, respectively, while achieving the comparable level of statistical accuracy.

Keyword:

Edge-enabled smart manufacturing

Decentralized optimization

Industrial prediction

Inexact ADMM algorithm

Machine learning

期刊

Future Generation Computer Systems-The International Journal of eScience

IF:

6.1

论文数:

6.9K

被引数:

2.3W

机构

Yanshan University

学者数:

1.7W

论文数: 1.1W

被引数: 1.3W

引用论文

The invitro activation of cyclic AMP production by either forskolin or isoproterenol in the syrian hamster pineal during the day is not accompanied by an increase in melatonin production

Biochemical and Biophysical Research Communications

IF0

1988-12-01

PREAI

Celsa Santana; Juan M. Guerrero; Russel J. Reiter; Maureen E. Troiani

Three Years into the Pandemic: Results of the Longitudinal German COPSY Study on Youth Mental Health and Health-Related Quality of Life

SSRN Electronic Journal

IF0

2023-01-01

PREAI

Ulrike Ravens-Sieberer; Janine Devine; Ann-Kathrin Napp; Anne Kaman; Lynn Saftig; Martha Gilbert; Franziska Reiss; Constanze Löffler; Anja Simon; Klaus Hurrelmann; Sabine Walper; Robert Schlack; Heike Hölling; Lothar H. Wieler; Michael Erhart

Dynamic scheduling of tasks in cloud manufacturing with multi-agent reinforcement learning 基于多agent强化学习的云制造任务动态调度

JOURNAL OF MANUFACTURING SYSTEMS

IF14.2

2022-10-01

PREAI

Wang, Xiaohan; Zhang, Lin; Liu, Yongkui; Li, Feng; Chen, Zhen; Zhao, Chun; Bai, Tian

Affordable federated edge learning framework via efficient Shapley value estimation 通过高效的Shapley值估计实现经济实惠的联合边缘学习框架

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE

IF6.1

2023-10-01

PREAI

Dong, Liguo; Liu, Zhenmou; Zhang, Kejia; Yassine, Abdulsalam; Hossain, M. Shamim