With the advent of Industry 4.0 era, intelligent manufacturing is becoming increasingly important. In this context, industrial PON (passive optical network) transmission technology, as an innovative communication solution, is gradually becoming an important force to promote the development of intelligent manufacturing.

This afternoon, the “Industrial PON Technology Seminar” co-sponsored by CIOE, C114 and the Research Institute of China Telecom (601728) Co., Ltd. was successfully held as the fifth online forum of the 2024 China High-quality Development Forum. In the keynote report, Wu Zhiwei, associate professor of Northeastern University, analyzed the shortcomings of the existing industrial control network and introduced the industrial PON control network solution of “one network to the end”, that is, the industrial PON is used as the industrial control network connection architecture, the bottom control network is integrated downward, and the intelligent control is enabled upward.

Industrial control network requirements and evolution

1771-OX Wu Zhiwei introduced that the typical industrial automation system architecture is a pyramid structure, which mainly includes the upper layer of enterprise resource planning ERP, the middle layer of manufacturing execution system MES, and the lower layer of workshop level monitoring, basic loop control and production process composed of PCS.

The traditional industrial control network mainly adopts a “two-layer and three-level” network structure, among which, “two-layer” refers to the IT network and OT network, and “three-level” refers to the factory level, workshop level and field level.

Wu Zhiwei pointed out that the current industrial automation system has two major problems: first, it is highly dependent on human experience and knowledge, and can not realize the optimization decision of the whole production process; Second, the PLC/DCS control system cannot achieve industrial intelligent algorithms driven by industrial big data.

The industry demand of industrial control network, on the one hand, is the industrial control network system and solution for intelligent systems, supporting the transformation of the traditional automatic control system with PLC/DCS as the core to intelligent control system, on the other hand, is to realize the evolution of the control network from “two layers and three levels” to “one network to the end”.

Facing the challenges of how the industrial control network can realize the interconnection of industrial big data under multi-protocols, how the industrial control network can support the in-depth mining of data value, and how the industrial control network can realize the closed-loop intelligent control driven by heterogeneous industrial big data, Wu Zhiwei proposed the “one network to the end” industrial PON control network solution.

Industrial control network solution based on PON

Wu Zhiwei introduced that the overall architecture of the new industrial control network solution based on industrial PON is divided into three aspects: end, edge and cloud. Among them, the end side of the production process control and working condition perception; Side is the data transmission base, based on industrial PON to build the network fusion system; Cloud side for industrial big data convergence and intelligent decision control.

Wu Zhiwei further introduced that the design idea of the new industrial control network based on industrial PON is “one network to the end”, the first is to realize the interconnection of industrial big data and internalize PON into the industrial control network connection architecture; Secondly, the deterministic transmission capability of the network is enhanced, and PON downintegrates the underlying control network. In addition, the integration of network and control is realized, and the intelligent control of PON is enabled upward. The design idea of control function is to realize intelligent perception of operating conditions, prediction and backtracking of production factors, and high-performance control. And the intelligent optimization decision and intelligent autonomous control of human-computer interaction and cooperation.

The key technologies include the network transmission technology and network connection architecture for closed-loop control, and the adaptability of industrial control scenarios. For example, network technologies with deterministic data transmission capabilities guarantee intelligence within a sampling period determined by the designer

1771-OX Data acquisition, transmission and processing involved in modeling and intelligent optimal decision algorithms; PON is a bearer technology for industrial Ethernet and fieldbus, and is compatible with existing industrial networks. Industrial network system unified connection architecture model based on the existing industrial PON architecture; The protocol stack, function module, information base, program and interface calling method of the unified access network of fieldbus, industrial Ethernet, industrial wireless network and other types of network protocols.

Wu Zhiwei based on industrial PON electric fused magnesia production process as an example, depending on the solution, the single ton energy consumption of electric fused magnesia products reduced by 4%, the output increased by 5%, improving the market competitiveness of enterprises. In the future, as technology continues to mature and innovate, it is expected that industrial PON will be more powerful and better able to serve the field of intelligent manufacturing, driving industrial automation and digital transformation.