Explore the future development trend of industrial automation in China
With the development of control technology, computers, communications, networks and other technologies, computer-controlled information exchange is rapidly covering all levels from factory field equipment to production and management. Industrial automation is generally reflected in the general term for automation technology (including automatic measuring instruments and control devices) for measuring and controlling industrial production processes and their electromechanical equipment and process equipment.
Today, the simple understanding of automation has also changed to: using computer technology to develop robotic equipment to partially replace or completely replace or surpass human physical strength. The author of this article will lead you to understand the current status and development direction of my country's industrial technology, as well as the key technologies: time-sensitive networking (TSN), real-time M2M, and 5G.
The current status of industrial automation in China
a. Changes from traditional manufacturing to intelligent manufacturing
We can express the components of smart manufacturing with the following formula:
Smart manufacturing = automation + informatization + intelligence.
Smart manufacturing is to produce personalized products in batches. It has the following characteristics:
-
Automation (Hardware): Automation hardware equipment constitutes the intelligent factory;
-
Informatization (software): Use big data statistics and artificial intelligence to filter useful information for intelligent production;
-
Intelligence (Internet): Application of 5G, artificial intelligence and cloud computing in manufacturing big data analysis.
In short, traditional manufacturing uses machines to replace human physical labor, while intelligent manufacturing combines the Internet and cloud computing technologies to achieve output that humans cannot achieve or tasks that cannot be completed.
b. Transformation from traditional manufacturing to intelligent manufacturing
my country's industrial technology is facing a transformation from traditional to high-tech technology, which involves a series of detailed changes.
-
Human resources: Traditional workers may not be able to quickly adapt to the use of high-tech equipment. Factories need to speed up the recruitment of new employees and provide training for old employees.
-
Introduction of supporting equipment: If the equipment cannot keep up with the pace of the times, it will be impossible to completely transform the industry. For example, the introduction of 5G and cloud computing technologies is inseparable from the construction of cloud platforms, the application of IoT sensors, and the deployment of databases and high-performance computers.
-
Security: In the Internet age, data encryption becomes extremely important, and maintaining information security also requires investment in manpower and funds.
Main technical pain points
a
Wired to wireless protocol conversion
Connectivity is an essential element for data exchange between the factory floor, edge nodes, and the cloud. The advantages of wireless network link protocols are reflected in greater flexibility, the ability to relocate and reorganize equipment in the production area, the opportunity to access information from previously unconnected devices, and the ability to expand production capacity by introducing new equipment without complex and expensive rewiring. Wireless technology can fully leverage the advantages of current use cases of multiple wireless standards and protocols, such as ultra-small sensors, large distributed wireless sensor networks, and factory and process automation.
b
Reliable transmission/anti-interference capability
There is a huge demand for industrial-grade network transmission equipment in various industries, including transportation, electric power, aviation fuel, petrochemicals, metallurgy, coal, water treatment, etc. Reliability and anti-interference capabilities are reflected in the accuracy and efficiency of network transmission.
c
From Data to Intelligence – Data Collection
The main technical difficulties include the fact that massive amounts of data cannot be used for analysis if stored directly, that there is no unified standard for industrial data protocols, that the bandwidth required for video transmission is huge, that data collection from the original system is difficult, and that security considerations are insufficient.
Main technical changes and features
a. From wired to wireless.
b. From analog to digital.
c. Application of AI and edge computing.
Point a has been mentioned above. Here we will explain point b in detail, and in the following part we will talk about edge computing at point c.
Internet data mainly comes from network devices such as servers and Internet users, including large amounts of text data, social data, and multimedia data; industrial data mainly comes from machine equipment data, industrial information data, and industrial chain-related data.
Industrial data mainly includes the following:
-
Key-value industrial data. Only machine data that has been deeply granularized can analyze massive amounts of industrial data. Therefore, the characteristics of this part of data are that each piece of data has very little content, but the key value frequency is extremely high.
-
Document data. Including engineering drawings, simulation data, etc.
-
Information data: Data generated by industrial information systems are generally stored in databases, and this part of data is very easy to collect.
-
Interface data: Interface type data provided by the built industrial automation or information system, including txt format, JSON format, XML format, etc.
-
Video data: There are a large number of video surveillance devices at industrial sites, which will generate a large amount of video data.
-
Image data. This includes images taken by various imaging devices at industrial sites (for example, image data taken by inspectors using handheld devices).
-
Audio data. Includes voice and sound information (e.g., operator conversations, volume of equipment operation, etc.).
Future development trends and prospects of automation technology
As mentioned above, industrial data is mainly composed of the following categories: massive key-value industrial data, document data, information data, interface data, video data, image data and audio data). Smart industry mainly analyzes these types of data. Therefore, we must use artificial intelligence and other technologies more flexibly and intelligently to be able to deal with these complex data types proficiently and contribute to industrial production. The details are as follows:
a. Take advantage of wireless sensor networks
Wireless Sensor Networks (WSN) is a distributed sensor network with the advantages of low cost, low power consumption and multi-function. Sensor networks have a wide range of application prospects, covering many fields such as medical, military and home. For example, the rapid deployment, high availability, scalability and fault tolerance of sensor networks make them more powerful than traditional networks in military command, control, communication, computing, intelligence, monitoring and other application fields.
Compared with traditional networks and other types of sensors, the characteristics of wireless sensor networks include flexible construction methods, unstable network topology that can be increased or decreased at any time according to needs, and reliable network strength.
b. Data analysis based on artificial intelligence technology
With the development of smart industry, the next generation of smart factories will apply advanced robotics and machine learning technologies to software services and industrial IoT to improve production capacity and efficiency. Edge computing and artificial intelligence technologies that use local sensors to control and manage outputs can significantly improve efficiency and reduce errors.
c. Time-Sensitive Networking (TSN)
“TSN is a new generation network standard based on Ethernet, which has functions such as time synchronization and delay guarantee to ensure real-time performance.”
Industrial Internet of Things is one of the most widely used applications of TSN in the future. All industrial fields that require real-time monitoring or real-time feedback need TSN networks. For example: robotics industry, deep-sea oil and gas exploration and development, and banking industry, etc. TSN can not only solve the uncertainty problem of traditional Ethernet, but also solve the complexity problem of industrial buses.
d. Real-time M2M
M2M literally means "machine to machine", which refers to a mode or system in which machines and devices communicate directly through the network and complete tasks on their own without human intervention. Such a mode or system provides services for the Internet of Everything.
These computing-enabled devices and machines can capture data about the world around them and share it with other connected devices, creating an intelligent network of “things.” This means machines can communicate and share information without the need for human intervention. Some time-consuming or boring industrial manufacturing processes can be automated, allowing people to perform other more useful or interesting activities.
e. 5G
It is the fifth generation of mobile networks. It is the new global wireless standard after 1G, 2G, 3G and 4G networks. It uses a brand new wireless infrastructure, with peak download speeds more than 10 times faster than 4G and is expected to eliminate almost any delay. 5G technology will start the Internet of Things (IoT) era, connecting billions of machines, devices and sensors at a low cost. At present, due to the limitations of deployment conditions for industrial equipment, a large amount of computing resources are mostly configured in the cloud. In contrast, the development of terminal equipment technology is relatively slow, resulting in a large amount of idle cloud resources and terminal resources being overwhelmed. This requires that the data transmission pipeline from the terminal to the cloud must be wide enough to respond quickly to demand. The characteristics of 5G's ultra-large bandwidth, ultra-low latency, ultra-high reliability, and ultra-dense connection just eliminate this problem and accelerate the development and application of intelligent manufacturing.
f. Application and technology promotion of edge computing in automation systems
If cloud computing is the centralized processing of big data in the "cloud", then edge computing can be understood as big data processing located at the edge, close to the terminal (such as a mobile phone).
Edge computing has several distinct advantages:
-
Low bandwidth dependence: Edge devices process some temporary data and no longer need to upload all data to the cloud. Only valuable data needs to be transmitted, which greatly reduces the pressure on network bandwidth and the demand for computing and storage resources.
-
Low latency: Processing data close to the data source can greatly reduce system latency and improve service response time.
-
Economical: If an application uses a cloud platform, it can not only technically solve the bandwidth and latency problems, but edge computing may also be more cost-effective.
-
Reliability: The network connection of the cloud platform and the cloud is not always reliable, but the application may need to operate all the time. For example, if the network connection of the face recognition door lock is disconnected, the door lock can still work normally with edge computing.
-
Privacy: When using the cloud, many applications require local processing for privacy reasons. Edge computing provides infrastructure for the storage and use of critical privacy data, improving data security.
g. Troubleshooting
Fault diagnosis combines AI technology with real-time data analysis of edge computing technology. When IoT sensors trigger alarms, they respond to machine failures in real time and quickly, thus avoiding disasters to a large extent. Edge systems can respond to inputs within milliseconds, either making adjustments to fix the problem or shutting down the production line to prevent serious problems. AI technology can analyze the condition of the machine in combination with real-time data, and can also achieve the effect of avoiding failures to some extent.
in conclusion
This article introduces the current status of industrial automation in China, and analyzes the current status and development trends of industrial automation in China from four aspects: the transformation from traditional manufacturing to intelligent manufacturing, the main technical pain points, changes and characteristics, and the future development trends and prospects of automation technology.
After a century of development, China's industry has a solid foundation. With the development of high technology in recent years, industry and manufacturing have followed closely and accelerated transformation. Combined with the development of artificial intelligence, it has effectively improved work efficiency, improved production quality, reduced workers' physical efforts, and achieved precision, efficiency, intelligence, environmental protection and convenience in the production process. However, industrial intelligence does not mean unmanned industry. To a certain extent, workers will also benefit from intelligent industry.
In the future, my country's industrial automation technology will continue to combine multiple technical fields and move towards a more intelligent, networked and convenient tomorrow.
This article is an exclusive original article. Please indicate the source if you reprint it. We reserve the right to pursue legal liability for unauthorized copying and non-compliant reprinting.