The global manufacturing industry is on the eve of a new productivity leap. Guangdong, China’s major manufacturing province, has keenly captured the core of this – when artificial intelligence 2.0 is deeply integrated with advanced manufacturing technology, the manufacturing technology system, production model and industrial form will be reconstructed at an accelerated pace.
This is not a simple technological superposition, but a paradigm upgrade related to the fate of the industry.
“The breakthrough and widespread application of new mechatronics (new generation of intelligent manufacturing) will drive the fourth industrial revolution to its climax.” On June 16, Zhou Ji, an academician of the Chinese Academy of Engineering, said at the “Guangdong Forum on Accelerating the Development of New Mechatronics Industry”.
The value of this symposium is not just a collision of opinions, but more importantly, a key signal released: Guangdong’s industrial upgrading is shifting from “single point breakthrough” to “system integration”, and AI has become the core variable in reshaping the order of the mechatronics industry.
Guangdong is aiming at the new track of “new mechatronics” and trying to seize a new position in the “fourth industrial revolution”.
AI redefines the boundaries of capabilities
AI is redefining the capabilities of mechatronics.
In the past, this industry pursued the precise coordination of machinery and electricity; now, equipment has the ability to sense, make decisions, execute and learn, and the industry is moving from “preset execution” to “autonomous decision-making and intelligent collaboration.”
In the most basic welding link of industrial manufacturing, this change has already occurred.
Headquartered in Guangzhou, Shugen Internet has its self-developed welding robot doing “big work” on the crane production line: the robot can independently search for welds and complete the welding tasks of more than 1,200 load-bearing arms. The pre-welding preparation time has been reduced by 80%, and the flaw detection pass rate has increased to over 98%, directly addressing a long-standing problem in the industry.
Liang Zaizhong, chairman of the company, said: “This robot can weld more steadily and better than a master who has been working for a lifetime.” This company, which was born out of the Sany Group, is injecting AI into capillaries made in Guangdong.
In the more complex process industries – petrochemical, metallurgy, electric power, pharmaceutical and other industries, the changes brought by AI are equally profound.
Chu Jian, the founder of SUPCON Technology, said that the time series pre-trained large model TPT they developed, which is suitable for the characteristics of the process industry, has been applied on a large scale in many large-scale petrochemical projects. A certain oil refining project optimizes the heat exchange network through TPT, and the annual benefit is expected to reach 40 million yuan.
As an important technical service provider in Guangdong, SUPCON Technology has more than 2,000 customers in the province, including Maoming Petrochemical, Guangdong Petrochemical, and CGN.
Chu Jian’s plan is very clear – during the “15th Five-Year Plan” period, Guangdong will be a top priority and the self-developed TPT large model will be promoted to more than 4,500 sets of process industry devices in the province to save energy and reduce emissions for Guangdong enterprises.
Embodied intelligence is another frontier.
Zhou Jian, chairman of the world’s number one humanoid robot company that emerged from Shenzhen, said: “Guangdong humanoid robots are very good products of mechatronics, and mechatronics is a solid foundation for intelligent manufacturing.” Through the integration of “mechatronics hardware + AI software”, the robot can not only execute with high precision, but also autonomously perceive and make decisions.
Today, UBTECH’s humanoid robots have entered the production lines of BYD, Foxconn, and Dongfeng Liuzhou Automobile, and have accumulated more than 10 million pieces of real industrial scene data.
Tao Jingwen, Vice Chairman of Huawei’s Board of Supervisors, is optimistic about three subdivided scenarios: from passive response to active prediction to achieve early warning of faults; from experience-driven to data-driven, with large industrial vision models widely replacing manual quality inspection; from single-point optimization to global collaboration, using the industrial Internet to break data barriers, realize intelligent scheduling of the entire process of orders, production, and delivery, and significantly enhance flexible production capabilities.
Create a benchmark for “AI + electromechanical” integration
At this symposium, one voice emerged repeatedly: Guangdong has the most complete electromechanical industry chain, the most diverse application scenarios and the densest communication infrastructure in the country, and is fully equipped to pioneer and be at the forefront in the field of “AI + electromechanical” integration.
Guangdong’s advantages are structural.
From the perspective of the industrial chain, Guangdong has a number of advantageous industrial clusters such as intelligent equipment, robots, and new energy vehicles, which are essentially the product of the deep integration of machinery, electronics, networks, and artificial intelligence.
From the perspective of scenarios, Guangdong’s manufacturing industry has a wide range of categories, from automobiles, home appliances to process industries, and from 3C electronics to medical equipment. Each segment provides real, complex, and large-scale verification scenarios for AI empowerment.
In terms of infrastructure, Huawei has built a computing power cluster of over 30,000P in Guangdong, Tencent Cloud’s large model service platform has exceeded 5 trillion tokens per day, and Pengcheng Cloud Brain’s large scientific equipment continues to provide computing power support for the industry.
How to turn advantages into winning momentum and promote the new mechatronics industry to become bigger and stronger?
The first is to create demonstration areas and demonstration clusters.
Gao Wen, academician of the Chinese Academy of Engineering and director of Pengcheng National Laboratory, suggested: “Guangdong can rely on the software and hardware advantages of the Greater Bay Area and use the new Internet as the core base to create a new industrialization demonstration zone; promote deep integration verification and application demonstration of industrial large models, embodied intelligence and electromechanical equipment on real production lines
to provide a replicable and scalable demonstration path for new industrialization.”
In addition, Li Qiang, Vice President of Tencent and President of Government and Enterprise Affairs, proposed: “Select a group of ‘AI intelligent demonstration enterprises’ to create benchmark cases in the fields of advanced manufacturing, government affairs, medical care, transportation, and retail.”
The driving effect of leading enterprises is emerging.
Huawei is working in depth with a number of national laboratories and leading industry companies in Guangdong to create a number of benchmark application scenarios for mechatronics and artificial intelligence, forming a transformation path that can be replicated, promoted, and scaled.
Midea exports AI algorithms, industrial software and solutions to the industry, uses AI to predict demand, optimize inventory and logistics routes, and has empowered more than 3,000 companies to complete intelligent transformation.
Focus on the subdivisions of manufacturing.
Minluo Equipment is a high-end equipment company that grew up in Huangpu District, Guangzhou. It focuses on providing digital intelligent manufacturing solutions for automobile manufacturing and other industries.
The magnetic levitation super production line developed by the company last year can increase the efficiency of automobile body manufacturing by more than 10 times and is expected to reduce unit manufacturing costs by more than 80%.
Minluo Equipment’s revenue has increased tenfold in the past eight years, and it has delivered 1,300 production lines in about 20 countries around the world, with more than 90% serving leading international companies.
Mingluo Equipment Chairman Yao Weibing proposed that the company will be based on the industrial foundation of the Greater Bay Area and build six world-class manufacturing service platforms, including pilot testing and evaluation, green manufacturing and carbon management, and industrial Internet of Things production line asset management, benchmarking against international benchmarks such as McKinsey and Accenture, and cultivating new business entities for the robot automation and manufacturing service industry.
The “second half” of new mechatronics
If the first half of the development of new mechatronics is about technology, what about the second half?
A broad consensus is that it depends on talent.
Yan Yijun, vice president of Xiyujizhi, believes that the difference in AI quality of talents is ultimately reflected in the competitiveness of the real economy.
“General AI is available to everyone, but the popularity of productivity AI is far from keeping up.” He suggested converting the original AI+ education and information construction costs into productivity AI subscription fees to improve the AI capabilities and quality of college students across the province.
Building an “AI native” talent highland has become the consensus of many participants.
Yang Zhongmin, president of South China University of Technology and academician of the Chinese Academy of Engineering, proposed “walking on two legs”: It is necessary to have a global perspective for “introduction” – setting up special talent introduction funds to introduce industry leading talents and industrial technology backbones to the world; it is also necessary to consolidate local “education” – to increase the per-student funding for mechatronics talent training, and use it for multi-disciplinary and multi-scenario construction.
Behind this suggestion is South China University of Technology’s special weight in Guangdong’s manufacturing talent map – the actual controllers of more than 200 listed manufacturing companies in the Greater Bay Area are alumni of Huagong, and nearly two-thirds of the leaders and core technical backbones of high-end manufacturing companies in the Pearl River Delta graduated from Huagong.
The demand for talent also comes from the front lines of the industry.
Tencent, Midea and other companies have expressed their urgent need for comprehensive talents who “understand both mechanical and electrical technology and AI” and expressed their willingness to open up their platforms and increase the training and introduction of AI talents.
Judging from the signals released by the symposium, Guangdong’s new mechatronics industry is standing at a critical turning point. It has the world’s rare dual advantages of mechatronics and digital intelligence. How to transform it into a replicable benchmark, an exportable standard, and a scalable increment will determine the position of Guangdong’s manufacturing in future global competition.
