China’s self-developed intelligent welding robot system for flexible manufacturing in offshore engineering has completed system integration in Tianjin and entered the testing and acceptance phase, according to Offshore Oil Engineering Co., Ltd. (COOEC) under China National Offshore Oil Corporation.
The vision-enabled robotic system features high flexibility, environmental adaptability, and intelligent control, making it suitable for diverse, customized welding tasks in complex environments. Its development involves interdisciplinary technologies spanning mechanical engineering, artificial intelligence, and sensing, with strict requirements in intelligent decision-making, environmental perception, and control systems. At present, only few countries around the world possess this core technology.
As one of the key intelligent manufacturing technology projects under the 14th Five-Year Plan for Scientific and Technological Innovation in the offshore oil and gas engineering sector, the three integrated systems are capable of intelligent welding for pipe node prefabrication, module node assembly, and reinforcement and collapse rings in deepwater jackets. Each system can handle workpieces weighing up to 30 tons, has a designed service life of 20 years, achieves 100 percent domestic development of core software and process libraries, and boasts a hardware localization rate of over 80 percent, filling a domestic technological gap in intelligent welding equipment for offshore engineering.
The system achieves more than 10 domestic technological innovations, such as AI-based weld seam recognition, 3D laser vision-guided intelligent assembly, and multi-layer, multi-pass intelligent welding. It can complete component welding with a one-click start, and is also equipped with intelligent path correction, intelligent root pass welding, and double-sided four-wire intelligent filling functions.
In the next step, the project team will focus on technology incubation and industrial application, carrying out system debugging and acceptance, and accelerating the “R&D–verification–promotion” cycle.
(Executive editor: Yuan Ting)
