Ling Yunguang: Machine vision + AI innovation helps 'intelligent' manufacturing of lithium batteries

Recently, the 2023 Machine Vision Technology and Application Summit hosted by GaoGong Robot and GaoGong Robot Industry Research Institute (GGII) was grandly held in Shenzhen. Ling Yunguang delivered a keynote speech on "Innovative Vision Solutions Light up Intelligent Manufacturing" and won the "2023 Machine Vision Industry Chain Top 30" award.

Yao Mingru, Director of Lingyun Optical Solutions, introduced that the needs of intelligent manufacturing in the "TWh" era are intelligent production, intelligent inspection, and intelligent management; the world-class problems in the field of industrial vision are unstable defect detection and "saving" The "people" strategy is not obvious and the value of data is difficult to reflect; the intelligent manufacturing of power batteries in the "TWh" era puts forward new requirements for visual inspection that are more accurate, more reliable, and more intelligent. By using big data and cloud computing technology, future industrial production will be data-based, knowledge-based and intelligent, which will greatly improve production and quality management levels.

The market continues to improve, new demands for intelligent manufacturing in the TWh era

In recent years, driven by the global net-zero carbon emission goal, the road transportation field has accelerated the electrification transformation, and electric vehicle power batteries are moving from GWh manufacturing to TWh era intelligent manufacturing. Under the new industry cycle, power battery companies have put forward higher requirements:

• High-speed production: Taking prismatic batteries as an example, the production speed requirement is moving towards 40PPM;

• Control costs: production costs should be reduced by 42%;

• Lean management: production efficiency should be increased by 50%;

• Quality requirements: The defect rate is reduced from ppm level to ppb level.

These requirements are the core of intelligent manufacturing of power batteries. The industry is developing rapidly, calling for the acceleration of intelligent manufacturing empowerment.

Ling Yunguang: Machine vision helps intelligent manufacturing of lithium batteries

"In 2000, Ling Yunguang cooperated with the Shanghai Banknote Printing Factory to develop a system that can conduct online quality inspection of large-format RMB gravure printing, thus starting the road to machine vision research and development." In the course of more than 20 years of development, Ling Yunguang has continuously increased investment in research and development, gradually accumulated technical capabilities such as algorithm platforms and GMQM quality management systems, and has won one first prize for national technological invention and two national scientific and technological progress awards. second prize. Facing the new demand for intelligent manufacturing in the lithium battery industry, Ling Yunguang used its own strength to submit a "machine vision AI" solution, using a more precise, more accurate, more stable and smarter method to solve the problem of defect detection in the industrial vision field. The problems of stability, lack of obvious "people-saving" strategy, and difficulty in reflecting the value of data will help the intelligent manufacturing of lithium batteries.

More precise: precise imaging

The premise of quality inspection is that the visual system should photograph various defects. Lingyun Optical Vision Inspection System has high-precision defect identification capabilities and can accurately capture pinholes, bright spots and other tiny defects. The detection accuracy can reach 0.05*0.05mm and can be 100% identified. Innovative double-sided leakage identification technology can identify single and double-sided leakage defects at the same time. In terms of stereoscopic imaging, the 2.5D system is used to take defect photos with 3D information, which can identify wrinkles, bulges, pits and other defects, with a classification accuracy higher than 99%. In a highly vibrating environment, the imaging is accurate without fear of jitter. After testing the vibration amplitude and frequency, the optimal installation point solution is obtained through anti-seismic simulation design and verification. In highly vibrating environments such as die cutting, winding, welding, and cell appearance inspection, The detection accuracy can be guaranteed to be 0.02-0.05mm.

More accurate: defect identification

Scientific measurement is the guarantee of quality management. Classification accuracy is determined by both imaging effects and deep learning. In order to ensure the scientific nature of quality management, Lingyunguang has developed a series of AI-based intelligent detection and classification technologies. These technologies can realize functions such as defect detection, intelligent classification, interference classification, and over-inspection suppression, and can accurately evaluate and classify The accuracy rate provides a strong guarantee for quality management.

More stable: stable and reliable

Full representation imaging system needs to cooperate with high-speed embedded edge computing processing platform to take advantage of it. Lingyun Optical high-speed embedded edge computing processing platform uses parallel pipeline computing and deep learning traditional algorithm fusion computing to achieve an image throughput of more than 2GB/s. The throughput is equivalent to the throughput of four traditional industrial computers, and high computing power returns can be achieved with low investment.

In addition, Lingyun Optical has realized the localization of core components in the entire process from image acquisition to image processing, including CMOS chips, cameras, lenses, processors, boards, light sources, etc., and can also quickly carry out various application scenarios. Customized development and ensuring a stable supply chain system.

At the same time, Lingyun Optical reserves the self-developed algorithm framework VisionWARE and continues to update and optimize it. It can select the required modules or customize development according to actual application scenarios to achieve efficiency improvements.

Smarter: Quality AI Empowerment

Lingyun Guang’s machine vision quality system uses big data and cloud computing technology to realize the digitalization, knowledgeization and intelligence of industrial production, and improve the level of production and quality management. By establishing a human vision model, the quality standard machine can formulate scientific and effective quality standards for equipment based on human subjective qualitative standards. Using the GMQM quality management system, one-click unified standards can be easily realized throughout the entire production line and the quality standards of different production lines can be coordinated. GMQM real-time quality monitoring and analysis system can be applied to visual inspection and quality inspection to provide product quality assurance, quality management, quality improvement and quality cost optimization.

Machine vision AI helps industrial artificial intelligence truly come to fruition

Since 2012, artificial intelligence technology based on deep learning has been widely used and re-innovated in industrial application scenarios. Facing the new demand for intelligent manufacturing in the lithium battery industry, Lingyunguang has comprehensively assisted the development of intelligent manufacturing in the new energy industry. Related vision system products with fully independent intellectual property rights have been widely used in the upstream and downstream production lines of the lithium battery industry.

2023 Machine Vision Industry Chain TOP30

At the meeting, the "2023 Machine Vision Industry Chain TOP30" was officially announced. Ling Yunguang ranked on the list with its advanced technical strength and significant market influence in the machine vision industry.

In the face of new changes and new demands in the lithium battery industry, Lingyunguang will use its solid technical strength to deeply explore customer needs in the field of lithium battery machine vision, help customers achieve the goal of zero complaints with innovative products and solutions, and assist machine vision and intelligent manufacturing. Industry flexible upgrade!

The above is the detailed content of Ling Yunguang: Machine vision + AI innovation helps 'intelligent' manufacturing of lithium batteries. For more information, please follow other related articles on the PHP Chinese website!