In the production process of new energy batteries, the quality of tab welding is crucial to battery performance and safety. WeLinkirt's AI vision technology provides an efficient inspection solution for it.
User scenario: The tab welding process line of a leading new energy battery manufacturer. Its products are various types of new energy batteries. In the tab welding process, strict inspection of the welding quality is required. The main inspection objects are the burrs and false welding at the welding joints to ensure the safety and reliability of the batteries.
Pain points: In the production mode of multiple varieties and small batches, traditional inspection methods face many difficulties. On the one hand, manual inspection has a relatively high missed detection rate and false alarm rate. The average missed detection rate is about 3%, and the false alarm rate is about 10%. This not only increases the subsequent re-inspection workload but also may allow batteries with quality problems to enter the market. On the other hand, a large amount of time is required for programming and debugging during model change. The average model change time is as long as 30 minutes, which seriously affects the release of production capacity. In addition, traditional inspection methods are difficult to meet the inspection requirements of micron-level accuracy, while the inspection of burrs and false welding in tab welding requires extremely high accuracy.
Technical principle
WeLinkirt uses advanced AI algorithms and imaging technologies to solve the problem of tab welding quality inspection. At the algorithm level, a feature recognition algorithm based on the visual basic model is used. Through learning from a small number of positive samples (1-20 good products), it can quickly and accurately identify the normal and abnormal features at the welding joints. This is because the algorithm can deeply analyze the images and extract representative feature vectors, thus achieving accurate judgment of burrs and false welding. In terms of imaging, the self-developed 3D camera can obtain the three-dimensional morphology information of the welding joints. Through the three-dimensional morphology reconstruction technology, the real morphology of the welding joints is presented in a high-precision three-dimensional model. This can clearly detect hidden solder joints, coplanarity, and micron-level morphological changes, providing more comprehensive and accurate data support for inspection.
- The feature recognition algorithm based on the visual basic model has strong learning ability for positive samples and few samples.
- The self-developed 3D camera can obtain three-dimensional morphology information with high accuracy in three-dimensional morphology reconstruction.
- It can comprehensively detect hidden solder joints, coplanarity, and micron-level morphology.
WeLinkirt solutions and products
WeLinkirt provides the DaoAI AI AOI software system and the DaoAI 2D / 3D AI AOI equipment to solve the problem of tab welding quality inspection. The DaoAI AI AOI software system has powerful programming and learning capabilities. It only takes 5 minutes to achieve zero-code automatic programming for one good product. It also uses APDT positive sample / few-sample learning (1-20 good products) and has a semantic false-alarm filtering function, which can significantly reduce the false-alarm rate. The DaoAI 2D / 3D AI AOI equipment is equipped with a self-developed 3D camera. Through the three-dimensional morphology reconstruction technology, it can accurately detect hidden solder joints, coplanarity, and micron-level morphology, providing high-precision hardware support for tab welding quality inspection. During the implementation process, the equipment is installed on the tab welding process line, and the software system is seamlessly connected with the equipment to detect the welding joints in real-time and feed back the detection results to the production line control system in a timely manner.
WeLinkirt's AI vision technology provides an efficient and accurate solution for the quality inspection of new energy battery tab welding.
Quantitative results: By adopting WeLinkirt's solution, the new energy battery manufacturer has achieved remarkable results. In terms of inspection accuracy, the detection rate has reached 99.2%, and the missed detection rate has been reduced to <0.8%, greatly improving the product quality. In terms of the false-alarm rate, compared with the traditional inspection method, it has been reduced by -70%, reducing a large amount of re-inspection work. In terms of model change time, zero-code model change can be achieved in 5 minutes, significantly improving the production capacity and bringing higher economic benefits to the enterprise.