Battery Winding/Stacking Alignment: 3D ACI Zero-Code Retooling

DaoAI 3D ACI equipment (proprietary 3D camera + 3D morphology reconstruction/point cloud, detecting hidden solder joints/coplanarity/micron-level morphology/pores and other 2D optical blind spot defects, 2D-3D fusion) significantly enhances the flexibility and efficiency of high-mix low-volume production by offering zero-code rapid retooling capabilities, reducing product changeover time in new energy battery winding/stacking processes from several hours with traditional manual parameter adjustment to under 5 minutes.

In new energy battery manufacturing, the winding or stacking process is a critical step determining cell performance and safety. Precise alignment of anodes and cathodes directly impacts battery internal resistance, energy density, cycle life, and even thermal runaway risks. With the rapid development of the new energy vehicle market, diverse and customized battery models are increasingly in demand, especially for high-end battery products, where high-mix low-volume production has become the norm. However, traditional inspection solutions often require significant time and human resources for parameter adjustment and model retraining when facing frequent changeovers, severely limiting production efficiency and cost control.

Pain Points: Why This Hurdle Is Difficult to Overcome

In winding/stacking alignment inspection, traditional methods face multiple challenges. First, traditional rule-based 2D AOI solutions are ineffective against micron-level morphological deviations, edge curling, tab deformation, and other 3D defects, leading to false negative rates exceeding 5% in some complex defect scenarios. Second, manual visual inspection is inefficient and susceptible to subjective factors, resulting in significant re-inspection time, potentially reaching 3-4 hours per shift during peak periods, with inconsistent results. The core pain point lies in the high-mix low-volume production model, where frequent product changeovers require engineers to spend several hours or even a full day adjusting parameters and deploying new models for traditional vision systems. For instance, production data from a mid-sized battery manufacturer showed that a single product changeover typically required 2-4 hours of downtime for vision system debugging, directly reducing Overall Equipment Effectiveness (OEE) by 15-20%. This high retooling cost and inefficient debugging cycle hinder companies in responding to market changes and launching new products. Furthermore, the growing industry demand for generalized defect detection highlights the limitations of traditional solutions, which require extensive annotated data and lengthy training for unknown or rare defects, contrasting sharply with the zero-shot segmentation capabilities of large vision models.

From a process perspective, the thickness of anode and cathode materials, coating uniformity, and edge cutting precision all affect the final alignment. These subtle morphological changes are difficult to capture in 2D images. In terms of imaging, reflective surfaces, uneven textures of battery materials, and shadows from stacked electrodes can interfere with 2D images. Material characteristics, such as subtle color or reflectivity differences between batches, further increase the false alarm rate of traditional rule-based algorithms. These root causes collectively contribute to the 'difficulty' of inspection.

Technical Principles

DaoAI 3D ACI equipment fundamentally solves the challenge of detecting 2D optical blind spot defects by combining proprietary high-precision 3D cameras with 3D morphology reconstruction technology. This equipment utilizes structured light or laser triangulation principles to acquire complete 3D point cloud data of the workpiece surface with sub-micron accuracy. By processing and analyzing the point cloud data in real-time, the DaoAI system can precisely quantify micron-level morphological defects such as electrode edge coplanarity, curling, surface bumps, and pores. Compared to traditional 2D AOI systems based on grayscale or color information, DaoAI 3D ACI directly captures height information of objects, enabling precise identification of 3D defects 'hidden' in 2D planar projections. For example, a slight warp at the edge of an electrode might appear as a blurry edge in a 2D image, while DaoAI 3D ACI can clearly reconstruct the height and angle of the warp and accurately determine if it is out of tolerance.

The DaoAI ACI OS operating system is at the core of achieving 'zero-code rapid retooling.' It incorporates visual foundation models with powerful feature recognition capabilities and an APDT (Adaptive Pre-trained Deep Transfer) few-shot learning mechanism. This means that when a new product or defect type needs to be inspected, users only need to provide 1-20 good sample images, and the system can automatically complete model training and deployment within 5 minutes, without writing any code or performing complex parameter adjustments. This capability stands in stark contrast to traditional methods: rule-based AOI requires engineers to write numerous rules and manually adjust thresholds for each defect, which is time-consuming and lacks generalization; traditional deep learning AOI requires vast amounts of defect samples for annotation and hours or even days of training. DaoAI 3D ACI, combined with its AI-driven operating system, reduces the false alarm rate to less than 1/5 of traditional methods, while shortening product changeover time to under 5 minutes, significantly enhancing production line flexibility.

Typical Application Scenarios

• **Anode and Cathode Edge Alignment Inspection**: Precisely measures the relative offset and coplanarity of anode and cathode edges during winding or stacking. DaoAI 3D ACI equipment can capture micron-level alignment deviations, preventing internal short circuits or capacity degradation caused by misalignment, which 2D methods often struggle with due to edge blur and shadows.
• **Tab Deformation and Warping Detection**: Detects minor bends, wrinkles, or warps in battery tabs before welding or pressing. These defects are often hard to identify or are obscured by the background in 2D images, but DaoAI 3D cameras can accurately reconstruct their 3D morphology, ensuring reliable subsequent connections.
• **Coating Defect Morphology Inspection**: For minor bumps, depressions, or particles in the electrode coating process, these defects might be overlooked in 2D images due to insufficient contrast. DaoAI 3D ACI provides precise height information, assisting in evaluating coating quality and ensuring uniform distribution of active materials.
• **Separator Damage and Wrinkles**: Detects minor tears, holes, or wrinkles in the separator during transport or alignment. These defects can lead to internal short circuits in the battery, and 3D morphology inspection can more reliably identify their 3D features, avoiding false negatives from traditional 2D optical inspection.
• **Solder Joint Morphology and Pore Detection**: For internal or external cell connection solder joints, DaoAI 3D ACI can inspect their coplanarity, height, width, and the presence of micro-pores or cold solder joints. These defects are blind spots for 2D optical inspection and are crucial for battery conductivity and safety.

Case Study

A leading new energy battery Tier-1 supplier in East China, whose production lines cover multiple cell models, had extremely high demands for winding/stacking alignment inspection accuracy and retooling efficiency. In the past, they primarily relied on traditional 2D AOI combined with manual re-inspection for quality control. However, with the expansion of product lines and increasing customer customization demands, frequent changeovers led to excessive production line downtime, severely impacting capacity. According to the manufacturer's internal assessment, traditional solutions required 3-5 hours for vision system parameter adjustment during each of the 3-4 product changeovers per week, accumulating over 600 hours of annual downtime, resulting in significant direct economic losses. Furthermore, the false negative rate for complex defects could not be effectively reduced, with actual data showing a false negative rate close to 0.8% for certain battery models, leading to considerable rework costs and customer complaint risks. After introducing DaoAI 3D ACI equipment, the situation improved significantly. Initially, the DaoAI engineering team collaborated closely with the client, leveraging the APDT few-shot learning capability of DaoAI ACI OS to deploy a defect detection model for a new product model within 30 minutes, using only 10 good samples. Following deployment, production line data showed that product changeover time was reduced to an average of under 4 minutes, a reduction of over 95% compared to traditional solutions.

"The DaoAI 3D ACI system not only improved detection accuracy but, more importantly, it completely transformed our production line's retooling efficiency bottleneck, enabling us to respond more flexibly to market demands and truly achieve efficient high-mix low-volume production." — Production Director, a leading new energy battery Tier-1 supplier.

DaoAI Solutions and Products

DaoAI's 3D ACI solution for the new energy battery industry, centered on its proprietary 3D camera and DaoAI ACI OS operating system, has achieved dual breakthroughs in detection accuracy and production flexibility. Our 3D ACI equipment not only provides micron-level defect detection capabilities, covering blind spots of traditional 2D optical inspection, but also thoroughly addresses the pain points of high-mix low-volume production through its AI-driven 'zero-code rapid retooling' mechanism. The DaoAI ACI OS operating system integrates feature recognition from visual foundation models and APDT few-shot self-training capabilities. Users only need to perform minimal operations on a user-friendly graphical interface to create and deploy new product models. Deployment methods are flexible, supporting SDK/API/Docker, and can achieve 100% on-premise private deployment, ensuring customer data security without leaving the factory. Concurrently, DaoAI 3D ACI also supports integration with existing MES/SCADA systems, uploading inspection data in real-time to build a comprehensive quality traceability system.

Through the deployment of DaoAI 3D ACI equipment, production line data from the client shows that the false negative rate for alignment defects decreased from 0.8% to <0.1%, and the false alarm rate was reduced by over −75%, significantly reducing manual re-inspection workload. More importantly, product changeover time was reduced from several hours to an average of 4 minutes, saving the company millions in annual downtime costs and greatly enhancing market responsiveness and overall competitiveness. DaoAI is committed to helping clients maintain leadership in fierce market competition through innovative technologies.

Full solution for this scenario: the full inspection solution for 3D ACI Equipment

This article was generated by AI. Customer cases are simulated scenarios based on real product capabilities and figures are illustrative; see product pages for official benchmarks.