Enhancing Quality Control in Battery Component Manufacturing

Enhancing Quality Control in Battery Component Manufacturing: Comparison

Please note this is a comparison between Version 2 by Fanny Huang and Version 1 by Thi-Thu-Huyen Vu.

battery manufacturing
quality
components

1. Introduction

In manufacturing and production lines, improving the quality control of products is an essential task prior to delivering them to consumers. Several factories and production companies strive to eliminate all defects from their products, aiming to bring the success of products to the market. To achieve zero defect manufacturing strategies [1,2,3^[1][2][3][4],4], many factories employ various techniques and processes to detect and prevent any defects that may appear in the final products [5,6,7,8]^[5][6][7][8].

With the current trend of outstanding electronic development, rechargeable batteries has become indispensable parts in electronic devices such as laptops, phones, and electric vehicles. In these devices, the rechargeable battery is the most important part that affects the speed, price, and performance of the device. Furthermore, the rechargeable battery also affects the safety and potential risk of using the devices. In order to create quality rechargeable battery products and minimize risks such as potential explosions, the quality of each part of the rechargeable battery must be ensured during the production process. Among these components are fastening parts used in battery assembly. These parts are shaped through the physical process of forging metals or alloys at high pressure and temperature. During the forging process, defects such as cracks or scratches on the surface cannot be avoided. So, in order to prevent these defects from affecting the final product, researchers have studied different approaches to identify defects in the parts of the battery [9,10,11]^[9][10][11] or to improve battery manufacturing processing ^[12].

For several years, using workers to detect defects on battery surfaces has been a widely adopted practice in manufacturing. This approach laboriously relies on the expertise and visual acuity of workers to identify any abnormalities or defects that appear in the products. This method requires skilled workers who have undergone thorough training to identify and categorize various types of defects accurately. Additionally, workers may employ their finger to identify surface defects that may not be clear. It is inevitable that these methods are subjective in person and can require significant time to implement. With the fast development of current deep learning algorithms, more and more methods [13,14]^[13][14] have been proposed, introduced, and applied to automatically identify defects on the surface of products in manufacturing during the production process. Some recent methods are as follows: using a probabilistic defect model ^[15], using contextual features and multi-view ensemble learning ^[16], using computed tomography ^[17], using ultrasound acoustic measurements ^[18], and using deep learning approaches [19,20,21,22,23]^{[19][20][21][22][23]}.

2. Enhancing Quality Control in Battery Component Manufacturing

Some recent research on effects of defects and defect detection in manufacturing, especially related to the rechargeable battery manufacturing sector are discussed. By examining the results acquired from related studies, researchers can accumulate effective defect detection methods and hence give the appropriate methods to detect and predict defects.

In ref. ^[24], Cannarella et al. investigated the effects of defects in the form of localized plating in lithium-ion batteries. They developed a model of the defect containing coin cell geometry to gain further insights into this the effects of these defects. The simulations demonstrated that the closed pores acted as “electrochemical concentrators”, resulting in elevated currents and overpotentials in the adjacent electrodes in lithium-ion batteries. The research also analyzed the impact of other factors, such as materials, geometry, and operating parameters, on the localized plating behavior in batteries. The study established a link between electrochemical degradation/failure and internal mechanical stress in lithium-ion batteries. Defects like separator pore closure, which can be induced or exacerbated by mechanical stress, generate localized regions of heightened electrochemical activity that lead to lithium plating. The findings in this study indicated that it is important to detect defects that occur in battery manufacturing, and defects greatly affect battery performance, as well as being a high risk of fire and explosion.

In ref. ^[17], Yi Wu et al. introduced a computed tomography (CT)-based non-destructive approach to evaluate the quality, identify defects, and assess structural deformation in batteries. The main ideas of the approach were that they used a CT system to scan the battery and then visualized the information in the 3D structure. After that, the 3D structure provided insight into its internal structure as well as material composition. The deformed structural and defects could be detected by directly observing the CT images from different angles. In this study, the authors proposed a useful method using the CT system to scan and visualize the structure of the battery. However, detecting defects on the surface or inside the product was still performed by direct observation.

In ref. ^[9], Changlu xu et al. presented an approach using multi-feature fusion and particle swarm optimization for a support vector machine model to detect defects in lithium batteries. The author explained that their approach reached an accuracy of 98.3% on a dataset containing 840 images of the battery surface. The methodology of their approach was that first, they preprocessed the defect image with image subtraction and contrast adjustment. After that, the Canny algorithm with the AND logical operation was used to extract the defect area on the image. Consequently, the features of the defect area were extracted using texture, edge, and HOG features. Finally, the particle swarm optimization method was used to optimize the support vector machine model to train and detect the defect on the image.

In ref. ^[10], Choudhary et al. introduced an autonomous visual detection method for detecting defects from battery electrode manufacturing. In particular, the YOLOv5 architecture was proposed in their approach for identifying the visual defects on the coated electrode from the battery. In the experiment, they captured 882 images of battery electrodes containing four types of defects in some images (agglomerate, bubble, foil, and scratch) and used these images for training and prediction to evaluate the model. The study showed that the model achieved 9.5 ms inference time and 88% mAP for predicting all combined classes. Their study presented evidence of effectively using the YOLO architecture to predict defects in battery manufacturing and efficient use in the web environment.

In ref. ^[25], Badmos et al. proposed a method using convolutional neural networks (CNNs) for detecting defects in lithium-ion battery electrodes. In the study, the authors attempted to achieve the best parameters for creating precise models to predict battery defects. They compared obtained inference results with different CNN models, i.e., Baseline, Sigmoid, SoftMax, VGG19, InceptionV3, and Xception. A dataset containing 3286 original images of battery electrodes was used to train and infer the four types of defects in the battery. Through experimental results, the VGG19 model performed best (with an F1 score of 0.99), followed by the InceptionV3 (with an F1 score of 0.97). Using the VGG19 architecture could be an effective solution for creating an accurate model to detect defects in images of lithium-ion battery electrodes.

References

Psarommatis, F.; Sousa, J.; Mendonça, J.P.; Kiritsis, D. Zero-defect manufacturing the approach for higher manufacturing sustainability in the era of industry 4.0: A position paper. Int. J. Prod. Res. 2022, 60, 73–91.
Powell, D.; Magnanini, M.C.; Colledani, M.; Myklebust, O. Advancing zero defect manufacturing: A state-of-the-art perspective and future research directions. Comput. Ind. 2022, 136, 103596.
Psarommatis, F.; May, G.; Dreyfus, P.A.; Kiritsis, D. Zero defect manufacturing: State-of-the-art review, shortcomings and future directions in research. Int. J. Product. Res. 2019, 58, 1–17.
Eger, F.; Coupek, D.; Caputo, D.; Colledani, M.; Penalva, M.; Ortiz, J.A.; Freiberger, H.; Kollegger, G. Zero defect manufacturing strategies for reduction of scrap and inspection effort in multi-stage production systems. Procedia CIRP 2018, 67, 368–373.
Liu, K.; Ma, Z.; Liu, Y.; Yang, J.; Yao, Y. Enhanced defect detection in carbon fiber reinforced polymer composites via generative kernel principal component thermography. Polymers 2021, 13, 825.
Fu, Y.; Downey, A.R.J.; Yuan, L.; Zhang, T.; Pratt, A.; Balogun, Y. Machine learning algorithms for defect detection in metal laser-based additive manufacturing: A review. J. Manuf. Process. 2022, 75, 693–710.
Wan, P.K.; Leirmo, T.L. Human-centric zero-defect manufacturing: State-of-the-art review, perspectives, and challenges. Comput. Ind. 2022, 144, 103792.
Chen, Y.; Ding, Y.; Zhao, F.; Zhang, E.; Wu, Z.; Shao, L. Surface Defect Detection Methods for Industrial Products: A Review. Appl. Sci. 2021, 11, 7657.
Xu, C.; Li, L.; Li, J.; Wen, C. Surface Defects Detection and Identification of Lithium Battery Pole Piece Based on Multi-Feature Fusion and PSO-SVM. IEEE Access 2021, 9, 85232–85239.
Choudhary, N.; Clever, H.; Ludwigs, R.; Rath, M.; Gannouni, A.; Schmetz, A.; Hülsmann, T.; Sawodny, J.; Fischer, L.; Kampker, A.; et al. Autonomous Visual Detection of Defects from Battery Electrode Manufacturing. Adv. Intell. Syst. 2022, 4, 2200142.
Schoo, A.; Moschner, R.; Hülsmann, J.; Kwade, A. Coating Defects of Lithium-Ion Battery Electrodes and Their Inline Detection and Tracking. Batteries 2023, 9, 111.
Niri, M.F.; Liu, K.; Apachitei, G.; Román-Ramírez, L.A.; Lain, M.; Widanage, D.; Marco, J. Quantifying key factors for optimised manufacturing of Li-ion battery anode and cathode via artificial intelligence. Energy AI 2022, 7, 100129.
Benbarrad, T.; Kenitar, S.B.; Arioua, M. Intelligent machine vision model for defective product inspection based on machine learning. In Proceedings of the 2020 International Symposium on Advanced Electrical and Communication Technologies (ISAECT), Marrakech, Morocco, 25–27 November 2020.
Adibhatla, V.A.; Chih, H.C.; Hsu, C.C.; Cheng, J.; Abbod, M.F.; Shieh, J.S. Applying deep learning to defect detection in printed circuit boards via a newest model of you-only-look-once. Math. Biosci. Eng. 2021, 18, 4411–4428.
Zhang, X.; Tang, L.; Fox, E.; Qiao, Z.; Fan, W.; Wang, G.A. Identifying product defects from user complaints: A probabilistic defect model. In Proceedings of the 22nd Americas Conference on Information Systems (AMCIS 2016), San Diego, CA, USA, 11–14 August 2016; pp. 1–10.
Liu, Y.; Jiang, C.; Zhao, H. Using contextual features and multi-view ensemble learning in product defect identification from online discussion forums. Decis. Support Syst. 2018, 105, 1–12.
Wu, Y.; Saxena, S.; Xing, Y.; Wang, Y.; Li, C.; Yung, W.K.; Pecht, M. Analysis of manufacturing-induced defects and structural deformations in lithium-ion batteries using computed tomography. Energies 2018, 11, 925.
Robinson, J.B.; Owen, R.E.; Kok, M.D.; Maier, M.; Majasan, J.; Braglia, M.; Stocker, R.; Amietszajew, T.; Roberts, A.J.; Bhagat, R.; et al. Identifying Defects in Li-Ion Cells Using Ultrasound Acoustic Measurements. J. Electrochem. Soc. 2020, 167, 120530.
Silenzi, A.; Castorani, V.; Tomassini, S.; Falcionelli, N.; Contardo, P.; Bonci, A.; Dragoni, A.F.; Sernani, P. Quality Control of Carbon Look Components via Surface Defect Classification with Deep Neural Networks. Sensors 2023, 23, 7607.
Lu, B.; Xu, D.; Huang, B. Deep-learning-based anomaly detection for lace defect inspection employing videos in production line. Adv. Eng. Inform. 2022, 51, 101471.
Kaji, F.; Nguyen-Huu, H.; Budhwani, A.; Narayanan, J.A.; Zimny, M.; Toyserkani, E. A deep-learning-based in-situ surface anomaly detection methodology for laser directed energy deposition via powder feeding. J. Manuf. Process. 2022, 81, 624–637.
Fang, Q.; Ibarra-Castanedo, C.; Maldague, X. Automatic defects segmentation and identification by deep learning algorithm with pulsed thermography: Synthetic and experimental data. Big Data Cogn. Comput. 2021, 5, 9.
Sampath, V.; Maurtua, I.; Martin, J.J.A.; Rivera, A.; Molina, J.; Gutierrez, A. Attention Guided Multi-Task Learning for Surface defect identification. IEEE Trans. Ind. Inform. 2023, 19, 9713–9721.
Cannarella, J.; Arnold, C.B. The Effects of Defects on Localized Plating in Lithium-Ion Batteries. J. Electrochem. Soc. 2015, 162, A1365–A1373.
Badmos, O.; Kopp, A.; Bernthaler, T.; Schneider, G. Image-based defect detection in lithium-ion battery electrode using convolutional neural networks. J. Intell. Manuf. 2020, 31, 885–897.