This study proposes a systematic data preprocessing algorithm tailored for AI-based modeling of manufacturing data from a roll-to-roll (R2R) lithium iron phosphate (LFP) battery electrode coating process. The preprocessing strategy specifically addresses process characteristics and spatiotemporal inconsistencies in sensor data, significantly improving data quality for machine learning applications. Utilizing the refined dataset, machine learning models were created to predict coating-related characteristics, resulting in high explanatory power and low prediction errors. This framework effectively illustrates the potential of data-driven modeling for reliable predictions and quantitative analysis of coating uniformity in battery manufacturing.

Flexible electronics have been to the fore because it is believed that flexibility can add incredible value such as light weight and mobility into the existing electronic devices and create new markets of large-area and low-cost electronics such as wearable eletronics in near future. Offset printing processes are regarded as major candidates for manufacturing the flexible electronics because they can provide the patterning resolution of micron-size effectively in large-area. In view of mechanics, the most important viewpoint in offset printing is how to achieve the synchronized movement of two contact surfaces in order to prevent slip between two contact surfaces and distortion of the blanket surface during ink transfer so that the high-resolution and good-overlay patterns can be printed. In this paper, a novel low-cost measurement method of the synchronization error using the motor control output signals is proposed and the compensation method is presented to minimize the synchronization error.

Solution processed conjugated molecules enable to manufacture various electronic devices by unconventional and cost effective patterning methods as screen or gravure printing. Spin-coating is the most popularly used method to form conjugated polymeric film for various electronic devices. The coating method has certain disadvantages such as a large amount of unwanted wastes, difficulty forming a film with a large area, and impossible to apply roll-to-roll manufacturing. We present here a promising alternative coating method, bar-coating for conjugated polymer film and OLED with the bar coated light emitting layer. In this papers, we show atomic force microscope images of spin- and bar-coated Poly[(9,9-di-n-octylfluorenyl-2,7- diyl)-alt-(benzo[2,1,3]thiadiazol-4,8-diyl)] (F8BT) films on substrate. The bar-coated film showed a slight lower RMS roughness (1.058 [nm]) than spin-coated film (1.767 [nm]). It means the barcoating is suitable method to form light emitting layers in OLEDs. By using bar-coating process, an OLED obtained with 4.7 [cd/A] in maximum current efficiency.

Recently, there are many issues about R2R printing technique for mass production of electronic devices. Among the various Roll-to-roll based printing techniques such as gravure, off-set, flexo and so on, “Gravure off-set printing technique” has an advantage of higher printing resolution. The printing unit of gravure off-set printing technique usually consists of plate roll, blanket roll and impressure roll whose. Linear velocities should be synchronized each other for fine pattern printing. However, roller’s manufacturing error and printing variations such as pringting pressure, printing speed, roll stroke and so on actually affected their synchronization anf thus the quality of fine fattern. In this paper, we analyzed the effective of synchronization error on printing quality. Also, this paper reviews the relative motion with each roll. And, this paper studys the synchronization error about its generation problem.

Ink transfer process is very important to determine quality of printed pattern, therefore its mechanism should be understood to control printing quality. Although there have been many attempts to understand ink transfer mechanism by numerical simulation and experimental studies, their model was too much simple to model realistic printing process and our understanding is not enough yet. In this paper we designed ink transfer visualization system to present flow visualization of ink transfer process for gravure offset printing. We considered rotational effect of blanket roll which is related with printing speed and used non-Newtonian fluid as working fluid such as Ag paste. For printing unit, cantilever-type blanket roll is used for convenient visualization of ink transfer. Serial images were captured continuously by using highspeed CMOS camera and long range microscope. We investigated the effects of various design parameters such as printing speed and pattern angle on the ink transfer process. We found more stretched ink filament for non-Newtonian fluid than Newtonian fluid. As increasing printing speed, length of stretched ink filament and height of break-up point are also increased. We also compared ink transfer process between CD and MD pattern and its relationship with ink transfer mechanism.

This paper presents how to design and fabricate the gravure offset printing system for enhancement of register precision. Factors of precision error are caused by imprecision of gravure plate, deformation of substrate, printing quality change due to the change of ink viscosity, Imprecision of printing machine, and so on. This study suggests concept design of gravure offset printing system which is able to minimize or remove these error factors.

The distortion of printed pattern is frequently observed in gravure offset printing process, which can be a serious problem in printing process for printed electronics. The mechanism of pattern distortion is studied and the factors which affect the amount and shape of distortion are found using FEM. The amount and shape of distortion is influenced by material properties of the roller, thickness of roller, applied load, and so on. As the printing pressure increases and Possion ratio increases, the degree of the image distortion increases. And the increase of the thickness of rubber roller brings a large distortion of image, too. In some cases, the distortion of printed pattern can reach a few hundred micro­millimeters. The comparison of the experiment result and the simulation result shows good agreement in their quantitative tendency.