Droplet evolution prediction in material jetting via tensor time series analysis

Review of Droplet Printing Technologies for Flexible Electronic Devices: Materials, Control, and Applications

“…Soon Wook Kwon [ 41 ] improved the predictive accuracy of the material printing process by introducing physical constraints into neural networks, as shown in Figure 6 b. Huang et al [ 50 ] studied the evolution behavior and process dynamics of ink droplets in the inkjet printing process using unsupervised learning methods. By using video data instead of images to study droplet evolution during inkjet printing, the experimental results demonstrated the high accuracy of the proposed method in predicting droplet evolution and understanding the dynamics of the inkjet printing process, as shown in Figure 6 c. Segura [ 67 ] studied the evolution of droplet behavior with different materials and process parameters through tensor time-series analysis of experimental data. The author successfully predicted the evolution behavior of droplets with different materials and process conditions using this method, as shown in Figure 6 d. Siemenn [ 17 ] proposed a method for optimizing the droplet-generation process using Bayesian optimization algorithms, effectively improving the efficiency and accuracy of the droplet-generation process and thereby enhancing the performance of droplet arrays, as shown in Figure 6 e. Mea [ 68 ] utilized a glass capillary microfluidic device to achieve programmed entrapment of droplets.…”

“…( d ) Schematic diagram of the NET scheme applied to tensor time series. Reproduced with permission from [ 67 ], published by Elsevier, 2023. ( e ) Bayesian optimization and computer-vision feedback-loop diagram.…”

“…By using video data instead of images to study droplet evolution during inkjet printing, the experimental results demonstrated the high accuracy of the proposed method in predicting droplet evolution and understanding the dynamics of the inkjet printing process, as shown in Figure 6c. Segura [67] studied the evolution of droplet behavior with different materials and process parameters through tensor time-series analysis of experimental data. The author successfully predicted the evolution behavior of droplets with different materials and process conditions using this method, as shown in Figure 6d.…”

Application of machine learning in polymer additive manufacturing: A review

“…140 Additionally, Segura et al evaluated droplet morphology using different ANNs. 253 Machine configurations such as dwell and echo voltage, dwell and echo time, and material properties such as density, surface tension, and viscosity were considered as model inputs for both works. Both frameworks aimed for greater adaptability beyond the materials they tested, ensuring the feasibility of analyzing new materials.…”

“…Brishty et al modeled droplet velocity, radius, and jetting regime (single drop/multiple drop/no ejection) using traditional ML models such as ensembles of DTs (boosted DTs and RF), KNN, and DNN 140 . Additionally, Segura et al evaluated droplet morphology using different ANNs 253 . Machine configurations such as dwell and echo voltage, dwell and echo time, and material properties such as density, surface tension, and viscosity were considered as model inputs for both works.…”