LLM-Inspired Vision Transformer Framework for Intelligent Quality Recognition in Aerosol Jet Printing

Haining Zhang; Joon Phil Choi; Xingchen Liu; Nannan Liang; Yangwen Yu; Yongrae Kim; Dongwoon Shin; Seung Ki Moon; Yong-Jin Yoon

doi:10.1016/j.eng.2026.03.021

Engineering ›› :202603021 DOI: 10.1016/j.eng.2026.03.021

research-article

LLM-Inspired Vision Transformer Framework for Intelligent Quality Recognition in Aerosol Jet Printing

Author information +

History +

PDF (8614KB)

Abstract

Aerosol jet printing (AJP) is an advanced direct-ink writing technique within additive manufacturing (AM) that enables the high-resolution fabrication of micro-scale electronic components through the controlled deposition of aerosolized functional inks. Despite its potential, AJP faces significant challenges in maintaining consistent printing quality due to dynamic process variability and material heterogeneity. Existing approaches are often insufficient to address these challenges. In particular, traditional data-driven models relying on numerical inputs inherently omit critical visual morphology features, while conventional supervised vision methods struggle to generalize under practical variability. To address this, we propose an intelligent quality recognition framework that adapts large language model (LLM)-inspired Transformer architectures to manufacturing vision tasks. Specifically, Vision Transformers (ViT), which translate the self-attention mechanisms of LLMs into computer vision, are employed to achieve robust and generalizable defect detection in AJP. The proposed approach leverages a dataset composition of approximately 46 000 unlabeled images for self-supervised pre-training to enhance feature representation, followed by fine-tuning on 9369 labeled images generated through randomized printing experiments. Afterwards, comparative studies demonstrate the effectiveness of the proposed system, achieving classification accuracy of 98.1% with high robustness. By leveraging LLM-derived architectures in the context of smart manufacturing, this work introduces a scalable, adaptable, and unified machine vision framework for intelligent quality control across diverse material deposition systems. The integration of self-attention-based ViT models constructs a synergistic system that not only advances the precision of defect recognition in AJP but also contributes to the broader realization of an autonomous, data-driven manufacturing ecosystem.

Keywords

Large language models / Vision transformer / Machine learning / Self-supervised representation learning / Anomaly detection / Aerosol jet printing / Smart manufacturing

Cite this article

Download citation ▾

Haining Zhang, Joon Phil Choi, Xingchen Liu, Nannan Liang, Yangwen Yu, Yongrae Kim, Dongwoon Shin, Seung Ki Moon, Yong-Jin Yoon. LLM-Inspired Vision Transformer Framework for Intelligent Quality Recognition in Aerosol Jet Printing. Engineering 202603021 DOI:10.1016/j.eng.2026.03.021

登录浏览全文

4963

注册一个新账户忘记密码

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Zeng P , Tian B , Tian Q , Yao W , Li M , Wang H , et al. Screen—printed, low—cost, and patterned flexible heater based on Ag fractal dendrites for human wearable application. Adv Mater Technol 2019; 4(3):1800453.

[2]	He P , Cao J , Ding H , Liu C , Neilson J , Li Z , et al. Screen—printing of a highly conductive graphene ink for flexible printed electronics. ACS Appl Mater Interfaces 2019; 11(35):32225-34.

[3]	Xiong J , Chen J , Chen W , Yue X , Zhao Z , Yin Z . Intelligent path planning algorithm system for printed display manufacturing using graph convolutional neural network and reinforcement learning. J Manuf Syst 2025; 79:73-85.

[4]	Zhang Z , Yang H , Chen J , Yin Z . Multi—scale conditional diffusion model for deposited droplet volume measurement in inkjet printing manufacturing. J Manuf Syst 2023; 71:595-608.

[5]	Yue X , Chen J , Li Y , Li X , Zhu H , Yin Z . Intelligent control system for droplet volume in inkjet printing based on stochastic state transition soft actor—critic DRL algorithm. J Manuf Syst 2023; 68:455-64.

[6]	Elsersawy R , Rahman A , Sakib—Uz—Zaman C , Khondoker MAH . Multifunctional inks in aerosol jet printing: performance, challenges, and applications. Front Manuf Technol. 2025; 5:1558209.

[7]	Fisher C , Skolrood LN , Li K , Joshi PC , Aytug T . Aerosol‐jet printed sensors for environmental, safety, and health monitoring: a review. Adv Mater Technol 2023; 8(15):2300030.

[8]	Hines DR , Gu Y , Martin AA , Li P , Fleischer J , Clough—Paez A , et al. Considerations of aerosol—jet printing for the fabrication of printed hybrid electronic circuits. Addit Manuf 2021; 47:102325.

[9]	Guo Z , Yu P , Li B , Xuan F , Zhao J . Direct wire writing technique benefitting the flexible electronics. Virtual Phys Prototyp 2024; 19(1):e2286514.

[10]	Lim HR , Kim HS , Qazi R , Kwon YT , Jeong JW , Yeo WH . Advanced soft materials, sensor integrations, and applications of wearable flexible hybrid electronics in healthcare, energy, and environment. Adv Mater 2020; 32(15):1901924.

[11]	Agarwala S , Goh GL , Dinh Le TS , An J , Peh ZK , Yeong WY , et al. Wearable bandage—based strain sensor for home healthcare: combining 3D aerosol jet printing and laser sintering. ACS Sens 2019; 4(1):218-26.

[12]	Han T , Kundu S , Nag A , Xu Y . 3D printed sensors for biomedical applications: a review. Sensors 2019; 19(7):1706.

[13]	Huang B , Wu S , Liu J , Liu J , Peng B , Zhou Z . Aerosol jet printing of polyelectrolyte—modified MXene ink for a multifunctional humidity and temperature flexible sensor. Chem Eng J 2025; 519:165403.

[14]	Karipoth P , Chandler JH , Lee J , Taccola S , Macdonald J , Valdastri P , et al. Aerosol jet printing of strain sensors for soft robotics. Adv Eng Mater 2024; 26(1):2301275.

[15]	Khan Y , Garg M , Gui Q , Schadt M , Gaikwad A , Han D , et al. Flexible hybrid electronics: direct interfacing of soft and hard electronics for wearable health monitoring. Adv Funct Mater 2016; 26(47):8764-75.

[16]	Serpelloni M , Cantù E , Borghetti M , Sardini E . Printed smart devices on cellulose—based materials by means of aerosol—jet printing and photonic curing. Sensors 2020; 20(3):841.

[17]	Min SH , Lee GY , Ahn SH . Direct printing of highly sensitive, stretchable, and durable strain sensor based on silver nanoparticles/multi—walled carbon nanotubes composites. Compos Part B Eng 2019; 161:395-401.

[18]	Seifert T , Sowade E , Roscher F , Wiemer M , Gessner T , Baumann RR . Additive manufacturing technologies compared: morphology of deposits of silver ink using inkjet and aerosol jet printing. Ind Eng Chem Res 2015; 54(2):769-79.

[19]	Mahajan A , Frisbie CD , Francis LF . Optimization of aerosol jet printing for high—resolution, high—aspect ratio silver lines. ACS Appl Mater Interfaces 2013; 5(11):4856-64.

[20]	Zhang H , Moon SK , Ngo TH . 3D printed electronics of Non—contact ink writing techniques: status and promise. Int J Precis Eng Manuf—Green Technol 2020; 7(2):511-24.

[21]	Chen G , Gu Y , Tsang H , Hines DR , Das S . The effect of droplet sizes on overspray in aerosol—jet printing. Adv Eng Mater 2018; 20(8):1701084.

[22]	Zhang H , Huang J , Zhang X , Wong CN . Uncertainty quantification of aerosol jet 3D printing process using non—intrusive polynomial chaos and stochastic collocation. Adv Eng Inform 2025; 65(Part A):103175.

[23]	Liu Z , Liu Y , He L , Cui L , Liang N , Choi JP , et al. A comprehensive investigation of process parameters and material properties effects on printed line quality of aerosol jet printing based on coupled three—dimensional numerical models. Int J Precis Eng Manuf—Green Technol 2024; 11:727-42.

[24]	Yoo D , Mahoney CM , Deneault JR , Grabowski C , Austin D , Berrigan JD , et al. Mapping drift in morphology and electrical performance in aerosol jet printing. Prog Addit Manuf 2021; 6(2):257-68.

[25]	Smith M , Choi YS , Boughey C , Kar—Narayan S . Controlling and assessing the quality of aerosol jet printed features for large area and flexible electronics. Flex Print Electron 2017; 2(1):015004.

[26]	Zhang H , Choi JP , Moon SK , Ngo TH . A hybrid multi—objective optimization of aerosol jet printing process via response surface methodology. Addit Manuf 2020; 33:101096.

[27]	Secor EB . Guided ink and process design for aerosol jet printing based on annular drying effects. Flex Print Electron 2018;3(3):035007.

[28]	Zhang H , Choi JP , Moon SK , Ngo TH . A multi—objective optimization framework for aerosol jet customized line width printing via small data set and prediction uncertainty. J Mater Process Technol 2020; 285:116779.

[29]	Secor EB. Light scattering measurements to support real—time monitoring and closed—loop control of aerosol jet printing. Addit Manuf 2021; 44:102028.

[30]	Divakaran N , Das JP , PV AK , Mohanty S , Ramadoss A , Nayak SK . Comprehensive review on various additive manufacturing techniques and its implementation in electronic devices. J Manuf Syst 2022; 62:477-502.

[31]	Kim S , Kim EH , Lee W , Sim M , Kim I , Noh J , et al. Real—time in—process control methods of process parameters for additive manufacturing. J Manuf Syst 2024; 74:1067-90.

[32]	Feng JQ . Multiphase flow analysis of mist transport behavior in aerosol jet® system. Int J Comput Methods Exp Meas 2017; 6(1):23-34.

[33]	Salary R , Lombardi JP , Samie Tootooni M , Donovan R , Rao PK , Borgesen P , et al. Computational fluid dynamics modeling and online monitoring of aerosol jet printing process. J Manuf Sci Eng 2017; 139(2):021015.

[34]	Salary R , Lombardi JP , Weerawarne DL , Rao P , Poliks MD . A computational fluid dynamics investigation of pneumatic atomization, aerosol transport, and deposition in aerosol jet printing process. J Micro Nano—Manuf 2021; 9(1):010903.

[35]	Zhang H , Xu H , Cui L , Pan Z , Lee PH , Jung MK , et al. An extensive study of the influence of key flow variables on printed line quality outcomes during aerosol jet printing using coupled three—dimensional numerical models. Materials 2024; 17(13):3179.

[36]	Ramesh S , Mahajan C , Gerdes S , Gaikwad A , Rao P , Cormier DR , et al. Numerical and experimental investigation of aerosol jet printing. Addit Manuf 2022; 59(Part A):103090.

[37]	Zhang H , Liu Z , Yin S , Xu H . A hybrid multi—objective optimization of functional ink composition for aerosol jet 3D printing via mixture design and response surface methodology. Sci Rep 2023; 13(1):2513.

[38]	Akhatov IS , Hoey JM , Swenson OF , Schulz DL . Aerosol focusing in micro—capillaries: theory and experiment. J Aerosol Sci 2008; 39(8):691-709.

[39]	Li M , Liu Z , Yin S , Choi JP , Zhang H . Comparison and identification of optimal machine learning model for rapid optimization of printed line characteristics of aerosol jet printing technology. Int J Precis Eng Manuf—Green Technol 2024; 11(1):71-87.

[40]	Zhang H , Choi JP , Moon SK , Ngo TH . A knowledge transfer framework to support rapid process modeling in aerosol jet printing. Adv Eng Inform 2021; 48:101264.

[41]	Zhang H , Hong E , Chen X , Liu Z . Machine learning enables process optimization of aerosol jet 3D printing based on the droplet morphology. ACS Appl Mater Interfaces 2023; 15(11):14532-45.

[42]	Zhang H , Moon SK . Reviews on machine learning approaches for process optimization in noncontact direct ink writing. ACS Appl Mater Interfaces 2021; 13(45):53323-45.

[43]	Choi E , An K , Kang KT . Deep—learning—based microfluidic droplet classification for multijet monitoring. ACS Appl Mater Interfaces 2022; 14(13):15576-86.

[44]	Choi E , Choi S , An K , Kang KT . Deep learning—based inkjet droplet detection for jetting characterizations and multijet synchronization. ACS Appl Mater Interfaces 2024; 16(14):18040-51.

[45]	Yao C , Wang L , Wang Q , Liu Z , Liu G , Zhang M . Deep—learning—guided evaluation method for the high—volume preparation of flexible sensors based on inkjet printing. ACS Appl Mater Interfaces 2024; 16(10):13326-34.

[46]	Zhang H , Huang J , Zhang X , Wong CN . Autonomous optimization of process parameters and in—situ anomaly detection in aerosol jet printing by an integrated machine learning approach. Addit Manuf 2024; 86:104208.

[47]	Liang X , Wang Z , Liu J . A survey of large language model—augmented knowledge graphs for advanced complex product design. J Manuf Syst 2025; 80:883-901.

[48]	Jeon J , Sim Y , Lee H , Han C , Yun D , Kim E , et al. ChatCNC: conversational machine monitoring via large language model and real—time data retrieval augmented generation. J Manuf Syst 2025; 79:504-14.

[49]	Shi D , Liedl P , Bauernhansl T . Interoperable information modelling leveraging asset administration shell and large language model for quality control toward zero defect manufacturing. J Manuf Syst 2024; 77:678—96.

[50]	Wu T , Li J , Bao J , Liu Q . ProcessCarbonAgent: a large language models—empowered autonomous agent for decision—making in manufacturing carbon emission management. J Manuf Syst 2024; 76:429-42.