An Automated Machine Learning Approach to Inkjet Printed Component Analysis: A Step Toward Smart Additive Manufacturing

TL;DR

The paper addresses the challenge of extracting multiple inkjet-printed material parameters from limited microwave measurements. It introduces an AutoML regression framework that automatically evaluates and selects the best algorithms for each parameter, trained on data generated from on-wafer S-parameters and EM simulations. The study demonstrates that XGBoost and LightGBM yield accurate estimates of the substrate and ink properties, with good agreement against independent measurements and FE validation, enabling rapid, automated characterization. This approach holds practical potential for process monitoring and control in smart additive manufacturing, reducing manual effort and enabling real-time parameter extraction.

Abstract

In this paper, we present a machine learning based architecture for microwave characterization of inkjet printed components on flexible substrates. Our proposed architecture uses several machine learning algorithms and automatically selects the best algorithm to extract the material parameters (ink conductivity and dielectric properties) from on-wafer measurements. Initially, the mutual dependence between material parameters of the inkjet printed coplanar waveguides (CPWs) and EM-simulated propagation constants is utilized to train the machine learning models. Next, these machine learning models along with measured propagation constants are used to extract the ink conductivity and dielectric properties of the test prototypes. To demonstrate the applicability of our proposed approach, we compare and contrast four heuristic based machine learning models. It is shown that eXtreme Gradient Boosted Trees Regressor (XGB) and Light Gradient Boosting (LGB) algorithms perform best for the characterization problem under study.

Figures (8)

• Figure 1: Measurement setup for the inkjet printed CPWs. GSG probes are connected to measured the small signal S-parameters of the printed CPWs. An LCR meter is connected in series to measure the dc resistance of the THRU line.
• Figure 2: Crosssection of CPWs in a CAD environment. Silver nanoparticle inks of conductivity ($\sigma_{ink}$) are deposited on a ($t_{FS}$) thick flexible substrate to print CPWs with a ground plane of width ($W_{g}$), center conductor of width ($W_{c}$) and gap of width (g). The CPWs are supported by a dielectric spacer of thickness ($t_{DS}$) underneath.
• Figure 3: Architecture of the proposed automatic machine learning (AutoML).
• Figure 4: eXtreme gradient boosted trees regressor with early stopping (learning rate = 0.02.
• Figure 5: Light gradient boosting on elasticnet predictions.