A Dual-Tree Complex Wavelet Transform Simulation Model for Improved Noise Modeling and Prediction of Real-Time Stencil-Printing Process

Abstract

This article presents a dynamic simulation model for the stencil-printing process (SPP) in surface mount technology (SMT) assembly lines, focusing on accurately replicating the real-time stencil printing while allowing adjustments to printer settings. The model offers a time and cost-effective alternative to the experiments and a reliable testing environment for researchers and technologists investigating advanced algorithms and strategic methodologies in SMT printing. SPP is influenced by various controllable factors, such as printer parameters. However, an additional challenge arises from uncontrollable environmental noise that affects the printing quality, leading to uneven solder paste application and machine precision that brings randomness to the results. Recognizing the need to mitigate the effects of this environmental noise and enhance the accuracy of the simulator, the proposed simulation model incorporates a dual-tree complex wavelet transform (DTCWT) algorithm. DTCWT used in this model addresses the challenge of environmental noise affecting the printing quality, showcasing an enhanced capability in noise reduction and signal clarity. The noise from the SPP data is modeled and extracted from the DTCWT model and introduced into the simulation model to improve the prediction accuracy. The simulation model demonstrated an improvement of 36% in Volume AVG and 62% in Volume STD accuracy on root-mean-squared error (RMSE), marking a significant advancement over the statistical simulator.