Coefficient Extraction of SAC305 Solder Constitutive Equations Using Equation-Informed Neural Networks

• Published in: Materials Vol. 16; no. 14; p. 4922
• Main Authors: Yuan, Cadmus, Su, Qinghua, Chiang, Kuo-Ning
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 10.07.2023; MDPI
• Subjects: Accuracy; advanced electronic packaging; Analysis; Approximation; Back propagation; Bayesian analysis; Coefficients; Constitutive equations; Constitutive models; Constitutive relationships; Datasets; Efficiency; Equation-Informed Neural Networks; Graphical representations; Lead free; Neural networks; Normal distribution; numerical Bayesian Inference; Parameter identification; Pb-free SAC305 solders; Solders; Statistical inference; Strain rate; Temperature; Temperature dependence; Tin base alloys; constitutive equations; advanced electronic packaging; numerical Bayesian Inference; Equation-Informed Neural Networks; Pb-free SAC305 solders

Equation-Informed Neural Networks (EINNs) are developed as an efficient method for extracting the coefficients of constitutive equations. Subsequently, numerical Bayesian Inference (BI) iterations were applied to estimate the distribution of these coefficients, thereby further refining them. We coul...