Inverse prediction of capacitor multiphysics dynamic parameters using deep generative model

TL;DR

This paper introduces a deep generative model approach to inverse prediction of dynamic parameters in capacitor simulations, enabling efficient modeling of structural changes without repeated simulations.

Contribution

It presents a novel application of deep generative models for inverse prediction in capacitor multiphysics simulations, improving accuracy over traditional baselines.

Findings

Outperformed baseline methods in inverse prediction accuracy

Achieved better visual and quantitative results

Demonstrated effectiveness on electrostatics field data

Abstract

Finite element simulations are run by package design engineers to model design structures. The process is irreversible meaning every minute structural adjustment requires a fresh input parameter run. In this paper, the problem of modeling changing (small) design structures through varying input parameters is known as inverse prediction. We demonstrate inverse prediction on the electrostatics field of an air-filled capacitor dataset where the structural change is affected by a dynamic parameter to the boundary condition. Using recent AI such as deep generative model, we outperformed best baseline on inverse prediction both visually and in terms of quantitative measure.