Inverse prediction of capacitor multiphysics dynamic parameters using deep generative model

2026-02-25

AI Summary

This work addresses the computational inefficiency in multiphysics simulation of capacitors, where minor structural modifications necessitate repeated finite element analyses. To overcome this bottleneck, we propose a deep generative model–based inverse prediction framework, which, to the best of our knowledge, is the first application of such models to the inverse mapping from electrostatic field responses to dynamic structural parameters in capacitor design. By directly learning the inverse relationship between field responses and design parameters, the method circumvents the need for iterative forward simulations. Experimental results demonstrate that our approach significantly outperforms existing baselines in both quantitative metrics and visual fidelity, achieving high-accuracy parameter inversion with markedly improved efficiency and thereby transcending the limitations inherent in conventional forward simulation paradigms.

Technology Category

Machine Learning: Deep Generative Models & AutoencodersSearch and Optimization: Sampling/Simulation-based SearchCognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applicationsUser Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendationGraph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

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.

Problem

• inverse prediction
• capacitor
• multiphysics
• dynamic parameters
• structural change

Innovation

• inverse prediction
• deep generative model
• multiphysics simulation
• capacitor design
• boundary condition optimization