A 3D advanced package optimization method for three-dimensional stacked chips

Through a collaborative design framework combining multiphysics simulation, machine learning proxy models, and intelligent optimization algorithms, global automatic optimization of the 3D packaging structure of three-dimensional stacked chips was achieved, solving the problem of low efficiency in existing technologies and providing a more efficient design solution.

CN122133451APending Publication Date: 2026-06-02HUAXIN MICRO SEMICONDUCTOR (TANGSHAN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
HUAXIN MICRO SEMICONDUCTOR (TANGSHAN) CO LTD
Filing Date
2026-02-05
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

Existing technologies are inefficient in optimizing 3D advanced packaging designs for 3D stacked chips, making it difficult to effectively balance thermal, mechanical, and electrical performance, and failing to systematically provide the best balance design solution between overall performance, reliability, and cost.

Method used

A collaborative design framework combining multiphysics simulation, machine learning proxy models, and intelligent optimization algorithms is adopted. By establishing parameterized models, multi-objective intelligent optimization, and high-fidelity simulation verification, global automatic optimization of 3D packaging structures is achieved.

Benefits of technology

It significantly shortens the design cycle, explores design solutions with better overall performance and higher reliability, solves the problem of low efficiency of traditional methods, and provides efficient design support for high-density heterogeneous integration in the post-Moore era.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122133451A_ABST
    Figure CN122133451A_ABST
Patent Text Reader

Abstract

This invention discloses a 3D advanced packaging optimization method for three-dimensional stacked chips, belonging to the field of packaging process optimization technology. The method includes: establishing a parameterized model of the target 3D advanced packaging structure; defining a design space consisting of multiple design variables, multiple optimization objectives, and at least one design constraint; selecting several sample points within the design space based on the design of experiments; performing coupled multiphysics simulations on the packaging structure corresponding to each sample point; establishing a corresponding machine learning surrogate model for each optimization objective; using a multi-objective intelligent optimization algorithm to solve for the Pareto optimal solution set satisfying the design constraints; and outputting the final optimized packaging design scheme. The advantages of this invention are: by constructing a collaborative design framework integrating multiphysics simulation, machine learning surrogate models, and intelligent optimization algorithms, it achieves a systematic trade-off and global automatic optimization of the thermal, mechanical, and electrical performance of 3D advanced packaging.
Need to check novelty before this filing date? Find Prior Art