An automatic calibration method for photolithography models and computer equipment

By automatically adjusting the weights and correction terms of the measurement pattern and dynamically optimizing the parameters of the lithography model, the problem of poor model accuracy and stability caused by reliance on human experience in the existing technology is solved, and efficient lithography model calibration is achieved, improving model accuracy and generalization ability.

CN122131553APending Publication Date: 2026-06-02SHENZHEN JINGYUAN INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHENZHEN JINGYUAN INFORMATION TECH CO LTD
Filing Date
2026-04-03
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

Existing AI-to-RI calibration technology for lithography models relies on human experience, resulting in poor model accuracy and stability. This makes it difficult to meet the challenges of model complexity and accuracy requirements in advanced processes, and the calibration process is inefficient.

Method used

An automated lithography model calibration method is adopted. By iteratively adjusting the weights and correction terms of the measurement pattern, the model parameters are dynamically optimized, achieving automated adjustment of weights and correction terms, eliminating manual dependence, and improving the consistency and reproducibility of the model.

Benefits of technology

It improved the accuracy and generalization ability of the model, shortened the calibration time and manpower input, ensured the consistency and reproducibility of model development, and improved calibration efficiency and R&D progress.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122131553A_ABST
    Figure CN122131553A_ABST
Patent Text Reader

Abstract

This invention provides an automatic calibration method and computer equipment for photolithography models. The automatic calibration method includes: performing an initial round of parameter optimization based on the initial weights and an initial set of correction terms of the measurement patterns; and adding and deleting correction terms in a cyclic iterative manner based on the initial set of correction terms to optimize the target indicators after model calibration. During the cyclic iterative process of adjusting weights, adding / removing correction terms, and optimizing model parameters, the weights of each measurement pattern are redistributed according to the preset priority and critical dimension errors of each measurement pattern, and correction terms to be added or deleted are determined based on the redistributed weights. This achieves automatic dynamic adjustment of weights and automatic addition / removal of correction terms, eliminating the reliance on manual weight setting and correction term selection, and ensuring the consistency and reproducibility of model development. By automatically executing iterative calculation tasks, the time cost and manpower investment in model calibration are reduced, calibration efficiency is improved, and the R&D cycle is shortened.
Need to check novelty before this filing date? Find Prior Art