首页>
外国专利>
MACHINE LEARNING-BASED METHOD FOR DETERMINING FOCAL PLANE POSITION OF PHOTOLITHOGRAPHY SYSTEM
MACHINE LEARNING-BASED METHOD FOR DETERMINING FOCAL PLANE POSITION OF PHOTOLITHOGRAPHY SYSTEM
展开▼
机译：基于机器学习的光刻系统焦平面位置的基于机器学习方法
展开▼
页面导航
摘要
著录项
相似文献
摘要
A machine learning-based method for determining a focal plane position of a photolithography system, comprising the following steps: S1, selecting a measurement mark of a focal plane position, performing exposure in a calibrated photolithography system at different known focal plane positions, recording the positions and using a photolithography aerial image sensor to acquire and record two-dimensional distributions of photolithography aerial images; S2, dividing valid results of the photolithography aerial images into a training set and a test set; S3, using a machine learning algorithm and the training set to establish a focal plane position detecting model; S4, using results in the test set to test the detecting model, if the test passes, going to the next step, and if the test fails, going back to S2; and S5, measuring the variation of the focal plane position of a photolithography system to be measured and feeding back to a control system. The method eliminates the negative effects of a measurement mark manufacturing error and three-dimensional diffraction effects on an internal mathematical model, and a measurement signal is upgraded from one-dimensional information of a graphical position offset to two-dimensional information of a graphical distribution. Thus, the precision and use range of the focal plane position measurement technique are improved.
展开▼
