Unlock instant, AI-driven research and patent intelligence for your innovation.

Virtual metrology method for multi-stage semiconductor process based on Gaussian process and convolutional neural network

A technology of convolutional neural network and Gaussian process, which is applied in the field of probabilistic virtual metrology of multi-stage semiconductor process, can solve the problems of unreliable predicted value and model cannot reflect the credibility of predicted value, etc., and achieve the effect of reducing calculation load

Active Publication Date: 2021-11-19
ZHEJIANG UNIV
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] On the other hand, past methods can only obtain scalar predictions of outcomes, which means that the model cannot reflect the confidence of the predicted values
In theory, the higher the uncertainty, the less reliable the predicted value at this time

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Virtual metrology method for multi-stage semiconductor process based on Gaussian process and convolutional neural network
  • Virtual metrology method for multi-stage semiconductor process based on Gaussian process and convolutional neural network
  • Virtual metrology method for multi-stage semiconductor process based on Gaussian process and convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0059] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be noted that the following embodiments are intended to facilitate the understanding of the present invention, but do not limit it in any way.

[0060] Taking the estimation of the deposition process results of a domestic factory as an example, the virtual measurement of the wafer height value through the multi-stage chemical process is carried out.

[0061] During production, the chemical vapor deposition process is similar to the process of applying solid thin film coatings on surfaces often used in the semiconductor industry. This process is complex because it involves many chemical reactions, and the reactors in a multi-reactor system are independently controlled to allow the film to be deposited in the process chamber under various conditions. Chemical vapor deposition equipment is equipped with a considerable number of sensors. Th...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a multi-stage semiconductor process virtual metering method based on a Gaussian process and a convolutional neural network, which comprises the following steps: (1) collecting the output signal of a process variable sensor related to the variable to be detected for the control process to be detected; (2) Preprocess the signal data of the collected process variables to remove outliers; (3) Rearrange the preprocessed data and retain the stage information of the data; (4) Extract features from all data and establish a regression model ; (5) save the current parameter weights, calculate the final maximum posterior value, if the stop condition is not satisfied, update the parameters and repeat step (4) until the stop condition is reached; (6) save the parameter values ​​of each layer, for the new prediction point Perform the recalculation and obtain the probability distribution of the geometric mass. Utilizing the present invention, higher-precision virtual measurement results can be obtained, and the uncertainty of prediction results can be calculated to provide a numerical basis for further improvement of the model.

Description

technical field [0001] The invention relates to the field of data mining in industrial systems, in particular to a multi-stage semiconductor process probabilistic virtual metering method based on Gaussian processes and convolutional neural networks. Background technique [0002] Semiconductor manufacturing involves many stages. For example, in the production of electronic chips, a wire saw first cuts a silicon ingot into segments, then goes through several flat stages, including cleaning, polishing and grinding, and then transfers the treated wafer to the front and back where the final chip is formed. end process. Due to the high-throughput nature of semiconductor manufacturing and the high cost of measuring wafers, it is not possible to measure all quality variables of produced wafers at every stage. Due to the limitations of physical measurements, wafer-to-wafer modeling is increasingly being used to predict the quality of the final product, allowing timely adjustment of...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F30/27G06N3/04G06N3/08
CPCG06N3/08G06N3/048G06N3/045
Inventor 谢磊吴小菲陈启明苏宏业
Owner ZHEJIANG UNIV