Detecting an excursion of a CMP component using time-based sequence of images and machine learning
Monitoring operations of a polishing system includes obtaining a time-based sequence of reference images of a component of the polishing system performing operations during a test operation of the polishing system, receiving from a camera a time-based sequence of monitoring images of an equivalent component of an equivalent polishing system performing operations during polishing of a substrate, determining a difference value for the time-based sequence of monitoring images by comparing the time-based sequence of reference images to the time-based sequence of monitoring image using an image processing algorithm, determining whether the difference value exceeds a threshold, and in response to determining the difference value exceeds the threshold, indicating an excursion.
1. A polishing system, comprising:
a platen to support a polishing pad;
a carrier head to hold a substrate against the polishing pad;
a component selected from a group of components including the platen, the carrier head, a conditioner arm, a load cup, or a robot arm;
a camera positioned to capture a time-based sequence of monitoring images of the component; and
a controller configured to
receive from the camera a time-based sequence of monitoring images of the component performing operations during polishing of a substrate, and
analyze the time-based sequence of monitoring images with a machine learning model trained by training examples to detect and indicate excursions of the component from expected operations, wherein the training examples comprise a time-based sequence of reference images of a reference component of a reference polishing system performing operations during a test operation, and
indicate an excursion based on the analysis.
2. The system of claim 1 , wherein the machine learning model is configured to, using an object tracking algorithm, determine motions of the component based on input of the time-based sequence of monitoring images.
3. The system of claim 1 , wherein the machine learning model is configured to predict a physical field profile of the component based on input of the time-based sequence of monitoring images.
4. The system of claim 3 , wherein the physical field profile of the component comprises: a temperature field of a rotating polishing pad, or a pressure field applied on a wafer.
5. The system of claim 3 , wherein the excursion comprises departure of the component from an expected path of motion.
6. The system of claim 1 , wherein the controller is configured to generate an alarm in response to an indication of an excursion.
7. The system of claim 1 , further comprising a display, and wherein the controller is configured to generating a user presentation on the display, wherein the user presentation comprises the time-based sequence of monitoring images and one or more overlays each representing a respective characteristic of the time-based sequence of monitoring images.
8. A computer program product, tangibly embodied in computer-readable media, comprising instructions to cause one or more computers to:
receive from a camera a time-based sequence of monitoring images of a component of a polishing system performing operations during polishing of a substrate; and
analyze the time-based sequence of monitoring images with a machine learning model trained by training examples to detect and indicate excursions of the component from expected operations, wherein the training examples comprise a time-based sequence of reference images of a reference component of a reference polishing system performing operations during a test operation.
9. The computer program product of claim 8 , wherein the machine learning model is configured to, using an object tracking algorithm, determine motions of the component based on input of the time-based sequence of monitoring images.
10. The computer program product of claim 8 , wherein the machine learning model is configured to predict a physical field profile of the component based on input of the time-based sequence of monitoring images.
11. The computer program product of claim 10 , wherein the physical field profile of the component comprises: a temperature field of a rotating polishing pad, or a pressure field applied on a wafer.
12. The computer program product of claim 8 , further comprising instructions to:
receive from a camera a time-based sequence of monitoring images encompassing a plurality of components of the polishing system performing operations during polishing of a substrate;
input the time-based sequence of monitoring images analyzing into a machine learning model trained by training examples to detect excursions of the plurality of components from expected operations; wherein the training examples comprise a time-based sequence of reference images of a plurality of reference components of a reference polishing system performing operations during a test operation, wherein the plurality of reference components include at least components equivalent to the plurality of components; and
receive from the machine learning model an indication of an excursion of the component from expected operation.
13. The computer program product of claim 8 , wherein the excursion comprises departure of the component from an expected path of motion.
14. The computer program product of claim 8 , wherein the excursion comprises an overspray of fluid from a fluid dispenser.
15. The computer program product of claim 8 , wherein the excursion comprises a wafer slippage when undergoing polishing.
16. The computer program product of claim 8 , comprising instructions to generate an alarm in response to an indication of an excursion.
17. The computer program product of claim 8 , further comprising instructions to generate a user presentation on a user interface component, wherein the user presentation comprises the time-based sequence of monitoring images and one or more overlays each representing a respective characteristic of the time-based sequence of monitoring images.
18. A method for monitoring operation of a polishing system, the method comprising:
receiving from a camera a time-based sequence of monitoring images of a component of a polishing system performing operations during polishing of a substrate;
inputting the time-based sequence of monitoring images analyzing into a machine learning model trained by training examples to detect excursions of the component from expected operations, wherein the training examples comprise a time-based sequence of reference images of a reference component of a reference polishing system performing operations during a test operation and
receiving from the machine learning model an indication of an excursion of the component from expected operations.
19. The method of claim 18 , wherein the excursion comprises departure of the component from an expected path of motion.
20. The method of claim 19 , wherein the component comprises one of a carrier head, a conditioner arm, a load cup, a platen or a robot arm.