Plasma Emission Virtual Sensing for Chamber State Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Installing physical sensors in substrate processing apparatuses to measure chamber conditions is technically challenging and costly, increasing manufacturing costs and complexity.
Innovation Solution
A substrate processing system that utilizes a virtual sensor to predict chamber states based on plasma emission data, using a trained model to estimate conditions like residual moisture content and parts wear rate without requiring additional hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical sensors are installed in the chamber to measure chamber conditions, then measurement precision is improved, but device complexity and manufacturing cost increase
Solution Approach 1:
The patent creates a virtual copy of the physical sensor measurement capability through a trained prediction model. Instead of installing physical sensors in the chamber, the system uses plasma emission data (which is already being collected for process monitoring) as input to a machine learning model that predicts chamber conditions such as residual moisture content and parts wear rate. This virtual sensor approach replicates the measurement function without the physical hardware complexity.
Solution Approach 2:
The patent replaces the mechanical/physical sensor installation system with an information-processing system. Rather than using physical sensors that require installation, calibration, and maintenance in the chamber, the system substitutes this with a software-based prediction model that processes plasma emission spectra data to infer chamber conditions, thereby eliminating the need for complex physical sensing infrastructure.
2Measurement precision
If physical sensors are installed in the chamber to measure chamber conditions, then measurement precision is improved, but manufacturing cost increases
Solution Approach 1:
The patent creates a virtual copy of the physical sensor measurement capability through a trained prediction model. Instead of installing physical sensors in the chamber, the system uses plasma emission data (which is already being collected for process monitoring) as input to a machine learning model that predicts chamber conditions such as residual moisture content and parts wear rate. This virtual sensor approach replicates the measurement function without the physical hardware complexity.
Solution Approach 2:
The patent employs a software-based prediction model that can be deployed and updated without physical hardware installation. This software solution is significantly cheaper than installing, calibrating, and maintaining physical sensors in the chamber environment, while providing comparable measurement capabilities for chamber condition monitoring.
3Device complexity
If plasma emission data is used to predict chamber state with a trained model, then device complexity is reduced, but measurement precision may be compromised
Solution Approach 1:
The patent performs preliminary training of the prediction model using chamber state data collected under various conditions before deployment. The model is trained offline using historical data that includes plasma emission spectra and corresponding actual chamber conditions (measured by physical sensors during calibration phases). This preliminary training ensures the model achieves high prediction accuracy before being used in production, eliminating the need for complex real-time sensor systems while maintaining measurement precision.
Solution Approach 2:
The patent implements a feedback mechanism where the prediction model continuously refines its accuracy by comparing predicted chamber states with actual measurements during periodic calibration phases. The system uses feedback from physical sensor measurements (when available) to update and retrain the prediction model, ensuring that measurement precision is maintained or improved over time while keeping the operational system simple.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables efficient and cost-effective estimation of chamber conditions, allowing for timely maintenance and process optimization without the need for dedicated sensors.
Implementation Method 1
acquire plasma emission data inside a chamber
Data Source
AI summary
A substrate processing system acquires plasma emission data inside a chamber, and predicts a state inside the chamber by inputting the acquired plasma emission data to a trained model that has learned a relationship between the plasma emission data and information indicating the state inside the chamber.