Plasma Processing Control Using State-Based Prediction Model Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing plasma processing technologies face challenges in accurately predicting processing result indices due to changes in the state of the plasma processing apparatus, leading to decreased prediction accuracy when using a single prediction model.
Innovation Solution
A plasma processing apparatus and system that includes a control unit capable of selecting from multiple prediction models based on the state of the plasma processing unit, using spectroscopic data to adjust plasma processing conditions and improve prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single prediction model is used to predict processing result indices, then the device complexity is reduced, but the prediction accuracy decreases when the plasma processing apparatus state changes
Solution Approach 1:
The patent divides the single prediction model into multiple segmented prediction models, each optimized for specific plasma processing apparatus states. This segmentation allows the system to select the appropriate model based on current conditions, maintaining high prediction accuracy without requiring an overly complex unified model.
Solution Approach 2:
The patent implements a dynamic model selection mechanism that adapts the prediction system to changing plasma processing apparatus states. By dynamically switching between different prediction models based on real-time state monitoring, the system maintains accuracy without the complexity of a single static model.
2Measurement precision
If multiple prediction models are used to maintain prediction accuracy, then the prediction accuracy is improved, but the device complexity increases
Solution Approach 1:
The system dynamically selects among multiple prediction models based on the current plasma processing apparatus state, using state monitoring data to determine which model is most appropriate. This dynamic approach maintains high prediction accuracy while avoiding the complexity of simultaneously managing all models.
Solution Approach 2:
The patent changes the parameter of model selection based on plasma processing state parameters. By monitoring state parameters and selecting models accordingly, the system achieves high accuracy without the complexity of a fixed multi-model structure.
3Ease of operation
If the prediction model is fixed, then the ease of operation is improved, but the adaptability to plasma processing apparatus state changes deteriorates
Solution Approach 1:
The system automatically adapts to plasma processing apparatus state changes by dynamically selecting the appropriate prediction model, eliminating the need for manual intervention. This maintains ease of operation while achieving high adaptability through automated model switching based on state monitoring.
Solution Approach 2:
The prediction system performs self-adjustment by automatically selecting the most appropriate model based on monitored plasma processing states, without requiring external control or reconfiguration. This self-service capability maintains operational simplicity while achieving adaptability.
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
This approach allows for accurate prediction and control of processing result indices by switching between prediction models, thereby maintaining high accuracy even with changes in the plasma processing apparatus state, reducing variation in control results.
Implementation Method 1
data obtained by measuring light emission by plasma
Data Source
AI summary
The plasma processing apparatus includes a plasma processing unit that performs plasma processing of a sample and a control unit that controls the plasma processing. The control unit selects one of a plurality of the prediction models for predicting a result of the plasma processing based on a state of the plasma processing unit, and predicts the result of the plasma processing by using a selected prediction model.