Substrate Processing Sensor Models for Mixed Sampling Rates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing substrate processing systems face challenges in efficiently creating and managing models for various edge devices due to differences in sensor sampling periods, leading to ineffective data processing and model management.
Innovation Solution
An information processing method that acquires time series data from sensors with different sampling periods, performs individual learning for each sensor, and uses the learned models to output estimation results, enabling efficient creation and management of inference models for edge devices in the substrate processing system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If time series data from sensors with different sampling periods are processed using a unified model, then model management is simplified, but data processing effectiveness deteriorates due to incompatible sampling rates
Solution Approach 1:
The patent divides the model management system into separate first learning models, each dedicated to processing time series data from a specific sensor type with its own sampling period. This segmentation allows each model to be optimized for its specific sensor's data characteristics while maintaining overall system manageability through modular architecture.
2Reliability
If individual learning models are created for each sensor type with different sampling periods, then data processing effectiveness is improved, but device complexity increases due to multiple models
Solution Approach 1:
The patent applies local quality by creating specialized first learning models tailored to each sensor type's specific characteristics and sampling period. Each model is locally optimized for its designated sensor, ensuring high processing effectiveness while the overall system maintains structure through this targeted specialization approach.
3Speed
If data from sensors with different sampling periods are aggregated without individual processing, then processing speed is maintained, but measurement precision deteriorates due to data incompatibility
Solution Approach 1:
The patent segments the data processing workflow into separate first learning models for each sensor type, allowing each model to process its specific sampling rate data at optimal speed while maintaining high measurement precision through specialized processing. The segmentation prevents data incompatibility issues while preserving processing efficiency.
Data Source
AI summary
To provide an information processing method, an information processing apparatus, and a substrate processing system. Acquiring time series data from a plurality of types of sensors having different sampling periods provided in a substrate processing apparatus, performing learning of first learning models that output information relating to the substrate processing apparatus in a case where the time series data from the sensors are input, using each of the pieces of time series data having different sampling periods for each of the sensors individually, and inputting the time series data from the sensors into the corresponding first learning models after learning to output an estimation result based on information obtained from the first learning models are included.