Nozzle Speed Learning for Stable Substrate Treatment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing substrate treatment apparatuses require redevelopment of speed profiles for nozzle movement during etching treatments due to changes in pretreatment processes, which burdens operators with additional development tasks.
Innovation Solution
A substrate treatment apparatus equipped with a nozzle, moving mechanism, storage portion, and control portion that uses a learned model generated from learning target speed information and treatment data to control the nozzle movement, allowing for adaptive treatment based on pre-learned patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a predefined speed profile is used for nozzle movement during etching treatment, then the treatment process is simple to operate, but the system lacks adaptability when pretreatment processes change
Solution Approach 1:
The system performs preliminary learning by executing etching treatments with various predefined speed profiles and measuring the resulting film thickness. The learned relationship between speed profiles and treatment outcomes is stored in advance, enabling the system to adapt to different pretreatment conditions without requiring operators to manually redevelop speed profiles when processes change.
2Manufacturing precision
If the speed profile is redeveloped whenever pretreatment changes, then the treatment precision is maintained, but the operator burden increases
Solution Approach 1:
The system performs self-learning by automatically executing treatments with multiple speed profiles, measuring results, and building its own knowledge base of relationships between nozzle speeds and treatment outcomes. This eliminates the need for operators to manually adjust speed profiles when pretreatment changes, maintaining precision while reducing their burden.
3Adaptability or versatility
If manual speed profile development is performed, then the treatment process is flexible, but the productivity decreases
Solution Approach 1:
The system replaces the manual mechanical process of speed profile development with an automated learning system. The control unit automatically executes treatments, measures results, and computes optimal speed profiles, eliminating the need for manual operator intervention and significantly improving productivity while maintaining adaptability.
Data Source
AI summary
A substrate treatment apparatus includes a nozzle, a moving mechanism, a storage portion, and a control portion. The learned model is generated by learning, as learning data, learning target speed information indicating a moving speed of the nozzle and the amount of treatment acquired by executing a treatment on a substrate that is a learning target while causing the nozzle to move at a speed based on the learning target speed information. The control portion causes speed information at the time of treatment to be outputted from the learned model by inputting a target amount of an amount of treatment to the learned model. The control portion controls a moving mechanism such that the nozzle moves at a speed based on the speed information at the time of treatment when the treatment is executed on a substrate that is a treatment target.