Dicing Machining Error Prediction From Temperature History Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing dicing apparatuses face challenges in maintaining machining accuracy across varying temperature environments, as temperature factors like room temperature, blade supply water temperature, and heat source supply water temperature are often interrelated, making it difficult to implement accurate corrections.
Innovation Solution
A machining state detection apparatus that acquires and processes temperature history data from these sources to predict machining errors using a learning model, allowing for informed corrections based on environmental and operational temperature changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If correction maps are used for each parameter independently, then correction can be performed for individual parameters, but it becomes difficult to cope with cases where temperature factors are related to each other
Solution Approach 1:
The patent combines multiple independent correction maps (spindle displacement correction map, work table displacement correction map, and blade position correction map) into an integrated correction system. The correction value calculating unit simultaneously accesses and applies corrections from all three maps based on the current temperature, enabling coordinated correction of related temperature factors while maintaining manageable system complexity through modular map structures
Solution Approach 2:
The correction system is designed to handle multiple types of temperature-related errors universally through a single framework. The same correction maps and calculating unit that correct spindle displacement also correct work table displacement and blade position, making the system adaptable to various temperature factor interactions without requiring separate correction mechanisms for each parameter
2Manufacturing precision
If multiple temperature parameters are monitored and corrected, then machining accuracy under varying temperature conditions improves, but the complexity of temperature management increases
Solution Approach 1:
The patent segments temperature correction into three distinct correction maps, each handling a specific component's thermal behavior: spindle displacement correction map for the spindle, work table displacement correction map for the work table, and blade position correction map for the blade. This segmentation allows each map to be optimized independently while collectively addressing multiple temperature parameters, thereby improving machining accuracy without overwhelming complexity
Solution Approach 2:
The correction maps are pre-calculated and stored based on advance temperature measurements and error data. During machining operations, the correction value calculating unit simply retrieves the appropriate pre-computed correction values from the maps based on the current temperature, avoiding complex real-time calculations. This preliminary preparation maintains high machining accuracy while keeping the operational system simple and efficient
Data Source
AI summary
A machining state detection apparatus acquires an environmental temperature history, a blade supply water temperature history, and a heat source supply water temperature history, and derives an environmental temperature history feature amount, a blade supply water temperature history feature amount, and a heat source supply water temperature history feature amount, and when a machining error is predicted based on a temperature history feature amount including the environmental temperature history feature amount, the blade supply water temperature history feature amount and the heat source supply water temperature history feature amount, with a learning model which is trained using the temperature history feature amount and the machining error as learning data and which outputs a prediction value of the machining error, when the temperature history feature amount is input.