Manufacturing Process Estimation Using Object-State Regression Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing manufacturing processes for materials like Nd—Fe—B sintered magnets and ferrite sintered magnets face challenges in achieving stable production due to variations in semi-finished products across multiple processes, leading to inconsistent quality, as conventional methods do not adequately consider the state of the processing object before processing.
Innovation Solution
A process estimation device and method using machine learning to create a regression model that correlates processing object data, processed object data, device condition data, and process data, allowing for more stable manufacturing by estimating process parameters based on the state of the processing object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional process adjustment methods are used based on manufacturing conditions and control values, then some process control is achieved, but manufacturing stability and quality consistency deteriorate due to insufficient consideration of processing object state variations
Solution Approach 1:
The system performs preliminary analysis of the processing object state before processing by extracting features from images and spectral data. This preliminary characterization enables the regression model to predict optimal process parameters in advance, ensuring consistent quality outcomes while accounting for variations in the processing object state that conventional methods overlook.
Solution Approach 2:
The system implements a feedback mechanism where the actual state of the processing object (captured through imaging and spectral analysis) is fed into the regression model to adjust process parameters dynamically. This closed-loop approach ensures that manufacturing conditions are continuously optimized based on real-time object state, improving both quality consistency and manufacturing stability.
2Adaptability or versatility
If multiple manufacturing processes are used to create complex materials, then product functionality is improved, but process variation and quality inconsistency increase
Solution Approach 1:
The regression model serves as a universal tool that can handle multiple manufacturing processes and material types. By training the model on diverse process data and object state variations across different processes, it learns generalized relationships that enable consistent quality prediction and control regardless of the specific process being used, thereby maintaining quality consistency while supporting process versatility.
Solution Approach 2:
The system dynamically adjusts process parameters based on the processing object state predicted by the regression model. For each manufacturing process, the model identifies optimal parameter settings that account for object variations, enabling consistent quality outcomes across multiple processes by adapting parameters rather than using fixed settings.
3Manufacturing precision
If detailed analysis of processing object state is performed, then quality control is improved, but measurement and detection difficulty increases
Solution Approach 1:
The system introduces imaging devices and spectral analysis tools as intermediaries to objectively capture and quantify processing object state. These intermediaries transform complex physical and chemical properties into measurable data (images, spectra) that can be processed by the regression model, thereby improving quality control while reducing the difficulty of direct measurement and analysis.
Solution Approach 2:
The system replaces manual or conventional mechanical measurement methods with optical and spectral analysis. By using imaging devices and spectral data to characterize the processing object state, the system achieves detailed quality analysis without the complexity and subjectivity associated with traditional measurement approaches, thereby improving quality control while simplifying detection.
Data Source
AI summary
A process estimation device using a computer is provided with a regression model creation processing unit that, in manufacturing a product through a plurality of manufacturing processes, defines a given manufacturing process, excluding a first manufacturing process, as a target process to be subject to process estimation, learns a relationship between at least a processing object data indicating a state of a processing object to be processed in the target process, a processed object data indicating a state of a processed object processed in the target process, a device condition data indicating a state of a device used for processing in the target process before processing, and a process data indicating a set value of manufacturing conditions of the target process by machine learning, and creates a regression model representing a correlation between the data, and a process estimation processing unit that estimates the process data to be estimated using the regression model created by the regression model creation processing unit. A computer-performed process estimation method includes defining the manufacturing process as the target, learning the relationship between at least the processing object data, the device condition data, the process data, creating the regression model, and estimating the process data using the regression model.