AI-Based Crimp Connection Quality Control From Process Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for evaluating crimp connections are inefficient and require extensive manual sampling and documentation, lacking advanced monitoring capabilities to ensure consistent quality.
Innovation Solution
Implementing AI and machine learning models to analyze process data from crimping operations, using sensors to determine mechanical properties and classify crimp connections, enabling real-time quality assessment and control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional sampling and visual inspection methods are used to assess crimp connection quality, then manual documentation and cross-sectional imaging are required, but this increases monitoring effort and reduces productivity
Solution Approach 1:
The patent replaces manual sampling, visual inspection, and physical cross-sectional analysis with an automated optical measurement system that uses image processing and evaluation algorithms to assess crimp connection quality, thereby eliminating manual documentation work while maintaining reliable quality assessment
Solution Approach 2:
The patent creates digital copies (images) of the crimp connection region and processes these copies through evaluation algorithms to determine quality characteristics, replacing the need for physical cross-sectional imaging and manual sample analysis
2Reliability
If extensive sampling and cross-sectional imaging are performed for quality documentation, then quality verification is thorough, but this increases device complexity and monitoring effort
Solution Approach 1:
The patent extracts only the essential quality-determining features from crimp connection images using evaluation algorithms, rather than requiring complete cross-sectional imaging and analysis of all sample characteristics, thereby simplifying the monitoring system while maintaining verification thoroughness
Solution Approach 2:
The patent introduces an intermediate evaluation layer that processes images and extracts quality characteristics through algorithms, serving as a mediator between raw imaging data and final quality decisions, thereby reducing the complexity of direct monitoring system design
3Reliability
If crimp force monitoring with limit checks is used during production, then quality control is maintained, but this lacks advanced monitoring capabilities for predictive intervention
Solution Approach 1:
The patent implements a feedback system where measured crimp connection characteristics are compared against reference values or target ranges, and this feedback information is used to adjust process parameters or trigger alerts, enabling continuous quality improvement and predictive intervention
Solution Approach 2:
The patent uses measured quality characteristics to predict potential quality issues before they occur by comparing current measurements against historical data and reference values, enabling preliminary corrective actions to be taken before defects manifest
4Reliability
If manual sampling and testing are performed for each new order and machine setting, then quality verification is comprehensive, but this increases loss of time and reduces productivity
Solution Approach 1:
The patent enables continuous quality measurement and evaluation during normal production operations rather than requiring separate sampling and testing steps, thereby maintaining comprehensive quality verification while eliminating the time loss associated with interrupting production for quality checks
Data Source
Figure 1
AI summary
A method for evaluating and controlling the quality of crimp connections is provided, wherein process data is acquired during a crimping operation using one or more sensors, and at least one model, in particular a machine learning model, is provided that simulates the execution of a crimping operation, combining the process data with the model. Based on this combination, at least one mechanical property and/or a classification of the crimp connection is determined.