Bonded Structure Failure Prediction Using Variable Element Dimensions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting failure initiation and propagation in bonded structures, such as the virtual crack closure technique (VCCT), require manual analysis and extensive engineering judgment, are labor-intensive, and lack the ability to predict the onset of failure without assuming a known damage state.
Innovation Solution
A method using finite element analysis models with different characteristic dimensions for analysis elements to predict failure initiation and propagation, where the bonding layer is modeled as a cohesive element, allowing for the scaling of material parameters based on element lengths to reduce computational resources and time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the virtual crack closure technique (VCCT) is used to perform failure analysis on bonded structures, then failure propagation can be determined based on fracture mechanics, but the method cannot predict the onset of failure and requires extensive manual analysis and engineering judgment
Solution Approach 1:
The bonding layer is divided into multiple analysis elements with different characteristic dimensions. A first portion uses smaller elements for failure initiation analysis, while a second portion uses larger elements for failure propagation analysis. This segmentation allows the system to predict both onset and propagation of failure without requiring extensive manual intervention at each stage.
Solution Approach 2:
The method changes the characteristic dimension parameter of analysis elements based on the stage of failure being analyzed. By using smaller characteristic dimensions for initiation analysis and larger dimensions for propagation analysis, the system automatically transitions between different failure stages, reducing the need for manual analysis sweeps and engineering judgment.
2Reliability
If deterministic analysis sweeps are used to create response surfaces and design spaces, then failure analysis can be performed, but the process becomes extremely manual and labor-intensive
Solution Approach 1:
The analysis method transitions from static, manual determination of failure locations to a dynamic approach where the failure state evolves naturally through the simulation. The system automatically progresses from initiation to propagation by changing material parameters and characteristic dimensions, eliminating the need for labor-intensive manual analysis sweeps while maintaining reliability.
Solution Approach 2:
The finite element analysis model performs self-service by automatically determining failure initiation and propagation without requiring continuous manual intervention. The system uses different characteristic dimensions and material parameters for different portions of the bonding layer to autonomously identify failure locations and progression, significantly improving analysis efficiency.
3Measurement precision
If small characteristic dimensions are used for analysis elements to accurately predict failure initiation, then prediction accuracy improves, but computational resources and time requirements increase significantly
Solution Approach 1:
The bonding layer is segmented into a first portion for failure initiation analysis and a second portion for failure propagation analysis. Only the first portion uses small characteristic dimensions requiring intensive computation, while the second portion uses larger dimensions that reduce computational burden. This segmentation maintains accuracy where needed while reducing overall computational time and resource requirements.
Solution Approach 2:
Different characteristic dimensions are applied to different portions of the bonding layer based on local requirements. The first portion near potential failure initiation points uses small elements for high accuracy, while the second portion uses larger elements for efficient propagation analysis. This local differentiation optimizes the balance between prediction accuracy and computational efficiency.
Data Source
AI summary
A method includes obtaining failure initiation characteristics of a bonding layer of one or more bonded structures and determining, based on the failure initiation characteristics, a first characteristic dimension for each analysis element of a first portion of a finite element analysis model. The method includes obtaining failure propagation characteristics of the bonding layer and determining, based on the failure propagation characteristics, a second characteristic dimension for each analysis element of a second portion of the model. The method includes assigning a first set of material parameters to analysis elements of the first portion of the model and assigning a second set of material parameters to analysis elements of the second portion of the model. The method includes evaluating failure modes of the one or more bonded structures based on a solution to the model, the first set of material parameters, and the second set of material parameters.