Predicting Fiber Composite Shape Before Deformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current design technologies cannot accurately predict the material shape before deformation processing for thick fiber-reinforced composite materials with multiple layers, leading to discrepancies in shape and strength between the final product and the 3-dimensional shape model.
Innovation Solution
A design support apparatus and method that creates a predicted shape model by separating a 3-dimensional shape model into fiber layers, setting a correspondence relationship, and generating an orientation vector field to develop the shape on a flat surface, allowing for accurate prediction of the material shape before deformation processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the technology in PTL 1 is used to predict the shape of woven fabric, then the surface shape of flat woven fabric can be predicted, but the shape changes of thick woven fabric with multiple fiber layers cannot be predicted
Solution Approach 1:
The invention divides the thick fiber-reinforced composite material into multiple fiber layers (first fiber layer, second fiber layer, etc.). Each layer is processed independently to calculate its contribution to the overall shape change, then the results are integrated. This segmentation allows the system to handle complex thick structures by breaking them down into manageable individual layers that can be analyzed separately and combined.
Solution Approach 2:
The invention extends the prediction methodology from two-dimensional flat surface prediction to three-dimensional thick structure prediction by adding the thickness dimension. The system calculates shape changes not only on the surface but also through the entire thickness of the composite material by processing multiple fiber layers at different positions (first fiber layer at first position, second fiber layer at second position, etc.).
2Measurement precision
If the technology in PTL 1 is used to predict the shape after weaving, then the product shape can be predicted from the organizational chart, but the material shape before deformation processing cannot be predicted
Solution Approach 1:
The invention inverts the conventional prediction direction. Instead of predicting the product shape from the material shape (as in PTL 1), it predicts the material shape before deformation from the product shape after deformation. This reverse prediction is achieved by calculating the deformation amount and applying it in reverse to the product shape model, thereby recovering the original material shape information that would otherwise be lost.
3Ease of operation
If expert skills and know-how are used to predict the material shape, then predictions can be made, but discrepancies in shape and strength occur between the actual product and the 3-dimensional shape model
Solution Approach 1:
The invention replaces the subjective expert judgment system with an objective computational system. Instead of relying on expert skills and know-how to estimate the material shape, the system uses automated calculations based on the organizational chart, fiber layer configurations, and deformation mechanics to compute the predicted shape model. This substitution eliminates human subjectivity and provides consistent, reproducible results with quantifiable accuracy.
Solution Approach 2:
The invention introduces a computational model as an intermediary between the organizational chart and the final shape prediction. This intermediary process systematically processes the organizational chart data, calculates deformation amounts for each fiber layer, and generates the predicted shape model. This intermediary computation layer ensures that the transition from input data to output prediction follows consistent mechanical principles rather than subjective expert judgment.
Data Source
AI summary
A design support apparatus supporting designing of a product which uses a fiber material includes a processor that creates a predicted shape model by predicting a shape of the product before a deformation processing. The processor: creates a 3-dimensional shape model of the product; creates curved shape models by separating the 3-dimensional shape model into two or more fiber layers; sets a correspondence relationship between the curved shape models; creates an orientation vector field in the curved shape models; and predicts the shape of the product before the deformation processing by developing the curved shape models on a flat surface based on the correspondence relationship between the curved shape models and the orientation vector field in the curved shape models, and creates the predicted shape model based on the predicted shape.