Digital Reconstruction of Composite Microstructure via Geometric Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for digital reconstruction of representative elementary volumes (REV) of composite materials, particularly those with long discontinuous fibers, face challenges in achieving high filling rates and accurately predicting mechanical properties due to randomness in fiber entanglement and geometry, leading to variability in material performance.
Innovation Solution
A method for digital reconstruction that involves defining an elementary volume and filling it with digital elements that can adapt geometrically to each other and the volume's walls, allowing for deformations beyond inclination, such as rotation, to achieve high fiber volume filling rates close to 100%, enabling more realistic representation of composite materials' microstructures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If random drawing of rigid geometric shapes is used to model fiber elements, then the implementation is quick, but the filling rate is very low (not exceeding 40%)
Solution Approach 1:
The patent applies the dynamics principle by allowing fiber elements to rotate and reorient themselves during the random drawing process. Instead of rigid fixed-orientation shapes, the fiber elements can change their orientation dynamically to achieve higher packing density, resolving the contradiction between quick implementation and high filling rate.
Solution Approach 2:
The patent changes the parameter of orientation freedom during the random drawing process. By allowing fiber elements to rotate and adjust their orientation angles randomly rather than being fixed, the system achieves both computational efficiency and high filling rates (exceeding 80%), resolving the contradiction between speed and quantity.
2Quantity of substance
If additional prints of equivalent geometric shapes with gradually reduced size are used, then the filling rate increases to around 80%, but the technique is not relevant for composite materials with long discontinuous fibers of similar sizes
Solution Approach 1:
The patent applies universality by creating a single random drawing algorithm that can handle various fiber configurations including long discontinuous fibers of similar sizes. The algorithm universally handles different fiber length distributions and orientations without requiring separate techniques, making it adaptable to real composite materials while maintaining high filling rates.
Solution Approach 2:
The dynamic orientation adjustment allows the model to accommodate long discontinuous fibers of similar sizes by randomly orienting them in three-dimensional space. This dynamic approach naturally handles the size similarity constraint while achieving high filling rates, resolving the contradiction between filling rate and adaptability.
3Manufacturing precision
If fiber elements are positioned to achieve high filling rates, then the representation of reality improves, but the randomness of microstructure makes prediction difficult
Solution Approach 1:
The patent applies preliminary action by pre-defining statistical distribution laws for fiber element positions and orientations before the random drawing process. This allows the generation of microstructures that are both representative of real materials and suitable for reliable predictions, as the randomization is controlled within predefined statistical frameworks that capture essential material characteristics.
Solution Approach 2:
The patent uses parameter changes by statistically distributing fiber element orientations and positions according to controlled probability laws. This allows the microstructure to represent real material randomness while maintaining predictable statistical characteristics, resolving the contradiction between representation accuracy and prediction reliability.
Data Source
Figure 1~4
Figure 5
Figure 6A~6B
AI summary
The invention relates to a method and device for digital reconstruction of an elementary volume representing a microstructure of composite material. The method includes: defining (F10) an elementary volume; filling the volume with digital elements modelling elements of fibres of the composite material, and extending longitudinally along a main axis, which includes: associating (F30) with each element a position in a plane of the space and an orientation of the main axis thereof; and consecutively positioning each element in the volume, in accordance with the position and the orientation which have been associated with same, said positioning including placing the element in contact (F40) with at least one wall of the elementary volume and/or at least one previously positioned element, and geometrically adapting (F50) the element to said at least one wall and/or to said at least one previously positioned digital element with which same is placed in contact, at least one portion of an element used to fill the volume undergoing, during the geometric adaptation, a deformation other than a tilting of the longitudinal axis thereof relative to the main axis of said element.