Tian Ye

Bio: Tian Ye is an academic researcher from Huazhong University of Science and Technology. The author has contributed to research in topics: Interpolation & Fiber (mathematics). The author has an hindex of 3, co-authored 10 publications receiving 22 citations.

Shepard Interpolation Based on Geodesic Distance for Optimization of Fiber Reinforced Composite Structures with Non-Convex Shape

Abstract: In the present paper, we propose to define the Shepard interpolation by using the geodesic distance. The geodesic distance between a pair of points is the length of the shortest geodesic line, and geodesic line is the generalization of straight line in the Euclidean geometry to general spaces, for instance 3D surfaces. When the shape of structure is non-convex, the geodesic distance is more reasonable than the Euclidean distance to define influence domain for interpolation points. In view of the fact that the length of a geodesic line can be considered as the time it takes to go with a certain velocity from one point to another, the fast marching method is used to compute the geodesic distance. In the design optimization of fiber reinforced composite structures, the proposed Shepard interpolation based on geodesic distance is used to construct a continuous global function that represents the fiber angle arrangement. The design variables to be optimized are the angles at scattered design points. Several examples with in-plane load are investigated. In the simple representative numerical examples, the proposed method shows good performance.

Shepard-interpolation-based curve fiber composite structure design waterfall type multi-level optimization method

Abstract: The invention belongs to the field of composite structure design optimization methods, and discloses a Shepard-interpolation-based curve fiber composite structure design waterfall type multi-level optimization method. The method comprises the following steps of building parameterization layers; for each layer of the layering structure, evenly laying out a series of discrete design points inside the structure as design variables, and on the basis of fiber angle values of the discrete points, utilizing Shepard interpolation for constructing a continuous global function for expressing the whole design domain fiber angle; utilizing finite element analysis for building the relation between a rigid matrix and the design variables; updating the design variables to achieve the purpose of the minimum structure flexibility; obtaining the fiber angle of the thick layer with the minimum structure flexibility, and then working out the design initial value of the adjacent thin layer; executing the steps repeatedly, and obtaining the optimal fiber angle space continuous change layout of the curve fiber composite structure. According to the optimization method, the design variables are reduced, the optimization efficiency is high, and the calculation cost of the optimization process is lowered.

Optimization of Composite Structures with Curved Fiber Trajectories

Abstract: This paper presents a new approach to generate and optimize parallel fiber trajectories on general non planar surfaces based on level-sets and the Fast Marching Method. Starting with a (possibly curved) reference fiber direction defined on a (possibly curved) meshed surface, the new method allows defining a level-set representation of the fiber network for each ply, and so defining the fiber trajectories. This new approach is then used to solve optimization problems, in which the stiffness of the structure is maximized (minimum compliance problem). The design variables are the parameters defining the position and the shape of the reference curve. The shape of the design space is discussed, regarding local and global optimal solutions. The possibility to include in the optimization problem a limitation on the curvature of the trajectories is also addressed.

Optimization of parts manufactured using continuous fiber three-dimensional printing technology

Abstract: In this paper, a topology optimization framework for continuous composite fiber three-dimensional (3D) printing is presented. The problem of compliance minimization with respect to the design parameters of density and orientation is formulated and solved numerically using the proposed filtration procedure, ensuring a smooth fiber orientation change and fast convergence. The optimization algorithm is based on the gradient method and implemented using Abaqus finite-element analysis software. It allows optimization of parts within complex computer-aided design models and considers different types of nonlinearities in the finite-element model. The efficiency of the proposed method is evaluated on examples of two-dimensional (2D) and 3D cantilevers, which reveals the general similarity of the resulting shapes in 2D and 3D. Nevertheless, spatially oriented trusses are clearly defined in the result part of the 3D problem. The pedal support of a race car is optimized using the developed framework with an optimization domain based on the geometry of the conventional metallic structure. The resulting topology and fiber orientation are postprocessed using special software for printing trajectory generation and produced by a continuous composite fiber 3D printer.