Search results

This study aims to propose and numerically assess different ways of discretising a very weak formulation of the Poisson problem.

We use integration by parts twice to shift smoothness requirements to the test functions, thereby allowing low-regularity data and solutions.

Various conforming discretisations are presented and tested, with numerical results indicating good accuracy and stability in different types of problems.

This is one of the first articles to propose and test concrete discretisations for very weak variational formulations in primal form. The numerical results, which include a problem based on real MRI data, indicate the potential of very weak finite element methods for tackling problems with low regularity.

Bayesian cubature (BC) has emerged to be one of most competitive approach for estimating the multi-dimensional integral especially when the integrand is expensive to evaluate, and alternative acquisition functions, such as the Posterior Variance Contribution (PVC) function, have been developed for adaptive experiment design of the integration points. However, those sequential design strategies also prevent BC from being implemented in a parallel scheme. Therefore, this paper aims at developing a parallelized adaptive BC method to further improve the computational efficiency.

By theoretically examining the multimodal behavior of the PVC function, it is concluded that the multiple local maxima all have important contribution to the integration accuracy as can be selected as design points, providing a practical way for parallelization of the adaptive BC. Inspired by the above finding, four multimodal optimization algorithms, including one newly developed in this work, are then introduced for finding multiple local maxima of the PVC function in one run, and further for parallel implementation of the adaptive BC.

The superiority of the parallel schemes and the performance of the four multimodal optimization algorithms are then demonstrated and compared with the k-means clustering method by using two numerical benchmarks and two engineering examples.

Multimodal behavior of acquisition function for BC is comprehensively investigated. All the local maxima of the acquisition function contribute to adaptive BC accuracy. Parallelization of adaptive BC is realized with four multimodal optimization methods.

The purpose of this paper is to investigate the free vibration response of a laminated honeycomb sandwich panels (LHSP) for aerospace applications. Higher order shear deformation theory (HSDT) was simplified for the dynamic analysis of LHSP. Furthermore, the effects of honeycomb parameters on the value of natural frequency (NF) of vibration were explored.

This paper applies HSDT to the analysis of composite LHSP to derive four vibration differential equations of motion and solve it to find the NF of vibration. Two analytical models (Nayak and Meunier models) were selected from literature for comparison of the NF of vibration. In addition, a numerical model was built by using ABAQUS and the results were compared. Furthermore, parametric studies were conducted to explore the effect of honeycomb parameters on the value of the NF of vibration.

The present model is successful in simplifying HSDT for the analysis of LHSP. The first five natural frequencies of vibration were calculated analytically and numerically. In the parametric study, increasing core height or young’s modulus or changing laminate layup will increase the value of NF of vibration. Furthermore, increasing plate constraint (using clamped edge boundary condition) will increase the value of NF of vibrations.

The current analysis is suitable for all-composite symmetric LHSP. However, for isotropic or non-symmetric materials, minor modifications might be adopted.

The application of simplified HSDT to the analysis of LHSP is one of the important values of this research. The other is the successful and complete dynamic analysis of all-composite LHSP.

Accurate and automatic clothing pattern making is very important in personalized clothing customization and virtual fitting room applications. Clothing pattern generating as well as virtual clothing simulation is an attractive research issue both in clothing industry and computer graphics.

Physics-based method is an effective way to model dynamic process and generate realistic clothing animation. Due to conceptual simplicity and computational speed, mass-spring model is frequently used to simulate deformable and soft objects follow the natural physical rules. We present a physics-based clothing pattern generating framework by using scanned human body model. After giving a scanned human body model, first, we extract feature points, planes and curves on the 3D model by geometric analysis, and then, we construct a remeshed surface which has been formatted to connected quad meshes. Second, for each clothing piece in 3D, we construct a mass-spring model with same topological structures, and conduct a typical time integration algorithm to the mass-spring model. Finally, we get the convergent clothing pieces in 2D of all clothing parts, and we reconnected parts which are adjacent on 3D model to generate the basic clothing pattern.

The results show that the presented method is a feasible way for clothing pattern generating by use of scanned human body model.

The main contribution of this work is twofold: one is the geometric algorithm to scanned human body model, which is specially conducted for clothing pattern design to extract feature points, planes and curves. This is the crucial base for suit clothing pattern generating. Another is the physics-based pattern generating algorithm which flattens the 3D shape to 2D shape of cloth pattern pieces.