Title: | The Physics of Natural Ventilation: From Wind Towers to Football Stadiums |
Speaker: | Dr. Ben Richard Hughes, School of Civil Engineering, University of Leeds, UK |
Time/Place: | 11:30 - 12:30 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | Natural ventilation relies on the movement of air without mechanical aid, it is the simplest form of space conditioning and as such the physics of this phenomena are often ignored. However with HVAC systems currently utilising 60% of global energy consumption, natural ventilation is now the focus of much research. This discussion will focus on the basic physical properties and present the tools and techniques available to investigate thoroughly, namely CFD and low speed wind tunnel experimentation, before moving onto current applications and how the methodology is being applied to new challenges such as the FIFA 2022 World Cup hosted in Qatar. |
Title: | Finite element methods for elastic multi-structure problems |
Speaker: | Prof. Jiangu Huang, Department of Mathematics, Shanghai Jiao Tong University, China |
Time/Place: | 15:30 - 16:30 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | Elastic multi-structures (EMS) are usually composed of elastic substructures of the same or different dimensions, which are coupled together by some junction conditions (rigid or contact connection). They are widely used in the fields of aviation, aerospace, civil engineering, mechanical manufacturing, etc. The remarkable features of elastic multi-structure problems include that (1) the geometric domains occupied by individual elastic members may have different dimensions; (2) the PDEs governing the deformation over individual elastic members are heterogenous; (3) the interface conditions among different elastic members are heterogeneous. Therefore, Prof. Ciarlet mentioned in his monograph “Mathematical Elasticity II” that “A challenging program consists in numerically approximating the mathematical models of elastic multi-structures that comprise “many” substructure.” In this talk, we will give a brief survey on finite element methods for such problems, mostly based on our work developed in the past decade. This is a joint work with C. Chen, L. Guo, X. Huang, J. Lai, Z. Shi and Y. Xu. |
Title: | Assessing Movement Coordination Post-stroke via Dynamic Factor Analysis |
Speaker: | Dr. LU Ying, Department of Humanities and Social Science, New York University, USA |
Time/Place: | 14:30 - 15:30 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | Assessment of the quality of functional movements of stroke patients provides important information about where patients are in the course of their recovery and how it changes with treatment. Traditional approaches focus on single dimensional measures such as the time to accomplish a task, the range of motion of single joints etc. These measures tend not to correspond very well with the actual functional recovery of the patients. For example, patients can use compensatory strategies by un-naturally moving other parts of the body in order to get a larger range of motion or faster speed. In this paper we assess the quality of functional movements by simultaneously utilizing kinematic information from multiple joints and EMG activity of related muscles. With the aid of multivariate statistics (dynamic factor analysis, DFA), we studied the movement pattern for Elbow Extension (a simple task in the Wolf Motor Function Test) for both healthy subjects and stroke patients. |
Title: | Multi-phase Texture Segmentation Using Gabor Features Histograms based on Wasserstein Distance |
Speaker: | Mr. Motong QIAO, Department of Mathematics, Hong Kong Baptist University, Hong Kong |
Time/Place: | 10:00 - 12:00 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | Image Segmentation is one of the hottest issues in the area of digital image processing. It is usually viewed as an important preprocessing for the other applications, such as image coding, image pattern recognition, etc. We present a multi-phase image segmentation method based on the histogram of the Gabor feature space, which consists of a set of Gabor-filter responses with various orientations, scales and frequencies. Our model replaces the error function term in the original fuzzy region competition model with squared 2-Wasserstein distance function, which is a metric to measure the distance of two histograms. The energy functional is minimized by alternative minimization method and the existence of closed-form solutions is guaranteed when the exponent of the fuzzy membership term being 1 or 2. We test our model on both simple synthetic texture images and complex natural images with two or more phases. Experimental results are shown and compared to other recent results. |
Title: | A Quadratic Inverse Eigenvalue Problem in Finite Element Model Updating |
Speaker: | Prof. Chu Delin, Department of Mathematics, National University of Singapore, Singapore |
Time/Place: | 14:30 - 15:30 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | Updating a system modeled as a real symmetric quadratic eigenvalue problem to match observed spectral information has been an important task for practitioners in different disciplines. It is often desirable in the process to match only the newly measured data without tampering with the other unmeasured and often unknown eigenstructure inherent in the original model. Such an updating, known as no spill-over, has been critical yet challenging in practice. Only recently, a mathematical theory on updating with no spill-over has begun to be understood. However, other imperative issues such as maintaining positive definiteness in the coefficient matrices remain to be addressed. This talk highlights several theoretical aspects about updating that preserves both no spill-over and positive definiteness of the mass and the stiffness matrices. In particular, some necessary and sufficient conditions for the solvability conditions are established in this investigation. |
Title: | Applications of stochastic reconstruction in model-free tests of spatial point processes |
Speaker: | Ms. Wong Ka Yiu, Department of Mathematics, Hong Kong Baptist University, Hong Kong |
Time/Place: | 10:30 - 11:30 FSC1217 |
Abstract: | Spatial point processes are useful tools for modeling irregularly distributed objects in space. In statistical analysis of spatial point patterns, it is often required to simulate similar point patterns. However, in some applications only certain aspects such as stationarity and isotropy, of the underlying model are known/assumed, and so Monte Carlo simulation is not feasible. Stochastic reconstruction is a simple method of generating point patterns with similar summary characteristics as the observed realizations. It requires only the knowledge of some prescribed summary characteristics without specifying any theoretical models for the point processes. In this study, a test for isotropy is developed using reconstruction. A test statistic measuring the extent of anisotropy of a point pattern is formed by considering the reduced second-order measure of sectors. Point patterns with similar summary characteristics are generated using stochastic reconstruction. The null distribution of the test statistic is then approximated from the reconstructed point patterns, which play the role of parametric bootstrap samples. In order to test for isotropy, the prescribed summary characteristics used for reconstruction assume stationarity and isotropy so that the reconstructed point patterns remain isotropic. Other potential applications of stochastic reconstruction in point process statistics will also be mentioned. |
Title: | Approximation methods for boundary integral equations on contours with corners |
Speaker: | Prof. Viktor Didenko, Department of Mathematics, Universiti Brunei Darussalam, Brunei |
Time/Place: | 11:30 - 12:30 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | The stability of the Nystroem method for the Muskhelishvili and Sherman-Lauricella equations is studied. If the initial contour has corner points, the corresponding integral operators are not compact and usual methods to study stability are not very efficient. However, an algebraic approach can be employed. This allows us to show that the method is stable if and only if certain operators are invertible.These operators depend on the opening angles of the corner points and on parameters of the approximation method. Numerical experiments show that there is only finite number of opening angles where the such operators are not invertible. Correspondingly, if the contour possesses at least one corner with such opening angle, the method is not stable. Otherwise the method is always stable, and examples show its excellent convergence. |
Title: | Difference-based methods for the residual variance in nonparametric regression |
Speaker: | Mr. Dai Wenlin, Department of Mathematics, Hong Kong Baptist University, Hong Kong |
Time/Place: | 10:00 - 11:30 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | Variance estimation is an important topic in nonparametric regression. Among the existing methods for the residual variance, the difference-based estimators do not require an estimate of the mean function and thus are very popular in practice due to the ease of implementation. In the first part, I propose a new difference-based estimator using the variance and bias reduction techniques. Both asymptotic results and simulations indicate that the proposed estimator outperforms the existing competitors significantly. In the second part, I extend the studies of difference-based estimation from the simple regression model to more complicated regression models. In particular, for nonparametric regression with jump discontinuities, I have proposed several methods for estimating the residual variance and for testing the existence of jump points simultaneously. Finally, I will finish the talk by mentioning some on-going projects, including the variance components testing in linear mixed effect models. |
We organize conferences and workshops every year. Hope we can see you in future.
Learn MoreProf. M. Cheng, Dr. Y. S. Hon, Dr. K. F. Lam, Prof. L. Ling, Dr. T. Tong and Prof. L. Zhu have been awarded research grants by Hong Kong Research Grant Council (RGC) — congratulations!
Learn MoreFollow HKBU Math