Skip to main content


Dr Yang Liu



Date and Time: 13:30 - 14:15 am, Wednesday 6 November 2019

Venue: Peter Chalk 2.5

Speaker: Mr Yiwei Han, School of Mechatronic Engineering, Southwest Petroleum University, Chengdu, China

Title: Nonlinear dynamic modeling of drill string in horizontal well - a geometrically exact approach

Abstract: A new nonlinear dynamic model of drill string in horizontal well is established based on geometrically exact beam theory and quadrature element method. The model proposed in this work takes into account the geometric nonlinearity of the drill string and the nonlinear contact between the drill string and the borehole. Benefiting from the accuracy of the geometric description of the geometrically exact beam, the model uses a geometrically exact method to describe the contact between the drill string and the borehole. Comparison with experimental results shows that the geometrically exact model can accurately describe the nonlinear characteristics of the drill string. Finally, the characteristics of the movement of drill string under different drilling pressure and rotation speed will be analyzed. Drilling pressure and rotational speed have an effect on the buckling of the drill string. Before buckling, the drill string moves periodically at the bottom of the wellbore. After buckling, the movement of the drill string becomes disordered. The model presented in this work provides a new method for dynamic analysis of drill string.


Date and Time: 13:30 - 14:15 am, Friday 30 August 2019

Venue: Harrison 170

Speaker: Dr Moi Hoon Yap, Manchester Metropolitan University, Manchester, UK

Title: Translating Computer Vision Research to Real-World Applications

Abstract: The Human-Centred Computing Group undertakes research in computer vision, games, HCI and AI across the themes of healthcare, medical imaging and sport. In facial analysis, we create new algorithms for face wrinkles detection, micro-expressions detection/recognition, facial inpainting and denoising algorithms. We are working with external partners to deliver our research to real-world applications. This talk covers some of our recent works in facial analysis and medical image analysis. I will also share our journey of diabetic foot ulcer project, beginning from image processing to adoption of deep learning models, and to partnership with Oracle to create cloud-enabled prototype.


Date and Time: 11:00 - 12:00 am, Thursday 2 August 2018

Venue: Harrison 102

Speaker: Dr Yao Yan, School of Aeronautics and Astronautics, University of Electronic Science and Technology of China, Chengdu, China

Title: Unsafe cutting

Abstract: To analyse the multi-stability in the non-linear time-delayed metal cutting process, we employ a classical model of regenerative cutting and introduce the Stribeck effect and process damping to the tool’s rake and flank surfaces. From a practical point of view, the multi-stable regions are considered as unsafe due to its sensitivity to the initial conditions of cutting, which could lead the system to an undesired behaviour. So, it is vital to evaluate cutting safety through estimating its unsafe zones. For this purpose, we have attempted to define the basins of attraction and basin's stability by using this cutting model in order to get insight into the issues of cutting safety. Special attention will be paid to the influences of delayed initial conditions, starting points, and the states at time zero on the long-term dynamics of such a time-delayed cutting system. Finally, by using the analytical perturbation methods, our new model shows that the Stribeck effect affects the estimation of the unsafe zones excessively, so that only a very small part of the estimation is acceptable.


Date and Time: 11:00 - 12:00 am, Monday 30 July 2018

Venue: Harrison 102

Speaker: Professor Joseph Páez Chávez, Department of Mathematics, Faculty of Natural Sciences and Mathematics, Escuela Superior Politécnica del Litoral, Guayaquil, Ecuador

Title: A model for Dengue transmission dynamics with seasonal effects and impulsive control

Abstract: In this talk we will introduce an SIR epidemiological model describing the vector-to-host and host-to-vector transmission dynamics of Dengue. The proposed model includes vector control in terms of pesticide applications, mathematically described as impulsive perturbations in the system, thereby accounting for reductions of the vector population in very short time intervals. In addition, seasonality is incorporated into the model via a sinusoidal forcing mimicking periodic regimes of the intensity of vector-to-host and host-to-vector infections, as well as vector growth. A detailed numerical analysis of the model is carried out via path-following techniques for non-smooth systems, implemented via the software COCO. The numerical study also considers the optimization of the times at which pesticide is applied.


Date and Time: 10:30 - 11:30 am, Monday 18 December 2017

Venue: LSI Seminar Room A

Speaker: Dr Luigi Manfredi, Institute for Medical Science and Technology, School of Medicine, University of Dundee, Dundee, UK

Title: Design and Construction of a Mini Snake-Like Medical Robot

Abstract: This talk presents the design and construction of a novel mini snake-like robotic endoscope developed within the ERC Proof of Concept grant “Compliant Actuation Robotic Platform for Flexible Endoscopy” (CARPE), based on a novel and patented “Active Tether Concept” (ATC). In contrast with a traditional “push colonoscope”, the smart control provides an active locomotion reducing the force exchanged with the colonic wall, and therefore, the pain and discomfort in the procedure. The robotic frame is designed by finite element analysis to achieve a weight of only 0.7 grams for each joint, and a high force/torque to weight ration. ATC has an external diameter of 13 mm and a central hollow space of 5 mm, which can be used for essential services, such as suction/irrigation and biopsy channel. The head of the robot can carry a HD camera and the flexible active robotic endoscope will be controlled from an external console.