**Keynote Speaker**

** Alam Md. Mahbub **

Professor

Harbin Institute of Technology , China

**Speech Title:** Intrinsic mechanism of thrust production of pitching hydrofoil

**Abstract:**
Theoretical and numerical analysis of the origin of the thrust on a forced pitching hydrofoil is done in this work. The non-dimensional frequency (Strouhal number, Std) and amplitude ratio (A*) of the hydrofoil pitching are varied as 0.21 ≤ Std ≤ 0.33 and 0.55 ≤ A* ≤ 0.8, respectively. While the numerical solution of the problem is obtained using unsteady Navier-Stokes equations, a theoretical model is developed to understand the origin of thrust produced by oscillating hydrofoil. Coefficients of thrust, power, and efficiency of the hydrofoil are calculated and presented on the Std – A* plane. The evolution of the flow structure around the pitching hydrofoil is clarified for the conditions corresponding to both drag and thrust generations. A flow model is hypothesized. A mathematical analysis of the flow model, involving Euler, Coriolis and centrifugal accelerations in a non-inertial frame, is developed to assimilate the physical insight into the thrust generation and power input. The analysis provides theoretical relationships of thrust, power, and efficiency as functions of Strouhal number and/or amplitude ratio. The data from the numerical simulation tangibly support the relationships. This subject would be handy for undergraduate and postgraduate studies.

** Wanyang Dai**

Professor

Nanjing University, China

**Speech Title:** Sharing and Competition with Dynamic Pricing and Probabilistic Machine Learning in Quantum-Cloud Federations

**Abstract:**
We will present game-theoretic “win-win” fair resource-sharing and “win-loss” competition policies with dynamic pricing for quantum-cloud federations such as the future Internet of quantum blockchains. These policies are computed and implemented via establishing dynamic evolving stochastic models and probabilistic machine learning. Their asymptotic optimality is proved through functional and diffusion approximations. In these stochastic models, the associated data noises can be statistically characterized by big data streams such as high-dimensional Brownian motions, Fractional Brownian motions, and even fractional Levy processes. They commonly appear in physical and blockchain management systems with different uncertainties of quantum particle movements such as velocity uncertainty, acceleration uncertainty, quantum entanglement uncertainty, and even particle fusion and fission uncertainty.

** Junhui Hu**

Professor

Nanjing University of Aeronautics and Astronautics, China

**Speech Title:** BAW Catalyzed Gas Sensors & Single-Sensor E-Noses

**Abstract:**
We report an ultrasonic catalysis method to enhance the performance of a gas sensor, which was proposed by the author’s group recently, and the efforts made by his group to apply this method in the implementation of single-sensor e-noses. In the method, the air borne bulk-acoustic-wave (BAW) is employed to catalyze the redox reaction at the sensing surface. Different from the conventional ultrasonic catalytic effect which occurs in liquid or at the solid-liquid interface, which is caused by the acoustic cavitation effect, the ultrasonic catalysis in this work happens at the gas-solid interface. The contents of this report include two parts. In the first part, principle, structure and characteristics of the BAW catalyzed gas sensors are given and explained. In the second part, working mechanism, structure and characteristics of the single-sensor e-noses catalyzed by air borne BAW are given and dsicussed. It shows that by employing the ultrasonic catalytic effect, not only the sensing limit and sensitivity of existing gas sensors can be improved greatly, but also the single-sensor e-nose can be implemented in a very different way from the conventional method.

** El Mostafa Kalmoun**

Associate Professor

Qatar University, Qatar

**Speech Title:** Multilevel optimization for inpainting-based image compression

**Abstract:**
A growing interest has recently centered around applying image inpainting for compression and coding purposes. In image inpainting, one aims at restoring missing or damaged regions using interpolation of known image data. The success of the inpainting-based image compression does not only rely on the inpainting method but also on the optimal selection of the encoding mask. This task is formulated as a bilevel optimization problem in which the upper level decision aims to minimize data similarity and mask sparsity, while the lower-level objective involves the inpainting method. In this paper, we solve this problem using a penalty functional approach. As such a bilevel optimization problem is known to be computationally challenging, we address the use of multilevel optimization techniques to solve the resulting large-scale unconstrained optimization problem.

**Yingxu Wang**

Professor

University of Calgary, Canada

**Speech Title:**On Intelligent Mathematics (IM): What's Missing in General AI and Cognitive Computing

**Abstract:**
Mathematics is the most fundamental and indispensable abstraction means underpinned almost all science and engineering disciplines. Intelligent Mathematics (IM) is a category of contemporary denotational mathematics extending classic analytic mathematics as defined in the domain of real numbers (R) to that of hyperstructures (H). IM represents a collection of novel mathematical structures that formalizes rigorous expressions and manipulations on complex entities in sciences, linguistics, knowledge representation, and machine intelligence generation. Paradigms of IM encompass AI inference algebra, concept algebra, semantic algebra, real-time process algebra (RTPA), system algebra, fuzzy probability algebra, big data algebra, image frame algebra, relation algebra, and causal probability, etc., as created in my lab. Each of them addresses a challenging abstract entity in H that had been out of the traditional domain of classic mathematics. This keynote speech presents a theoretical framework of IM and applications in general AI and the next generation of cognitive computers. Real world applications of IM toward rigorous AI and machine reasoning beyond traditional data-driven technologies will be demonstrated based on brain-inspired autonomous systems. IM will lead to the emergence of mathematical engineering (ME), which addresses the challenges in formal structural and functional modeling of complex cognitive objects and their rigorous manipulations in a wide range of contemporary applications.

**C.W. Lim**

Professor

City University of Hong Kong, China

**Jiabin Sun**

Associate Professor

Dalian University of Technology, China

**Speech Title:**Theory and Mathematical Solution Framework for Nonlinear Post-buckling of Moderately Thick Cylindrical Shells

**Abstract:**
For effective applications of light-weight structures, it is of great importance to establish a theory and mathematical framework for stability of circular cylindrical shells. A nonlinear post-buckling equilibrium state exists for a compressed cylindrical shell when the critical loads are far below the solution obtained by the common linear eigenvalue theory. This nonlinear stability state is often suggested as the load-carrying capability of cylindrical shells in practical engineering applications. After decades of continuous research, the nonlinear post-buckling for isotropic thin cylindrical shells with accurate solution procedure has been well established. However, owing to the inherent mathematical difficulty and physical complexity, precise solution and analysis for these nonlinear systems are still far from sufficiently satisfactory to reveal more intricate post-buckling behaviors of cylindrical shells made of smart and carbon-based composite materials. The situation is particularly intricate mathematically when the cylindrical shells are characterized with varying wall thickness. Motivated by this reason, a unified shell theory consisting of classical (TWST), first-order shear deformation theory (FSDT), and higher-order shear deformation theory (HSDT) is first proposed. The post-buckling equilibrium paths with mode-jumping phenomenon are obtained by an accurate solution framework established on the basis of Galerkin’s method that involves a set of newly defined displacement functions for both symmetric and asymmetric post-buckling modes. A comparison study is presented to verify the accuracy of the present analysis and excellent agreement is reported. Finally, the effects of key influencing parameters on the nonlinear post-buckling behavior of some fashionable cylindrical shells represented by graphene platelets reinforced composite (GPLRC) cylindrical shells and magneto-electro-elastic (MEE) composite cylindrical shells are investigated. The discrepancy among different shell theories and their scopes are also discussed in detail.

**Ding-Geng Chen**

Professor

University of North Carolina-Chapel Hill, USA

**Speech Title:**Stochastic Cusp Catastrophe Model and its Applications

**Abstract:**
Within the catastrophe theory, cusp catastrophe model is the most used in mathematical and statistical modeling because of its capability to characterize both rational and irrational behavioral processes simultaneously. Contributions by a number of great methodologists make it possible to analyze empirical data with this cusp catastrophe model-based methods. This talk will discuss the recent development in cusp catastrophe modeling and its applications to social, behavioral and public health sciences. We will focus on the mathematical and statistical development to provide a mathematic connection between observed data and the deterministic cusp catastrophe at its equilibrium as well as the stochastic cusp catastrophe model for time series data based on Maxwell and Delay conventions.