Research Interests & Highlights (Last update: 01/2019)
1. Damage Resistant Topology Optimization for Energy Absorbing Structures
Energy absorbing structures, which are able to transfer external energy into plastic work dissipated by the material, have been widely used in practical applications. One of the key requirements in these applications is to obtain cost-efficient designs with high energy absorption capacity. As the energy absorption increases during the plastic dissipation process, the plastic/inelastic strains are accumulated at the material points. In real materials, this high accumulated plasticity may cause initiation of damage that will eventually lead to material failure with associated loss of structural strength and stiffness. Thus, increasing the energy absorption capacity while preventing the damage and failure due to accumulated plastic strains are conflicting design requirements. In this category of study, we aim at developing systematic design approach based on topology optimization, which can expedite the design process is needed to identify efficient topologies that can satisfy and balance these two conflicting design requirements.
2. Topology Optimization Considering Geometrical and Material Nonlinearities
Despite numerous topology optimization studies were conducted based on linear elastic structures, it is widely recognized that linear elasticity often has poor performance when subjected to large deformation and large strain. In these situations, geometrical and material nonlinear analyses for structural behavior should be considered in the optimization process. In this broad category of research, we focus on incorporating geometrical and material nonlinearities into discrete truss and continuum density-based topology optimization.
3. Topology Optimization Considering Length-scale Effects
Length-scale effects, also known as size effects, are usually prominent when the physical structural size is comparable to the characteristic size of its underlying material microstructure. They are prevailingly observed in various micro-structured materials such as granular, cellular, biological, and composite as well as small-scale systems such as micro-devices in microelectromechanical systems (MEMS). In these applications, classical continuum theories are no longer sufficient to describe the correct associated physics. So far, most studies on topology optimization were carried out using classical continuum theories, which are based on the assumption that the material microstructural influence on the macroscopic mechanical behavior due to size effect is negligible. The purpose of this study is to extend the application of higher-order elastic theories in topology optimization of micro-structured materials where length-scale effects are prominent. The considered theories include: Elasticity with Microstructures (EM), Gradient Elasticity (GE) as well as Staggered Gradient Elasticity (SGE).
4. Dual Sequential Approximation (DSA) Methods for Topology Optimization
For large-scale nonlinear topology optimization, the efficiency of the optimization algorithm plays an important role in reducing the overall iteration number and computational time. Since topology optimization typically has large number of design variables and far less number of constraints, dual methods based on sequential approximation (DSA) (e.g. MMA, OC, etc.) were frequently used for updating the optimization problems. The efficiency and accuracy of DSA algorithms highly depend on the quality of their approximated sub-problems. In this study, we proposed two novel approximations based on the history gradient information and the number of constraints as follows:
(a) Single constraint: This case is often seen in the classic minimum compliance problem formulation, in which the linear volume constraint is the only constraint. In this case, an improved MMA approximation is proposed which increases the convergence dramatically compared to the original MMA. The proposed algorithm (TGMMA) is demonstrated to be better than most of the current popular algorithms.
(b) Multiple constraints: If more than one constraint is considered, special algorithms should be developed since the dimension of dual space changes. To this end, an adaptive quadratic approximation (AQA) was proposed to adaptively choose the accurate approximation based on the historic information. AQA algorithm is demonstrated to greatly outperform other state-of-art algorithms such as MMA, T2:R and T2:E in solving structural and topology optimization problems.
(a) Single constraint: This case is often seen in the classic minimum compliance problem formulation, in which the linear volume constraint is the only constraint. In this case, an improved MMA approximation is proposed which increases the convergence dramatically compared to the original MMA. The proposed algorithm (TGMMA) is demonstrated to be better than most of the current popular algorithms.
(b) Multiple constraints: If more than one constraint is considered, special algorithms should be developed since the dimension of dual space changes. To this end, an adaptive quadratic approximation (AQA) was proposed to adaptively choose the accurate approximation based on the historic information. AQA algorithm is demonstrated to greatly outperform other state-of-art algorithms such as MMA, T2:R and T2:E in solving structural and topology optimization problems.
5. Experimental Test and Numerical Modeling of Creep Behavior of Glubam Material
Bamboo is a marvelous plant on our planet. Although the natural properties of bamboo are quite comparable to those of wood, the original geometrical shape of bamboo culms made it not so easy to be used in modern construction. Using industrialized manufacturing process, a new type of laminated bamboo that can be cost-effectively used as structural elements are invented by Yan Xiao et al. with a trademark of GluBam, imitating the well-known GluLam or glue laminated timber. The structural glubam elements are made by laminating glubam sheets, which are typically 20 to 40 mm thick produced using existing production process for bamboo veneers or so called plybamboo sheets. The glubam or plybamboo sheets follow the original technology of plywood production to laminate mats of thin layer bamboo strips with a thickness of about 2 mm. As a newly emerge structural material, this research aimed at comprehensively calibrating the mechanical properties of GluBam by experimental testing. Specifically, the long-term creep behavior of GluBam material and GluBam-CFRP composite was investigated by testing. Constitutive model for creep of GluBam was also proposed based on the collected data, which helps to deeply understand the creep mechanism in GluBam.
Teaching Experiences
Starting from the Fall 2012, I served as a Teaching Assistant in Department of Civil & Environmental Engineering and Earth Sciences, University of Notre Dame, and lecturer in Department of Structural Engineering, University of California, San Diego for the classes on both graduate level and undergraduate level listed below
Lecturer, Department of Structural Engineering, UCSD
Teaching Assistant, Department of CEEES, University of Notre Dame
- Graduate Course - Design Optimization for Additive Manufacturing SE 286 Fall 2018
(Guest Lecturer for Prof. Alicia Kim) - Short Course - Topology Optimization for Additive Manufacturing Summer 2019
(Guest Lecturer for Interdisciplinary Training and Networking in Engineering and Next Generation in Simulation and Experimentation, CEER ) - Graduate Course - Structural Optimization SE 285 Fall 2019
(Guest Lecturer for Prof. Alicia Kim) - Graduate Course - Design Optimization for Additive Manufacturing SE 286 Fall 2019
(Guest Lecturer for Prof. Alicia Kim)
Teaching Assistant, Department of CEEES, University of Notre Dame
- Undergraduate Course - Statics CE 20150 Fall 2012
(Teaching Assistant for Prof. David Kirkner) - Undergraduate Course - Structural Analysis CE 30210 Spring 2013
(Teaching Assistant for Prof. Kapil Khandelwal) - Undergraduate Course - Statics CE 20150 Fall 2013
(Teaching Assistant for Prof. Ashley Thrall) - Undergraduate Course - Structural Steel Design CE 40280 Spring 2014
(Teaching Assistant for Prof. Brian Smith) - Undergraduate Course - Reinforced Concrete Design CE 40270 Fall 2014
(Teaching Assistant for Prof. Yahya Kurama) - Graduate Course - Advanced Structural Analysis I CE 60260 Spring 2015
(Teaching Assistant for Prof. Kapil Khandelwal) - Undergraduate Course - Reinforced Concrete Design CE 40270 Spring 2018
(Teaching Assistant for Prof. Yahya Kurama)