The Advanced Manufacturing and Design Laboratory is a collaborative research group founded by Dr. Ilya V. Avdeev under the department of Mechanical Engineering in the College of Engineering and Applied Science at the University of Wisconsin – Milwaukee.
Areas of Focus
- Advanced modeling techniques for solving multiphysics problems, including Finite Element Analysis (FEA), Computational Fluid Dynamics (CFD), Numerical Analysis
- Methods for real-time coupled-field simulations
- Design of energy generation and storage systems
- Industrial innovation and technology entrepreneurship
Coupled-Physics Modeling and Design of Energy Storage Systems (Johnson Controls, Inc.)
Increased usage of lithium-ion batteries in automotive applications makes it necessary to understand their mechanical behavior under extreme loading conditions, such as mechanical impact. One of the key design aspects of any energy storage system, including batteries, is safety, which can be improved by: (a) reducing the probability of an event and (b) lessening the severity of the outcome should an event occur. The goal of the project is improving the crash safety of lithium-ion batteries. There have been some studies on failure modes, fault tree analysis, safety and reliability of lithium based batteries. In the case of lithium-ion batteries, thermal stability is probably the most important parameter that affects safety in cells, modules and battery packs. Although various safety mechanisms have been implemented in individual lithium-ion cells as well as entire battery packs based on the thermal state of the battery, there is still a need for better understanding of battery response to extreme loading, which increases a risk of short circuits and the following thermal runaway or fire.
Design and Modeling of Novel MEMS Systems for Drug Discovery
The main objective of this project is the prototyping, testing, and ultimately commercialization of the E-Trap, a novel hardware platform for trapping a variety of charged micro- and nanoscale particles as well as single molecules in aqueous solution using electrostatic fields (see Figure). The underlying trapping method is based on patented technology developed by one of the PIs (JCW; U.S. Patent 8,465,967, UWM Research Foundation). The work is carried out in close collaboration with Isopoint Technologies LLC, a new startup company founded by UWM graduate student Alex Francis whose E-Trap entry was one of the winners of UWM’s 2013-14 Student Startup Challenge competition and the Grand-Prize winner of the 2013-14 New Venture Business Plan Competition sponsored by UWM’s Sheldon B. Lubar School of Business.
The E-Trap device takes fundamental research on molecular, nanoscale, and microscale objects and applications in fields such as drug discovery, disease control, and biomedical diagnosis/analysis to the next level by not only allowing for the observation but also the direct interaction with the system under investigation. For example, it would enable the confinement of a single molecular or biological system in solution over extended periods of time so that a detailed biological, biochemical, or biophysical analysis can be carried out. The novel trapping technique has been shown to allow the manipulation of individual, charged microspheres (diameter: 2 μm), spherical nanoparticles (diameter: 21 nm), and single-stranded, 800-nucleotide DNA molecules.
Experimental, Numerical and Analytical Characterization of Torsional Disk Coupling Systems (Rexnord)
Torsional couplings are used to transmit power between rotating components in various power systems while allowing for small amounts of misalignment that may otherwise lead to equipment failure. When selecting a proper coupling type and size, one has to consider three important conditions: (1) the maximum load applied to the coupling, (2) the maximum operation speed, and (3) the amount of misalignment allowable for normal operation. There are many types of flexible couplings that use various materials for the flexible element of the coupling. The design of the coupling and the materials used for the flexible portion will determine its operating characteristics. In this project, investigation of a disk coupling that uses a stack of metallic discs to counter the misalignment effects is performed. Benefits of this type of coupling include: ease of replacement or repair, clear visual feedback of element failure, and the absence of a need for lubrication. The torsional stiffness of a coupling is a major factor relative to the amount of misalignment allowable. Currently, flexible couplings are tested by manufacturers to experimentally determine the torsional stiffness; a process which requires expensive equipment and more importantly employee time to set-up and run. The torsional coupling lumped characteristics, such as torsional- and flexural stiffness, as well as natural frequencies are important for design of the entire power system and have to be as precise as possible.
Distributed Algorithms for Embedded Thermo-Mechanical Control (GE Healthcare)
The research objective of this project is to develop algorithmic solutions and an efficient parallel code for distributed finite element solvers specifically designed for embedded high-performance energy-efficient computing (HPEEC) architectures, which exploits data-parallelism and multiple cores. Developed static, modal and transient solvers will offer scalability and will be designed to satisfy required deadline constraints for real-‐time thermo-‐mechanical monitoring and control of critical CT detector components. Real-‐time monitoring and control of a CT detector’s thermal state and subsequent mechanical deformations based on sparse thermal sensor data input can lead to improved scan quality through software compensation or more robust thermal management. Achieving fast, reliable, and robust performance of the simulation parallel algorithms and the embedded signal processing and finite element code will depend on a specific class of addressed problems and the type of architecture used to solve the problem. The key issues that will be addressed in this project are: (a) decomposing the task of solving a coupled-field thermo-mechanical system of partial differential equations into independent blocks of work to avoid inter-block communication; (b) balancing the number of computations with the efficiency of the data flow; and (c) determining a data flow that minimizes global memory transactions.
Other Recent Projects:
- Multiscale Modeling and Design of Concrete Composite Materials and Structures
- Multiphysics Modeling of Metal/Rock Cutting ProcessesModel Order Reduction in Dynamic Systems