Engineering
Designing a Code for Private Machine Learning
Liu, Xintong (University of Utah)
Faculty Advisor: Ji, Mingyue (College of Engineering, Department of Electrical and Computing Engineering)
One of the significant challenges of the machine learning faces today is how to deal with the privacy constraint of the user in a large-scale and distributed communication network. A myriad of data produced by billions of distributed devices need to be sent into the central cloud and to be managed, but what happens if the user does not want to send his/her data to the central cloud. It is reasonable that many users expect the data they send is being protected and maintain privacy. So, we are thinking about whether it is possible to create an individual machine learning in the application of Federated Learning so that the user's data will be protected from the privacy constraints. In this case, the raw data will not be known by anyone except the owner of these data. So, there would be all unknown input data pass through the private machine learning model, and the generated result, which is still hidden data will be sent back to the user. The main topic of the presentation is the designed codes which produces a private configuration with non-linear computation for the learning model and enable privacy constraints for the user's data.
Faculty Advisor: Ji, Mingyue (College of Engineering, Department of Electrical and Computing Engineering)
One of the significant challenges of the machine learning faces today is how to deal with the privacy constraint of the user in a large-scale and distributed communication network. A myriad of data produced by billions of distributed devices need to be sent into the central cloud and to be managed, but what happens if the user does not want to send his/her data to the central cloud. It is reasonable that many users expect the data they send is being protected and maintain privacy. So, we are thinking about whether it is possible to create an individual machine learning in the application of Federated Learning so that the user's data will be protected from the privacy constraints. In this case, the raw data will not be known by anyone except the owner of these data. So, there would be all unknown input data pass through the private machine learning model, and the generated result, which is still hidden data will be sent back to the user. The main topic of the presentation is the designed codes which produces a private configuration with non-linear computation for the learning model and enable privacy constraints for the user's data.
Controlled Flight Through Morphing Wing Aircraft
Moulton, Benjamin (Utah State University)
Faculty Advisor: Hunsaker, Doug (College of Engineering, Mechanical and Aerospace Engineering Department)
A morphing allows for more efficient controlled flight. Morphing wings induce a continuous deflection of control surfaces. Deflection can be caused by compliant mechanisms and composite materials. Factors contributing to efficient morphing wings range from a continuous morphing trailing edge to stiffness and flexure. Wing stiffness supports aerodynamic loading. Wing flexure supports transverse deflection, or twist of the wing trailing edge. Graduate students in the USU Aerolab have written an algorithm to optimize where these deflections should occur on the wing. The student seeks to build a morphing wing to demonstrate the success of the optimization code. Different manufacturing methods are explored. 3D printing provides the most promising results. The 3D printing of thermoplastic materials allows for shear and deflection.
Faculty Advisor: Hunsaker, Doug (College of Engineering, Mechanical and Aerospace Engineering Department)
A morphing allows for more efficient controlled flight. Morphing wings induce a continuous deflection of control surfaces. Deflection can be caused by compliant mechanisms and composite materials. Factors contributing to efficient morphing wings range from a continuous morphing trailing edge to stiffness and flexure. Wing stiffness supports aerodynamic loading. Wing flexure supports transverse deflection, or twist of the wing trailing edge. Graduate students in the USU Aerolab have written an algorithm to optimize where these deflections should occur on the wing. The student seeks to build a morphing wing to demonstrate the success of the optimization code. Different manufacturing methods are explored. 3D printing provides the most promising results. The 3D printing of thermoplastic materials allows for shear and deflection.
Creating the Digital Pathologist
Boyce, Cassandra; Runyan, Josh (Brigham Young University)
Faculty Advisor: Wingate, David (Brigham Young University, Computer Science)
India has the world's highest rate of mortality due to cervical cancer. Despite this variant's high treatability, there aren't enough pathologists to read the pap smear slides. In order to streamline this process, we developed a low-cost, digital pathologist using deep learning to read pap smear results as a form of preliminary testing in order to decrease mortality rates. Deep learning alone cannot provide a solution because a housing is required for the hardware. The industrial design aspect of this project is also important to create a medical device that is not only functional and robust but accessible and unintimidating for those in rural India.
Faculty Advisor: Wingate, David (Brigham Young University, Computer Science)
India has the world's highest rate of mortality due to cervical cancer. Despite this variant's high treatability, there aren't enough pathologists to read the pap smear slides. In order to streamline this process, we developed a low-cost, digital pathologist using deep learning to read pap smear results as a form of preliminary testing in order to decrease mortality rates. Deep learning alone cannot provide a solution because a housing is required for the hardware. The industrial design aspect of this project is also important to create a medical device that is not only functional and robust but accessible and unintimidating for those in rural India.
Design of Modular Dynamic Charging Primary Coils Compatible with SAE J2954 Secondary Coils
Zane, Regan; Kamineni, Abhilash; Nimri, Reebal (Utah State University)
Faculty Advisor: Kamineni, Abhilash (College of Engineering, Electrical and Computer Engineering Department); Zane, Regan (College of Engineering, Electrical and Computer Engineering Department)
Transportation electrification will bring a positive effect on sustainable environments and robust economies. Electric Vehicles (EV) are emerging in today's market as a solution. However, the battery technologies on EV cannot compete with fuel and diesel cars in terms of energy storage capacity, and time needed to recharge (equivalently refuel). These limitations directly reflect on the consumers' convenience, the max miles the Vehicle can perform for, and hence EV adoption. Wireless power transfer (WPT) systems are seen as a solution to ease consumers' transition to EV. A high-level diagram of WPT is shown in Figure (**).
Major advancements in WPT technology has enabled the commercialization of stationary WPT solutions — materials technology has been a major impediment. Hence, academia and industry are jointly considering more advanced solutions in WPT, namely Dynamic Wireless Power Transfer (DWPT). Implementing DWPT systems will permit on the go charging for EV the upper hand over present charging (equivalently fueling) methods and encourage EV adoption. This document reflects on some of the challenges of realizing an effective DWPT that maintains power transfer between the primary pad and secondary pad, and a proposed solution to allow dynamic charging. Also, the proposed DWPT offers compatibility with SAE J2954 (Wireless Power Transfer for Light-Duty Plug-in/Electric Vehicles and Alignment Methodology) WPT3Z3 pad.
A comprehensive approach was taken in the design of the primary pad to validate the power transfer requirements for the designed pads. The proposed solution consists of a custom primary pad and a custom secondary pad for dynamic charging. This document will refer to the custom primary pad and a custom secondary pad as DGA and DVA, respectively. DGA offers compatibility with WPT2Z3 for dynamic charging and DVA offers compatibility with the Universal Ground Assembly (UGA) for stationary charging.
Faculty Advisor: Kamineni, Abhilash (College of Engineering, Electrical and Computer Engineering Department); Zane, Regan (College of Engineering, Electrical and Computer Engineering Department)
Transportation electrification will bring a positive effect on sustainable environments and robust economies. Electric Vehicles (EV) are emerging in today's market as a solution. However, the battery technologies on EV cannot compete with fuel and diesel cars in terms of energy storage capacity, and time needed to recharge (equivalently refuel). These limitations directly reflect on the consumers' convenience, the max miles the Vehicle can perform for, and hence EV adoption. Wireless power transfer (WPT) systems are seen as a solution to ease consumers' transition to EV. A high-level diagram of WPT is shown in Figure (**).
Major advancements in WPT technology has enabled the commercialization of stationary WPT solutions — materials technology has been a major impediment. Hence, academia and industry are jointly considering more advanced solutions in WPT, namely Dynamic Wireless Power Transfer (DWPT). Implementing DWPT systems will permit on the go charging for EV the upper hand over present charging (equivalently fueling) methods and encourage EV adoption. This document reflects on some of the challenges of realizing an effective DWPT that maintains power transfer between the primary pad and secondary pad, and a proposed solution to allow dynamic charging. Also, the proposed DWPT offers compatibility with SAE J2954 (Wireless Power Transfer for Light-Duty Plug-in/Electric Vehicles and Alignment Methodology) WPT3Z3 pad.
A comprehensive approach was taken in the design of the primary pad to validate the power transfer requirements for the designed pads. The proposed solution consists of a custom primary pad and a custom secondary pad for dynamic charging. This document will refer to the custom primary pad and a custom secondary pad as DGA and DVA, respectively. DGA offers compatibility with WPT2Z3 for dynamic charging and DVA offers compatibility with the Universal Ground Assembly (UGA) for stationary charging.