Jason Nezvadovitz

Mentor: Dr. Carl Crane
College of Mechanical and Aerospace Engineering
 
"As part of the Machine Intelligence Lab, I work on SubjuGator, an ongoing student design project where an autonomous submarine is developed for entry in the annual RoboSub competition hosted by AUVSI and the Navy. This year, we found extremely nice brushless thrusters to use for our new mechanical system. However, they come with propellers that are not ideal for our application. After extensive reading on marine propulsion, I realized that designing better propellers is more of a research endeavor (with empirical testing and analysis) than a typical engineering design task (where well understood mathematical methods drive you to the solution). Thus I decided to turn this task into a study. The results of this research will also provide useful information to the huge industry surrounding marine propulsion."

Major

Mechanical Engineering

Minor

Electrical Engineering

Research Interests

  • autonomous systems
  • machine learning
  • optimization methods

Academic Awards

  • Assorted MAE Undergraduate Scholarships for academic success in mechanical engineering
  • Assorted ECE Undergraduate Scholarships for academic success in electrical engineering

Organizations

  • Machine Intelligence Lab
  • UF IEEE

Volunteer

  • N/A

Hobbies and Interests

  • Designing and implementing autonomous vehicles.
  • Talking about math.
  • Weight training. General fitness.
  • Playing music (guitar and piano).

Research Description

Optimal Propeller and Nozzle Design for the Thrusters of an AUV
Background ----- SubjuGator is an ongoing autonomous-underwater-vehicle (AUV) project within the Machine Intelligence Lab at UF (http://subjugator.org/). The eighth generation vehicle is currently under development, and among many other advancements, it will make use of high-performance brushless electric motors in its thrusters for propulsion. Unfortunately, the motor manufacturer can only provide these thrusters with an outdated propeller and nozzle that were designed for one of their older product models. Performance tests run by the manufacturer have shown that not only are the old propeller and nozzle incapable of properly loading the new motors, they are also highly asymmetric in thrust output in the forwards and backwards directions. Proposal ----- The objective of this research project is to develop an optimal propeller and nozzle design that will replace the outdated propeller and nozzle for these thrusters. The new thruster will meet the following requirements: • The propeller must load the electric motor such that it draws roughly 80% of its maximum rated power when it is spinning at its peak efficiency RPM. o The motors have been dynamometer tested by the manufacturer, and that data will be used to spec the performance characteristics of the new propeller and nozzle. Namely, the operating region (80% of max power draw) for the thruster should be placed around the peak efficiency RPM of the motor, which is known. • The thrust at any given RPM in the "forward" (clockwise) direction should be within 5% of the thrust at the same RPM in the "backward" (counterclockwise) direction. o Symmetry of output force is highly desirable for autonomous systems with fixed actuator orientations; it provides increased stability and decreased likelihood of actuator saturation. Most AUV’s fall into this category, and thus thrust symmetry is also desirable to the manufacturer. • The outermost diameter of the nozzle is no more than 5 inches. o This geometric constraint has been provided by the manufacturer. • The thrust generated (in either direction, since they should be roughly the same) is maximized. Like most design problems in fluid mechanics, propellers do not have a simple set of equations from which the performance characteristics can be accurately derived. Instead, this research will entail a significant amount of computational fluid mechanics (CFD) techniques, as well as rapid-prototyping and empirical testing as the design converges to an optimal solution. Design, analysis, and testing will be carried out primarily with STAR-CCM+ CFD software, additive and subtractive rapid-prototyping machines, and a custom designed static-thrust testing-rig.