USC Logo 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Machinery Diagnosis and Prognosis

Text Box: MDPL Main Page

Text Box: About Myself
Text Box: RCS

By: Andrew Jarvie

The Machinery Diagnosis and Prognosis Laboratory (MDPL) is a group of highly dedicated Professors, Graduate, and Undergraduate students that are working closely with several major companies to utilize new technologies to better understand mechanical wear and fatigue. The goal is to develop a system to analyze data taken through various sensors and report a full diagnosis of the mechanical system. Utilizing previous data and real time data, this system will also be able to predict the remaining life in the system. This phase is the prognosis portion of our efforts.

     There are two major projects the MDPL group is currently working on. First is the primary project that is working joint with the SC National Guard and the BF Goodrich Corporation. In this project, the study of mechanical wear to developed the diagnosis and prognosis system will be explored on military craft. The project is currently in the beginning phases, which includes building the test stand which data will be taken from, "mining" (categorizing) previously found data to develop a computer program to read the data and provide simulations, and preparing the experimental setup.

    My particular task on this project is to work closely with a graduate student to assist in the design and construction of the test stand. This involves the design and setup of a test stand to study mechanical diagnosis and prognosis of specific drive-trains. The design must take into account the ability to easily remove and reinstall both sensors and drive train components for different tests. The various sensors measure vibrations, strain, temperatures, etc. In addition, the design must allow for easy maintenance, since the test stand will be studying mechanical wear and progressive damage. We are currently working with several companies to decide what motors, dynamometers, etc. will work best, and how to set them up based on our designs. I also do allot of work with designing in a software package commonly known as Pro/E.

    The second project the MDPL group is working on consist of vibrational analysis of a triple cantilevered beam of unique composition. This project is funded by the GulfStream Aerospace Corporation, and the beam that is undergoing tests is proprietary; therefore, we do not have privileges to post actual dimensions, pictures, or composition of the beam. On this project I am assisting a graduate student in running the tests on the beam.