BMW_Leipzig_MEDIA_050719_Download_Karosseriebau_max.jpg
Nellis_AFB_Solar_panels.jpg
Internet_map_4096.png
BMW_Leipzig_MEDIA_050719_Download_Karosseriebau_max.jpg

Nasco-NHK


Nasco-NHK

America Suspension

Components Inc.

Electrical Engineer

SCROLL DOWN

Nasco-NHK


Nasco-NHK

America Suspension

Components Inc.

Electrical Engineer

I started working as an Electrical Engineer at Nasco-NHK in May of 2015. As only one of three Electrical Engineers, my role in the company was extensive, covering maintenance to design and programming of entirely new production lines and equipment. Nasco-NHK is a Japanese owned company located in Bowling Green, KY which manufactures automotive coil springs, disk springs, electric motor cores, and torsion bars. My primary work was with Mitsubishi PLC's which I became quickly adept at programming. I was personally involved in programming and designing novel countermeasures for a laser marking system, full implementation of three brand new conveyors with part tracking, creating full network communication between PLC's on several production lines, factory-wide productivity and runtime tracking, and performing routine maintenance and feature additions on all sorts of mechanisms factory-wide.

 
mainpic.jpg

Institute for Combustion Science & Technology


Institute for Combustion Science & Environmental Technology

Electrical Engineer

Institute for Combustion Science & Technology


Institute for Combustion Science & Environmental Technology

Electrical Engineer

 

 

Dr. Yan Cao, ISCET Director

Dr. Yan Cao, ISCET Director

Amplifier Design

For some time I did design work for the Institute for Combustion Science and Environmental Technology (ICSET) developing high speed, high accuracy, high gain amplifier technology for interfacing a LabVIEW DAQ and a Piezo Electric micrometer actuator. Much of the project is still in development and unable to be published, but my aspect of the project was completed in Summer of 2015 with great success. The design of the PCB was based on an example board that came unable to meet our design demands, and our model eventually allowed for signal gains from micro-volts to 120V AC, at very low current, using 60V power supply and a board measuring only a few inches in height and width.

Applied Physics Institute


Applied Physics Institute

Electrical Engineering Student

Applied Physics Institute


Applied Physics Institute

Electrical Engineering Student

CHEMICAL DETECTION

Photo by Geo-grafika/iStock / Getty Images
Photo by Geo-grafika/iStock / Getty Images

MACHINE LEARNING in APPLICATION

We used K-Nearest Neighbor techniques to classify chemical activation of various sensors along a chemical detection array as differing chemical compounds, along with estimating their intensity in the atmosphere. This, effectively created an electronic nose which could sense and classify gaseous chemicals in it's environment. We worked on developing consumer, automotive, and academic applications of this technology, and developed several prototypes for mass marketing.

Nellis_AFB_Solar_panels.jpg

Chemistry Department


Solar Panel Testing and Implementation

WKU Chemistry Department

Chemistry Department


Solar Panel Testing and Implementation

WKU Chemistry Department

WKU Chemistry Department

LabVIEW Solar Panel Testing

        I was hired by the Chemistry Department at Western Kentucky University over the winter of 2014 to create for them a LabVIEW interface to test prototype solar panel's IV characteristics using a Keithley 2400 SourceMeter. It was important that the application be able to use these IV measurements in a variety of ways, such as to calculate the solar panel conversion efficiency and fill factor, as well as the IdVg and IdVd characteristics of the device. The task itself was not exceedingly complicated, libraries were provided for the device on the company website (examples were provided as well), I had extensive experience programming in LabVIEW, and the technology was familiar to me. However, we did run into several challenges along the way which made this relatively simple project a unique experience in understanding the challenges of real world engineering.

        The program itself was simple. I decided to use a state-machine layout for the design of the main interface, giving the queue an enum type-definition that contained each step of the process needed to make the measurements and display them to the user. In the end, the user was only required to select the type of measurement he/she wanted from a menu on the main interface, and click measure; the rest was handled by the program itself. The program internally handled minute details of the user interface, such as changing the x and y labels on the graph to measurement appropriate values, and handling multiple plots along with a proper legend. Since all data was graphed neatly onto the main interface, saving the data was easy. Property and invoke nodes were a highly necessary component of the program, as multiple loops, event structures, and queues handled how data would flow throughout the code at the users command.

2014-01-06 09.51.35.jpg

        Two issues came up along the way that could have potentially delayed the project past its expected two week deadline. First, the GPIB card that was installed in the computer provided by the department was obsolete, and unable to be recognized by the new operating system our campus IT was switching too. This resulted in an inability to interface with the necessary device via the computer, and a new card had to be ordered before we could begin debugging the code. Second, the Keithley SourceMeter itself was experiencing common hardware difficulties that myself and others at the engineering lab were unable to successfully predict and identify, and thus repair proved unfeasible for the time being. The hardware difficulties, however, only rarely interfered with the implementation of the code, and due to the extreme cost of the specialized equipment, we chose to ignore it and implement safe shutdown procedures into the code instead.

        Commenting and documenting the code is always the last step. With a full library of assisting code, and a fully working unified interface for the source meter, I believe this project was a complete success! It was a pleasure working with LabVIEW for these two to three weeks, and I am glad to have learned a lot about solar technology and data acquisition in the time I spent on this project.

Dates: 01/06/2014 - 05/01/2014

Internet_map_4096.png

SALSA @ PNNL


"SALSA" SociAL Sensor Analytics

Pacific Northwest National Laboratory

SALSA @ PNNL


"SALSA" SociAL Sensor Analytics

Pacific Northwest National Laboratory

SALSA @ PNNL

An illustration of how social media posts involving different themes sometimes intersect. (Credit: Image courtesy of DOE/Pacific Northwest National Laboratory)

An illustration of how social media posts involving different themes sometimes intersect. (Credit: Image courtesy of DOE/Pacific Northwest National Laboratory)

        While working for PNNL, I was given the task of constructing a mathematical model to increase the detection of twitter-traffic spikes in a large web-traffic time-series. Over the course of several months, working with a mathematician PHD and two undergrads of differing specialties, I learned how to write fluently in Python, using several different mathematical and graphical libraries, and wrote thousands of lines of well documented code for every task imaginable. My primary line of research was in the area of Fast Fourier Transforms (FFT), which required a lot of rigorous mathematical study and code manipulation on my part. I spent months writing and testing highly complex algorithms for mathematical analyses, and even more time interpreting the data and trying to improve the graphical outputs my code would generate. In the end, the experience was highly rewarding, and it taught me how to learn and research independently, as well as how to function well in a business environment.