Graduate Student Dedicates Research to Detecting Gas Leaks in Vehicles
“I want to continue developing new cool stuff that can affect everyday life!” said Ruochen Yang, who is doing just that at the Center for Automotive Research (CAR).
Yang is a doctorate student in the Department of Electrical and Computer Engineering at Ohio State. Her research is based on the application of fault diagnosis for automotive systems. Specifically, she is using physics based and data driven models to more accurately detect evaporative vehicle emissions.
“My research is designed to help reduce the amount of hydrocarbons vented or evaporating into the atmosphere, for regulations such as the Environmental Protection Agency (EPA) and other organizations that monitor these emissions,” Yang said.
In a production vehicle, companies use evaporative emissions (EVAP) systems with On-Board Diagnosis that play very conservatively to EPA standards of gas emissions to keep them safely compliant with regulations. These systems will sometimes detect small leaks when the problem is not severe. Upon detection, a driver might receive a warning such as the check engine light on their dashboard.
“You bring it to a dealership or however you want to fix your car, and the technician will use a scan tool on the On-Board Diagnostics port in your vehicle to see where the fault is from. They will see that it’s from the EVAP system and that it’s a small leak problem. Usually the technician will then try to pinpoint the location of the leak, but if they can’t find anything that is reproducible, based on their experience, they will replace the most likely culprit- even if it’s a false alarm. It is highly possible that the same thing will happen again, and you’ll have to go back to the dealership. That reduces customer satisfaction and any part that is replaced during the warranty is a cost to the company,” Yang said.
A more accurate system for small leak detection would not only save vehicle manufacturers money, but it would also save customers time, and help to reduce overall evaporative emissions which is better for the environment.
Yang has worked on projects revolving around signal processing since her time as an undergraduate electrical engineer at Ohio State. Her undergraduate thesis revolved around implementing echolocation on a Lego robot. During her master’s at UCLA she worked on speech recognition, which is based in audio signal processing.
“Now I’m working in fault diagnosis but underlying it is signal processing of sensor measurements. So, looking back, it’s very consistent even though I worked in different areas, but it all builds up to where I am today,” Yang said. “In the future, I would like to go into research orientated but still application-based kind of work.”
Written by Muhammed Al Refai, CAR Writing Intern
