ECE News

Fall 2003
Power Monitor

 BioMed Option



 Bulleted Slides

R&D 100 Award

 FDA Symposium



 Night Clouds

 Head Letter

 Exemplary Award

 Faculty Retirements

 Faculty News

 Tucker Inducted

2003 Annual Report

Spring 2003

(Back to Student Projects)
(Back to Undergraduate Research Projects)
Fall 2003

Undergraduates Build Equipment to Monitor Power Grid

A team of undergraduates designed and built frequency recording equipment that will be used in an internet-based network to monitor the health of the nation's power grid. From the left are: Daniel Nash, Charles Lewis, Patrick McDougle, Stephen Nash, and Jonathan Perry.
A team of ECE undergraduates have spent the past year designing equipment for a nationwide, internet-based power frequency monitoring network (FNET) that may help prevent cascading blackouts like this summer's event.

Working with a research team of power engineering faculty, the students are developing inexpensive frequency recording units (FRU) for 110V single-phase outlets. The FRUs will be placed in universities, schools, and offices across the country. Using the internet for communications, the system will allow continuous, wide area collection of GPS time-stamped frequency data.

Measuring Frequency to Monitor Power Disruptions
The real-time, wide-area, synchronized measurement system is based on the concept that system frequency remains constant regardless of voltage level. However, when a significant disturbance occurs in the power system, the frequency varies in time and space.

The system is being deployed with funding from the National Science Foundation (NSF), ABB, and the Tennessee Valley Authority and will make observation of the country's entire power network possible for the first time, according to Yilu Liu, the principal investigator on the FNET project.

Replacing Incomplete Models with Data
"A complete picture of how the entire U.S. power behaves during disturbances is not well understood," she said. "Any information we have is from computer simulations that are subject to model errors and simplifications. Complete models for nonlinear loads are not available despite the critical roles they play in system dynamics. Most of all, no model could capture the time varying nature of power system operational conditions over long time spans — we need some real-time measurements," she explained.

Real-Time Data for Operators and Researchers
The FNET system is expected to be a boon to both power researchers and utility operators. FNET will give researchers data for studies of inter-area oscillations, disturbance analysis, scenario reconstruction, and model verification. Researchers hope to use data from a nationwide FNET system to forecast incipient system breakups and recommend remedial actions. FNET's continuous synchronized information will provide system operators all over the country with close to real-time system status so that they can better manage the grid and for post-mortem analysis after a system disturbance.

FNET will provide a common time stamp, which will help operators reconstruct what happened after a major event. One of the delays in understanding this summer’s blackout has been the use of different timing by the companies involved.

Inexpensive, Practical Monitoring
Power engineering researchers have worked for decades to develop the tools for the measurements needed. In the early 1980s, a Virginia Tech ECE team led by Arun Phadke developed a real-time measurement device, the synchronized phasor measurement unit (PMU). PMUs measure voltage and current at power stations, and use the GPS system for time-stamping. Although the device was commercialized in 1985, its installation is measured in the hundreds, compared to the thousands that would be needed for system-wide measurement and analysis.

"Utilities operate with tight profit margins, and the cost of installing computers and devices at every power station can seem daunting," said Phadke, who is one of the faculty members of the FNET team. PMU's can cost from $10,000 to $25,000 per unit, not including substation installation and dedicated communication channels. By comparison, the equipment for the FNET system, will cost $500-$800 with almost no installation cost, he said.

Liu said that although the FNET system gives up certain parameters, such as tie line voltage angles and flows offered by PMUs, it is a practical way to quickly deploy a cost-effective, real-time, wide-area synchronized measurement system.

Deployment this Fall
Deployment of a prototype FNET system is expected to begin later this fall. Initially, FRUs will be installed in 50 office buildings that are electrically close to major generation centers and major transmission tie-lines, and will measure the Eastern region of the national grid. Being portable, the FRUs will be relocated according to the needs of the project.

Undergraduate Research/Design Experience
The FNET research team selected a group of advanced undergraduates to develop the equipment. "This project provides an excellent opportunity for undergraduates to use their design skills in a real-world, state-of-the-art application," Richard Conners, a faculty team member, who worked closely with the students. "We are very pleased with their work. This was a very talented group of students and it was an honor and pleasure to work with them."

Stephen Nash (BSCPE '03), was on the project since summer 2002 and served as team leader until this summer. "We met with the professors last summer," he said. "They told us, 'this is the FRU. Here is how it fits into the FNET. Go build one.'" They spent the first summer researching and ordering the components, then the past academic year integrating the components and developing a working prototype.

Design, Components, Integration
"It is amazing what is out there," he said. "We kept thinking we were going to have to invent things, then would find that somebody else already was selling it. We just had to find it." In one case, the team thought they would need to purchase a $20,000 license and develop proprietary code, then found hardware that did the job for $150. "We just had to plug in a wire and read the manual," he said.

Finding the optimal components was a big challenge, agreed teammate and brother, Daniel Nash (BSCPE '03). "We thought at first we would find one chip that would have all the subsystems we needed. It didn't fall in our lap that way, as a lot of chips did not have ethernet incorporated."

Other technical issues involved getting the GPS to communicate with the chip, interfacing the network with the board, coding the algorithm that calculates the frequency, and integrating the filter transformer that plugs into the wall and measures the voltage. "We didn't initially realize how computationally intensive the measuring and analysis would be," he said. "In future generations we hope to use a faster processor."

Stephen Nash said one of the biggest lessons he learned from the experience is to make more realistic time estimates. "I would initially say that a task would take two weeks, and find that it took more than two months," he explained.

Joining Stephen and Daniel on the undergraduate team were: Patrick McDougle (BSEE '03), who worked with the GPS system; Jonathan Perry (BSCPE '03), who worked on the analog and digital converters and served as team leader this summer in developing the commercial units; and Charles Lewis (BSCPE '03), who worked with the ethernet communications. Sarin Phillip (BSEE '04) joined the team this summer and is responsible for the PCB design. Conners and Virgilio Centeno serve as faculty advisors for the FRU design. Engineers Rick Cooper and Bob Lineberry, from the ECE Shop serve as technical advisers

Privacy Statement | Contact Webmaster

© 2006 Virginia Tech Department of Electrical and Computer Engineering
Images on this site are the property of Virginia Tech.
They may not be used for commercial purposes.
Last updated: Tue, Nov 4, 2003