Back to ECE News

2004 Annual Report

Head Letter

Chair Letter

Campus Tansformation

Beyond Imaging

Photonic Biomedicine

Cell Behavior

No-Kill Sensors

Laser Scanning

Hokie Suit

Software Design

ECE Research Update

Research News

Cognitive Radio

Network Game Theory

Hardware Middleware


High-Temp Sensor

Cell phone detector

Optical Cryptography

Shoot-Through Failures



R&D 100 Award

Electronic Noses

Distributed Generation


$1000 Elevator

Embedded Systems

Defect Tolerance

Efficiency Tools

Pervasive Networks

Video Networks

Networked Testbeds

Real-Time Solutions

2002/2003 Ph.D.s

2003 Patents




Special Report:
ECEs and Biomedicine

April 2004

Photos courtesy of Detroit Edison

Small generators for Detroit Edison are running control software developed by Virginia Tech researchers to control their generation output to solve peak load problems in a cost-effective manner. The software is the first of its kind to incorporate distributed generators with existing large power plants. “When we first started up the Detroit system, field measurements were within one volt of the calculations,” said Robert Broadwater, who served as the project leader.

Real-time control of distributed generation

ECE researchers have developed the first real-time control software for power generation firms to control distribution circuit low voltages and overloads using distributed generators to supplement large generating plants. The software is currently operating 24/7 on six distributed generators at Detroit Edison.

Distributed power generation is increasingly popular and is expected to represent 20 percent of all power generated by 2020, according to Robert Broadwater, the project leader. “Even though the fuel is more expensive, distributed generation is economical for many utilities because overall transmission and distribution costs are lower,” he explained. He noted that transmission lines can cost $1 million per mile and distribution lines can cost $200,000 per mile.

“Many circuits need additional generation capacity to accommodate peak loads, which can be as minimal as 50 hours per year,” he said, “and firms are installing small generators physically close to the demand to meet the peak loads.”

The ECE team, which includes Ph.D. candidate Dan Zhu, developed software to incorporate the smaller generators into existing system-wide controllers. The software takes measurements from the circuit, and using load research statistics, calculates the needed generation. Previous work has shown that the load statistics can be used to predict within 4 or 5 percent of the actual load on the circuit.

“We don’t want to over generate, because the fuel from the small generators is more expensive, so the software calculates just enough power to remove the load problem,” Broadwater said. “This can save the electric utilities significant investment dollars, since it allows small generators to be smoothly integrated into their existing systems.”

The circuit models, which may include more than 5000 components, have been verified by several utilities. Without this model-based load prediction capability, the installation of the small generators would be much more expensive due to additional instrumentation and communication equipment required.

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: Thu, Jun 10, 2004