Alex Flueck, Illinois Institute of Technology
From 2004-06, Illinois Institute of Technology (IIT) experienced 12 power outages. Each cost up to $500,000. Because IIT owns the 4.16 kV on-campus distribution system, it decided in 2006 to solve its costly reliability problems by exploring advanced technology. At the same time, the university was planning to launch a major sustainability program. These issues led to an opportunity to merge reliability and energy efficiency upgrades as it investigated smart grid technology for IIT’s distribution system.
Four electrical and computer engineering department faculty joined experts from the facilities department and several outside organizations, including the Galvin Electricity Initiative (GEI), Commonwealth Edison Co. (ComEd), S&C Electric Co. and a consulting group that became Intelligent Power Solutions. The team developed a conceptual model for the IIT distribution network called Perfect Power based on S&C Electric’s High-Reliability Distribution System (HRDS).
In 2007, the team submitted a $12 million proposal to the U.S. Department of Energy’s Renewable and Distributed Systems Integration program. The project launched in fall 2008. By the end of 2009, the team had achieved the first year’s milestones, as well as additional milestones.
In summer 2006, the GEI hosted a workshop on microgrids spurring innovation in the heavily regulated electric power delivery sector. IIT power engineering faculty attended to see if we could help with long-term research. At the time, IIT was experiencing three or four outages per year, with some outages costing the university roughly $500,000. In addition, the university’s load demand was increasing, and the facilities department was exploring constructing a $5 million substation on the east side of campus.
Based on the GEI microgrid workshop, an IIT follow-up meeting discussed the possibility of implementing advanced technology on IIT’s campus to address rising energy use and costs, the school’s rising carbon footprint, reliability problems and the integrating renewable sources into the campus power grid. New technology required capital. Preliminary economic analysis showed that the payback period was about five years because of the high outage costs. Hence, the Perfect Power project was launched.
Perfect Power Goals, Technology
The Perfect Power project has several goals in three categories: technical, financial and leadership. The technical goals include demonstrating Perfect Power’s key capabilities regarding reliability, demand response load reduction, energy efficiency load reduction and renewables integration. The financial goals include deferring major capital costs, reducing energy and outage costs and the influx of ancillary services’ revenues. The leadership goals include reducing the university’s carbon footprint, creating a living laboratory and the opportunity to lead smart grid development through the Perfect Power project.
Perfect Power incorporates advanced technologies including a few being developed through the research thrusts in parallel with the commercial equipment deployment.
- Smart grid and technology-ready infrastructure (HRDS, IPPSC, smart meters and demand response, ZigBee load control, on-site generation and storage, renewables)
- Intelligent distribution system and system controllers (HRDS, IPPSC)
- On-site electricity production and storage (turbines, backup gensets, uninterruptible power supply (UPS))
- Renewable energy sources (rooftop photovoltaic (PV), on- and off-campus wind)
- Demand response capability (consumer-driven load control—Siemens controllers and wireless ZigBee controls for large and small air conditioning, lighting, major loads, office and laboratory plug loads)
- Intelligent Perfect Power system controller (coordinate demand response actions with local utility and independent system operator (ISO); eventually real-time markets; smart meter application for a commercial customer; energy analytics)
- Sustainable energy systems and green buildings and complexes (PV, wind; carbon dioxide reduction; improved efficiency—insulation, building envelope upgrades, major mechanical upgrades).
Perfect Power—Deployment Timeline
The Perfect Power project will span five years. The first two phases are complete. In addition, the first loop of the HRDS has been installed and commissioned. Four of the core academic buildings have equipment supplying power to the building loads. The next loop of the HRDS will be installed this summer. Major construction will begin immediately after the university’s spring commencement ceremony.
Phase I (complete)
- Energy efficiency upgrades (lighting, high-efficiency HVAC, hot water building loops) (reduced energy usage: five- to 10-year payback)
- Utility supply reliability repairs, upgrades (provided by ComEd)
Phase II (complete)
- Upgrade two existing 4-MW turbines for fast start capability (capital cost)
- Enroll in demand response programs (manual operation) (annual revenues to IIT)
Phase III (in progress)
- High-reliability distribution system (permissive overreaching transfer trip with backup directional comparison blocking; primary faults cleared in fewer than six cycles; fiber communication loop)
- Intelligent Perfect Power system controller (demand response coordination and electricity market monitoring)
Phase IV (future)
- Solar PV
- IPPSC (full implementation)
- Optimal efficiency, reliability, demand response, ancillary services
Perfect Power—High Reliability Distribution System
The HRDS is a major improvement over the typical, radial distribution system with manual switches. The figures illustrate the advantages of the HRDS compared with a traditional, radial distribution system.
First, a typical radial system appears in Figure 1. Feeder 1 serves two loads, shown by the switchgear blocks in white and the light bulbs in yellow. Feeder 2 serves one load. A de-energized alternate feeder is connected to two open switches, one at each end of Feeders 1 and 2.
If a fault occurs, as in Figure 2, then the substation breaker must open to clear the fault. All of the loads on the faulted feeder are out of service, shown by gray light bulbs. The outage could last anywhere from a few hours to half a day, assuming that the alternate feeder has sufficient capacity all the way back to the substation.
Following power loss, an electrician would be notified of the outage. In a typical distribution system, the steps required to restore power to all customers, and the time required for each step, are shown below:
- Respond: one to four hours,
- Locate fault: one to four hours,
- Isolate fault: manual—one hour,
- Close tie: manual—one hour,
- Total outage time: four to 10 hours.
Figure 3 shows the final arrangement of switches with the loads restored.
Figure 4 illustrates a major improvement to the traditional radial system presented above. An HRDS solution has the following features:
- Closed loop=single feeder,
- Simultaneous dual feeds to loads,
- Circuit breaker protection within the loop,
- Faults on main feeder cleared without outage.
If the same fault occurs in an HRDS (see Figure 5), then the event is over in less than one-tenth of a second, and no load experiences any outage. The sequence of events is as follows:
- Breakers isolate fault to only one section,
- Location: nearly instantaneous
- Isolation: 0.1 seconds
- Restoration: instantaneous
- Total outage time: zero seconds.
The key innovation in the HRDS is that the entire loop remains energized during the fault scenario presented. Additional equipment is required in an HRDS beyond the typical equipment in a traditional radial system with manual switches.
Figure 6 presents the smart grid components necessary for the HRDS. The components include:
- High-speed relaying for fault detection,
- High-speed communications between breakers,
- Coordinated operation: the right breakers open,
- High-speed interruption: fault cleared without outage.
Although the HRDS solution has an upfront capital cost, those costs are earned back as the university avoids the high cost of outages, such as lost productivity, food spoilage, ruined experiments, and off-site student housing.
Perfect Power—Research Thrusts
While the bulk of the Perfect Power project focuses on the four deployment phases, a significant portion of the funding targets several research thrusts. Three research areas will be presented briefly, including advanced distribution automation, ZigBee load control and the intelligent Perfect Power system controller.
Advanced Distribution Automation
Beyond available products, the Perfect Power team is exploring advanced, autonomous, agent-based controls for high-level distribution automation functions, including loss reduction and integration of distributed resources such as renewable solar PV generation and electronic converter storage. An example of an electronic converter storage system is a power electronic inverter system interfaced with a chemical energy conversion device, e.g., a battery or fuel cell.
To intelligently and safely incorporate distributed resources in a distribution system, an autonomous, agent-based control requires an appropriate model of the generation or storage device, as well as a communication network and framework for decision-making. If the distributed resource coordinates with the rest of the distribution feeder, then the opportunity exists to leverage the resource during normal and emergency operation, such as when a radial feeder segment is de-energized because of a fault.
Figure 7 illustrates a fault scenario on the IEEE 34 node three-phase unbalanced distribution feeder test system with additional smart switches and electronic converter storage (ECS) modules. The substation is at the left end of the feeder at bus 800. If a permanent fault occurs between nodes 828 and 830, then once the fault is cleared, switches SWI_28 and SWI_52 can be opened to isolate the fault. For the de-energized, yet unfaulted portion of the network, it might be possible for ECS 860 to supply power to some portion of the local load. Personnel safety is the No. 1 concern, so the coordination of ECS 860 and the rest of the distribution equipment is critical.
ZigBee Load Control
A second research thrust focuses on wireless sensor networks applied to load control. This work is led by professor Chi Zhou of IIT. The motivation for this research comes from our demand response objectives. If we need to reduce load on campus suddenly during a demand response event, then we have two options: Increase on-campus generation to reduce the overall load on the utility’s distribution grid, or curtail load on campus. If we need to curtail load, there are two methods: Shut down entire buildings via smart switches in the HRDS, or control individual loads within buildings.
Figure 8 presents a MeshBean ZigBee Mote from MeshNetics, now part of Atmel. The board contains a ZigBee radio for communication and several sensors and expansion connectors. The first applications of the ZigBee Mote include lighting and HVAC control. A research goal is to deploy a low-power wireless sensor or actuator network to interface with smaller loads distributed throughout the campus buildings. A few large loads, such as large air conditioner compressors, will be connected to a Siemens energy management building controller. Hundreds of smaller loads, however, must be monitored and controlled, which the ZigBee wireless network can do effectively and efficiently.
Intelligent Perfect Power System Controller
A third research thrust explores the creation and deployment of a hierarchical Intelligent Perfect Power System Controller (IPPSC). John Kelly and Greg Rouse of Intelligent Power Solutions lead this work. The controller must supervise the entire on-campus energy system by communicating with the upstream utility, ComEd; and ISO, PJM; as well as the campus distribution system, HRDS; on-site generation, fast-start natural gas turbines and renewable sources such as solar PV and wind; on-site storage, UPS; and individual building load controllers, Siemens energy management building controllers and wireless ZigBee sensor network controllers.
Some of the IPPSC’s tasks will include:
- Starting and stopping local generators and storage devices,
- Controlling local loads based on a predetermined sequence of operations and a load-reduction priority scheme,
- Automatic switching of loads to alternate transformers, campus feeds and substations,
- Placing a building or the entire campus in island mode.
The overall site energy system control scheme will consider economic, environmental, comfort, disturbance threats and other end-use objectives to decide the proper operating modes and sequences. The IPPSC enables perfection by anticipating system needs and taking action to mitigate system reliability and performance threats.
In the words of Robert W. Galvin, “Perfection is a journey, not a destination.” Transforming the U.S. power grid will be a journey with monumental change. As Perfect Power enters its second year, IIT continues looking for ways to align engineering and technical talent with society’s changing needs. The U.S. work force must be prepared to implement and sustain this change.
The Perfect Power project’s advances are helping impel the eventual power grid transformation. As it grows, Perfect Power at IIT will provide a living laboratory for advanced technology and a training ground for power engineers and technical personnel. IIT is expanding its energy programs by establishing the newly funded Smart Grid Education and Workforce Training Center. In April IIT received $5 million in American Recovery and Reinvestment Act funding to support a total $12.6 million collaborative effort to establish the center, which will offer smart grid technology courses and certification for people of all ages via on-campus and distance-learning classes.
IIT’s Perfect Power, smart grid and power engineering infrastructures will help train groups from utilities, corporations, labor unions, veterans, kindergarten through 12th-grade educators, universities, and community colleges to be the strongest work force in the world, taking on global smart grid challenges in technology, energy independence, clean tech and sustainable energy. The center will be complete within three years and is expected to train nearly 49,000 people in its first three years of operation. As IIT continues its journey, we invite you to track our progress and push for perfect power.
Alex Flueck is a professor in IIT’s Department of Electrical and Computer Engineering. Perfect Power is funded by the DOE, IIT and S&C Electric under Award DE-FC26-08NT02875 and led by IIT professor Mohammad Shahidehpour. Founded in 1890, IIT has more than 7,700 students.