ECE: Electrical & Computer Engineering
ECE News

Resilient, Sustainable Infrastructures

Print Version

image of article

.pdf Print Version
164 KB

Power lines and power grids

While smart grids, microgrids, and even dc-power grids are being discussed to improve the nation’s infrastructure, a team led by ECE researchers is exploring how a complex systems perspective can make power and communications networks less vulnerable to catastrophic failures. They hope to ultimately develop a safety net that incorporates local alternative energy generation and energy retail markets.

The $2-million effort, headed by ECE Professor Lamine Mili, is funded by the National Science Foundation (NSF) under its Emerging Frontiers in Research and Innovation program. ECE’s Sandeep Shukla is a co-principal investigator (PI), along with Michael von Spakovsky of mechanical engineering, Benjamin Hobbs of Johns Hopkins and Arnold Urken of Stevens Institute of Technology. ECE Professor Yilu Liu is also on the project.

Infrastructures team

From left to right: Lamine Mili, Yilu Liu, Sandeep Shukla

“Our goal is to improve the resilience of interdependent complex networks, such as power and communications, when facing catastrophic failures and natural disasters,” Mili said.

“Currently, the monitoring, protection, and control of electric power systems rely heavily on computer-based communications networks. Consequently, the failure of one can affect the functioning of the other. A series of cascading events may have a catastrophic impact on the whole society,” he explained. The risk of such events is increasing, he said, because of the current trend in operating infrastructure systems closer to their stability and capacity limits.

Simple rules grow complex systems

When viewed locally, complex systems seem to have grown by basic simple rules, explained Shukla, who is a co-PI. “For example, a network grows by simply connecting another node to existing nodes, or a forest grows by having its seed grow more trees in its neighborhood,” he said. “Such simple local phenomena give rise to very complex dynamics.” Fractals are an example of such a system.

Complex systems that develop without central planning are usually tolerant of defects and faults that are expected or planned for, but are not resilient to unanticipated problems. “Such systems are susceptible to cascading failure – like a wide area blackout on the power grid – where a failure in one location spreads through domino effect to cover larger and larger parts of the system,” Shukla said.

Extending the HOT model

The problem gets more complicated when two systems are intertwined, as with the power and communications networks. The team has selected the highly optimized tolerance (HOT) model, which was developed for a single complex system, and plans to extend it to accommodate the interdependencies of the power and communications systems subject to cascading events within and across these infrastructures. They will validate their new models with data from the North American Electric Reliability Corporation (NERC) and from Central Florida and Southern Brazil testbeds.

“The reason for such modeling,” Shukla said, “is that by understanding the dynamics, we can take safety measures against the unanticipated failures.” Since both the communication network and the power grid are complex systems and have cascading failure possibilities, “we have to optimally place guards to contain the failure locally. This must be done in a way that the cost of guarding for fault containment does not exceed the projected cost of failure. In our costs, we include resources, dollars, and human lives,” Shukla said.

Once the possibilities and vulnerabilities can be measured and protective actions identified, a computerized control system will be developed. The team from Stevens Institute of Technology is working on software agents and decision theories that will enforce the protection mechanism, look for vulnerabilities and make decisions about emerging problems.

Microgrids with alternative energy sources

Fractal Fractal

Fractals are well known representations of complex systems. The two shown here describe the dynamics and complexity of the power system. Top: A low resolution representation of a true fractal, showing that the computations in the power system are not necessarily clear cut. Bottom: A truncated fractal color coded according to where the energy flows.

The team anticipates that local distribution grids incorporating non-traditional energy sources – called microgrids – may be part of their strategy, according to Shukla. “In a cascading blackout, if a part of the power grid gets disassociated from the rest of the grid, the local microgrids could tap the smaller amounts from alternative energy sources to keep the lights on locally,” he said. Furthermore, they will provide a power system with an enhanced level of resiliency that will enable it to recover from a blackout in a speedy and less costly way.

A sustainable power grid involves not just reliability and resiliency issues, but also environmental and economical issues. Michael Von Spakovsky of mechanical engineering is exploring various scenarios of fuel mix and technologies that are more efficient and generating less pollutants and greenhouse gas emissions.

Once the possibilities and vulnerabilities can be measured and protective actions identified, a computerized control system will be developed. The team from Stevens Institute of Technology is working on software agents and decision theories that will enforce the protection mechanism, look for vulnerabilities and make decisions about emerging problems.

“...the local microgrids could tap the smaller amounts from alternative energy sources to keep the lights on locally...”

The revenue model for such locally generated energy provided by local energy sources should be economically viable without public subsidies. So, economists and environmentalists from Johns Hopkins are on the team, working on new environmentally friendly electric energy market models. “We expect the communication network to play a crucial role, as the price of microgrid-enabled energy will be based on auction strategies in virtual markets and the price negotiations will be done over the microgrid’s associated communication network,” Shukla said.


Meeting the grand challenge

All of the efforts, from network stability to auction markets and environmental assessment, are needed to meet what Mili calls the grand challenge. “We need to discover how to design, manage, and operate in a robust and sustainable manner such a complex multilayered system of systems, where each layer contains millions of components switching on and off in a random manner and spreads over a continental or global scale,” he said.

“This work will provide a basis for using the communications and power infrastructure control mechanisms to regulate the stability of water, transportation, and other infrastructure systems. These critical infrastructures can be regarded as the backbone of a country’s economy, since they provide material support for the delivery of basic services to all segments of a society,” Mili said.