May 05, 2010

Electrical network design is a complex, continuously evolving process. Energy Central's Sierra Energy Group says that as many as 150,000 people are working on this arcane pursuit.

Many of these people have engineering degrees that enable them to deal with the complex mathematical and scientific calculations necessary to deal with America's complex and constantly growing electrical grids. Whenever a new generation source comes online, engineers have to design the system that will take the power from the station, get it to the correct voltages for the transmission system, specify all the proper equipment and supervise the construction. At the other end of the grid, when a new subdivision -- or even one house -- is added, similar steps are taken at lower voltages.

If all this work isn't done correctly, the grid can become unbalanced and fail at any one of millions of points along the pathways that electricity must follow from the point where it is generated to the home or business where it is used. Failures are not tolerated very well.

As politicians and others push for what is called the smart grid, they are envisioning a new system that doesn't always involve a straightforward electrical flow from central generation to end user. Smart grid is expected to embrace intermittent resources, which at times may be drawing power from the grid and at other times reversing the flow -- through devices such as electric cars, solar cells and wind turbines, some of them not yet in production, but envisioned.

In addition to intermittent, two-way flows, design engineers also are going to see an increasing overlay of communications system and information technology on the grid.

Most of these communications systems are being designed for Internet Protocol (IP) communications, which do not flow very well over power lines. CenterPoint Energy, for example, abandoned an experimental pilot involving power line carrier communications to read smart meters. Thus, a separate communications system to reach smart devices on the grid, substations, transformers and transmission systems is going to have to be designed and overlaid on the grid.

In addition, consider in-home devices that must be communicated with to enable demand response -- reducing residential consumption to compensate for a decline in central generation. "Most of the technology is going to be able to cope with the new challenges. Smart grid doesn't do much until you get to end-use consumer products," said Dan Walter, senior manager, energy markets with Tri-State Generation and Transmission Association. "We need to start getting washing machines ready to turn on/turn off. Smart grid with electric cars is going to be tricky. There will be a real challenge with intermittent resources. And if you believe some of the draft legislation on where they want to be on carbon in another 20 years, that will be a huge challenge."

Expensive Endeavor

Price, both of electricity in the future, and of the systems and networks that have to be built out, is the major issue. At a meeting in New York City, Dian Grueneich, commissioner of the California public utility commission, said publicly that building out the smart grid is going to "cost trillions of dollars." From a technological perspective, all this is doable, if complex.

Baltimore Gas & Electric Co. is struggling with the cost issue on smart grid, according to Barry DeBald, a senior information analyst with the utility. "Right now, everything is pending public service commission approval," he said. "We've done some preliminary rollouts on smart meters and they worked well. But going system-wide is a $500 million to $600 million proposition. Whether we're going to be approved is a coin flip. As far as how network design will change, as we see people starting to shed load, that should take some strain off the system. As we start to see more plug-in vehicles, that will shift load significantly. The question is will we ever get to where we balance the load or unbalance the load? It's hard to say, but there are going to be impacts on design because we're going to be using more load off-peak."

The grid itself has become smarter from generation to generation with SCADA, distribution automation and other advances coming online over the past 20 to 25 years. What is missing is the communications overlay to enable smart meters, electric cars, intermittent renewable electricity and two-way flows.

Already control room operators can look at the grid in real-time on computer screens and see what circuits are functioning properly and which have outages. Already they can manipulate electric flows from their workstations. Already they can see the location of service vehicles as they move along streets and highways. Not all utilities have all of these capabilities, but many of them do and more become so-equipped every year.

Some utilities can already update grid design systems from the field with as-builts. Most utilities don't do this extensively because they like to have system designs checked and double-checked before they are placed into service. But the capability already exists in many places. Engineers can already make live updates from the field if they want to.

The smart grid really is not that great of a challenge from the technological or design perspective and new capabilities become available almost daily. As Grueneich and DeBald point out, however, the issue is going to be costly. It took untold trillions of dollars to build the grid as it is today. It will take additional trillions to rebuild it.

 

Respond to the editor.
Warren Causey