Modern metering, old grid

If the utility industry ever adopts terminology from the military, they might begin calling smart meters the point-of-the-spear when it comes to the smart grid. That's because smart meters are how most consumers experience smart grid features. The problem is smart meters alone do not constitute a smart grid. They are of course only one part of a more complex system of energy distribution. Unfortunately, a truly smart grid will never materialize without the automation of the less glamorous energy transmission and distribution components.

And there are problems galore with getting the grid to a point where it can accommodate a significant amount of capacity in renewable energy sources, which are typically intermittent. Then there is the coming challenge of an electric-vehicle (EV) infrastructure comprising thousands of EV charging stations. To make matters even more complicated, these stations can potentially act as tiny energy providers if need be, sucking power from EVs that are garaged and idle.

All in all, grid resources and demands are likely to grow increasingly uncertain and hard to plan for. Of course, these problems are widely appreciated. Typical of the efforts to deal with such difficulties are those of the Power Systems Energy Research Center (PSERC), a research consortium that is investigating areas which include dynamic grid reserve requirements and hierarchical coordinated control of grid resources. To do so, they are “leveraging existing digital technologies that can enable effective end-to-end adaptation of renewable resources into the electric grid system,” says Vijay Vittal, director of PSERC and Ira A. Fulton Chair in Electrical Engineering at Arizona State University. The school is working with universities nationwide in pursuit of this goal.

Of course, the automated meter reading (AMR) environment made possible by smart meters should let utilities more easily base their rates on “time of use” rather than the flat rates residential users are accustomed to. The DoE says some 3 million homes already have smart meters. About 65 million such meters will be installed in the U.S. by 2020 according to the utility industry's Institute for Electric Efficiency.

But some consumers are going into the age of AMR kicking and screaming. A recent analysis by the Electric Power Research Institute (EPRI) estimated an intelligent smart-metered grid could reduce electricity use by more than 4% a year by 2030. Nevertheless, many consumers think these benefits accrue to the utilities, not to them. In pilot projects, some smart-meter households report the net result of smart metering was a higher electric bill. Such incidents lead many to question meter accuracy, security, and potential for compromised privacy.

To be fair, utilities have some blame in this situation because they haven't clearly explained how smart metering benefits the public. “Beyond the fact that the smart meter itself is a relatively simple tool, the customer relationship is a delicate thing,” says Richard W. Caperton, a policy analyst and climate expert with the Center for American Progress, a liberal policy group. “It needs to be handled with that in mind,” he adds.

Consumers have security concerns about data gleaned from their usage habits, and with good reason. After all, smart grid data is protected by the same kinds of facilities - firewalls, intrusion detection, network access authentication - used to guard IT resources. And there are successful hacks of these defenses all the time.

One suggestion coming from Dept. of Defense officials is that infrastructure resources such as the smart grid be put on their own isolated networks. Others advocate using security measures that meet high international standards. The standard covering IT security is called ISO/IEC 15408. The highest operating system protection profile under this scheme is dubbed the Separation Kernel Protection Profile (SKKP).

SKKP incorporates a variety of techniques to head off trouble. For example, operating systems certified under SKKP must have ways of isolating malicious applications so they don't corrupt others. SKKP also spells out software development practices and even demands that operating systems withstand attacks from NSA experts who have access to the source code.

There are already commercial packages incorporating such extreme measures. One is from embedded software provider Green Hills Software which has an operating system meeting the NSA high-robustness Common Criteria EAL 6+, the highest level available within SKKP. Green Hills says it is working with other companies on more bullet-proof smart grid security architectures.

Page 2 of 3

A failure to communicate

The Global Smart Grid Federation, which helps companies address technical and policy issues associated with smart grid, says that smart meter vendors are growing more concerned with security and now regularly scan their codes for potential problems, this in the wake of live demonstrations at high-profile security conferences where security experts were able to exploit vulnerabilities in smart meter code. Similarly, vendors of controls for electric plants are reevaluating their products' security functions in light of recent attacks such as Aurora, wherein users would unknowingly download malicious programs that opened remote backdoors to their computers, which then established encrypted covert channels that masqueraded as SSL connections. This gave attackers access to the computers and to other parts of the networks.

There are also efforts underway to address privacy concerns. In particular, the Office of the Information and Privacy Commissioner of Ontario, the Canadian utility Hydro One, and several other companies working in smart grid technology have come up with general practices for ensuring that user information stays private. The parties involved have enough clout to make smart grid suppliers pay attention. The best-practices they devised are general principles rather than specific to technology platforms. For example, they say smart grid systems should feature privacy principles in their overall project governance framework and proactively embed privacy requirements into their designs. They also advise manufacturers to build their systems such that privacy is the default mode, as in, “no action required.”

Standards in addition to those for privacy are in the works as well, simply because even basic communication schemes among grid components are still up in the air. That's why the governing board of the Smart Grid Interoperability Panel (SGIP), which coordinates the development of the smart grid framework, recently addressed the need for wireless communications among grid-connected devices as well as the ability to upgrade household electricity meters as the smart grid evolves.

The SGIP's “Guidelines for Assessing Wireless Communications for Smart Grid Applications” and “Meter Upgradeability Standard” are now among 17 other standards development projects called “Priority Action Plans,” or PAPs.

The wireless PAP recommends standards for wireless communications between all devices connected to the smart grid - not just residential meters, but also components in power plants, substations and transmission systems.

“Technologies like Wi-Fi and Bluetooth were not designed with Smart Grid in mind,” said NIST manager of emerging and mobile network technologies Nada Golmie. “What [the wireless PAP] does is ensure that any technologies that we use - whether off-the-shelf or not - will provide the features the grid needs.”

Golmie says that - to give one example - there can be far less tolerance of delays between transmission and reception or interruption of signals among grid devices than there is among general data communication devices, such as cell phones.

Similarly, the meter upgradeability PAP makes it possible to upgrade any meter as the standards evolve, and to do so remotely. “We would like vendors and standard-setting organizations to become aware of the features a grid-worthy technology will have,” she said. “We're trying to help facilitate a conversation between technology developers and grid operators, to ensure they are all on the same page. It's hard to do that without hard numbers about how devices must perform, and [the upgradeability PAP] provides these numbers.”

The Home Area Network (HAN) is a concept that is acting as a de facto standard for letting home appliances work together toward saving energy. Several organizations have adopted it for interconnecting line-powered home appliances with the smart grid using the Internet Protocol (IP). HAN was developed by the ZigBee Alliance with inputs from organizations that include NIST, the HomePlugPowerline Alliance, and the Wi-Fi Alliance. The main application envisioned for HAN-enabled devices is implementation of time-of-use pricing methods.

It looks as if this concept is ready to get some traction. General Electric's new GeoSpring hybrid water heater will be ZigBee Smart Energy certified. (See Here comes the hybrid water heater, EE&T Nov.-Dec., 2009)

Another important step in the communications arena has been made by On-Ramp Wireless Inc. and GridSense Inc.. These two have developed what they call the first affordable utility distribution point monitoring system using On-Ramp's Ultra-Link Processing (ULP) technique and GridSense's TransformiQTM platform. Their communication scheme uses spread spectrum methods to transmit half-duplex signals between grid components.

On-Ramp says this method gives 40 dB of additional receiver sensitivity compared with other free spectrum radios, and thus is advantageous in areas characterized by a lot of radio noise. The system uses what's called Random Phase Multiple Access (RPMA) - an access method similar to CDMA but with a different way of identifying devices on the network. On-Ramp says an RPMA receiver decodes messages more efficiently than CDMA systems thanks to clever algorithms. The ULP over-the-air protocol is also simple, optimized for transmission of many small data packets from dispersed devices to minimize overhead and boost throughput.

Page 3 of 3

Schemes like ULP envision traditional grid components such as transformers becoming not just a means of distributing electricity, but also hubs for the collection of smart grid data. Yet few of the millions of transformers on the electric grid have any intelligence or communications capabilities. That is changing as smart grid initiatives pick up steam.

At the North Carolina State University, Alex Huang, professor of electrical engineering, heads up an effort to reinvent utility-scale transformers. In the Future Renewable Electric Energy Delivery and Management Systems (FREEDM) program there, smart transformers under development are built to quickly change power, voltage and frequency as they communicate with the rest of the grid. Such facilities become necessary in scenarios where power flows not just out from the utility to substations, but potentially back from distributed sources such as solar arrays and wind farms.

Other work at FREEDM includes a new initiative to develop transformerless intelligent power substations, which will enable the direct interconnection of renewable energy resources and energy storage systems to the smart grid with bidirectional power flow control. Also undergoing development are power semiconductors based on silicon carbide technology that can handle thousands of volts.

In the same vein, the Electric Power Research Institute (EPRI) is working on technologies that should help make smart grid operations more economical. Notable among these efforts is robotic inspection of electric transmission lines.

It turns out that a lot of overhead transmission assets — including towers, conductors, insulators, and other components - can be difficult to get to because they sit in remote, rugged environments, or span wide expanses of water. Frequently, inspection workers conduct helicopter surveys or must climb towers to get a close look.

EPRI is now working on a robotic system originally conceived by British Columbia Transmission Corp. and Hydro-Québec which propels itself along transmission lines. The device is still in the prototype phase. EPRI expects to try the system in 2014 on the Potomac-Appalachian Transmission Highline, a new 765-kV, 275-mile-long circuit.

If everything goes according to plan, electromagnetic interference detectors, high-definition cameras, and other equipment installed on the robots will identify changes in component condition, vegetation management, and right-of-way encroachment, flagging potential problems.