MDM applications support the loading, validation, editing, and estimation of meter data. Due to the very high volumes of data logged by smart meters (as often as 15 minute intervals), new MDM applications must achieve high levels of scalability. In the Austin Energy roll-out of 500,000 meters with 15 minute sampling, the annual storage requirement went up 10X to 200TB. This is roughly 400MB per meter per year. Pacific Gas & Electric is sampling twice per day and is adding storage to accommodate 170MB per meter per year.

Smart metering can aid utilities in optimizing revenue by providing alerts that detect idle usage and energy theft. Theft costs utilities about 1-3 percent of revenue or about $6 billion across the industry. Analytical functions can extract information from interval data and translate it into reports for trigger appropriate proactive actions. MDM may include functions such as:

• Connections to meter systems
• Meter data validation, estimation, and editing (VEE)
• Error handling
• Meter data access
• Meter inventory management
• Usage analytics
• Information delivery to customer portals

Sources:
meter data volumes: Smart Grid News
smart meter estimate: FERC Aug-09 report on Demand Response

At the bottom of the stack are the hardware and OS. Google operates hundreds of thousands of machines in 30-40 data centers. Each machine uses a standard AMD/Intel x86 chipset running a customized version of Linux. Each server has its own 12-volt battery to supply power in case the main power supply goes down.

Google does not say how many servers they operate but observers have estimated that the power required for a half-million servers ranges upwards of 20 megawatts and would cost in the neighborhood of $2 million per month in electricity charges. Google has an obsessive focus on energy efficiency. Fortunately, they have recently started to share some of their knowledge with the rest of the world.

If you wonder what this has to do with smart grid, it's simple. The smart grid needs to be architected for massive scalability with millions of devices including meters that transmit readings as often as every 15 minutes. This particular note will appear in the book as a sidebar relating to our discussion of enterprise application architectures.

Sources: High Scalability, Wikipedia, and CNET
http://highscalability.com/google-architecture
http://en.wikipedia.org/wiki/Google_platform
http://news.cnet.com/8301-1001_3-10209580-92.html

• Generation portfolio management
• Multiple supply/demand scenarios
• Integration with demand response
• Integrated economic analysis with market pricing
• Decision support tools

Advanced capabilities such as these will be essential for managing the increased complexity of the smart grid's ability to integrate new ways for producing power and managing demand. It gives grid operators the forward-looking view required for making better decisions.

Sources: Areva ASPCON-09 paper by Cheung et al. 2009

Nova Scotia Power and its technology partner OpenHydro recently unveiled a 1-MW tidal turbine. It will be deployed in the Bay of Fundy this fall as part of Nova Scotia’s tidal power test facility. The Open-Centre Turbine was manufactured in Ireland by OpenHydro. The turbine will rest directly on the ocean floor using a subsea gravity base fabricated in Dartmouth by Cherubini Metal Works.

The 33-foot turbine will be deployed in the Minas Passage of the Bay of Fundy. Testing will last up to two years. Operational data will be collected and shared by Nova Scotia Power and OpenHydro to determine the environmental performance and future feasibility of tidal power in the Bay of Fundy. The testing will focus on the robustness of the turbine in the harsh environment of the Bay of Fundy, close monitoring of the environmental impacts of the turbine, and its energy production capabilities.

The Bay of Fundy sits on the northeast Atlantic coast of North America between New Brunswick, Nova Scotia, and Maine. It is significant for having one of the highest vertical tidal ranges in the world. This project is a big step forward from previous projects and proposals which mainly involve building a dam to hold back part of the bay and extracting power from water flowing through the sluice gates (similar to conventional hydro). Dams can severely disrupt the marine ecosystem, trapping fish and marine mammals. The OpenHydo project, by contrast will sit directly on the seabed floor. If the project reduces interference with the marine ecosystem, and it holds up in the icy currents of Fundy, it could move tidal power technology from the pilot stage into broader production. 

Sources: OpenHydro and Wikipedia.

DLR is best understood in comparison with static line ratings in which the capacity of a transmission line is determined based on worst-case assumptions of full sun, high air temperature, and no wind. Those factors reduce the capacity of a line because hotter lines are more likely to overheat when carry electricity at their rated capacity. If operators knew that clouds were blocking the sun, the weather was cool, and the wind was blowing, they could use the full capacity of the transmission line. It turns out that line tension is a robust predictor of the environmental factors affecting its capacity. The Valley Group produces the tension and environmental sensors along with the software to generate realtime capacity information for grid operators. The system uses data on average line temperature, sag, clearance, and a realtime rating-factor. The system interfaces with the operator's EMS/SCADA and is customized for the operator’s requirements.

Dynamic line ratings improve system reliability, as they allow operators to make fewer corrections in system dispatch, while providing advance warnings of impending thermal/capacity problems. They save money by fully utilizing transmission assets. Aivaliotis says that operators can transmit up to 30% more power over 90% of the time. This extra capacity can be used to manage delays in line construction and management of the network during major system disruptions. DLRs are an enabling technology for more effective use of renewable energy. For example, wind farm production is often limited by transmission constraints. Production is highest when the wind is blowing. That same wind increases the capacity of the transmission system. DLR is the key to unlocking its full capacity.

The plant will be almost At 950MWs, nearly the capacity of a nuclear power plant. It will use a high-efficiency combined-cycle technology and sit on the site of the retiring coal-fired plant. The new gas-fueled plant is expected to cost about $900 million and take two years to build. The plan also will involve the construction of a natural gas pipeline to the site. Progress Energy expects to announce a contract for the gas supply in the near future.

Progress Energy's plan contrasts with moving to next-generation nuclear plants for baseload power. The next generation of nuclear plants is expected to cost at least $6 billion each, take up to ten years to build, and provide about 1300MWs of capacity.

If analysts are correct that North America has at least 100 years of non-conventional natural gas reserves, this strategy makes a lot of sense. Natural gas plants emit about half the carbon-dioxide as coal plants. High efficiency combined-cycle designs such as this one emit about one-third the carbon of the older coal plants they will replace. We may need next-gen nuclear plants as a bridge to the future when renewables can provide baseload power, but this model suggests that natural gas can also provide part of that bridge. Diversity in the U.S. fuel mix makes sense as a hedge against future uncertainties.

Source: EnergyCentral, Progress Energy

To support the standards development process, the GridWise Architecture Council (GWAC) created an eight-layer model. The model illustrates the many levels of standards and technologies required to support end-to-end interoperability.

As you can see in the figure, end-to-end interoperability involves everything from the electrical connections at the bottom of the stack to the regulatory environment at the top. Levels 1-2 involve standard connections across the myriad of grid components and communication networks. Levels 3-4 insure that the meaning of data and information remains intact as it flows across different elements. Levels 5-6-7 deal with the business and organizational issues of interconnected systems. Level 8 points to the need for a regulatory environment that does not impede the interoperability of systems.

The GWAC Stack is a means to identify well-known interfaces. Unambiguous interfaces are essential to interoperability. This s demonstrated on the Internet which has a physical and data link layer that allows messages to cross any type of network (Ethernet, Wi-Fi, microwave, optical) and still be managed end-to-end by a common network layer. On the Internet, protocols such as TCP/IP handle the transport details making sure that data packets get to their destination correctly, while higher level protocols such as HTTP and XML structure the message so that it can be interpreted properly at each end of the network.

Source: GWAC Interoperability Context, March 2008
http://www.gridwiseac.org/pdfs/interopframework_v1_1.pdf 

Part Two of the book will get into smart grid architectural details. If you have deep technical expertise and are interested in reading a draft of part two, let me know. It should be ready by mid-October.

With any luck, I'll start filling the pages of this blog again next week. Thanks for hanging in there. --JCB

• switching power through fully automated substations
• re-routing power around bottlenecked lines
• detecting power outages
• proactively identifying outage risks
• automating three of four distribution substations
• automating four computer monitored power feeders
• monitoring 23 feeders for voltage irregularities

A grid monitoring system, installed by CURRENT Group has helped avert four power outages by alerting Xcel operators to transformers that were ready to fail. Not a bad start to Xcel's "grid optimization" efforts.

I live in Boulder and have applied to be a beta tester for Xcel's next phase of development which will include demand response programs, in-home displays, and other energy-saving devices. We'll see what happens. --JCB

My work activity was focused on finishing Part One of my book. I will have the first five chapters ready for reviewers during the coming week. If the editing goes smoothly, I will start posting to the blog again. My topical coverage will shift toward technical architecture, covering both networks and applications. The week of Sep 21, I am attending Grid Week in Washington D.C. My interviews there will be focused on Smart Grid architecture, the subject of the next two sections of the book. My blog posts that week will be opportunistic covering both product announcements and research findings.

Let me know if you are attending Grid Week and would like to schedule a meeting. --JCB

The Texas Interconnected System is not interconnected with the other two networks (except by certain direct current lines). The other two networks have limited interconnections to each other. Both the Western and the Texas Interconnect are linked with different parts of Mexico. The Eastern and Western Interconnects are completely integrated with most of Canada or have links to the Quebec Province power grid. Virtually all U.S. utilities are interconnected with at least one other utility by these three major grids. The exceptions are in Alaska and Hawaii. The bulk power system makes it possible for utilities to engage in wholesale (sales for resale) electric power trade. Wholesale trade has historically played an important role, allowing utilities to reduce power costs, increase power supply options, and improve reliability. Historically, most wholesale trade was between interconnected utilities within the continental United States. With open access and deregulation of wholesale markets cross-border trade has become more prominent in meeting domestic electricity requirements. U.S. international trade is mostly imports. Normally, most imports are from Canada, with a small portion coming from Mexico.

Source: Energy Information Administration (EIA)

While most storage systems remain very expensive, pumped hydro and compressed air are cost-effective today. Pumped hydro is well established and makes hydroelectric power one of the most dispatchable of all fuel sources. With a storage option, hydroelectric production can be turned up or down in minutes (see post). Compressed air energy storage (CAES) is in an earlier stage of development but is starting to see commercial pilots and deployments. Both have limitations in terms of location. Hydro needs a vertical drop and plenty of land for the storage reservoir. Compressed air needs favorable geological strata deep underground (see post).

The smart grid will provide the communication and software solutions to make large-scale storage systems work. Their charge and discharge cycles need to be tightly managed to match charging with periods of excess power and to match discharging with optimal periods of demand. In a dynamic pricing environment, this is no easy task.

Utility-scale storage is a natural match for utility-scale renewable generation.

In the case of wind power, excess power generated when the wind is blowing (often at night, off-peak) can be used to charge any of these systems (CAES is illustrated above). An effective storage system can make wind dispatchable, greatly increasing its value to utilities. The combination of dynamic pricing and intermittent generation requires sophisticated modeling tools. To meet future needs, advanced wind forecast models are under development at the National Center for Atmospheric Research (NCAR).

Utility-scale storage systems promise to increase efficiency by better matching supply with demand. They will increase reliability by smoothing out power fluctuations. And they will increase stability by providing ride-through during short power disruptions.

Pumped Hydroelectric Storage
Sources: Wikipedia, Tennessee Valley Authority (TVA)
http://www.tva.gov/power/pumpstorart.htm