4-page Case Study - Posted 1/7/2008

Views: 607

Rate This Evidence:

Exelon Corporation

Exelon Models Distribution Network to Predict and Prevent Outages using SQL Server 2005

Chicago-based Exelon, one of the largest electric utilities in the United States, uses automated meter reading (AMR) technology to monitor electric power usage of its 1.7 million customers in Philadelphia. Looking for additional ways to derive value from the AMR data, the company deployed a pilot project using a meter data analysis program from Microsoft® Utility Partner Itron, and hosted on Microsoft SQL Server® 2005 database software as the relational data store. Working from a subset of a 7 terabyte database to model its distribution network, including loads on unit substations, step-down transformers, local transformers, and power cables, the company found it could use the AMR information to predict equipment failures that could cause power outages. The ability to model loads on its distribution network is expected to help the company reduce outages and gain lower operational costs.

Situation

Exelon Corporation is one of the largest electric utilities in the United States with 2006 revenue of U.S.$15.6 billion, income of $3.5 billion, and one of the industry's largest portfolios of energy generation capacity. The company, which serves more than 5 million customers, was created in 2000 by the merger of PECO Energy Company of Philadelphia and Unicom of Chicago (owner of Commonwealth Edison).

In 2004, Exelon’s PECO Energy Company completed deployment of a fixed network automated meter reading (AMR) system from vendor Cellnet to provide daily automated meter reading services for some 1.7 million electric and 500,000 gas meters in Philadelphia. The AMR system is based on a radio-frequency fixed network. Each meter is equipped with an AMR module that every 5 minutes transmits data to local control units which relay the data to the control center.

Rather than sending human meter readers to homes and businesses, the Cellnet AMR system automatically transmits meter readings to the utility. The system also delivers data that can flag meter tampering, and is designed to receive and process special meter information packets, such as “last-gasp” outage messages when power is lost, and “power-up” restoration messages.

The AMR data is so rich in information, that Exelon sought additional ways to pull value from the system by using data from the electric meters to reduce outages, as part of its “smart grid” practices, an industry term referring to the automated gathering of distribution data that can be used to enhance service and conserve energy. To accomplish this it needed to create a database to help analyze information that might help it predict and prevent points of failure in the delivery system.

Had we been working in real-time mode, as we hope to upon completion of the study, the DAA meter data analysis running on the SQL Server database would have flagged these overloaded transformers for attention. Glen Pritchard, Consulting Engineer, Meter Reading Technologies, Exelon

Solution

Working with Microsoft® Utility Partner Itron, Exelon launched a pilot project to determine whether AMR data could be used to reduce the occurrence of and severity of power outages. Itron created its Distribution Asset Analysis (DAA) meter data analysis program, using Microsoft SQL Server® 2005 database software as the relational data store, running on the Windows Server® 2003 Enterprise Edition operating system. The solution was deployed on a single Intel-based computer.

“We have been pleased with the performance and ease of use of SQL Server 2005,” says Eric Miller, Vice President of Software at Itron. “The native capabilities of SQL Server, as well as its integration into the full Microsoft reporting and workflow tools make it a great choice for high-scale business intelligence applications like DAA.”

An enterprise-grade database is required because the AMR information transmitted every 15 minutes from 1.7 million electric customers generates more than 7 terabytes of data per year. To support richer modeling of the distribution network to support smart grid initiatives, the full DAA deployment will work with three years of data, generating database loads of some 21 terabytes and 750 million rows.

The application integrates with SQL Server 2005 to create complex analytical models that represent the entire power distribution network, with special emphasis on modeling the loads on four distribution elements: 

• Unit substations. Large stand-alone transformers serving between 1,000 and 2,000 customers.
• Step-down transformers. Step-down transformers are large transformers typically found on utility poles or in underground power vaults that reduce primary voltage from a typical 13,200 volts to 2,400 volts.
• Local transformers. Found on utility poles or in underground locations, local transformers provide neighborhood distribution for small collections of users.
• Power cables. The modeling included the cables used throughout the distribution system, from unit substations to business and residential delivery.

For the pilot project, a subset of one year’s worth of AMR readings that already had been collected by the AMR system, was analyzed. Predicted failures from the analysis were then compared against what actually happened.

“By creating a circuit model of the full distribution network, we’re hoping to gain better insight into how the distribution system is operating,” says Glen Pritchard, Consulting Engineer, Meter Reading Technologies, at Exelon. “Basically we’ve designed a family tree showing power distribution. The circuit starts at the substations and then it splits many times until it gets to the customer. If a transformer has five customers on it, the model sums up the load for each of the customers on that transformer. We can easily extend the model to include fuses, and look at the loads that five transformers behind a fuse are creating.”

Benefits

The pilot project demonstrated to Exelon that analyzing AMR data using Itron’s DAA meter data analysis application supported with SQL Server 2005 could provide the utility with a number of smart grid benefits, including the ability to predict and prevent power failures, avoid collateral damage, reduce operational costs, and help model customer needs.

Predict and Prevent Power Failures
The pilot project confirmed that new value could be drawn from the same AMR data already collected for billing, and that AMR data can bring value to an organization’s outage management system (OMS). Exelon found that integration of AMR data into OMS business processes should help the company to:

• Improve planning and engineering
• Increase crew productivity through predictive maintenance
• Reduce outages
• Enhance customer satisfaction

Collecting our AMR data on SQL Server for analysis enables us to gain a degree of system status that we’ve never had before. Glen Pritchard, Consulting Engineer, Meter Reading Technologies, Exelon

Examining the information collected on the SQL Server 2005 database, Exelon identified five transformers that appeared to be overloaded. When the predictions were tested against real-life incidents, two of the transformers had already failed.

“Had we been working in real-time mode, as we hope to be in the future, the DAA meter data analysis running on the SQL Server database would have flagged these overloaded transformers for attention,” Pritchard said. “This would have given us the warning we need to act proactively, avoiding the expense of doing maintenance in emergency mode, and relieving our customers from the disruption of a power failure. We’re also modeling cable sizes and loads to predict and prevent line failures.”

The modeling information will enable Exelon to improve planning and engineering, and to increase crew and dispatch productivity by reducing the disruptions that require crews to work in emergency mode.

The smart grid solution is flexible enough to support a spectrum of modeling options.

“We can use the DAA meter data analysis solution hosted on SQL Server to create rich scenarios, such as: What happens if we have five consecutive days of 95-degree temperatures?” says Pritchard. “We can use historic data and our modeling to create a wide variety of scenarios to gauge the sensitivity of different elements of the distribution system. We’re finding more and more innovative ways to draw value from the AMR information. All of this helps reduce the chance or duration of an outage, which greatly enhances customer satisfaction.”

Avoid Collateral Damage
Adding to the problems of a transformer, cable, or other element of the power distribution delivery system failing is the fact that such occurrences can sometimes cause collateral damage to unrelated utilities, such as telephone and cable infrastructure especially when the equipment is co-located in duct banks and other underground utility spaces.

“When a transformer or cable fails, you can end up with manhole fires that can cause considerable damage to unrelated cables or circuits,” says Pritchard. “Even if the neighboring cables aren’t damaged by flames, there can be collateral damage. Using DAA and our SQL Server database to predict and prevent equipment failures helps eliminate the collateral damage that can occur.”

Using predictive modeling of the power distribution system to prevent local power failures can also help prevent triggering wider outages that can occur when local failures trigger cascading events that can result in widespread blackouts.

Reduce Operational Costs
The DAA meter data analysis should help reduce operational costs by providing a clearer view of what is happening with the power distribution network. “Using SQL Server to analyze our AMR data is giving us the information we need to work smarter, with a much more precise model of what is happening with our power distribution,” says Pritchard. “We anticipate this will lead to lower operational costs as we are able to more precisely direct our maintenance efforts to avoid power failures rather than reacting to failures after the fact.”

Prior to storing the AMR information on a SQL Server database, it was a time consuming job to sit at a keyboard assembling information for specific trouble spots. “Without SQL Server we would have to gather data from the AMR system meter by meter, and then analyze what we found to gain a single result in time,” says Pritchard. “SQL Server and our DAA application automate the process and provide data on a system wide basis to provide a rich model of what is happening. This information helps us work smarter. And working smarter reduces operational costs.”

Model Customer Needs
The same data modeling that helps predict and prevent power failures can help Exelon plan for future growth. “Collecting our AMR data on SQL Server for analysis enables us to gain a degree of system status that we’ve never had before,” says Pritchard. “This gives us information that will help us monitor growth in customers and energy demands so that we can on a very granular basis, ensure that our infrastructure isn’t being overloaded. We have gained analytic information on customer usage patterns to create modeling assumptions that show how much growth can be accommodated, and at what point different parts of the distribution system need to be updated.”

Applying analytics to AMR data provides a spectrum of opportunities for utilities.

“Increasingly, our utility customer will be creating and implementing many new BI applications that leverage the wealth of data provided by advanced metering systems,” says Miller. “The Microsoft platform makes sense both for software providers, and for utility IT personnel developing specialized applications for their business users.”

In summary, Exelon has gained visibility into its operations by importing AMR data into its DAA meter analysis application supported by a SQL Server 2005 database, and creating distribution system models that help it to predict and prevent outages.

Microsoft Server Product Portfolio
For more information about the Microsoft server product portfolio, go to: www.microsoft.com/servers/default.mspx

Microsoft SQL Server 2005
Microsoft SQL Server 2005 is comprehensive, integrated data management and analysis software that enables organizations to reliably manage mission-critical information and confidently run today’s increasingly complex business applications. By providing high availability, security enhancements, and embedded reporting and data analysis tools, SQL Server 2005 helps companies gain greater insight from their business information and achieve faster results for a competitive advantage. And, because it’s part of Windows Server System, SQL Server 2005 is designed to integrate seamlessly with your other server infrastructure investments.

For more information about SQL Server 2005, go to: www.microsoft.com/sqlserver

For More Information
For more information about Microsoft products and services, call the Microsoft Sales Information Center at (800) 426-9400. In Canada, call the Microsoft Canada Information Centre at (877) 568-2495. Customers who are deaf or hard-of-hearing can reach Microsoft text telephone (TTY/TDD) services at (800) 892-5234 in the United States or (905) 568-9641 in Canada. Outside the 50 United States and Canada, please contact your local Microsoft subsidiary. To access information using the World Wide Web, go to: www.microsoft.com

For more information about Itron products and services, call (509) 924-9900 or visit the Web site at: www.itron.com

For more information about Exelon products and services, visit the Web site at: www.exeloncorp.com

This case study is for informational purposes only. MICROSOFT MAKES NO WARRANTIES, EXPRESS OR IMPLIED, IN THIS SUMMARY.

Document published January 2008