Baltimore Gas and Electric Company (BGE), a subsidiary of Exelon Corporation, has successfully deployed C3 Revenue Protection™ and C3 AMI Operations™ across the utility’s two million meter smart grid network.
Efficient and effective service delivery
The aim of the technology is to help the utility reach its goals of ensuring cost-effective, efficient, and reliable energy delivery.
The two C3 Energy Smart Grid Analytics™ solutions analyze over 35 billion rows of data from twelve Baltimore Gas and Electric Company and third-party data sources to provide critical, actionable insight to the utility’s revenue protection and grid operational teams. This will help teams operate more efficiently and effectively.
The data sources include MDMS, head end, CIS, work management, EAM, OMS, GIS, CMS, and meter installation vendor systems, among others.
Christopher Burton, vice president for smart grid and technology, Baltimore Gas and Electric Company, says that the company is already experiencing efficiencies. Customers are benefitting from improved service reliability at affordable rates.
The utility serves over 1.2 million electric customers and over 650,000 gas customers in a 2,300-square-mile area. C3 Energy and Accenture worked closely with Baltimore Gas and Electric Company’s IT and business teams through the process of integrating, implementing, user testing, and launching the solutions.
C3 Revenue Protection identifies load profile anomalies and outliers, while C3 AMI Operations validates the operational functionality and integrity of the automated metering infrastructure (AMI) network, helping the operational team ensure high levels of service reliability for the utility’s customers. Together, the two solutions reduce instances of unbilled energy.
The initial results from the two solutions put BGE on track to achieve its goals for reductions in unbilled energy, which is expected to deliver tens of millions of dollars per year in economic value to the utility and its customers.
"The BGE implementation of C3 Energy Smart Grid Analytics is one of the most sophisticated examples of machine learning in power delivery today. The system analyzes data from hundreds of thousands of devices, and then generates hundreds of analytic features per device, all in real time," said J. Zico Kolter, Assistant Professor of Computer Science, and Carnegie Mellon University. “The machine-learning algorithms then use this data to evolve automatically over time, ensuring the effective resolution of grid operational issues.”
Data analytics-not a quick fix
Baltimore Gas & Electric views data analytics as an ongoing process, not a once-off quick fix.
The company is constantly looking to see how advanced analytics can resolve very specific business challenges.
Here are a few:
1. Revenue protection. Picking up on unauthorized consumption or meter reconnections by assessing consumption factors.
2. Focusing on demand response. Identifying the right customers to target for Baltimore Gas & Electric’s peak rewards program, as well as locating demand response devices that are not operational. The utility is finding defective or disconnected devices with 60% accuracy.
3. Voltage analysis. Identifying transformer issues. In a couple of cases, transformers were found not to be supplying the right voltage. It can also be a network capacitor at fault or the meter was incorrectly reporting voltage.
4. Net meter and defective meter identification. Identifying likely net metering customers to ensure they are registered and receiving credit. And evaluating meter events to identify meters that might be defective. Since the utility has had some issues with overheating meters, it was eager to spot high-temperature meters before they caused an emergency
Tom Humphreys, Baltimore Gas and Electric Company 's Project Manager, Advanced Metering Capabilities, explains that it's important to have an executive sponsor, someone at the top with the vision who is engaged and driving execution.” He explains that often, other departments and business units have their own projects and deadlines. “It can be a challenge to get them to pause their normal activities to help identify issues and build value cases.”