Data analysis: what’s it telling us?
Using the data to provide new efficiencies
Published In: Intelligent Utility September / October 2011
THANKS TO THE INFORMATION SMART METERS and other smart grid technology is bring in, utilities are beginning to face an explosion of new data.
Now the question becomes: what can this data tell us, and how can we use it to better our services, understand our customer better, and make our utility operations more efficient. Aligning business intelligence, or BI, with an analytics strategy, provides a more complete picture. The electric utility industry has begun turning its eye to data analytics in order to best mine the new information that is now coming in in droves.
Looking at data differently
Strategies differ, depending upon the data and the need. For example, OGE Energy Corp.'s "Information Factory" is adopting data analytics tools to determine how best to use the new data it has available. "If you can capture and analyze that data, and use it in a predictive fashion, what can we do with it to improve our operations and enhance our customer experience?" Craig Johnston, the utility's vice president, corporate strategy and marketing, told Intelligent Utility earlier this year.
OGE looked to its internal teams with an initial focus: "What information do you want, and how would you use that data?"
Beyond that initial focus, it is also looking at how to start applying statistics on top of data, and use that to improve its forecasting. And then, Johnston said, "we can take that and do some `what ifs,' or predictive analysis."
Trend analysis and asset management
Detailed data analysis can also provide the structure for looking at trends, as well as at asset management based upon the history of the asset. More broadly, it can become a key strategic differentiator for the utility with the ability to apply quantifiable metrics (supplied by better data analysis) to a business case.
As another example, energy efficiency and demand response programs can also benefit from a more detailed analysis of the new data being provided by smart meters. The ability to leverage detailed energy consumption data to segment and target customers, and to provide them with valuable electricity usage feedback, is the basis of true energy efficiency, on a granular, customer-by-customer level.
San Diego Gas & Electric (SDG&E) is now leveraging the new data brought in from its Smart Meter Program to put more energy usage information into the hands of their customers, in a plan both to educate its customers about the role of the new technology and to better target energy efficiency, demand response, sustainability, and energy reliability.
"We are looking at several initiatives to help maximize the value of (the new) data," said Brendan Blockowicz, SDG&E's smart meter IT program manager. "As every utility before has confirmed, once you get in there and start to work with the data, you learn a lot more about how to use it. We are looking to centralize some of this capability."
For customers, the utility plans in the near term to provide online tool capability enhancement such as calculations for bill-to-date, bill forecast and alerts. As well, SDG&E will be expanding its home area network (HAN) pilots to a customer-wide base and integrating that data.
"Data analysis and business intelligence are foundational to putting more energy usage in the hands of our customers. Good data in a timely fashion allows both customers and operations to make good decisions that save energy, time and money, prevent losses and increase service levels. It's what the business case benefits rely on to deliver the long-term results," Blockowicz said.
Predicting asset health
SDG&E also uses real-time data to power its condition-based maintenance solution. The ultimate objective, of course, is to avoid potential catastropic asset failures by being able to better predict any asset's current health. Real-time data availability provides the potential for learning more about the asset, the ability to provide more timely identification of potential asset issues, and greater O&M cost savings.
Of course, all of this requires taking the new real-time data available, deciding which data is most valuable for the task at hand, and then "layering" or "factoring" that data for further analysis. In effect, it's deciding what to add and what to subtract for the most effective final solution.
"The objective of SDG&E's condition-based maintenance (CBM) project is to extend the useful life of and make greater utilization of transmission and distribution substation transformers. We use technology to measure the performance and condition of equipment to make better maintenance decisions," Blockowicz explained.
The streaming data is monitored in real-time for any irregularities by the CBM system. Data irregularities trigger alerts, ranging from levels one to four: "one" is business as usual, and "four" is the most critical.
"In creating the system subject matter experts from our construction and maintenance, asset management, engineering and operations departments worked together to define the elements that needed to be monitored, the alert levels and who should be notified in each instance," he said. "These alerts are disbursed via email and/or text, notifying the appropriate parties to review the data and make informed decisions. Engineers are able to analyze the monitor data in conjunction with data from our SCADA, Work Management and TOA4 (Dissolved Gas Analysis) systems.
"Earlier this year, the CBM system sent an alert notifying us that the gas levels inside one of our transformers had begun rapidly increasing. Based on some follow-up data analysis, it was decided to schedule an outage to perform an inspection of the transformer. Upon inspection, evidence of heating was found and some parts replaced. The alert from the CBM system and the timely follow-up helped avoid further damage to the transformer," Blockowicz explained.
Managing manpower requirements effectively
With real-time data and analytics, the utility is able to quickly spot situations that require immediate attention; perform asset maintenance in a better-planned fashion, when and where it is needed; operate assets closer to their operating limit; and enhance the ability to determine the optimum time to replace an asset.
In essence, this means cost savings and workforce efficiences across the board, as manpower requirements can also be managed more effectively, and just-in-time delivery of new assets can better be managed, rather than warehousing "spare parts" within the utility on a more ad hoc basis, as has happened in the past.
"This is not just another technology project - it's an overall change of how we do business internally, and is the foundation of a new relationship with our customers," Blockowicz said. "The technology changes are large and impact many other departments at the company - therefore we are integrating vertically like never before.
"Daily operations are more complex than ever before. We are creating new organizational structures to best address the changes we are implementing. We are identifying new skills needed by our workforce for the future to support these new operations."
Real-time and future predictions
In many ways, data analytics is all about prediction, about clearing the fog within each utility's crystal ball. Or, to adopt another analogy, it's about using the whole brain, with all of its intricate interconnections, rather than a sum of the information coming from its separate neurons.
Analytics-based utility applications can include superior customer service (because customer usage information can be tooled to provide better customer-focused solutions), operational efficiency (including real-time maintenance-based solutions), optimized delivery of power, smart energy procurement (based on known and anticipated energy needs on a day-ahead basis), demand response and dynamic pricing, and using pattern recognition to better detect energy theft, to name but a few.
And that's just structured analysis. Imagine, taking it one step further, the amount of new insight more unstructured analytics can provide. Or, more simply put, what happens when a lot of unstructured data is pulled into a database and exposed to new queries? New connections of previously unconnected data can be made, and utilities will, over time, be able to even more clearly understand its system as a whole... rather than a sum of its parts.
Asking new questions
It's also about asking new questions, as Boreas Group co-founder Robert Sarfi pointed out, and the ability to use all the new data to define an integrated resource plan, for example. "If I have demand response and renewables, how can I better leverage those to gain operational efficiencies at a distribution level?" he asked. It's all about the data.
And it's about innovation, as well. As Southern Company CIO Becky Blalock noted (see "A legacy left behind," page 40): "I think that there are going to be incredibly innovative ideas that come out about how you mine this data so that you can make the best possible decisions for how you run your business."
Data analytics is going to play a powerful role in how electric utilities move their businesses forward. In future issues of Intelligent Utility, Energy Central's new Utility Analytics Institute will explore the boundaries of the often complex world of utility data analytics here in a regular new column.