From Research to Action | Utility Data Analytics Project Makes Molehills out of Mountainsby Doug Dorr and Jared Green
Big data, little data, AMI data … all as good as the back end analyses.
Enter the Electric Power Research Institute’s (EPRI’s) Distribution Modernization Demonstration (DMD). The initiative aims to identify what can be done with existing data, gain insights from the data that were previously unknown or perhaps not yet conceptualized, and assist utilities as they become more versatile with big data analytics and associated activities. Through this initiative, EPRI and its collaborators will develop and maintain a data repository where analysts can get the data sets of interest. They will prioritize a wide variety of data-driven use cases, define the necessary data sets, define action plans, and identify valuable lessons learned in order to deliver the best possible value from big and little data.
The EPRI Data Repository and Data Mining Initiative has evolved out of the DMD. It’s a demonstration set forth to assist utilities with conquering three primary challenges associated with the mountains of data coming from intelligent communicating devices on the distribution system and other supporting data sets. The more than twenty utilities that are participating in this initiative are looking to receive actionable insights from disparate data sets; they want to move closer to predictive analytics – being able to meet a customers’ need before it becomes urgent – and they seek to better understand impact en masse – how changes in technology and end use are impacting their system as a whole.
Through the initiative participants will learn what can be done with existing data; they will identify insights from the data that were previously unknown, and they will become more versatile with big data analytics, strategy, and activities. As the concept of the “prosumer” continues to grow, as more renewable energy enters the grid, and as other distributed energy functions such as demand response and behind-the-meter energy storage take shape, EPRI’s Data Repository and Mining Initiative becomes ever more important as the industry drives from a “smart” to an “optimally integrated” electric power system.
Building the repository
The amount of time, effort, and financial investment required to evaluate data and derive meaningful analytics is considerable. Therefore it makes sense to coordinate a single place where data sets and data analytics cases can be vetted by EPRI and its research partners for the most benefit. EPRI is creating that portal and seeding the fertile soil of the data farm with rich sets of real power system data that can be used by researchers in the project.
The data repository is set up to contain an unprecedented diversity and quantity of time series data from member utilities. The data sets include data from advanced metering systems, supervisory control and data acquisition (SCADA) systems, geospatial information systems, outage management systems, distribution management systems, asset management systems, work management systems, customer information systems, and intelligent electronic device databases. Other supportive data sets, such as weather, social media, forestry, and imagery information will be included to bolster the primary data sets. Although the bulk of the data is expected to be loaded into the repository in the first quarter of 2016, a steady stream of data will continue to flow into the repository through 2016- 17 as the electric service providers complete the data collection phase OF their demonstration project(s).
Analyzing the data
Over the next two years, the initiative will provide a test bed for data exploration and innovation and seek to solve the top challenges, identified as data analytics cases in the initiative, faced by the utility industry.
The data analytics cases, also referred to as use cases, will help utilities answer some of the most common questions they face. EPRI is categorizing the data analytics cases by the type of issue addressed and ultimately the benefit provided by the aggregation of additional datasets: how can outage awareness be enhanced; what additional knowledge can be ascertained on the health of assets; how can situational awareness be improved; how can better awareness of loads augment present operational practices; and what impacts and benefits can be leveraged from emerging data analytics technologies and practices?
Once identified and developed, the data analytics cases are ranked based on a scale of low, medium, and high priority as designated by EPRI and the electric service providers. Each case has a supporting document that explains the challenge to be addressed; existing method(s) being used to tackle the problem; potential methods to improve upon current solutions; desired outputs of an upgraded or a new, innovative solution; and the likely costs and benefits of implementing the resulting algorithm or application. As new ideas are thought of and additional challenges identified, EPRI will work with the electric service providers and research partners to document each in a data analytics case report.
Research partners also will be invited to demonstrate their analytics capabilities, leveraging the data repository to attain data for their test cases. Each of them will be provided an opportunity to share their findings in order to demonstrate fully the present state of capabilities that exist in the market.
Driving no hyphen
Demonstrations of the creative data analytics tools will start in 2016 and continue through 2018 as EPRI and the research partners report on their successes in addressing the challenges documented in the data analytics cases. During this time project participants will have ample opportunity to explore, recommend, and identify additional data analytics cases for evaluation.
As more datasets are added and more participants join in the effort, this initiative will provide an opportunity for them to virtually move in and out of disparate data sets brought together in the repository and ascertain the 30,000-foot view of what’s going on in the power system today, such as how distributed energy resources impact reliability, delivery, and end use. Additionally, the data sets will support innovative solutions that have a specific objective, such as fault waveform identification, electric vehicle charging profile detection, and many other answers to common utility challenges.
Finally, this collaborative initiative will foster a better understanding of industry needs, capture leading data analytics practices, allow knowledge transfer from industry experts, and accelerate ideas (solutions) to the market.
To learn more about the initiative and becoming a research partner, or to find the latest on the Data Repository and Data Mining Initiative, visit the EPRI Data Mining Initiative at http://smartgrid.epri.com/DMD-DMI.aspx.
About the Authors
Douglas S. Dorr is a program manager at the Electric Power Research Institute (EPRI). His major area of focus is data acquisition and analytics for electric power systems. In this position he manages EPRI’s Distribution Modernization Demonstration research initiative. Mr. Dorr has 25 years of industry experience related to analytics for distribution power systems and energy utilization areas. Mr. Dorr is an IEEE Senior Member and a member of CIGRE. He holds two U.S. patents for power system sensor technologies and has published many industry documents on power system analytics, power quality, distributed resources, grounding, stray voltage, and lightning protection. He received a BSE degree in 1989 from Indiana Institute of Technology in Ft. Wayne, Indiana.
Jared Green is a technical leader in EPRI’s Information, Communication and Cyber Security team. In this role he addresses prevalent challenges associated with integrating Distributed Energy Resources (DER) and fully applying them in industry demonstrations for system-wide interoperability and integration. Green’s educational background includes a Bachelor of Science degree in electrical engineering from the University of Alabama. He is a registered professional engineer, a certified energy manager and a certified carbon & ghg reduction manager.