6 Steps to Becoming a Utilities IoT Ninja

By Alyssa Farrell, SAS

Can you remember a day in the past few months when you haven’t seen a headline about how the Internet of Things (IoT) is going to change utilities’ world?

IoT is one of the hottest topics for utilities today. Defining IoT, becoming its master, and effectively putting it to work to recognize business value represent an enormous advantage for utilities attempting to progress to full smart grid status. In simplest terms, IoT involves objects containing sensors and other connected devices. Collectively, the connected devices have the power to create an increasingly autonomous grid that can handle billions of endpoints on utility networks at speeds fast enough to protect vulnerable areas of the grid while integrating more renewables and simultaneously bringing utilities closer to digitally-savvy customers.

IoT provides the channels that can deliver more granularity and timely insight into electric grid operations. In the past, detailed diagnostic data about asset performance was captured but not transmitted unless there was an outage. Analysts then would pore over those artifacts looking for clues about what had gone wrong. With connected devices, data can be streamed back at appropriate intervals to increase visibility into the performance of assets and decrease the time between event detection and correction.

Survey Says Data-driven Decision Making is Top Priority.

If utilities can use real-time data through IoT to pinpoint areas of operational improvement, they could save on operations and maintenance costs. These savings could be allocated to capital spending in new areas such as renewables and distributed energy resources. Energy sources such as wind and solar power tend to be intermittent in their production, therefore, utilities must carefully manage how they bring that capacity onto the grid to maintain reliability and avoid unplanned outages.

A recent study asked utilities leaders to prioritize how they planned to use IoT (and the related technology machine learning). Even though the list of choices in the survey was common across both categories, the only common response that made it into the top five for both IoT and machine learning was “improved data-driven decision making.” The research revealed that executives perceive the top benefits associated with IoT to be customer-facing, such as customer service and energy efficiency. The benefits named for machine learning, which will be covered in part two of this series, are more grid-oriented, including areas such as service restoration and cybersecurity.

Meaningful Consumer Engagement is Easier with Iot

Utilities don’t yet segment consumers with the granularity achieved by some retailers, but IoT data can help provide better comparisons across energy consumers with similar usage behavior to improve energy efficiency campaigns. This helps utilities optimize marketing expenditures and improve the data provided to consumers about peer group consumption and potential home maintenance problems, such as inefficiencies in insulation, windows or HVAC.

The IoT provides utilities the opportunity to craft personalized energy services for consumers keen on managing their energy profile. That might be one of the most exciting disrupters associated with IoT, whether a consumer actively manages his or her energy profile for financial or environmental reasons. Consumers might choose to participate in time-of-use rate programs and receive smartphone alerts. More advanced technology might connect a consumer’s phone and thermostat to sense when to pre-heat or pre-cool a dwelling. Commercial customers who are vigilant about energy consumption might use IoT-enabled notifications to switch to on-premise generation during peak pricing.

How do You Become an Iot Ninja?

When thinking about ways to extract value from the IoT, the possibilities boggle the mind. The vast pool of use cases makes it important to start small and seek incremental successes. In doing so, a utility can try one use case, glean lessons learned, fail or succeed, adjust accordingly, then quickly move to the next use case. In no time, being an IoT ninja will become second nature.

Initial steps for getting started include:

1. Develop a strategy. Technology investments implemented without a strategic plan have a shorter shelf life. Edge analytics-analytical capabilities positioned within or near a connected device rather than only in an enterprise network-might be all the rage, but take the time up front to map edge analytic needs to what can be supported by new IoT gateways and existing computing nodes throughout the network.

2. Pick a project. Whether you’re increasing the model segmentation or pricing analysis, start with an area that is already data-rich. Implement on a smaller scale, but consider larger problems once you’ve had some success.

3. Don’t overlook change management. Engineers and customers both are notoriously difficult to influence. Put a plan in place to increase adoption of IoT data in decision-making. Much of utilities’ world is predicated on a centralized model with a human in the loop, but IoT and machine learning transforms that model. Utilities should prepare to make changes to their core business model and prepare employees for these changes.

4. Learn from other industries. Benchmarking is critical to keep the industry moving forward. Sometimes adopting new language can change mindsets. Intel, for example, uses the terms “tweeting” and “streaming” to describe the way it plans to use connected devices.

5. Plug into other utilities. Utilities continue to invest in research and have test beds of technology embedded in smart grid demonstration projects. Connecting with other utilities opens the door to best practices, what’s worked and what has not, vendor feedback, and future projects.

6. Invest wisely. Robust analytics make sense of all that IoT data. But the infrastructure around the analytics are equally important. An effective infrastructure includes a streaming engine coupled with a distributed computing architecture (including analytics at the edge). These help determine when system performance is outside expected boundaries and can pave the way to initial success.

Editor’s note: This article is Part 1 of a two-part series on IoT. Look for Part 2 in POWERGRID International’s September issue. For more information about IoT in the utilities industry, download the whitepaper, The Autonomous Grid: Machine Learning and IoT for Utilities at sas.com/whitepapers.

Alyssa Farrell is the Global Energy-Utilities Industry marketing manager at SAS. Follow her on LinkedIn and @alyssa_farrell on Twitter.

Authors

Previous articleThe Cloud in the Field is Taking Utilities by Storm
Next articleRMI: An Integrative Business Model for Net Zero Energy Districts

No posts to display