A Grid Reliability Paradigm Shift: Progressive Utilities and a New Approach

What Does Progressive Grid Reliability Look Like?

As with progressive organizations in other industries, progressive utilities seek to consistently improve upon key metrics. That focus on improvement, and the willingness to employ new methods to achieve it, also means these utilities are often early adopters of innovative technology solutions that boost reliability and outage preemption.

To be an industry innovator, leading the way in leveraging new solutions and achieving successful outcomes like consistent outage frequency and duration improvements, a utility must also be nimble and adept with change management.

This is the case with utilities that have embraced a paradigm shift in their efforts to gain impactful reliability results as they begin to predict and preempt impending system outages with Sentient Energy’s new, state of the art reliability solution.

The Next Grid Reliability Breakthrough

Sentient Energy’s outage prediction solution combines intelligent line sensor data, advanced analytics, and machine learning to identify precursor anomalies that indicate likely impending permanent faults due to equipment failure or vegetation contact.

Much like grid reliability breakthroughs of the past, outage prediction and preemption has the potential to be a reliability game-changer. Easily installed line sensors also provide immediate system visibility and situational awareness, enabling significant improvement of SAIDI, SAIFI, and MAIFI metrics.

The substantial grid reliability improvements associated with outage prediction make it an intriguing possibility for many utilities. However, it’s important to consider the complexities that go along with predicting impending equipment failure and vegetation contact outages.

Solid business transformation and change management practices, along with access to the right data, allow for these intricacies to be effectively addressed. Regardless of how forward-thinking or ready for transformation your utility is, three key areas should be considered before starting down an outage prediction path. They include:

Common causes of outages on your distribution system
Change management skills and resources within operations, planning, reliability, IT and leadership teams
Leadership support and buy-in

Power Outages That Can Be Successfully Predicted

Not all outages are created equal. Faults from events like a motorist going off the road and hitting a power pole, or an errant squirrel causing a short on the power system, are random and cannot be predicted. Power outages associated with random and isolated weather events like tornadoes, lightning, and ice storms are also very hard to predict.

Two common categories of faults are more predictable, however — those caused by equipment failure, and vegetation encroachment. These permanent fault categories produce incipient faults prior to a full-blown over current event.

How Power Outages Are Preempted – Why a Nimble Process Is Essential

Current and voltage waveform anomaly data is captured by intelligent line sensors. This data is run through advanced analytics to be classified, filtered, and matched against anomalies with signatures indicative of impending equipment failure or vegetation encroachment faults.

More sensitive than conventional protection devices that are deployed on feeders, line sensors are typically installed using a best practice approach of four sets deployed at quarter points along a feeder. When working to preempt an impending outage, responsiveness is key. This strategic placement of line sensors allows for quicker, more precise inspection, location, and preemptive action to address the problem as identified through precursor anomaly alerts.

Preemptively addressing impending faults related to equipment failure and vegetation encroachment involves three main steps:

Outage prediction via line sensing data, advanced analytics, and machine learning.
The feeder segment with the likely impending permanent fault is inspected.
The identified equipment failure or vegetation issue is addressed and the outage is preempted.

The timing of steps two and three above where action must be taken to investigate, locate, and mitigate is critical to successful outage avoidance.

It is for this reason that evaluating your utility’s ability to be nimble and support change management is so important — especially within planning, operations, maintenance, and vegetation management teams.

Responsiveness Is Key

When the daily Advance Outage Event Detection Report indicates numerous precursor anomalies associated with a high probability of impending equipment failure, a series of actions must be taken within one to two days of the alert. In the case of likely equipment failure, a utility field inspector must investigate to locate and identify the problem issue, such as a broken pin insulator or cracked cut-out, and a crew then needs to correct the issue to avoid a likely outage.

The process for handling the precursor anomaly alerts for vegetation encroachment is much the same as handling probable equipment failure faults, except that tree-trimming is required to avoid an outage rather than maintenance. After the indicated feeder segment is inspected and an encroachment issue is confirmed, the vegetation management crew trims the vegetation, preventing the outage.

Whether the precursor anomalies indicate probability of equipment failure or vegetation encroachment, prompt response is a must for the best grid reliability results.

Vegetation Management and Grid Reliability

With the need for a nimble process in mind, it’s important to assess how your utility currently manages tree trimming. Vegetation-related outage predictions may necessitate a higher degree of change management than the handling of equipment failure fault predictions.

Most utilities manage vegetation trimming with a focus on resource efficiency and use a calendar-driven approach. If your utility doesn’t already have data-driven vegetation management capabilities in place, a phased change management approach is an option to consider.

Starting with outage prediction, inspection, and then tracking the number of cases where outages do occur may make the most sense initially. This allows the vegetation management team and utility leadership to gain insight as to the need and value associated with implementing necessary data-driven tree trimming to prevent an outage.

When there is confidence in the accuracy of predictions, it is easier to justify implementing a new spot-trimming approach to realize grid reliability gains and as important if not more so — mitigate wildfire ignition risk.

Information Technology Capabilities to Preempt Outages

Outage prediction and preemption is data intensive and requires advanced analytics, making IT support and capabilities like cloud-hosted software essential for achieving grid reliability breakthroughs.

Although most utility operational technology (OT) systems run on-premises, cloud-hosted software offers critical advantages for predicting outages with intelligent line sensing and advanced analytics. Ease of scalability, reduced IT resource requirements, and simplified integrations with cloud-based analytics engines are some of the most important benefits of cloud hosting.

Support from Leadership in Grid Reliability Initiatives

Estimating return on investment (ROI) for solutions involving predictive analytics is not as straightforward as measuring ROI for a technology that has long been used in the industry, like line sensors or reclosers. This is largely due to the probabilistic nature of the outcomes and required change management which often may have to be rolled out in a phased approach.

To provide a full-value picture and a higher likelihood of leadership buy-in and support, both the realized and unrealized grid reliability benefits of outage prediction must be tracked.

Realized benefits happen when the outage prediction is acted upon promptly enough to avoid the outage. The true or realized value of an averted outage is estimated by calculating the value of customer minutes interrupted (CMI) that would have happened if the outage prediction had been ignored and nothing done.

This calculation utilizes feeder level Customer Average Interruption Duration Index (CAIDI) and estimates the closest upstream protective device that would have operated if the outage had occurred. For example, avoiding an estimated 3-hour outage for 55 people would yield a potential CMI savings of 9,900 minutes. If CMI is valued at $2 per minute saved, a utility could value the preempted outage as $19,800 of realized value.

Unrealized grid reliability benefits happen when the third step of the outage prediction process as described earlier doesn’t happen as needed, when preemptive action isn’t taken before the permanent fault happens. Example scenarios might be an equipment or vegetation problem that isn’t located or identified, action that for some reason wasn’t taken to address the predicted outage, or when the outage occurs before the relevant team could take preemptive action.

In cases like these where an outage does occur, the unrealized grid reliability benefit of outage prediction is a straightforward calculation of number of customers affected and CMI value. This represents the benefit that would have been realized if the prediction process had been followed successfully.

The value of both realized and unrealized benefits are important components of understanding just how impactful predicting outages can be. While realized value represents the current ROI of the reliability program, metrics on unrealized value can be used to drive change management and shift unrealized value into additional realized value over time.

Leveraging the Next Grid Reliability Breakthrough

Reliable delivery of power has long been a top priority for all utilities. But for progressive utilities that have undergone a paradigm shift in their approach to reliable electrification, leaning into innovation to materially move the grid reliability needle is the priority.

Predicting and preempting outages with line sensing and predictive analytics is a field-proven and cost-effective path to achieving impactful ROI for utilities ready to lead the way with the industry’s next grid reliability breakthrough. The key to this “grid reliability match” is determining whether your utility is well-positioned for business transformation and change management support.

If preempting distribution system outages is part of your utility’s future grid modernization plans, Sentient Energy would love to help you achieve your reliability goals with a unique predictive analytics solution.