By Don Leick
The U.S. has weathered many large storms since the turn of the century. Whether hurricanes, ice and snow storms, storm surges or strong winds, Americans seems to have seen it all and then some. As weather volatility has increased, utility outages, by consequence, have too. Major outages have increased six-fold in the past 20 years, according to IDC Energy Insights’ “Facing Down Weather Report, 2013.”
Outages caused by severe weather such as thunderstorms, hurricanes and blizzards account for 58 percent of outages observed since 2002 and 87 percent of outages affecting 50,000 or more customers, according to data collected by the U.S. Department of Energy on Form OE-417. Statistics released in 2013 by the National Climate Data Center, reveal that in the last decade, costs associated with extreme weather losses have totaled $476 billion. Part of that includes large amounts of unplanned dollars that many utilities spent on recovery and restoration efforts.
In addition to volatile weather conditions, utilities also must contend with aging infrastructure, increased regulatory demands and more technically astute and social media driven customers.
It is vital, therefore, that utilities move beyond simply looking at a weather report to predict impact. They must turn weather forecasts into data that can be used to predict asset damage and improve operational efficiencies.
Severe weather puts even greater challenges on utilities’ already aging infrastructures, most of which aren’t prepared for nature’s extreme and volatile elements.
A 2013 American Society of Civil Engineers (ASCE) report showed that overall electric grid investments are falling short of their 2020 goals. The industry is lagging by several billion dollars in both transmission and distribution investments. In direct response to the outdated infrastructure, utilities are beginning to work with leaner organizational structures, often contracting or sharing resources with neighboring utilities using the mutual assistance framework to manage significant weather events.
ASCE predicts that by 2020, the distribution investment gap will grow to $57 billion and the transmission gap will widen to $37 billion. Now is the time for solutions.
Storm Impact Analytics Models Combine Multiple Data Types
Customers are Getting Smarter and Harder to Please
As the world becomes increasingly connected, utilities face heat from customers and regulators. Social media allows people to be constantly in the loop, making the public more impatient and less tolerant, especially when it comes to major power outages. In addition, increasing regulatory and political scrutiny leads to fines and denied costs when recovery rates increase.
As social media awareness grows, so does the larger media attention from daily broadcasts and newspapers. Social media amplifies the concerned voices of customers, tracking the events caused by severe weather in real-time and closer detail than ever before. The attention utilities receive during volatile weather makes excellent customer communication strategies crucial, which means utilities must be armed with as much real-time data as possible. When it comes to weather, sometimes it’s not so much about what a utility is doing as it is about what they plan to do. Intuitive information, no matter how small, can be the difference between profit and loss.
The regulatory side is not that much different from the media. Utilities are heavily scrutinized and must justify their operational decisions during significant weather events. This makes data critical to optimal weather-influenced decisions.
How to Respond?
Like with all crises, response is critical to successfully managing severe weather catastrophes. Utilities must be equipped and prepared before an event occurs and they must respond quickly during and after a weather event. Technology that supports utilities before and during a storm has traditionally allowed them to:
• Track all relevant forecast metrics
• Provide meteorological consultations
• Provide alerts to both approaching and real-time conditions
• Track severe weather through customized risk indices
• Provide situational awareness throughout the duration of the storm
Even with these abilities, prepping for a response before and during a storm needs improvement.
Utilities can do Better
Weather intelligence that prepares a utility before and during the storm is crucial, but any intuitive information that allows utility operations teams to estimate an incoming storm’s initial impact can save time and resources during storm response efforts. In the wake of bad weather, time is incredibly valuable and limited. Some of the key questions that need to be quickly answered are:
• How bad is asset damage and where?
• What resources and materials are needed for repairs?
• What requests for help or commitments for helping others should be made through mutual assistance?
• How did the forecast change during the event?
• How can crew adjustments ensure a specific estimated time of restoration goal for my region or service territory?
• How can we improve communication between utilities?
• How can we plan a coordinated response to an event?
Enter predictive damage response. A storm impact analytics model applies a combination of weather parameters, such as high winds, lightning, ice accretion and more, to determine the strength of incoming storms and the associated risks at various times.
This information can be combined with non-weather intelligence, such as vegetation, utility asset records and other variables, to create an application that predicts and displays severe weather threats as they move through a service territory.
Statistical modeling and predictive analytics help determine the impact of weather on assets at specific geographical locations. Such an application can produce decision points, reports and predicted impact scenarios for any incoming severe weather event.
Finally, everything is translated into impact statistics, helping with damage and restoration effort estimates.
To reap the benefits of storm impact analytics and be truly prepared for any type of extreme weather scenario, a utility can shape the model in only three simple steps:
1. Identify historic storm-related outages (thunderstorms, ice and wind storms, heat, population density, etc.) that have impacted the service territory using corresponding outage data.
2. Gather as much data as possible that is relevant to the processes that influence outages.
3. Aggregate all the variables on a grid or geographic unit and use this to create an analytics model that predicts impacts on assets and outages per unit area.
A storm impact analytics model allows a utility to map all parts of its service territory impacted by predicted severe weather. Such insight can improve the speed of service restoration while reducing associated costs. Furthermore, this information can be used to enhance customer communications.
A solution like this can help a utility better prepare for future storms and identify at-risk areas within its service territory, supporting targeted hardening efforts. In addition, it allows utilities to prepare for multiple scenarios, simulating how different conditions can evolve during a severe weather event. This can help utilities optimize their response times and restoration efforts.
A New Frontier
Increasing situational and sophisticated awareness throughout the service territory will improve the ability to identify not only at-risk utility assets, but also at-risk facilities, such as mission-critical infrastructure locations, hospitals and schools.
Such comprehensive analytics will allow for better crew planning, including capacity and need, and can be used to help strategize and support the response, making shift schedules, hold overs and on-call decisions more intuitive. Working with first responder organizations and other community leaders will become the norm, supporting more informed decisions and training opportunities for emergency scenarios.
With the recent surge of severe weather, utilities have been forced to reassess protocols to preserve their bottom lines. Turning weather forecasts into data that can be used to predict asset damage can greatly improve operational efficiencies. New datasets will deliver new results, which will ultimately help the utility industry thrive and grow for years to come. UP
Don Leick is a senior product manager for Schneider Electric’s weather business. Leick leads the future direction and enhancement of online, mobile and alerting products. He’s been the product manager for the WeatherSentry product for most of his 12 years with the company. Leick has extensive experience working with many of Schneider Electric’s customer segments including utilities, wind farms, sports and winter road maintenance.