Vector uses The Weather Company from IBM and uses AI and data analysis to predict and prevent crashed poles and lines.
CES 2019: How business professionals can benefit from AI in the company
IBM Watson can help with marketing, supply chain, and commerce, especially as AI changes from general to broad data.
As storms and hurricanes have increased in number and strength over the last 30 years, the effects of climate change have been felt by everyone. This includes drastic increases in the length and width of power outages. According to IBM, 70% of disruptions worldwide are due to weather-related causes such as wind, rain, and ice.
But what if there was a way in which utility companies could know where the power cuts would strike before they occurred? What if utility companies were able to predict and shut down poles and lines that could remain powerless for thousands of days?
Thanks to The Weather Company from IBM, this is now a reality for Vector Energy in New Zealand. The partnership with IBM has led to a reduced number of power outages thanks to artificial intelligence and data-driven analysis, IBM said.
“All utilities are affected by bad weather; they are all worried and are trying to figure out how to better address this problem, how to prevent downtime by shortening the right branches at the right time, or restoring power based on faster mobilization of crews in the right places, “said IBM Manager Robbie Berglund.
“Many utility companies have used a traditional approach based on experience. What they realize is that the old way to solve problems when you are in a region with severe weather conditions is inefficient and not as effective. To continue with the old ways of just looking at the weather and deciding that the human brain is too much to process. That leaves a lot of room for calculations and analysis to help improve and increase what they are already doing, “he added.
Berglund explained that IBM worked with Vector for about a year and a half and that it took so long to make the tools fully operational. The IBM system mainly helps with two important things: preventing disturbances with vegetation management and anticipating where line errors will be so that crews can be deployed quickly and efficiently.
Such a prediction has been around for longer than vegetation management, according to Berglund, and IBM has worked on it with dozens of utilities around the world. IBM says this has led to immediate cost savings as utility companies can now prevent excessive mobilization in places that are not severely affected by severe weather.
IBM AI and weather data help energy companies predict power outages during heavy weather.
With forecasting, utility companies can also avoid a costly practice called ‘mutual aid’, namely when they borrow power from each other in the event of significant power problems. By knowing in advance which areas are likely to experience the highest number of disruptions, they can optimize their predictions and possibly prevent them from fully engaging in mutual assistance.
“If you put a truck with four people on the road and multiply that by many trucks, if they are not driven or assigned to where the disruptions happen quickly, it becomes expensive. If you can combine your decision-making with data-driven analysis, it will decrease the costs of your mobilization, and you will need to restore power faster. That’s what Vector is looking for, “Berglund said.
Vegetation management is something IBM recently started. With the help of aerial recordings from drones, planes, or satellites, the company’s AI system knows which lines are likely to be disturbed by tree branches or other natural barriers that can knock them down in the event of a storm.
“Before the storms ever arrive, we tell them,” here’s where you have to trim beforehand, “so that in 12 months we can tell where to trim trees and vegetation and where the worst spots are,” Berglund added.
“That may sound simple, but if you have a tool with a streamline of 10, 20, 30, 40, or 100 thousand miles, you have to do more than guess, and it is really expensive to drive people around and try the worst ones, so you can do that with AI and remote sensing with satellite images or aerial photos and tell a utility where their worst areas are, “he said.
IBM’s AI system analyzes the historical performance of the lines and poles of a utility to see how storms of certain intensities will affect them. The company now works with many utilities on these specific tasks, including Canada’s Hydro One.
Traditionally, utilities such as Vector would only rely on the expertise of their employees and decide what to do based on their past experiences. But Berglund said that nothing IBM offers can replace the kind of institutional knowledge that old utilities have.
“What I always tell utilities is to keep your approach to the man in the hall, who has been there for 30 years, and he knows that this storm will be a doozie. Keep that experience, but improve it with a data-driven approach that has no emotion,” “Berglund said.
“It is black-and-white modeling that performs quite well. The two together are the best approach.”
Vector and IBM say they see tangible results from their partnership. According to Berglund, the number and severity of power outages have decreased since the partnership started last year.
While some utilities choose to do all this alone without the help of a company like IBM, many may find it difficult and beyond their comprehension.
“Utilities usually find that they are really good at being utilities but not so good at handling large, complex data sets or analyses, and especially not at artificial intelligence,” Berglund said. “They know that to do better, they have to use data and analysis. If someone buys, I have seen so far that they absolutely change the process.”
For most utilities, he said, the process starts about three days before a storm hits. About a week before a storm strikes, Vector is notified with basic facts about the storm, such as whether there will be snow or how strong the wind will be. Based on this information, it is two days before a storm hits when utility companies decide how large their crew should be to handle downed poles and lines and which areas may be affected.
IBM plans to work with Vector on this project for years and hopes to expand its work with other utilities. Berglund said that almost any tool could benefit from a tool like that from IBM.
“We don’t see a tool that doesn’t need a data-driven analytical approach. Most tools recognize that,” he said. “I really believe it’s only a matter of time before most or all of the utilities use data and analysis, or AI, to address these two issues.”
