Disaster response and mitigation in an AI world
After the destructive California wildfires in 2019, the U.S. government hosted a White House Executive Forum to develop better ways to protect the nation and key infrastructure, such as the power grid, from wildfires and more. disasters. In 2020 alone, more than 10.3 million acres burned in the United States, a level three times the 10-year average of 1990-2000. Between the fire suppression costs, direct and indirect costs, wildfires in 2020 cost the United States more than $ 170 billion. Add to that floods, hurricanes and other natural disasters, and the toll of disasters on Americans’ livelihoods is astronomical.
Andre Coleman and his team of researchers at the Pacific Northwest National Laboratory (PNNL) are part of the First Five Consortium, a group of government, industry and academic experts committed to reducing the impact of natural disasters using technology. Coleman and his team are extending PNNL’s Rapid Analytics for Disaster Response (RADR) image analysis and operational modeling suite to mitigate damage to key energy infrastructure. Using a combination of image capture (satellite, airborne and drone imagery), artificial intelligence (AI) and cloud computing technologies, Coleman and his team work not only to assess damage, but also to predict it. .
Accurately forecasting the movement of natural disasters (forest fires, floods, hurricanes, windstorms, tornadoes and earthquakes) gives first responders a boost, enabling them to take action to reduce damage, conduct planning advance of resources and increase infrastructure restoration time. For example, if a fire hit an electrical substation or other network infrastructure, an entire community (homes, businesses, and schools) would experience a power outage that could take days to recover.
“This is an exciting and timely effort to apply artificial intelligence to reduce the impact of wildfires, protect energy infrastructure and ultimately save lives,” said Pamela Isom, Acting Director of the US Department of Energy (DOE) Artificial Intelligence and Technology Office. “The work has the potential to make a difference in what we anticipate to be a very difficult forest fire season. It has been a very productive collaboration between several partners, including our colleagues at the Joint Center for Artificial Intelligence in the Ministry of Defense, Ministry of Internal Security, and PNNL. “
Since 2014, Coleman and his team have been working with these technologies. The project originally started with the creation of a change detection algorithm, which analyzes different types of satellite images and determines what has changed in the landscape after a storm. Authorities use the tool to quickly assess the impact of physical damage from natural disasters, often before ground crews can enter. The first iteration of the tool was used during the 2016 hurricane season to assess hurricane damage and determine whether energy infrastructure – power grid, oil, and gas facilities – has been damaged or at risk.
Overall, RADR analytical products provide value, but Coleman and his team recognize opportunities to extend tool functionality and seek to improve RADR response time, damage assessment, visibility, capacity. prediction and accessibility of data.
To improve punctuality and ground ratings, the team incorporated new and different image sources. RADR can extract images from a variety of satellites with different detection capabilities, including national and international government satellites which are offered as open data as well as commercial satellites which are available through the International Disaster Charter. Having multiple aerial image sources reduces the response time to just a few hours, the main limitation being the latency of the aerial images, or the time between collecting the images and their availability for analysis. Once the images are received, the RADR software can generate an analysis in just over 10 minutes.
To observe forest fire smoke and cloud cover, the team added infrared images to the RADR. The new capability provides a clearer view of the landscape that was not previously available, giving responders information such as damage to key infrastructure or a safe location to mount relief efforts that responders may not have been able to afford. – to be otherwise unaware.
The team also incorporates publicly available images from social media. Often in times of disaster, social media networks like Twitter, Flickr, and Instagram provide a wealth of real-time data as users post photos of what’s going on around them. By combining aerial imagery with ground imagery, the team can provide a more comprehensive assessment. Satellite images, for example, can show damage to a generating resource, power lines, or the power grid; however, ground images may indicate otherwise. The tool takes all of these images, removes redundant ones, and stitch the images together to provide a more accurate view of changing conditions.
As with any computational model, it’s only as good as the data. The added image sources provide additional data to be interpreted by RADR, thus improving accuracy. To predict the possible outcomes of a forest fire, the team combines image analysis with weather, fuel, and forecast data. For example, wind, vegetation, and whatever else a fire can consume, all of which factor into the size of a fire and the direction it takes. By combining imagery with fuel data and forest fire models, the team hopes to be able to accurately predict the path a fire will take.
Of course, reviews need to be in good hands. Coordinating a response requires local, regional and national resources, each in different locations but requiring the data as quickly as possible in an easily accessible and interpreted format, especially in a limited data communications environment. A cloud-based system provides an end-to-end pipeline to retrieve available images, process analyzes, and disseminate data for use directly in a user’s own software, through desktop web browsers, and / or through applications mobile. The added visual analysis produces images and data sets that can be easily discerned for a large audience of respondents.
Recent years have resulted in an increase in the frequency and severity of forest fires, floods and other extreme weather events. Coleman and his team hope that at the very least, RADR’s additional capabilities will provide responders with information that can be used to make informed decisions, reduce or plan damage to key energy infrastructure, plan relief efforts, and save lives. .
Quote: Disaster Response and Mitigation in an AI World (2021, June 25) retrieved June 25, 2021 from https://phys.org/news/2021-06-disaster-response-mitigation-ai-world. html
This document is subject to copyright. Other than fair use for private study or research purposes, no part may be reproduced without written permission. The content is provided for information only.