How Neara is using AI to protect facilities from extreme weather

Image credits: near

Over the past few decades, extreme weather events have not only become more severe, but also occurring more frequently. Neara is focused on enabling utility companies and energy providers to create models of their power grids and anything that might affect them, such as wildfires or floods. The Redfern, New South Wales, Australia startup recently launched AI and machine learning products that create large-scale network models and risk assessments without the need for manual surveys.

Since its commercial launch in 2019, Neara has raised a total of AU$45 million (about US$29.3 million) from investors such as Square Peg Capital, Skip Capital, and Press Ventures. Its clients include Essential Energy, Endeavor Energy and SA Power Networks. It also partners with Southern California Edison Co. and EMPACT Engineering.

Neara's AI and machine learning features are already part of its technology stack and have been used by utilities around the world, including Southern California Edison, SA Power Networks, Endeavor Energy in Australia, ESB in Ireland, and Scottish Power.

Co-founder Jack Curtis told TechCrunch that billions are being spent on utility infrastructure, including maintenance, upgrades and labor costs. When something goes wrong, consumers are immediately affected. When Nera began incorporating AI and machine learning capabilities into its platform, the goal was to analyze existing infrastructure without performing manual inspections, which he says are often inefficient, inaccurate and expensive.

Neara then developed its own AI and machine learning features so it could create a large-scale model of the utility network and its surrounding areas. Models can be used in a number of ways, including simulating the impact of severe weather on electricity supplies before, after and during the event. This can increase the speed of power recovery, keep utility teams safe, and mitigate the impact of weather events.

“The increasing frequency and severity of extreme weather events motivates us to develop our products more than any other event,” Curtis says. “Recently, there has been an uptick in extreme weather events around the world and the grid is being affected by this phenomenon.” Examples include Storm Isha, which left tens of thousands without power in the United Kingdom, winter storms that caused widespread power outages across the United States, and tropical cyclone storms in Australia that left Queensland's power grid vulnerable.

Using artificial intelligence and machine learning, Neara's digital models of utility networks can prepare energy providers and utilities for them. Some of the situations Neara can predict include places where high winds could cause power outages and wildfires, floodwater levels that mean grids need to shut down their power, and ice and snow buildup that can make grids less reliable and resilient.

In terms of training the model, Curtis says that AI and machine learning “were built into the digital network from the beginning,” with LiDAR being crucial to Nera’s ability to accurately simulate weather events. Its AI and machine learning model has been trained “on more than a million miles of diverse network areas, helping us capture seemingly small but highly significant nuances with extreme accuracy,” he adds.

This is important because in scenarios such as flooding, a one-degree difference in elevation geometry can lead to inaccurate modeling of water levels, meaning utilities may need to energize power lines before they need to, or on the other hand, conserve energy. for longer than required. security.

Neara's founders are Daniel Danilatos, Karamveer Singh and Jack Curtis

LiDAR images are captured by utility companies or third-party capture companies, rather than LiDAR. Some customers scan their networks to continuously feed Neara new data, while others use it to gain new insights from historical data.

“One of the key outcomes of assimilating this LiDAR data is the creation of a digital twin model,” Curtis says. “This is where the power lies in contrasting with raw LiDAR data.”

Examples of Nera's work include: Southern California Edison, where her goal is “automatic prescription,” or automatically identifying where plants are likely to catch fire more accurately than manual surveys. It also helps inspectors tell survey teams where to go, without putting them in danger. Since utility networks are often huge, different inspectors are sent to different areas, which means multiple collections of personal data. Using the Neara platform makes the data more consistent, says Curtis.

In the case of this Southern California Edison company, Neara uses LiDAR technology and satellite imagery and simulates things that contribute to wildfires spreading through vegetation, including wind speed and ambient temperature. But some of the things that make predicting vegetation risk more complicated is that Southern California Edison needs to answer more than 100 questions for each of its utility poles because of regulations, and it's required to inspect its transmission system annually.

In the second example, Neara began working with SA Power Networks in Australia following the 2022-2023 Murray River flood crisis, which affected thousands of homes and businesses and is considered one of the worst natural disasters to hit South Australia. SA Power Networks captured LiDAR data from the Murray River region and used Neara to perform digital modeling of the flood impact and see how much damage had been done to its network and how much risk remained.

This enabled SA Power Networks to complete a report in 15 minutes that analyzed 21,000 power lines within the flood zone, a process that would have otherwise taken months. For this reason, SA Power Networks was able to reactivate power lines within five days, compared to the three weeks originally expected.

3D modeling has also allowed SA Power Networks to model the potential impact of different flood levels on parts of its electricity distribution networks and predict where and when power lines may breach approvals or be at risk of a power outage. After river levels returned to normal, SA Power Networks continued to use the Neara model to help it plan the reconnection of its electricity supply along the river.

Neara is currently doing more research and development in the field of machine learning. One goal is to help utilities get more value from their current live and historical data. It also plans to increase the number of data sources that can be used for modeling, with a focus on image recognition and photogrammetry.

The startup is also developing new features with Essential Energy that will help utilities evaluate every asset, including poles, in the grid. Individual assets are currently valued based on two factors: the likelihood of an event such as severe weather occurring and how well it will hold up under those conditions. This type of risk/value analysis is typically done manually, and sometimes doesn't prevent failures, such as in the case of power outages during the California wildfires, Curtis says. Essential Energy plans to use Neara to develop a digital network model that will be able to perform more accurate asset analysis and reduce risks during wildfires.

“Basically, we allow utilities to stay ahead of severe weather by understanding how it impacts their grid, allowing them to keep the lights on and their communities safe,” Curtis says.