Over the past few decades, extreme weather events have not only become more severe but also more frequent. Neara is focused on enabling utilities and energy providers to model their power networks and the things that could impact them, such as wildfires and floods. The Redfern, New South Wales, Australia-based startup recently launched an AI and machine learning product that creates large-scale models of networks and assesses risk without manual inspection.
Since launching commercially in 2019, Neara has raised a total of AUD 45 million (approximately USD 29.3 million) from investors including Square Peg Capital, Skip Capital and Press Ventures. Customers include Essential Energy, Endeavor Energy and SA Power Networks. We also partner with Southern California Edison Co and EMPACT Engineering.
Neara's AI and machine learning-based capabilities are already part of the company's technology stack and are used by power companies around the world, including Southern California Edison, SA Power Networks, Australia's Endeavor Energy, Ireland's ESB, and Scottish Power. used in
Co-founder Jack Curtis told TechCrunch that billions of dollars are spent on utility infrastructure, including maintenance, upgrades and labor costs. When something goes wrong, consumers are immediately affected. When Neara began integrating his AI and machine learning capabilities into its platform, it was to analyze existing infrastructure without manual inspection. This is often inefficient, inaccurate and can be expensive, he says.
Neara has since enhanced its AI and machine learning capabilities to create large-scale models of utility networks and their surroundings. The model can be used in a variety of ways, including simulating the impact of extreme weather events on power supply before, after, and during an event. This increases the speed of power restoration, ensures the safety of utility teams, and reduces the impact of weather events.
“The increasing frequency and severity of severe weather events is motivating our product development more than any other event,” says Curtis. “Severe weather events have increased rapidly around the world in recent days, and power grids have been affected by this phenomenon.” Examples include Storm Isha, which left tens of thousands of people without power in the UK, and caused widespread power outages across the United States. These include the winter storm and the Australian tropical cyclone storm that left Queensland's power grid vulnerable.
Using AI and machine learning, Neara's digital model of utility networks can prepare energy providers and utilities. Situations that Neara can predict include: strong winds that could cause power outages and wildfires, flood water levels that mean the network has to shut off energy, and reduce network reliability and resiliency. Possible ice and snow buildup, etc.
When it comes to training models, Curtis said AI and machine learning have been “incorporated into digital networks from the beginning” and that LiDAR is critical to Neara's ability to accurately simulate weather events. He added that the company's AI and machine learning models were “trained on over 1 million miles of diverse network regions, helping us to capture seemingly small but resulting important nuances with ultra-high accuracy.” .
This is important. This is because in scenarios like flooding, her one degree difference in elevation geometry can lead to inaccurate water level modeling. This means that the utility company may need to energize the power lines before they are needed, or conversely, may need to keep the power on longer than it currently does. safety.
Neara Co-Founders Daniel Danilatos, Karambir Singh, Jack Curtis
LiDAR images are captured by the utility company or a third-party capture company on behalf of LiDAR. Some customers scan their networks and continuously feed new data into Neara, while others use it to derive new insights from past data.
“A key outcome of this LiDAR data ingestion is the creation of a digital twin model,” says Curtis. “As opposed to his raw LiDAR data, that’s where the power lies.”
An example of Neara's work is at Southern California Edison. The project's goal is to “autoprescribe,” or automatically identify where vegetation is likely to ignite more accurately than manual surveys. It also helps the inspector direct where to go without putting the investigation team at risk. Utility networks are often large, so different inspectors are sent to different areas, requiring multiple subjective data. Curtis says using Neara's platform increases data consistency.
In this Southern California Edison case, Neara uses LiDAR and satellite imagery to simulate factors that contribute to the spread of wildfires through vegetation, such as wind speed and ambient temperature. However, factors that make predicting vegetation risk more complex include regulations that require Southern California Edison to answer more than 100 questions about each utility pole and require annual inspections of the transmission system. Masu.
In the second example, Nyala is a major disaster 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 southern Australia. We started working with SA Power Networks. SA Power Networks took his LiDAR data from the Murray River region and used Neara to perform digital flood impact modeling to see how much of its network was damaged and how much risk remained .
This enabled SA Power Networks to complete a report analyzing 21,000 power line spans in flood areas in 15 minutes. Without this, this process would have taken months. As a result, SA Power Networks was able to re-energize the power lines within five days, instead of initially expecting it to take three weeks.
3D modeling also allows SA Power Networks to model the potential impact of different flood levels on parts of the distribution network, determining when and where transmission lines could breach gaps or be at risk of power outages. It's now predictable. Once river levels returned to normal, SA Power Networks continued to use Neara modeling to help plan the reconnection of power supplies along the river.
Neara is currently conducting further machine learning research and development. One goal is to help utilities extract more value from their existing 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 capabilities with Essential Energy that will help utilities evaluate each asset, including utility poles, in their networks. Currently, individual assets are valued based on two factors. It is the probability that events such as extreme weather will occur, and how well they can withstand such conditions. Curtis said this type of risk-value analysis is typically done manually and sometimes fails to prevent failures, such as the power outages during the California wildfires. Essential Energy plans to use Neara to develop digital network models that can perform more accurate analysis of assets and reduce risk during wildfires.
“Fundamentally, we can help power companies stay one step ahead of extreme weather events by understanding exactly how they impact their networks, keeping their lights on and their communities safe. We try to maintain it,” Curtis said.