This energy tech startup is using AI to help electric utilities during natural disaster-like emergencies
Growing up in a middle-class family in Kolkata and Midnapore in West Bengal, India, Dr Sayonsom Chanda was no stranger to strong winds and relentless rain knocking down electricity for hours, days, and even weeks. He and his family lived in East Midnapur through the horror of the 1999 Odisha cyclone and Sidr cyclone in 2007.
It was one of the core reasons for him to start Sync Energy in 2017. The startup builds artificial intelligence (AI) tools that simplify emergency and disaster response planning for electric power distribution companies.
The platform helps electric utilities reduce customer downtimes and be better informed about the impact of a disaster before it actually strikes. This, in-turn, will help electric power companies to decrease costs associated with emergency-related power outages.
Today, Sync Energy is a global startup headquartered in New York, with its India office in the process of being established in Kolkata, India. It is powered by the passion of a 12 person team, scattered throughout the US, India, Denmark, and Ghana.
Starting up in the space
“I remember we had only one emergency lamp. When the lights would go out in the evenings, I would unplug it and place it on the kitchen counter so that my mother would have some of that radiance to cook, while I could use a part of it to finish my homework,” Sayonsom recalls.
Several years later, when he was pursuing his PhD in Washington State University, Sayonsom met Robert Kabera, a Stanford University graduate working for the same professor - Prof. Anurag K Srivastava.
When Hurricane Harvey and Hurricane Maria ripped across the southeastern states of the US and Puerto Rico in 2017, Sayonsom, with his newfound knowledge about the electric utility industry and familiarity with the capabilities of simulations and artificial intelligence, could spot the bottlenecks and inefficiencies within the electric power industry.
Looking at what was needed
Over the next few days, they tried to reason why such prolonged power outages were unavoidable. Discussing with one of Robert’s mentor at Stanford University, Michael Bernard, who spent his career on public safety and emergency response, they observed that field engineers, operators, practitioners, and planners are disconnected with the artificial intelligence and computation capabilities.
“That is why many problems (such as prolonged power outages) can be theoretically predicted and resolved on paper far more easily than on the field,” says Sayonsom.
He realised he had to build a tool that would predict the damage to the power grid so that it would give utilities ample time to prepare for the event.
“We saw an opportunity and a real advantage to automate much of these calculations and put the grunt-work behind the scenes so that engineers are less stressed and can focus on every utility engineer, linemen, crew, and first responders the power to do simulation,” explains Sayonsom.
“We provide a no-code interface for electrical engineers, field crew, and planners to perform advanced analytics, collaboration, and a rich data-discovery platform. Our product is designed such that every engineer, employee, or contractor in the electric utility industry can have access to AI-powered forecasts and insights when they need it,” says Sayonsom.
The team claims to have built the first and biggest power system incidents database called Events Intelligence (EI), which is a library built using thousands of historical datasets, and over 350,000 scientific papers, articles, and international standards.
“We are the first to do it for resilience, recovery, and restoration problems related to electric utilities. We provide a web-based interface along with smartphone apps to our analytics platform, so they can be used anywhere, anytime,” says Sayonsom.
The smart grid is not there everywhere yet, where the predictive technologies coupled with automation could bring back electricity to customers very easily.
In most parts of the world, and especially in rural parts of eastern India, which are extremely vulnerable to multiple powerful cyclones every year, if the power network falls in the path of a storm, all customers will be affected as there are very few opportunities to backup and restore power using redundant networks.
This leads to week-long power outages, which further cascades to affect agriculture, small industries, and the already fragile economics of the rural population.
In such circumstances, while the traditional power grid transitions into a smart grid - Sync has opened access to its damage predictive capabilities for first responders, emergency response NGOs, state disaster planning departments, block development officers of rural areas, and researchers to plan for how many emergency or solar powered lamps to distribute, how many power banks need to be distributed, if backup diesel generators will be required - when, how many, for how long, and where.
Sync Energy also predicts how much damage will be caused to the electric grid due to the cyclone, earthquake, wildfires or even a cyberattack. The information can be used both for near term planning and initiating long-term resilience projects.
What sets them apart?
“Unlike our competitors, our proprietary technologies provide component-by-component analytics of how each asset of the power grid infrastructure will respond to the disaster event. This allows us to offer analytics for the electric power grid at a very high resolution - giving it the industry’s highest accuracy in damage prediction,” says Sayonsom.
The key clients are electric power distribution companies - whether owned by the state or private. Sync charges them an annual licence fee to access its advanced cloud computing resources. The fee is set based on the size of simulations the electric utility typically has to undertake.
Recently, following the aftermath of Amphan in West Bengal, Sync decided to offer its asset damage prediction technologies for free to emergency response NGOs, rural block development officers, and research organisations.
“Our core mission is to light up people’s lives in any way we possibly can, and to make disasters less dark and the recovery less messy. We will do anything to stay true to our mission,” says Sayonsom.
The business model
Sync Energy works as a simulations-as-a-service provider. Sayonsom says the team defines its unit as the cost to perform simulations required to predict the probability of damages suffered to serve one customer. Citing an example, he says:
“Our revenue forecasts are based on the number of customers served by the utility and the number of planning and emergency response activities that are done.”
“Our analytics could potentially save the utility millions in disaster-related damages. We provide a great financial advantage for the electric utilities for a strategy stand-point,” he adds.
Since both the threats faced by the power grid and the demand for reliable electric power are growing year-on-year, the startup’s revenue forecasts are looking strong for the foreseeable future.
Market and future
There are several startups like Enheroes in the data and energy space and Blume Ventures-backed Zenatix, an IoT-driven energy data analytics company working in this sector.
A study by research firm IDC estimates that by 2020 there will be over 28 billion objects with data exchange capabilities.
“Our software is for mission-critical applications with no room for error. It took us four years of intense research and development and attention to detail to build the minimum viable product that was ready for testing at the industrial scale,” says Sayonsom.
The backend technology costs more than $300,00 over four years, including the time put in by the co-founders to build the predictive analytics brain of Sync.
The front-end development costs were around $15,000 and there are fixed ongoing costs associated with cloud computing for building and training the machine learning models. The startup has raised funding from a leading US-based accelerator.
For the last two years, Sync Energy worked solely with electric utilities, helping them make emergency response planning simpler, smarter, and safer.
“The coronavirus pandemic has shown us how interconnected and interdependent we are. Thus, we realised that if we are to achieve true resilience, it has to be inclusive, all-pervasive, and holistic,” says Sayonsom.
This year, the team is expanding to communities without smart grid technologies. It is helping local authorities and disaster response teams in those areas by predicting the disruption to their fragile supply chains, and providing them with planning and preparation resources to live through the difficult days.
Sayonsom says, “We are building a paradigm for modern, data-driven storytelling to prepare communities and companies to be resilient by predicting what’s the worst that can happen to them, and what’s our plan to help them recover -before the event happens. In addition to rapidly adding clients to our predictive analytics platform, we plan to grow by creating an optimum marketplace with vendors offering resilience-enabling products and technologies, and taking them to people and communities in need, and those who can support them.”
Edited by Megha Reddy