How This Mumbai-Based Transmission Company Managed Power During Amphan With AI-Powered Weather Intelligence Platform

How This Mumbai-Based Transmission Company Managed Power During Amphan With AI-Powered Weather Intelligence Platform
Image © How This Mumbai-Based Transmission Company Managed Power During Amphan With AI-Powered Weather Intelligence Platform


The recent Cyclone Amphan in Bay of Bengal and Nisarga in the Arabian Sea has disrupted millions of lives and killed hundreds of people. However, Mumbai-based power transmission company, IndiGrid, managed to overcome challenges arisen due to the strong cyclone with a weather intelligence platform.

Established in 2016, IndiGrid provides the infrastructure for the power transmission assets in India. With 20 transmission lines of ~5,800 km and four substations of 7,735 MVA capacity across 13 states in India, currently, IndiGrid is acting as one of the largest platforms for operational transmission assets in India. The institution has also signed framework agreements which shall allow IndiGrid to grow its AUM to approximately Rs. 18,000 Cr in the next two years. 


Sign up for your weekly dose of what's up in emerging technology.

The Challenges

Transmission lines, which are used for the transmission of electrical power to various distribution units, are usually vulnerable to extreme weather events like cyclones, and floods. The increasing natural calamities due to climate change has forced transmission companies to remain alert in such situations to respond in a timely manner. Such unplanned major breakdown events can create power outages or load shedding in affected areas, which, in turn, affects societies and industries. 

When asked, Satish Talmale, COO of IndiGrid, said, “As a transmission service provider, it’s of paramount importance to minimise the restoration time of the transmission lines during such catastrophic events. In the current scenario, without the access to accurate data and predictions, this becomes very difficult.” With the majority of these towers and conductors placed in remote locations, the meteorological data from the available sources are far away. Therefore, it gets even difficult to forecast the actual wind speed, which, in turn, extends the response and restore time in an emergency.

Alongside, in order to understand the failure modes while performing root cause analysis of such natural disasters, the transmission companies need to have precise microsite wind data on each tower. However, the available wind data with India meteorological department or Skymet Weather Services (private Indian company that provides weather forecast and solutions to Indians) becomes irrelevant as all their sensors are installed far away from the transmission towers, which again defeats the purpose. “Due to the lack of such precise wind data, it is difficult for design engineers to investigate the failure modes of tower collapse events. This challenge is also evidently reported in the CEA (Central Electricity Authority) tower event report,” said Talmale.

Also Read: How Deep Learning Is Mitigating Climate Change Threats

Apart from the delayed power outages, such delayed response and absence of the ability to manage such situations during natural calamities have impacted the industry at large. Alongside, it also significantly impacted the operating revenues of the transmission companies, which are again linked to the availability of transmission lines, calculated as per CERC regulations. 

Explaining further, Talmale said, though such natural disasters are eligible for force majeure claims, without accurate and reliable data it takes longer for the company to justify the force majeure event in front of concerned authorities. Without precise wind data to establish clauses, it also took the regulatory authorities longer in their decision-making process and insurance companies to approve the claims. Situations like this have always put the company at risk of losing operating revenues from such events which are beyond control due to lack of credible and accurate data.

The Solution & Benefits

In order to mitigate aforementioned risks and develop the capabilities for swift emergency restoration during such unplanned major breakdowns, the company decided to transform their work process and obtain the help of a weather intelligence platform for a better outcome. After severe evaluation, the company chose to partner with ClimaCell’s AI-powered Weather Intelligence Platform that assisted the company in analysing the historical data to determine the recent changes in climatic conditions compared to wind zones. Such analysis enabled the company to proactively work on tower monitoring and strengthening measures to help determine a robust risk mitigation plan during the Amphan Cyclone.

Also Read: Case Study – How This Madurai-Based Manufacturing Company Streamlined Its Logistics Process With Augmented Analytics

ClimaCell’s Bespoke Atmospheric Model (CBAM) works on the inputs from its proprietary “Weather-of-Things” which add millions of new observations that have not been previously used for weather forecasting — from wireless signals to cars sensors. The solution integrates the traditional data sources with the data from the connected world then analyses the data using models that provide a supreme level of accurate results, and thus, helped in improving forecasting for IndiGrid. This cloud-native numerical weather prediction model (NWP), CBAM operates in a cloud-based high-performance computing environment, thereby allowing seamless scaling of the solution as and when the NWP needs of the customers’ increase or decrease.

Alongside, the solution’s high-resolution analysis of rapid refresh worked in the resolution of three kilometres and performed hourly refresh cycles during the cyclone, whereas most other public models update every six hours. Also with its WoT approach, the solution worked from hundreds of millions of proprietary observations from the connected world including radiances, GPS radio occultation, reflectivity, velocity-azimuth display and doppler winds, and allowed the company to predict wind speed for 48-72 hours and enabled them to have proactive planning during Amphan.

During Amphan, this AI-powered solution by analysing the data equipped IndiGrid with information like the expected time of the cyclone, duration of the cyclone, intensity and location of the calamity, and impact of the storm, which gave the company a heads up to prepare. The initial configuration of this advanced AI solution for all the power transmission towers took 4-6 weeks and “post that it’s pretty seamless to operate,” said Talmale. “Also, the solution provided a forecast within 60 minutes or lesser resolution for a day, which enabled proactive planning and organising abilities to deploy resources like manpower, raw materials like tools and spares required for the restoration of towers, machines etc. which was critical during the cyclone.” 

Also Read: How Having Bigger AI Models Can Have. Detrimental Impact On Environment

The timely restoration of power transmission lines accelerated the overall restoration time and eliminated the challenges associated with it. The CBAM assimilation system leverages ensemble forecast data from NCEP’s global ensemble forecast system to provide critical model-uncertainty information using GSI’s hybrid 4-Dimensional Ensemble Variational (4DEnsVar) assimilation technique. During the cyclone, the company used the HyperCast AI-powered dashboard to control the power transmission in the city and make informed decisions. 

Furthermore, this technology, which operated independently with its data feed and quality assurance system also helped in mitigating enterprise risk by minimising operational revenue impact and also allowed the company to serve the nation continuously with reliable power during the catastrophe. Also, the solution is maintained by a staff of expert scientists and engineers to maintain forecast accuracy and to provide 99.9% uptime of the operational system.

“The solution also provides alert management for early warnings via mobile SMS or email and advanced insights from its analytical-based artificial intelligence technology,” said Talmale. “The only challenge could be the computer processing speeds at user end as the platform can manage a huge amount of data points, and thus the user needs to have desired devices with high-end configurations.”

Moreover, the solution helped IndiGrid to mitigate vegetation risks proactively with the early warning on the specific high wind tower and corridors. This proactive approach also prevented power system trips or damages to transmission towers/conductors which are observed during earlier cyclones. 

“Deploying this solution also helped us in designing a robust transmission tower which is resistant to high wind speed, which again ensured a committed performance to our shareholder and stakeholders,” said Talmale. “Such technology-enabled reduction in overall response time, leads to the earliest resolution time, thereby facilitating earliest power supply restoration to the nation.”

More Great AIM Stories

Sejuti Das
Sejuti currently works as Associate Editor at Analytics India Magazine (AIM). Reach out at

Our Upcoming Events

Conference, in-person (Bangalore)
MachineCon 2022
24th Jun

Conference, Virtual
Deep Learning DevCon 2022
30th Jul

Conference, in-person (Bangalore)
Cypher 2022
21-23rd Sep

3 Ways to Join our Community

Discord Server

Stay Connected with a larger ecosystem of data science and ML Professionals

Telegram Channel

Discover special offers, top stories, upcoming events, and more.

Subscribe to our newsletter

Get the latest updates from AIM