In India, despite regular policies announcing the shift to cleaner fuels, coal-fired thermal power plants still account for more than 70% of India’s electricity. For developing countries where coal is the cheapest fuel, this may not be surprising. Therefore, in India, thermal power will continue to occupy the top spot for a long time, rather than concerns about carbon emissions, pollution and climate change.

What the use of AI technology can reduce the damage to our health and the environment is that it reduces the consumption of thermal power plants. This requires power plants to increase efficiency by adopting new technologies. But state-owned power plants and power subsidies complicate investment in technological solutions to incentivize the transition to more efficient power plants. This is one of the reasons why industrial IoT startups could help power plants with data analysis, but found it difficult to sell their products in India in the past and focus on buyers in the global market.

In the past few years, this situation has changed due to a variety of factors. "This is a very difficult market to kill, but thermal power emphasizes the industry. Now, relying on AI technology can improve operational efficiency and predictive maintenance.

Key challenge

At present, one of the pressure points for thermal power plants in India comes from the increase in the share of renewable energy. This means that when solar energy reaches the peak level of the grid, thermal power plants must operate at partial capacity during daylight hours. This brings operational challenges. "When you suddenly lower the temperature, physical stress is generated. The plant operator does not have the mind map required to reach the optimal level. This is a flexible operation plan, and this is one of the things we want to solve for the power plant."

With the help of AI, start-ups can create digital control of the entire process, taking into account various conditions, and find the best configuration in the on-site environment, which helps improve efficiency and reduce fuel input, which in turn has an impact on carbon emissions.

"We have been able to increase the boiler efficiency by 1%, which is very important. For a 500 MW thermal power plant, this means that 10,000 tons of coal can be reduced each year." The use of AI technology will affect another A key area is to reduce expensive downtime caused by failures. For example, the failure of a large steam turbine may cause the entire production to stop. In a 500MW power plant, the daily cost is 3-4 crore rupees.

ExactSpace faces global competitors such as GE and its Predix platform provides industrial IoT solutions. However, unlike the GE platform built for multiple vertical industries, this Indian startup focuses on the special needs of thermal power.

The two co-founders Raghunathan and Arun Jose started visiting the power plant while working with Bangalore IoT service company. When they decided to establish an industrial IoT startup to meet the emerging needs of the thermal power industry, they realized that they needed insiders with deep domain expertise.

Therefore, they hired Boben Anto, a veteran in the thermal power industry, as the third co-founder. Anto carried out the conceptual design and commissioning of the NTPC Ltd's power plant, where he worked for 22 years. He later worked for the German power company Steag for 11 years, of which the last four years served as the executive director of India.

The startup has recently bid for companies such as GE, Siemens and Schneider Electric, and has signed contracts with Indian electric utility OEMs.

“We were able to defeat all these global players technically and commercially in an open tender. I cannot overemphasize the importance of knowing the area in which you are trying to solve the problem,” Raghunathan said.

New solution

The verification also came from ExactSpace ABB’s Synerleap accelerator program selected for the Swedish-Swiss multinational company, which bridges the gap between corporate innovation and start-ups.

"We have identified several areas where ExactSpace's AI technology can help us accelerate ABB's digital vision. Their technology provides us with new vacancies for solutions, some of which have been installed on customer sites," ABB Vice President and Synerleap Business Said Martin Olausson, development supervisor.

The partnership with ABB opens up possibilities for ExactSpace's platform, making it useful in other areas adjacent to thermal power generation, such as cement or steel. But it also wants to stay true to its original method.

"The value we provide to our customers remains unchanged. We will provide solutions for specific areas through the platform. We will not release a platform and say "you create value"," Raghunathan said.Editor/Luo Wentao