NIT Rourkela Develops AI System For Autonomous Solar Cleaning
POWER & RENEWABLE ENERGY

NIT Rourkela Develops AI System For Autonomous Solar Cleaning

Researchers at the National Institute of Technology Rourkela have developed an Artificial Intelligence (AI) powered autonomous system to monitor and clean solar panels only when required, a development designed to improve the efficiency of solar power plants while reducing maintenance costs and water consumption. The innovation, titled Federated Learning (FL) based Autonomous System and Method for Monitoring and Cleaning Solar Plant, has been granted an Indian patent and was developed by faculty and research graduates in the Department of Computer Science and Engineering.

The system employs Federated Learning to enable intelligent monitoring while preserving data privacy by sharing encrypted model updates rather than raw operational data, thereby reducing cybersecurity risk and bandwidth usage. It integrates edge computing, predictive maintenance and autonomous fault detection to optimise operations and to identify panels that require selective cleaning. The technology aims to address efficiency losses caused by dust, bird droppings and industrial pollutants, which researchers say can reduce panel output by up to 40 per cent in arid regions.

Existing cleaning regimes are commonly labour intensive, water intensive and scheduled irrespective of panel condition, whereas the new system detects dirt accumulation and orchestrates selective cleaning only where necessary, which is expected to conserve resources. The approach has been validated through simulations at Technology Readiness Level (TRL) three and the team plans to develop a hardware prototype with Internet of Things (IoT) sensors for pilot deployments with government agencies and industry partners.

Researchers estimate that, once commercialised, the system could offer advanced AI driven capabilities at nearly 10 per cent of the cost of comparable solutions, which would support India’s National Solar Mission and net zero aims by enabling smarter maintenance of solar infrastructure. The technology is intended for deployment across utility scale parks, floating and rooftop photovoltaic installations, industrial facilities, smart city infrastructure and remote off grid renewable energy systems.

Researchers at the National Institute of Technology Rourkela have developed an Artificial Intelligence (AI) powered autonomous system to monitor and clean solar panels only when required, a development designed to improve the efficiency of solar power plants while reducing maintenance costs and water consumption. The innovation, titled Federated Learning (FL) based Autonomous System and Method for Monitoring and Cleaning Solar Plant, has been granted an Indian patent and was developed by faculty and research graduates in the Department of Computer Science and Engineering. The system employs Federated Learning to enable intelligent monitoring while preserving data privacy by sharing encrypted model updates rather than raw operational data, thereby reducing cybersecurity risk and bandwidth usage. It integrates edge computing, predictive maintenance and autonomous fault detection to optimise operations and to identify panels that require selective cleaning. The technology aims to address efficiency losses caused by dust, bird droppings and industrial pollutants, which researchers say can reduce panel output by up to 40 per cent in arid regions. Existing cleaning regimes are commonly labour intensive, water intensive and scheduled irrespective of panel condition, whereas the new system detects dirt accumulation and orchestrates selective cleaning only where necessary, which is expected to conserve resources. The approach has been validated through simulations at Technology Readiness Level (TRL) three and the team plans to develop a hardware prototype with Internet of Things (IoT) sensors for pilot deployments with government agencies and industry partners. Researchers estimate that, once commercialised, the system could offer advanced AI driven capabilities at nearly 10 per cent of the cost of comparable solutions, which would support India’s National Solar Mission and net zero aims by enabling smarter maintenance of solar infrastructure. The technology is intended for deployment across utility scale parks, floating and rooftop photovoltaic installations, industrial facilities, smart city infrastructure and remote off grid renewable energy systems.

Related Stories

Gold Stories

Next Story
Infrastructure Energy

Asian Energy Services Q1 FY27 PAT Rises 129 Per Cent

Asian Energy Services Limited reported a 129 per cent year-on-year rise in net profit to Rs 128 million for Q1 FY27, compared with the corresponding quarter last year.Revenue increased 135 per cent year-on-year to Rs 2.71 billion, supported by continued momentum across its services business, disciplined execution and contributions from domestic and international operations. EBITDA grew 81 per cent year-on-year during the quarter.As of June 30, 2026, the company’s standalone order book stood at Rs 17.54 billion, with around 60 per cent coming from oil and gas services and 40 per cent from min..

Next Story
Infrastructure Urban

BioBTX to Build First Commercial-Scale Circular Chemicals Plant

Dutch circular chemistry technology developer BioBTX is building what it says will be the world’s first commercial-scale plant to convert mixed plastic waste into high-quality aromatic chemicals using its proprietary Integrated Catalytic Cracking Process (ICCP) technology.The facility will be built at Chemical Park Delfzijl on the northern coast of the Netherlands and is expected to create 35 jobs. Covestro, which has been a shareholder and strategic partner of BioBTX since 2024, holds a mid-single-digit million-euro investment in the company.BioBTX’s ICCP technology uses catalytic pyrolys..

Next Story
Real Estate

Awfis Q1 FY27 PAT Jumps 140% as Revenue Rises 27%

Awfis Space Solutions reported a 140 per cent year-on-year rise in consolidated profit after tax (PAT) to Rs 240 million for Q1 FY27, compared with Rs 100 million in the corresponding quarter last year.Revenue from operations increased 27 per cent to Rs 4.25 billion from Rs 3.35 billion, while EBITDA rose 28 per cent to Rs 1.62 billion. EBITDA margin improved to 38.2 per cent from 37.8 per cent. Profit before tax increased 135 per cent to Rs 240 million.The company's co-working business recorded 27 per cent year-on-year growth, supported by demand from enterprises, Global Capability Centres (G..

Advertisement

Subscribe to Our Newsletter

Get daily newsletters around different themes from Construction world.

STAY CONNECTED

Advertisement