Intellectual Property India Publishes Patent Application for 'Iot-Integrated Intelligent System For Solar Panel Condition Monitoring Using Deep Learning And Multimodal Data Analysis' Filed by Mr. Jonnalagadda Chetan Vikas; Mr. Divyansh Goswami; Mr. Katukuri Charanjeet Reddy; and Dr. J. Shobana

MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641048395 A) filed by Mr. Jonnalagadda Chetan Vikas; Mr. Divyansh Goswami; Mr. Katukuri Charanjeet Reddy; and Dr. J. Shobana, Chennai, Tamil Nadu, on April 16, for 'iot-integrated intelligent system for solar panel condition monitoring using deep learning and multimodal data analysis.'

Inventor(s) include Mr. Jonnalagadda Chetan Vikas; Mr. Divyansh Goswami; Mr. Katukuri Charanjeet Reddy; and Dr. J. Shobana.

The application for the patent was published on May 1, under issue no. 18/2026.

According to the abstract released by the Intellectual Property India: "The present invention discloses an IoT-integrated intelligent system for automated monitoring and classification of solar panel conditions using deep learning and multimodal data analysis. The system combines visual data acquired from cameras or drones with real-time environmental and operational data collected through IoT sensors, including temperature, dust, light intensity, and voltage/current sensors. The objective is to enhance the efficiency, reliability, and automation of solar panel inspection and maintenance. The invention employs advanced deep learning models based on transfer learning architectures such as MobileNetV2, ResNet50, EfficientNetB0, and DenseNet121 to classify solar panel conditions into categories including clean, dusty, snow-covered, and physically damaged. To improve prediction accuracy, the system incorporates techniques such as test-time augmentation, class balancing, and heuristic refinement methods, including HSV color analysis and edge detection. For video inputs, the system performs frame extraction and applies temporal smoothing and majority voting to ensure stable and consistent predictions. A multimodal data fusion approach integrates sensor data with visual analysis to provide more accurate and context-aware decision-making. The system is deployed through a web-based interface that enables real-time monitoring, automated report generation, and maintenance recommendations. The proposed invention offers a scalable and efficient solution for solar panel condition monitoring, reducing manual intervention, minimizing operational costs, and improving overall energy generation efficiency."

Disclaimer: Curated by HT Syndication.