Publications

Abstract : Landslides in India pose a significant threat due to the large areas prone to them. An early warning system based on rainfall data collected by rain gauges is essential to avoid such catastrophic events. However, during the rainy season, solar power that typically powers these gauges in remote areas can be scarce, leading to system shutdowns. To solve this issue, we propose a system that combines wind, hydro, and solar power sources, ensuring a steady power supply throughout the year. The system uses rain gauge sensors to collect rainfall data, which is then analyzed using edge analytics and machine learning algorithms to detect potential landslides. Early detection is critical as it allows authorities to take necessary evacuation measures, potentially saving numerous lives. Our paper provides a detailed analysis of the hybrid power system’s architecture, which utilizes a combination of energy sources to ensure a continuous power supply. Additionally, we outline the algorithm used to collect data from the rain gauge and the machine learning model employed to process the information for early detection of landslides. Combining multiple energy sources allows the system to maintain a consistent power supply, enabling the rain gauge sensors to collect critical rainfall data for analysis. The advanced analytics and machine learning algorithms then process the data for early detection of potential landslides.