An IoT-based real-time sensing model for rainfall-runoff using big data analytics
T. Saravanan, S. N. Deepa, P Sasikumar, Mohamed Saber Simum, Haider Sharit
Abstract
Millions of dollars’ worth of infrastructure has been destroyed as a result of floods' destructive effects in the past. It is a serious issue that leads to crop degradation, population decline, loss of infrastructures, and the deconstruction of many public utilities. An efficient strategy to handle this is to warn the neighborhood of an impending flood and give them plenty of time to leave and safeguard their property. There is still no global, universal system that can gather, store, evaluate huge information and produce flood prediction results despite extensive research. Using the convergence of big data and the Exponential Smoothing Technique (EST), forecasting architecture and a social collaborative IOT-based smart flood monitoring are suggested in this research. Based on flood events tracked by rainfall sensors in the catchment, we implemented an EST-based IoT network model to simulate the rainfall-runoff process. The findings demonstrate that, as opposed to physical-based and conceptual models, the suggested network is more appropriate for rainfall-runoff models. Data is retrieved from the open-source Kaggle meteorological database, and experimental evaluation is done. The results demonstrated the viability of the 23 suggested architectural designs.