Forecasting Flash Floods with Optimized Adaptive Neuro-Fuzzy Inference System and Internet of Things

Over the past few decades, global warming and climate change have resulted in unpredictable floods in many regions of the world, which has the potential to cause a wide range of catastrophes. The primary objective of this work is to design a system for early flood prediction using soft computing and the Internet of Things (IoT). Predicting heavy rainfall with extreme precision is critical for saving people from flooding and minimizing property damage. There are numerous methods for predicting rainfall available today, but all of them are worthless due to drastic climate change. This study proposes an hybridized adaptive neuro-fuzzy inference system to reduce the mistakes in rainfall forecasts caused by climate change. ANFIS has been hybridized by fire fly algorithm. Weather big data was collected from the Chennai metrological region from 2010 to 2020 and analyzed using an upgraded adaptive neuro-fuzzy inference system. Additionally, IoT technology is being used to automate flood alarms and monitor flood parameters regularly. Finally, the proposed method is implemented experimentally to demonstrate the proposed early flood prediction model's accuracy.