Novel Smart Rainfall Prediction System

Due to climate change, extreme weather events such as heavy rainfall are becoming a common occurrence. In a tropical setting, rain can be highly localized and changes quickly. Rainfall tends to appear and disappear quickly within an hour causing flash floods in urban areas. This technology relates to a novel smart rainfall prediction system utilizing data from multiple X- band rainfall monitoring radars. The X- band radar scans the sky for rain, feeding data into the system every 2 minutes to create an accurate map of the real-time rainfall intensity though the use of a nowcast model. This allows the relevant municipal authorities to improve on existing flood management measures and react quickly in case of flash floods. The technology owner is currently looking to license the technology to logistic companies and city governments.

• Compact, cost-effective and energy-efficient X-band monitoring radars in multiple locations
• Best for monitoring rainfall within a range of 30km to 50 km
• Machine Learning techniques applied to improve the accuracy of rainfall observations
• Rainfall observations are input for a nowcast (localized rainfall forecast of up to 2 hours) model that predicts flash flood
• Feeds data into a nowcast model that uses the data and past observations to forecast the movement, growth, and decay of rainclouds
• Allows rainfall intensity forecasts to be generated 30 minutes in advance with an accuracy of 65%
• Predict movement, growth, and decay of rainclouds, enhanced with Deep Learning techniques to better forecast heavy rain events 30 minutes before they actually occur

• Urban areas to predict rainfall and impose necessary flood prevention measures 

• System that produces forecasts 30 minutes ahead of a rainfall event with 65% accuracy 
• Overlapping fields of radar signal coverage makes data more robust (i.e. blockages caused by buildings can be overcome)
• Improves flood prevention measures in urban areas