Obtain data on previously difficult locations
Big data and AI activities are expected in various fields, and the same is true for flood damage prediction. However, data itself is currently lacking. This is because there are no power or communication lines in sewage systems that show signs of flooding, or in mountain rivers that affect downstream areas and show signs of landslides, where solar power generation and battery replacement are difficult.
HydroVenus can be realized in any size, but its maintenance-free characteristics, which are unaffected by drifting debris, and its design flexibility, which allows it to operate in shallow or weakly flowing areas, make it possible to acquire energy from various parts of the river for sensing and wireless communication. No installation work is required, and a sensing network can be established inexpensively simply by mooring the small device.
From real-time hazard maps to sluice gate control
By collecting measurement data and precipitation data from various river locations, as well as control data from sluice gates and dams, in the cloud, and then using AI to learn over time, it will be possible to construct region-specific learning models and make predictions. It is expected to be able to predict future flooding risks based on weather forecasts and sluice gate conditions, as well as to provide navigation of sluices and dams to minimize damage.
The control of reticulated channels in the plains, including sluice gate control, and the prediction of the speed of rising water from mountainous areas downstream, which have been difficult to achieve so far, are areas where physical simulation is difficult, and can only be realized through multi-point data acquisition and AI.
