MatrixFlow, which provides an AI utilization platform that does not require programming, has partnered with Sanshin Denki Co., Ltd., an “electronics general trading company”, to provide local governments with “water level prediction” using electronic devices and AI utilization platform “MatrixFlow”. AI Solution” has been developed and will be available from December 2021.
By using MatrixFlow technology, it is now possible to “predict” river water levels, which was previously limited to “measurement”.
This development makes it possible to detect an increase in water level in advance and to take early measures against possible flood risks in the future.
In recent years, serious damage has occurred frequently due to floods caused by heavy rains and typhoons. In municipalities that mainly manage secondary rivers*1 or lower and directly issue evacuation orders to residents, there is a growing need to use not only weather forecasts but also water level forecasts to grasp flood risks at an earlier stage than ever before. I’m here.*1 The river law stipulates that rivers designated by prefectural governors pertaining to water systems other than first-class water systems that are important to the public interest (second-class water systems).
In response to the above issues, the Ministry of Land, Infrastructure, Transport and Tourism provides water level forecast information up to 6 hours ahead for 298 flood forecast rivers .
However, since there are 7,000 rivers nationwide, it is necessary to predict water levels six hours or more in advance, including rivers other than flood forecast rivers, for early detection of flood risk.
In addition, the Ministry of Land, Infrastructure, Transport and Tourism is promoting the use of new digital technologies in its Flood Damage Report 2020 , which calls for the promotion of digitalization, etc. to efficiently implement measures related to national resilience.
Sanshin Electric and MatrixFlow have jointly built a “water level prediction AI solution” through machine learning using real-time water level data sent from water level sensors, actual weather data such as past rainfall, and weather forecast data. .
This makes it possible to predict the water level of almost all rivers after one week, and to grasp the risk of flooding at an early stage. In the future, we will start providing it to local governments and companies.
In recent years, as the number of heavy rains and damage caused by water disasters are increasing, flood control works such as the development of drainage channels are being carried out at great expense. However, rivers and areas that can be treated are limited due to the high cost of flood control work. However, by introducing the “water level prediction AI” developed this time, it will be possible for local governments to deal with flood damage early at a lower cost than existing solutions, effectively suppressing human and economic damage caused by flood damage. You will be able to