Using IoT Data and AI-Based Flood Control
Discover how Alibaba Cloud assisted a city in Zhejiang province of China to combat floods using AI with ET Environment Brain.
Join the DZone community and get the full member experience.Join For Free
In August 2018, many regions in the northern hemisphere of China were hit by floods, which caused agricultural damage and affected the lives of millions of Chinese citizens. Desperate for a solution, Jinhua, a city in Zhejiang province of China, started tackling this issue by using Artificial Intelligence (AI) for flood control.
The flood and drought control command center of Jinhua city uses Alibaba Cloud ET Environment Brain to analyze water resources and weather data in the city, which helps to improve the decision-making efficiency during the flood season. According to statistical data, AI provides a much higher speed and accuracy than traditional methods in predicting extreme weather such as heavy rain and typhoons. Through these predictive analyses, cities can then proactively respond to adverse weather, saving precious time on decision-making and logistics for flood control and rescue missions.
As Jinhua is located in a basin-and-hill area in the central part of Zhejiang, this city is subject to small watershed mountain torrents due to the unique terrain. The city often encounters heavy rainfall and floods during the "plum rain" season from May to June every year. Whereas from July to September, Jinhua may suffer from droughts due to the shortage of rainfall or even unexpected storms brought by typhoons.
Zhu Shenghuo, chief scientist of decision intelligence in Alibaba machine intelligence lab, gives an introduction to the technology: "Based on data of Jinhua flood and drought control command center, the AI algorithm analyzes the real-time high water mark and weather in comparison with historical high water marks and weather in flood periods, to figure out the relations between high water marks of rivers and reservoirs. In addition, AI can predict the flood control conditions of rivers and reservoirs."
In the past, most flood predictions are made based purely on experience, historical data, and knowledge of professional flood control personnel. However, the growth of the Internet of Things (IoT) technology led to the explosive growth of hydrological and meteorological data. Traditional flood control models are unable to process the massive amounts of data because these methods are labor-intensive, slow, and have limited computing power. Modern-day AI systems such as ET Environment Brain help to address these challenges, allowing cities to make better-informed decisions through accurate predictions.
In the future, the flood control system of Jinhua can not only predict the high water marks and trends of floods, but also assess the economic and sociological effect brought about by the floods. This can help Jinhua to make more effective search and rescue plans and enable automatic dispatching of relief resources and evacuation plans.
Jinhua has started IT-based flood control since as early as 2002. The city has collected a large amount of data about water flow, rain, and disasters and has established multiple database systems including the real-time water flow database, meteorological database, and historical water flow database. All these lay a solid foundation for data analysis.
Published at DZone with permission of Leona Zhang. See the original article here.
Opinions expressed by DZone contributors are their own.