摘要
通过智能物联网技术实时获取积水监测实况数据,利用天津市气象精细化格点预报产品和城市自动雨量观测站实况数据,以机器学习、神经网络模型和天津市城市内涝风险等级划分原理为基础,研究基于用户实时位置的城市内涝预报预警技术,研发天津市城市自动化积水监测预警系统。结果表明,该系统具备一定的城市内涝风险监测预警预报能力,并在2018—2020年多次重大天气过程中应用,积水深度预报结果与监测结果基本一致,应用数据表明验证结果良好,系统可以为政府防灾减灾决策、指挥调度提供精准、及时的气象数据支撑。
This paper used intelligent Internet of Things technology to obtain real-time monitoring live data of stagnant water,and obtained Tianjin’s fine-grained meteorological grid forecast products and live data from the city’s automatic rainfall observation stations.Based on the machine self-learning,neural network model and the classification principle of urban waterlogging risk classification in Tianjin,the paper studied the technology of urban waterlogging prediction and early warning based on the real-time location of users,and developed the automatic water logging monitoring and warning system in Tianjin.The results showed that the system had a certain capability of monitoring,early warning and forecasting of urban waterlogging risks.In the application of multiple major weather processes from 2018 to 2020,the results of the depth prediction of the accumulated water were basically consistent with the monitoring results,and the application data showed that the verification results were ideal.The system can provide accurate and timely meteorological data support for government disaster prevention and mitigation decision-making,command and scheduling.
作者
侯天宇
梁好
霍凯
赵敏
陈子煊
张春莉
苑超
Hou Tianyu;Liang Hao;Huo Kai;Zhao Min;Chen Zixuan;Zhang Chunli;Yuan Chao(Tianjin Public Emergency Warning Information Release Center,Tianjin 300000)
出处
《气象研究与应用》
2021年第1期85-89,共5页
Journal of Meteorological Research and Application
基金
安全天津、科技惠民与可持续发展实验区建设科技专项(17ZXCXSF00060)。
关键词
城市内涝
智能物联网
机器自主学习
神经网络模型
内涝预报预警
urban waterlogging
intelligent Internet of things
machine independent learning
neural network model
flood forecast and early warning
