摘要
为了应对城市积涝问题,通过深度学习模型结合遥感图像获取城市道路网络数据,结合高效的人工智能技术与水力学方法进行城市雨洪模拟。结果显示,该模型对城市各路网结构的识别准确率超过95%。在模拟管道流量峰值、液位峰值及总水量等指标时,偏差分别仅1.30%、1.11%和6.78%。融合水力学与人工智能技术的城市雨洪模拟,在管道流量和液位预测方面展现出高精准度。
In order to deal with the problem of urban waterlogging,the deep learning model combined with remote sensing images was used to obtain the urban road network data,and the urban rainwater simulation was carried out by combining efficient artificial intelligence technology and hydraulic methods.The results show that the accuracy of the model for the identification of urban road network structures is more than 95%.When simulating the peak flow rate,liquid level peak and total water volume,the deviations are only 1.30%,1.11%,and 6.78%,respectively.Urban stormwater simulation,which combines hydraulics and artificial intelligence technologies,shows high accuracy in pipeline flow and liquid level prediction.
作者
何楚翘
马高磊
HE Chuqiao;MA Gaolei
出处
《智能城市》
2025年第6期131-133,共3页
Intelligent City
关键词
水力学
人工智能
城市雨洪模拟
深度学习模型
hydraulics
artificial intelligence
urban rainfall and flood simulation
deep learning models
作者简介
何楚翘,硕士,高级工程师,研究方向为人工智能、航天飞行控制器、自动化控制等。
