期刊文献+
共找到3篇文章
< 1 >
每页显示 20 50 100
基于图像的深度学习降雨强度估计方法 被引量:2
1
作者 尹航 郑飞飞 +2 位作者 段焕丰 Dragan Savic Zoran Kapelan 《Engineering》 SCIE EI CAS CSCD 2023年第2期162-174,M0007,共14页
城市洪水是世界性的重大问题,造成巨大的经济损失,严重威胁公共安全。减轻其影响的一种有希望的方法是开发实时洪水风险管理系统;然而,由于缺乏高时空降雨数据,构建这样一个系统通常具有挑战性。虽然一些方法(即地面降雨站或雷达和卫星... 城市洪水是世界性的重大问题,造成巨大的经济损失,严重威胁公共安全。减轻其影响的一种有希望的方法是开发实时洪水风险管理系统;然而,由于缺乏高时空降雨数据,构建这样一个系统通常具有挑战性。虽然一些方法(即地面降雨站或雷达和卫星技术)可用于测量和(或)预测降雨强度,但使用这些方法很难获得具有理想时空分辨率的准确降雨数据。本文提出了一种基于图像的深度学习模型来估计具有高时空分辨率的城市降雨强度。进一步来说,一种称为基于图像的降雨卷积神经网络(image-based rainfall CNN,irCNN)模型是使用从现有密集传感器(即智能手机或交通摄像头)收集的降雨图像及其相应的测量降雨强度值开发的。随后使用经过训练的irCNN模型根据传感器的降雨图像有效地估计降雨强度。分别利用合成降雨数据和真实降雨图像来探索irCNN在理论和实际模拟降雨强度方面的准确性。结果表明,irCNN模型提供的降雨量估计值的平均绝对百分比误差在13.5%~21.9%之间,超过了文献中其他最先进的建模技术的性能。更重要的是,所提出的irCNN的主要特点是它在有效获取高时空城市降雨数据方面成本较低。irCNN模型为估算城市降雨强度提供了一种有前景的替代方案,可以极大地促进城市实时洪水风险管理的发展。 展开更多
关键词 卷积神经网络 时空分辨率 降雨强度 深度学习 CNN 替代方案 公共安全 智能手机
在线阅读 下载PDF
Spanning tree-based algorithm for hydraulic simulation of large-scale water supply networks 被引量:1
2
作者 huan-feng duan Guo-ping YU 《Water Science and Engineering》 EI CAS 2010年第1期23-35,共13页
With the purpose of making calculation more efficient in practical hydraulic simulations, an improved algorithm was proposed and was applied in the practical water distribution field. This methodology was developed by... With the purpose of making calculation more efficient in practical hydraulic simulations, an improved algorithm was proposed and was applied in the practical water distribution field. This methodology was developed by expanding the traditional loop-equation theory through utilization of the advantages of the graph theory in efficiency. The utilization of the spanning tree technique from graph theory makes the proposed algorithm efficient in calculation and simple to use for computer coding. The algorithms for topological generation and practical implementations are presented in detail in this paper. Through the application to a practical urban system, the consumption of the CPU time and computation memory were decreased while the accuracy was greatly enhanced compared with the present existing methods. 展开更多
关键词 large-scale networks hydraulic simulation graph theory fundamental loop spanning tree EFFICIENCY
在线阅读 下载PDF
Large eddy simulation of plunging solitary wave:Understanding the breaking and turbulent mechanisms along shoaling region
3
作者 Abbasali Rahmani Alireza Keramat +1 位作者 Jinghua Wang huan-feng duan 《Journal of Ocean Engineering and Science》 2025年第3期366-378,共13页
A large eddy simulation(LES)is conducted to investigate the distribution of turbulence kinetic energy(TKE)under a plunging solitary wave over a 1:15 slope.This study provides a novel contribution to the field by exami... A large eddy simulation(LES)is conducted to investigate the distribution of turbulence kinetic energy(TKE)under a plunging solitary wave over a 1:15 slope.This study provides a novel contribution to the field by examining the roles of resolved and sub-grid scale TKE in plunging solitary waves at the different stages of wave breaking.Furthermore,comparing the performances of two sub-grid scale(SGS)models in simulating the distribution of TKE was carried out to identify their performances.The separate investigation of these components in the context of wave breaking and recognizing the importance of an appropriate sub-grid scale model to consider the effects of small-scale eddies provide a significant advancement in understanding coastal morphological changes and nearshore sediment transport.Both the zero-equation and one-equation SGS models demonstrated acceptable performance in simulating water surface and kinematic properties.The one-equation SGS model,however,provided more accurate results on TKE transport during the breaking process and as the wave approaches its collapsing point.The study’s results reveal that an SGS model’s inability to simulate TKE transport(such as in the zero equation model)leads to inaccurate simulations of the TKE level and breaking location in the breaking zone.Additionally,the results of the one-equation model demonstrated that the maximum horizontal fluid velocity around the wavefront surface is a better predictor of breaking wave onset than the horizontal fluid velocity at the wave crest. 展开更多
关键词 Plunging solitary wave Breaking wave vortices Large eddy simulation Sub-grid scale models Turbulent kinetic energy HYDRODYNAMICS Shoaling zone
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部