基于ANN的地下排水管线系统知识深度学习与城市内涝水位模拟

Knowledge Deep Learning of Underground Drainage Pipeline System and Urban Waterlogging Water Level Simulation Based on ANN

城市地下排水系统水力建模技术能够很好地模拟城市内涝的演进状态,但这是一个持续性的过程,需要长时期的数据采集、处理、以及水力建模,且研究和模拟的范围受时间空间限制。人工智能技术的发展,为架构在城市排水地理信息系统基础上的城市内涝模拟提供了增强效率的可能。本文就使用人工神经网络(Artificial Neural Network,ANN)技术对地下排水管线知识进行学习,模拟针对2013年7月18日昆明城市多点内涝状态下河道的水位,实现人工智能技术在城市内涝研究中的应用,以寻找一种高效模拟城市地下排水系统内涝灾害的方法。

Hydraulic modeling technology of urban underground drainage system can well simulate the evolution of urban waterlogging,but it is a continuous process,requiring long-term data acquisition,processing,and hydraulic modeling,and the scope of research and simulation is limited by time and space.With the development of artificial intelligence technology,it is possible to enhance the efficiency of urban waterlogging simulation based on Urban Drainage Geographic Information system.In this paper,the artificial neural network(ANN)technology is used to learn the knowledge of underground drainage pipelines,simulate the water level of the river under the condition of multi-point waterlogging in Kunming city on July 18,2013,and realize the application of artificial intelligence technology in the study of urban waterlogging,in order to find an efficient method to simulate the waterlogging disaster of urban underground drainage system.