Page 47 - 2022年第53卷第3期
P. 47

Intelligent rapid prediction method of urban flooding based on BP neural network and
                                               numerical simulation model


                                1,2          1,2                1,2            1,2       1,2        1,2
                   LIU Yuanyuan   ,LIU Yesen    ,ZHENG Jingwei    ,CHAI Fuxin    ,LI Min   ,MU Jie
                              (1. China Institute of Water Resources and Hydropower Research,Beijing  100038,China;
                      2. Research Center on Flood and Drought Disaster Reduction of the Ministry of Water Resources,Beijing  100038,China)


                   Abstract: Numerical simulation model is the main technical method to analyze and simulate urban flood⁃
                   ing, forecast and early warning. However, the computational efficiency of numerical mode cannot meet the
                   needs of urban rainstorm risks. How to introduce artificial intelligence technology into the rapid prediction
                   of urban flooding with less training samples and labeled data is a key concern and urgent problem to be
                   solved. To solve this problem,this paper proposes a new method for rapid prediction of urban flooding by
                   combining the numerical model with good calculation accuracy and the artificial neural network model with
                   high computational efficiency. Based on the simulation results of numerical model as training samples, the
                   neural network prediction model is constructed to simulate the process of urban flooding. The results show
                   that this method has high prediction accuracy and fast calculation speed,which can meet the needs of dai⁃
                   ly flood control and emergency response, and provides a new idea for the application of artificial intelli⁃
                   gence technology in flood control and disaster reduction.
                   Keywords: artificial intelligence technology; BP neural network; numerical simulation; urban waterlogging;
                   rapid prediction
                                                                                   (责任编辑:李福田)
                (上接第 283 页)
                         Study on regional climate effect under water diversion in Northwest China


                                  1,2            3,4             1,2             1,2            1,2
                       ZHAO Yong    ,HUANG Ya      ,WANG Hejia     ,XIAO Weihua    ,WANG Hao
                    (1. Department of Water Resources,China Institute of Water Resources and Hydropower Research,Beijing  100038,China;
                         2. State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin,Beijing  100038,China;
                                    3. College of Oceanography,Hohai University,Nanjing  210098,China;
                        4. China Water Resources Pearl River Planning,Surveying & Designing Co.,Ltd.,Guangzhou  510610,China)


                   Abstract: Water diversion project is a strategic measure to optimize the allocation of water resources. Chi⁃
                   na has been studying water diversion project in the west since 1950s. However,the introduction of regional
                   off-transfer water will inevitably break the original water and energy balance of the water demand area,
                   change the process of water and energy exchange between local atmosphere and land surface,and then af⁃
                   fect the local climate. Therefore, it is necessary to carry out water diversion of regional climate effect re⁃
                   search. Using the output of MPI-ESM-MR global climate model in the CMIP5 as an initial and boundary
                   conditions of RegCM4,the influence of temperature,precipitation and energy balance processes from 2021—
                   2050 under two representative concentration pathways scenarios for water diversion irrigation in the north⁃
                   west region in the future are analyzed. The results show that diversion irrigation has an effect on the annu⁃
                   al temperature and precipitation, especially in summer. Specifically, evapotranspiration increased significant⁃
                                                               -2
                   ly in summer (latent heat flux increased by 35~45W·m ), and sensible heat flux decreased, resulting in
                   a decrease of nearly 2℃ in average surface temperature. The increase of summer precipitation is mainly in
                   the vicinity of the study area and the surrounding Tianshan,Qilian and Kunlun mountains,etc. The abnor⁃
                   mal change of low-altitude circulation and the enhancement of convective rising movement all provide dy⁃
                                                                                      -1
                   namic conditions for the change of summer precipitation (an increase of 0.5~1mm·day ),and increase the
                   convective precipitation potential of the water demand area and the surrounding alpine area. In addition,di⁃
                   version irrigation caused complex atmospheric circulation changes in this area, and the vertical gradient of
                   meteorological elements changed significantly at 600~925 hPa.
                   Keywords:water diversion;regional climate;Northwest China;RegCM4;representative concentration path⁃
                   ways (RCPs)
                                                                                   (责任编辑:李福田)
                                                                                               — 295  —
   42   43   44   45   46   47   48   49   50   51   52