摘要
针对SWMM原始动态链接库缺乏相关应用接口函数和优化模块无法进行参数自动率定的问题,提出了一种基于PySWMM并耦合遗传算法的SWMM参数自动率定模型,并以重庆悦来新城为研究对象,选取36场独立降雨事件对SWMM进行校准和评估。结果表明,雨型特征对模型的模拟性能有较大影响;校准后的模型对不同雨型的降雨过程均有良好的适应能力,决定系数R~2达到了0.79以上,对发生频率较高的单峰靠前(Ⅰ型)降雨事件的模拟效果最好,其纳什效率系数(NSE)值达到0.90,峰值相对误差(PE)仅为-0.07。
The SWMM original dynamic link library is lack of relevant application programming interface functions,and the parameter optimization module is unable to perform automatic parameter calibration.Therefore,a model for automatic calibration of SWMM parameters based on PySWMM coupled with genetic algorithm was proposed,and it was calibrated and evaluated by 36 independent rainfall events in Yuelai New Town,Chongqing.Characteristics of rainfall type had great influence on the simulation performance of the model.The calibrated model had good adaptability to the rainfall process of different rainfall types,and the determination coefficient R~2 was more than 0.79.The best simulation performance was achieved for the rainfall event with high frequency and single peak forward(typeⅠ),the Nash-Sutcliffe efficiency coefficient(NSE)reached 0.90,and the peak relative error(PE)was only-0.07.
作者
王菲菲
卿晓霞
杨森雄
崔忠捷
WANG Fei-fei;QING Xiao-xia;YANG Sen-xiong;CUI Zhong-jie(College of Environment and Ecology,Chongqing University,Chongqing 400045,China;School of Civil Engineering,Chongqing University,Chongqing 400045,China;PowerChina Guiyang Engineering Corporation Limited,Guiyang 550000,China)
出处
《中国给水排水》
CAS
CSCD
北大核心
2022年第21期124-130,共7页
China Water & Wastewater
基金
国家重点研发计划项目(2017YFC0404704)
重庆市教委科学技术研究项目(KJZD-K202100104)
重庆市科委社会民生类重点研发项目(cstc2018jszx-zdyfxmX0010)。
