Optimal Selection Of Typical LID Measures For Urban Waterlogging Control

Urban flooding and waterlogging have become a more serious issue due to rapid urbanization.And it is urgent to be managed and alleviated by Low Impact Development(LID)management.The study of optimal selection of typical LID measures for urban waterlogging control is carried out in Tianjin,North China.The effects of different design parameters and rainfall conditions on LID measures were studied based on the LID experiments.A Storm Water Management Model(SWMM)in the study area was established and the performance of runoff reduction of LID measures under different rainfall conditions is studied.Then the optimal selection suggestions of different research scales were given.A multi-objective optimization model was established by coupling SWMM and Non-dominated Sorting Genetic Algorithms(NSGA-Ⅱ).The model was used to carry out the synchronous optimization of the size and location of LID measures.The theoretical support and scientific basis for scheme optimization are provided.The main contents and results are as follows.(1)The LID experiments of single measures such as sunken greenbelt,vegetative swale,bio-retention cell and permeable pavement were carried out under various rainfall conditions.The results show that: The rate of runoff reduction of the sunken greenbelt decreases exponentially with the increase of the return period and increases with the increase of the sunken depth.The maximum value of it is 100% and the minimum value is 27.6%.The larger the longitudinal slope of the vegetative swale,the faster the runoff transport,but the smaller the rate of runoff reduction.The runoff reduction rate of vegetative swale decreased about 60% to about 20% with the increase of the return period.Bio-retention cell has a good runoff reduction rate which is about70%.The reduction rate is 100% in the return period of 1a.Fillers with larger porosity have a better reduction effect.Permeable brick is more suitable for runoff reduction than grass-planting brick.Besides,adding drainage pipe can further improve the effect.(2)The impact of rainfall characteristics on runoff reduction effects of single LID measures are analyzed by simulation of SWMM.The results show that: the sunken greenbelt is the most suitable for reducing the flood peak since it has the best peak-cut effect.The runoff reduction rate of it is 12.9%.The vegetative swale runoff reduction effect is 3.7%～5.6% and mainly used for runoff transport.Bio-retention cell has a better runoff reduction rate of 18.1%,but it is easily limited by the water storage capacity and is sensitive to the change of the rain peak coefficient.The permeable pavement has the best runoff reduction rate of 21.8% in the case of long-term rainfall.In the case of shortterm heavy rainfall,the runoff reduction effect of it is bad due to the infiltration-excess runoff production.(3)The surface runoff coefficient,the surcharge time of pipe network and the total investment cost are taken as the objective functions.Then a multi-objective optimization model is established by coupling SWMM and NSGA-Ⅱ,which avoids the complicated calculation and the manual subjective determination of the scheme planning.A decision space can be obtained for the planners to select schemes according to the requirements.The model is used in the study area and the optimal results show that: Compared with traditional development,the optimization designs reduce the surface runoff coefficient from 0.7 to about 0.5,the conduits surcharge time from 1.62 hours to 0.04～0.47 hours,and the total investment costs are only between 395,000 ￥ ～872,000 ￥.Compared with the results of single LID designs or LID combined designs,the optimization designs do synchronous optimization on three objectives.The objectives of the surface runoff coefficient and the conduits surcharge time need to be balanced according to the evaluation of results.