Research On Green Infrastructure Spatial Allocation Optimization Based On Multi-objective Evolutionary Algorithm

In order to solve the problems of urban waterlogging under the background of rapid urbanization,the concept and practice of rain and flood management are gradually rising.The key technology is green infrastructure(GI).The study of spatial allocation optimization of GI can achieve the optimal cost-benefit,and provide a reference for rain and flood management in highly urbanized areas and the optimal benefit of GI in the planning and construction of sponge cities.In this paper,because of the lack of consideration of the complex relationship between different GI,a new type of spatial allocation generator for green infrastructure(SAGGI)is proposed.By coupling SAGGI with multi-objective evolutionary algorithms and SWMM,a spatial allocation optimization framework for GI considering the allocation proportion and connectivity of GI is constructed.Based on the Guangzhou Tianhe Smart City,this paper analyzes the influence of algorithms and allocation factors on spatial allocation optimization of GI and puts forward the key techniques of spatial allocation optimization of GI.The main research work of this paper is as follows:(1)Based on SWMM,the stromwater model of Guangzhou Tianhe Smart City is constructed,and the parameters of the model are calibrated and verified by the measured data.The results show that the SWMM of each drainage district has high precision.(2)SAGGI,which is based on the sub-catchment,manhole,and the allocation of GI of SWMM,can systematically generate all feasible GI allocation proportion and connection relation scheme library in the study area,according to the superior-subordinate relation of GI and sub-catchment distance.(3)The framework of GI spatial allocation optimization was constructed,which based on the scheme library generated by SAGGI.And it took the total cost of GI,runoff reduction rate and pollutant load reduction rate of outlet as objective functions,taking SPEA2,NSGA-II,and MOEA/D as optimization engine.The optimization calculation is carried out after coupling SWMM.Then the performance of the algorithm is evaluated by four indexes.The results show that the optimization performance of the SPEA2 is the best.The total cost of pervious pavement(PP)accounts for the main part of the total cost of GI when the combined reduction rate is small.And when the combined reduction rate is large,the total cost of the PP and green roof(GR)account for the main part of the total cost of GI.(4)Taking the SPEA2 as an example,three scenarios are set up to analyze the changing law of cost-benefit of each GI in the optimal spatial allocation of GI in each scenario,and the results of the three scenarios are compared to explore the influence of GI allocation ratio and connectivity on the optimal spatial allocation.The results showed that the cost-effectiveness of sunken lawn(SL)and PP were higher than that of bioretention cell(BC)and GR.The optimal spatial allocation obtained by the joint action of GI allocation proportion and connectivity is much better than that obtained by the action of a single allocation factor.When the total cost of the project is small,the GI connection system can be built mainly with SL connection and supplemented with PP connection.When the total cost of the project is large,the GI connection system can be constructed,in which the connection between PP and SL is the main,and the connection between BC and GR is the auxiliary.