Framework Of Multi-objective Optimization And Comparison For Drainage Pumping Stations And Its Application

Urban flooding caused by urbanization has become a common problem faced by many countries and regions in the world,which has a serious impact on social life and economic production.Drainage pumping stations play an important role in urban drainage and flood prevention.Problems such as high energy consumption and serious pump wear in pumping station not only lead to great economic loss,but also not conducive to long-term operation.To ensure the operational economy and drainage capacity of pumping stations,multiple objectives need to be optimized.However,there is a lack of economic quantification of number of pump startups/shutoffs,and working hours of pumps.In addition,there are many multiobjective optimization algorithms,and different algorithms will have different impacts on the optimization results of a drainage pumping station.Due to the lack of comparison framework,how to choose a better optimization method is also facing challenges.To solve this issue,it is necessary to carry out the research on comparison framework of drainage pumping station,and carry out economic quantitative research on the objectives of number of pump startups/shutoffs and working hours of pumps,to provide theoretical support and scientific guidance for the quantification of related objectives of drainage pumping station and comparison and selection of optimization algorithm..Based on the Storm Water Management Model(SWMM)and MATLAB,this study provides a new comparison framework,which includes optimization module(different optimization algorithms)and comparison module(objectivity and subjectivity),and verifies the feasibility and reliability of this framework in different scale cases.The main results and conclusions are as follows:(1)From the perspective of operational economy and drainage capacity of a drainage pumping station,the four optimization objectives of number of pump startups/shutoffs(n),energy consumption of drainage pumping station(E),working hours of pumps(Th)and maximum reservoir/river depth(Hvmax)are integrated,and economic quantification formulas of number of pump startups/shutoffs(n)and working hours of pumps(Th)are proposed.These objectives can cover operational economics and drainage capacity of a pumping station,thus ensuring the practical significance of the proposed framework.(2)On the basis of optimization factors and objectives,two types of algorithms are proposed in the optimization module to ensure that the proposed framework is suitable for multi-objective optimization and can be popularized and applied:Type Ⅰ is the multi-objective algorithm containing subjectivity,that is,Particle Swarm Optimization algorithm(PSO)combines Linear Weighted Sum Method(LWSM),and obtains the optimal weight of each objective based on Analytic Hierarchy Process(AHP),called PSO-LWSM.Type Ⅱ includes multi-objective algorithms with objectivity,including Multi-objective Particle Swarm Optimization algorithm(MOPSO)combined with Technique for Order Preference by Similarity to an Ideal Solution(TOPSIS),called MOPSO-TOPSIS;Non-dominated sorting genetic algorithm-Ⅱ(NSGA-Ⅱ)combined with TOPSIS,called NSGA-Ⅱ-TOPSIS.(3)From the perspective of objectivity and subjectivity,the comparison module in the proposed framework contains two types of evaluation methods,namely:TOPSIS evaluation method,which represents objective evaluation,and E&C evaluation method,which represents subjective evaluation.Through these two comparative evaluation methods,it can not only find out the influence of human factors on the comparison of optimization results,but also determine the final optimization result more reasonably.(4)To explore the impact of case scales on optimization algorithms and optimization results,a drainage system as a first development area in Chizhou city(192 ha),and a Xiaohekou drainage system in Ma’anshan city(1320 ha)are taken as cases for verification.With the increase of drainage area,the convergence of the three optimization algorithms are the varying degrees of decline.From the optimization results,when the drainage area is small,MOPSO-TOPSIS is superior to PSO-LWSM and NSGA-Ⅱ-TOPSIS.When the drainage area is large,PSO-LWSM is superior to MOPSO-TOPSIS and NSGA-Ⅱ-TOPSIS.(5)To verify the reliability of the proposed comparison framework,the influences of design storms with different return periods and different expert scores in the AHP on optimization results were investigated.Under return periods of 5a,10a,30a,50a and 100a,the final optimization schemes obtained have good results.The results show that the generated optimization and comparison results are basically consistent with those of the previous schemes,which verify the reliability of the proposed framework.The weight of four objectives should be E>Th>Hvmax>n.