Research On Optimal Scheduling And Leakage Location Model Of Urban Water Supply Network Based On Genetic Algorithm

In order to meet the city water demand,the load of urban water supply network increases,the old pipe network leakage is serious,and it poses a serious threat to the city operation.It is difficult to meet the demand by traditional manually monitoring the state of the water supply network.At present,many cities have established dynamic hydraulic models to achieve the safe operation of water supply networks,and provide a basis for intelligent management of water supply networks.This paper established a dynamic hydraulic model,hydraulic optimization scheduling model and leakage positioning model for Chaohu City water supply network,and the water supply network information system have been optimized and improved in order to reduce leakage and energy consumption,and assure that economic and social benefits of water supply enterprises being improved.Firstly,combined with the current situation of Chaohu City water supply network,the existing network situation will be briefly introduced and analyzed.Information such as water pump,pool,pressure measuring points and pipe network survey map will be collected,and SCADA data will be combined to simplify and verify the pipe network information.The pipe structure is sorted and debugged.The water quantity and water loss of the users are distributed;the accuracy of the hydraulic model is checked to make sure that the pipe network system being accurately simulated and providing reliable support for subsequent optimization models.Secondly,the hydraulic optimization scheduling method of water supply network is proposed.Taking the actual pipe network as an example,the fitness function is designed according to the actual situation,and the risk of the model falling into the local optimal solution is reduced.Then the hydraulic optimization scheduling model based on genetic algorithm is established,and a reasonable scheduling scenario is proposed.The results show that using genetic algorithm to establish optimal scheduling model can reduce the total energy costs.In the optimization process,the solution of the population can quickly converge to the optimal one and all the pumps' start-stop scheme within one day can be obtained.The optimal scheduling model based on genetic algorithm is more energy-saving than traditional experience scheduling.Many pumping stations in China have adopted the pump combination with variable speed pump as part of their energy saving and consumption reduction measures.Combined with the domestic development situation,the optimal scheduling model of the pump with variable speed pump is proposed.This kind of scheduling method needs to simultaneously encode the speed of the fixed speed pump and the speed of the variable speed pump,and solve it by genetic algorithm.Since the fixed speed pump has only two operating states,open and closed,it can only be encoded by binary.For the speed ratio,the binary code will make the chromosome length too long.This paper proposes an optimal scheduling model with variable speed pump,but the optimization scheme with variable speed pump is not stable enough.Therefore,the optimal scheduling of variable speed pump is worthy of further discussion.Furthermore,the main influencing factors of water supply pipe network leakage are analyzed.The leakage of the pipe network is generally related to the node pressure of the pipe section.It can be simulated as a kind of injection flow,and the expressions related to pressure and leakage are proposed.In general,the pressure sensitivity index can be taken as 0.5.On this basis,the pressure-related leakage model is established,and the possible leakage nodes are searched through optimization calculation.Since the leakage can be expressed as the flow outflow at the point,the node with the injection coefficient greater than zero is regarded as the leakage or considering the pipe connected to this node being likely to leak.Aiming at the problem of the small number of monitoring points in urban water supply pipe network in China,according to the influence of the data of the flow and pressure monitoring points on the accuracy of the model,the pressure-related leakage positioning model is evaluated,and the conclusion that the flow monitoring points are more sensitive to the model positioning result compared with pressure monitoring points is obtained.In addition,the model is also applicable to medium and large water supply network systems,because the dimension of the model optimization search is determined by the maximum number of leaky nodes rather than the total number of pipe network nodes.The model can artificially set the maximum number of leaks.Therefore,the number of decision variables is much smaller than the summary points of the model,which can effectively avoid the occurrence of “dimension disaster”.The leakage location and optimization scheduling algorithms proposed in this paper are all written by C#.The EPANET hydraulic calculation engine and AForge function library under.NET platform are called.Combined with the pipe network case,the realization process of the algorithm and the feasibility,advantages and disadvantages of the algorithm are described.