Optimal Operation Scheduling Of Urban Water Supply Pumping Station Based On Artificial Electric Field Algorithm

In the urban water supply system,the pumping station is not only the hub of the whole system,but also the main energy consumption unit.At present,most of the water supply pumping stations in China still adopt the operation mode of manual experience.The unreasonable operation mode not only leads to a large amount of energy waste,but also causes other problems such as pipe network leakage and pipe explosion.Therefore,in order to reduce the energy consumption of the pumping station and reduce the accidents of the pipeline network,it is very necessary to study the optimal operation of the pumping station.In this paper,around the objective of the optimal operation of pumping station,taking the water supply area of M city as the research object,the urban water consumption prediction,the hydraulic model analysis of water supply network and the optimal operation of pumping station are studied.The main work is as follows:1.The water consumption prediction model based on BP neural network is established,taking the water consumption data of the past 24 hours as the input variable and the water consumption of the next hour as the prediction target.In addition,considering that BP neural network is easy to fall into local extremum,strong randomness of initial weight and threshold and other defects,taking full advantage of the strong global search ability of artificial electric field algorithm(AEFA),introduces AEFA to optimize the initial weights and thresholds of BP network,and constructs AEFA-BP water consumption prediction model.Compared with the single BP neural network model,the prediction error of the AEFA-BP combination model is controlled within 3% in most periods,and the prediction accuracy is improved.2.The hydraulic model of water supply network can be divided into micro model and macro model.By analyzing the principle,applicable conditions and application scope of the two models,it can be seen that the macro model is simple in modeling,fast in calculation and more suitable for real-time optimal operation of water supply system.Therefore,a macro hydraulic model of the water supply network based on the BP neural network is established with the water supply pressure of the pumping station as the prediction target.The example proves that the prediction error in all periods is less than 4%,and the simulation results can provide data support for optimal scheduling.3.As the research objective of this paper,the optimization of pumping station is a mathematical model that takes the minimum energy consumption of the pumping unit as the objective function,and restricts the water supply capacity of the single pump,pump station water supply and pressure,and the pump speed ratio as constraints.In view of the fact that most of the researches on the optimal operation of pumping stations only consider the power of pumps,but ignore the influence of the loss of variable frequency drive and motor,the optimal operation model of pumping stations considering the loss of variable frequency drive and motor is established.According to the city’s water consumption and water supply pressure changes,a scheduling cycle(usually 24 hours)is divided into 6 periods.The mathematical model is solved using the AEFA algorithm with strong optimization ability,and the optimized scheduling scheme obtained at each time period can effectively reduce the energy consumption of the pump unit.