Multi-condition Hydraulic Performance Optimization Design Of Seawater Desalination High Pressure-pump Based On NSGA-? Genetic Algorithm

Turbine-type energy recovery machine device integrates waste energy recovery and liquid pressurization.It is widely used in island development,ocean-going ships,drilling platforms,and other fields.It has the advantages of high efficiency,convenience,and energy saving.As the core equipment of the energy recovery machine,the high-pressure pump often runs under variable working conditions(variable flow rate and variable speed).However,the high-pressure pump has low efficiency at non-design operating conditions,a narrow high-efficiency zone,and unstable operation,which affects the overall performance of the integrated machine.Therefore,to obtain products that meet the design parameters and have the best performance,it is of great significance to optimize the design of the high-pressure pump.This article takes the high-pressure pump of the turbine-type energy recovery machine as the research object.Firstly,the maximum weighted efficiency of the three working conditions is taken as the optimization goal,and the Plackett-Burman screening test is used to determine the optimization variables.Secondly,based on the Isight multidisciplinary optimization platform,an intelligent CFD is built.The RBF neural network was used to fit the mapping relationship between the objective function and the optimization variables,and the NSGA-? genetic algorithm was used to optimize the design of the key geometric parameters of the high-pressure pump impeller.Finally,the external characteristics and internal flow field of the high-pressure pump before and after the optimization are optimized.in addition,a calculation model of the transition process of high-pressure pumps with variable speed is established based on the CFX-CEL language,and the transient parameters and internal flow field are numerically simulated and analyzed.The main research work and results of this paper are as follows:(1)Aiming at the pump unit of the high-pressure pump and energy recovery integrated machine of the daily production 1000t/a membrane seawater desalination system,the structured grid is used to conduct a preliminary hydraulic design of the high-pressure pump,and the grid independence verification is completed,and the integration of the high-pressure pump is combined.The machine test verifies the accuracy of the numerical simulation.(2)Through the Plackett-Burman screening experiment,select the inlet angle?₁,outlet angle?₂,blade outlet width b₂,and blade wrap angle?as optimization variables,and determine the variation range of each parameter,given geometric parameter constraints,and finally unified the multi-condition optimization form.Based on the multi-disciplinary optimization platform Isight,secondly,based on the multidisciplinary optimization platform Isight,By writing batch commands to integrate Cfturbo?ICEM?and CFX to build an intelligent hydraulic optimization platform to realize CFD automatic performance prediction of high-pressure pumps.(3)Using the optimal Latin hypercube experiment method,56 sets of experiments were designed for the 4 geometric parameters selected;based on the numerical simulation calculation results,the RBF artificial neural network was used to establish an approximate model between the geometric parameters and the target value.Finally,the NSGA-? genetic algorithm was used to optimize the RBF neural network with the maximum value.After 500 generations of iterations,the frontier distribution of the Pareto solution set of the optimization target is obtained and the individual with the best-weighted efficiency is selected as the final optimization plan.The external characteristic curve comparison of the impeller scheme before and after optimization shows that the weighted average efficiency of the three working conditions of the high-pressure pump has increased by 3.37%,and the efficiency has increased by 2.22%,3.59%,and 4.23 at 0.8Q_d,1.0Q_d and 1.2Q_d working conditions,respectively.(4)To deeply analyze the hydraulic performance of the high-pressure pump impeller scheme before and after the optimization,the intermediate section pressure,velocity streamline distribution,turbulent kinetic energy distribution,vortex core distribution inside the impeller,and blade load are compared between the initial impeller and the optimized impeller.The results show that after optimization,the pressure distribution in the high-pressure pump is more uniform,the velocity streamline distribution is more reasonable,the turbulent kinetic energy and the vortex core area are smaller,and the impeller is more functional,indicating that the overall performance of the optimized high-pressure pump is better than the initial impeller,and the hydraulic loss is smaller.(5)Aiming at the characteristics of the self-adaptive change of the speed of the integrated machine,based on the CFX-CEL language,the impeller speed change mode is defined,the calculation of the variable speed unconstant value of the high-pressure pump is completed,and the transient external characteristics and internal flow field of the high-pressure pump are analyzed..The results show that when the high-pressure pump changes from steady-state to unsteady state and unsteady state to steady-state during the transition process,the transient parameters of the high-pressure pump will fluctuate strongly.The design of the machine's high-speed rotor system provides a theoretical basis.