| Nowadays,the former extensive mode of economic development can no longer meet the need of environmental protection in modern society,which creates enormous pressure on the work of energy saving and emission reduction.As a result,the mixed flow pump,which is widely used in the national economy,has attracted great attention and the research on its high-efficiency hydraulic model has been one of the most important and basic research directions in the field of hydraulic machinery.In this thesis,a mixed flow pump with a specific speed of 336 is taken as the research object and the multi-parameter optimization method involving no change in impeller outlet diameter D2 is proposed and verified by the research method combining model test,numerical calculation and theoretical analysis.The main research contents and creative achievements are as below:1.Research method of the inner flow characteristics of vane pump and its optimization design method were systematically summarized,on the basis of which,the hydraulic optimization method of commonly used vane pump was investigated.Meanwhile,the concept and characteristics of Artificial Neural Network and the research status of the genetic algorithm were generally introduced.2.The performance including the external characteristics,pressure pulsation and vibration characteristics of the model pump was tested and after analyzing the performance of the model pump under different working conditions,some findings were concluded:(1)The pump head efficiency of the model does not meet the design requirements;(2)Compared with the design conditions,the pulsation of the non-design conditions is more turbulent,and the monitoring points have different degrees of oscillation in a certain frequency range(3)The vibration in the measurement point of impeller inlet flange T1 is mainly in the X direction,and the vibration in the Y and Z directions changes little with the increase of the working condition.The vibration law of the impeller export flange T2 in X,Y and Z directions is basically the same as that in the measurement point of T1 with the amplitude of vibration slightly larger than that of T1.3.The numerical simulation was carried out in different conditions of the model pump and then the results were compared with the experimental values.It shows that the calculation error of the head and the efficiency is less than 3%,which can meet the requirements of the CFD optimization design.This conclusion provides the basis for theoptimization in the next step,that is,the establishment of sample points and the analysis of the inner flow of the pump.4.A multi-parameter method is presented for the optimization performance of a mixed flow pump without changing the impeller outlet diameter D2:(1)utlet dip angle ?,wrap angle ?,inlet angle of the segment ?1,outlet angle of the segment ?2,width of impeller outlet l,dip angle of front shroud T1,dip angle of back shroud T2,number of blade z were selected as optimization variables;(2)The RBF neural network model is used to establish the performance prediction model of mixed flow pump with the training and testing samples determined by orthogonal test and CFD numerical method;(3)The multi-island genetic algorithm was applied to solve the prediction model to develop a multi-parameter optimization method for the performance of a mixed flow pump without changing the impeller outlet diameter D2.The results showed that the performance of the model was improved greatly.Not only the head but also the maximum efficiency was raised obviously,and the range of the high efficiency zone was also increased.5.The external characteristics,pressure pulsation and vibration characteristics of the optimized model pump were experimentally tested,and the internal flow change of the optimized model pump was compared based on the CFD results.The results showed that:(1)At design point,the pump head and the efficiency of the pump Compared with the unoptimization model were increased by 11.5% and 4.32 % respectively.(2)On the 1.0Qd condition,the amplitude of the pressure fluctuation at the inlet of the pump?outlet of the impeller?outlet of the guide vane?outlet of the pump were reduced by 62.9%?37.5%?28.57%?62.5%compared with that of the unoptimized one.(3)On the design condition,the vibration velocity amplitude of the optimized of the impeller export flange in X,Y,and Z direction was decreased by 12.2%?14.28% and 25% respectively compared to the model without optimization,vibration velocity amplitude and direction of the optimization of Z compared to before respectively decreased 4.71%,9.52%,35.29%.Similarly,the amplitude of vibration velocity of the direction of X,Y and Z in the optimized of the impeller import flange was decreased by 4.71%,9.52%,35.29%,respectively.After optimizing,the head loss in the guide vane was reduced by 1.23%.The pressure difference was reduced due to the increase of the pressure on the back of the impeller blade after optimizing,which effectively restrained the unstable flow caused by pressure difference. |
PDF Full Text Request