A numerical prediction method of cavitation-induced noise for a vane pump

Abstract

The application discloses a kind of blade pump cavitation induced noise numerical prediction method, the application provides a kind of blade pump cavitation induced noise numerical prediction method based on test, this method not only can be predicted the monopole noise generated by the cavity volume pulsation under the condition of blade pump cavitation, but also can predict the flow induced noise except monopole noise.The high-precision unsteady cavitation numerical calculation of the application can provide accurate sound source information for subsequent acoustic calculation, to improve the accuracy of acoustic numerical prediction, the prediction result is applied to the low-noise hydraulic design of blade pump, which can reduce the number of tests, shorten the development cycle, save development cost, thereby improve the efficiency of low-noise blade pump design;On the other hand, it can reveal the time-space evolution characteristics of vortex structure inside blade pump and the coupling mechanism of vortex and cavitation in the process of cavitation unsteady flow, to provide theoretical support for early-stage design or later-stage optimization of blade pump.

Description

TECHNICAL FIELD

[0001] The present application relates to a numerical prediction method, in particular to a blade pump cavitation-induced noise numerical prediction method which can predict monopole noise generated by cavity volume pulsation under blade pump cavitation conditions. BACKGROUND

[0002] The blade pump is widely used in various departments of the national economy, and plays an important role in the fields of fluid transportation such as national defense, aerospace, energy, ship, and household. It is a strategic equipment related to the people's livelihood and national security, and is equivalent to the human heart. Cavitation widely exists in the actual operation of the blade pump, especially in the blade pump of the military field such as ships and submarines. The collapse of cavitation and the change of cavity volume will release acoustic energy. With the continuous development of water weapons towards high precision and long distance striking characteristics, the blade pump cavitation-induced noise of the ship and submarine target vehicle greatly increases the probability of exposure and being hit, seriously affecting its survivability and combat effectiveness. Therefore, the noise problem induced by blade pump cavitation flow cannot be ignored. At present, the numerical simulation technology of flow-induced noise of blade pump under non-cavitation conditions has made great progress, but the research on blade pump cavitation flow-induced noise is mainly based on theoretical research and experimental research. The existing numerical simulation method of cavitation-induced noise does not consider the monopole sound source generated by the cavity volume pulsation, so the numerical calculation precision is not high. Therefore, it is of great academic value and engineering significance to study the mechanism of blade pump cavitation flow-induced noise. Therefore, it is an urgent practical problem to construct a blade pump cavitation flow-induced noise numerical prediction method. SUMMARY

[0003] In view of the above problems, the main purpose of the present application is to provide a blade pump cavitation-induced noise numerical prediction method which can predict the monopole noise generated by the cavity volume pulsation under the blade pump cavitation conditions.

[0004] The present application solves the above technical problems by the following technical scheme: a blade pump cavitation-induced noise numerical prediction method, which numerically calculates the blade pump cavitation-induced noise on the basis of experiments, solves the blade pump cavitation flow-induced noise based on large eddy simulation (LES-Large Eddy Simulation) and ZGB (Zwart-Gerber-Belamri foreigner's name) cavitation model. The blade pump cavitation-induced noise numerical prediction method comprises the following steps:

[0005] Step 1: Perform non-cavitation operating condition external characteristic test and cavitation operating condition external characteristic test of the model pump;

[0006] Step 2: Establish a three-dimensional model of the blade pump water body and perform structured grid division, perform non-cavitation steady, cavitation steady and cavitation unsteady numerical calculation, and compare with the test results;

[0007] Third step: based on the spherical cavity radiation theory, when the noise numerical calculation of the cavity volume pulsation is carried out, the last 10 rotation periods of the cavity volume are processed according to formula (1) based on the data processing software, so that the time domain sound pressure and the frequency domain sound pressure of the monopole noise generated by the cavity volume pulsation are obtained, and the expression of formula (1) is as follows:

[0008]

[0009] Fourth step: the flow field information in the computational fluid dynamics result is derived as the sound source information of the acoustic finite element calculation, the acoustic grid of the blade pump acoustic calculation domain is divided, and in order to ensure the accuracy of the acoustic finite element calculation, the maximum size of the acoustic grid should be less than 1 / 6 of the wavelength corresponding to the maximum calculation frequency, that is: L max c / 6f max , wherein L max represents the maximum size of the acoustic grid, c represents the sound speed, f max is the maximum frequency of acoustic calculation, the acoustic analysis model is created by using the method of calculating acoustics, the boundary conditions are set, the body sound source information is saved as a result file in the format of.nff, the time domain information of the sound source is extracted by using the iCFD function, the time domain information is subjected to fast Fourier transform, then the saved iCFD file is executed for calculation.

[0010] Fifth step: the field point and the sound pressure cloud picture output are set in the pre-processing, the frequency domain file in the format of.edat is saved, and the sound propagation calculation is executed by means of the solver.

[0011] Sixth step: in the post-processing, the sound pressure level and the sound source cloud picture distribution and the sound pressure level spectrum curve of the monitoring point are respectively viewed.

[0012] In the specific embodiment of the present application, the specific steps of the second step include:

[0013] (2-1) according to the specific geometric parameters, the three-dimensional modeling and the structured grid division of the blade pump water body are carried out, the control equation is established, the boundary conditions are set according to the external characteristic test, the numerical simulation of the non-cavitation condition is carried out first, the turbulence model selects the SST k-omega model, and the numerical calculation result is compared with the measured head, efficiency, shaft power and the like.

[0014] (2-2) if the error is greater than 5%, it is indicated that the non-cavitation numerical calculation is not accurate, and the related content is modified by returning to (2-1);

[0015] (2-3) If the error of the two is less than 5%, it is indicated that the non-cavitation numerical calculation is reliable, then the non-cavitation numerical calculation result is taken as an initial condition, cavitation steady numerical calculation under different cavitation numbers is carried out, the turbulence model is selected as SST k-omega model, the cavitation model is selected as ZGB model, and the obtained cavitation external characteristics are compared with the measured head, efficiency, shaft power and the like;

[0016] (2-4) If the error of the two is greater than 5%, it is indicated that the cavitation steady numerical calculation is not accurate, and the related content is modified in (2-3);

[0017] (2-5) If the error of the two is less than 5%, then unsteady cavitation numerical calculation under different cavitation numbers is carried out, the turbulence model is selected as large eddy simulation (LES), and the cavitation model is selected as ZGB model, the bubble volume and the like generated by the blade pump in multiple rotation periods are solved by unsteady cavitation numerical calculation, the data of the last 10 rotation periods are usually selected to carry out acoustic calculation; in order to solve the noise generated due to the bubble volume pulsation, the solution language of the bubble volume at each time is written in the user-defined function, the bubble volume at each time is obtained, and the transient pressure item, density item and velocity item are checked in the pre-processing.

[0018] In the specific embodiment of the present application, the rotation period in the above step (2-5) is set to 30 rotation periods.

[0019] The positive progress effect of the present application is that the blade pump cavitation induced noise numerical prediction method provided by the present application has the following advantages: the high-precision unsteady cavitation numerical calculation provided by the present application can accurately capture the bubble volume and the pressure item, density item and velocity item generated in the running process of the blade pump, on the one hand, accurate sound source information is provided for subsequent acoustic calculation, so as to improve the accuracy of acoustic numerical prediction, and the prediction result is applied to low-noise hydraulic design of the blade pump, so as to reduce the test times, shorten the research and development period, save the development cost, and improve the efficiency of low-noise blade pump design; on the other hand, the spatio-temporal evolution characteristics of vortex structure inside the blade pump and the coupling mechanism of vortex and cavitation in the cavitation unsteady flow process can be revealed in depth, so as to provide theoretical support for the early design or later optimization of the blade pump. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 The flow chart of the blade pump cavitation flow induced noise numerical prediction of the present application.

[0021] CFD: Computational Fluid Dynamics; LES: Large Eddy Simulation; ZGB cavitation model: Zwart-Gerber-Belamri cavitation model; Matlab and Excel: no corresponding Chinese; Actran-iCFD: no corresponding Chinese; Lighthill: no corresponding Chinese; Actran-aero acoustic: no corresponding Chinese; Actran-VI: no corresponding Chinese. DETAILED DESCRIPTION

[0022] The preferred embodiments of the present application are described below in detail with reference to the accompanying drawings, so as to explain the technical solutions of the present application.

[0023] Figure 1 The flow chart for numerical prediction of cavitation flow induced noise of the blade pump of the present application is shown in Figure 1

[0024] Based on large eddy simulation and ZGB cavitation model, the computational fluid dynamics method and the computational acoustic method are combined to solve the cavitation flow induced noise of the blade pump under the design condition, and the specific steps are as follows:

[0025] (1) Obtain the head, efficiency, shaft power, flow rate, pressure and other data of the blade pump under cavitation condition and non-cavitation condition under the design condition on the multi-functional test bench of the blade pump;

[0026] (2) Establish a three-dimensional model of the water body of the blade pump and perform structured grid division, perform non-cavitation steady, cavitation steady and cavitation unsteady numerical calculation, and compare with the test results;

[0027] For (2), the following specific steps are provided:

[0028] (2-1) According to the specific geometric parameters, a three-dimensional modeling and structured grid division of the water body of the blade pump is performed, control equations are established, boundary conditions are set according to the external characteristic test, the inlet boundary condition is set to total pressure inlet (1 atm), and the outlet boundary condition is set to mass flow outlet (q m According to the specific flow rate of the blade pump, the numerical simulation of the non-cavitation condition is first performed, the turbulence model is selected as SST k-ω model, and the numerical calculation results are compared with the test measured head, efficiency, shaft power and the like;

[0029] (2-2) If the error is greater than 5%, it indicates that the non-cavitation numerical calculation is not accurate, and the related content is modified by returning to (2-1);

[0030] ​(2-3) If the error is less than 5%, it means that the non-cavitation numerical calculation is reliable, then take the non-cavitation numerical calculation results as the initial condition, and perform cavitation steady numerical calculation under different cavitation numbers, the turbulence model is selected as SST k-omega model, and the cavitation model is selected as ZGB model, and the obtained cavitation external characteristics are compared with the measured head, efficiency, shaft power and the like;

[0031] (2-4) If the error is greater than 5%, it means that the cavitation steady numerical calculation is not accurate, return to (2-3) to modify the relevant content;

[0032] (2-5) If the error is less than 5%, then perform unsteady cavitation numerical calculation under different cavitation numbers, the turbulence model is selected as large eddy simulation (LES), and the cavitation model is selected as ZGB model, the bubble volume and.ensight format pressure item, density item and velocity item generated by the blade pump in multiple rotation periods (generally 30 rotation periods) are solved, and the data of the last 10 rotation periods are usually selected for acoustic calculation. In order to solve the noise generated by the bubble volume fluctuation, the data of the bubble volume changing with time are first obtained, the solving language of the bubble volume at each time is written in the user-defined function, and the bubble volume at each time can be obtained, and the transient pressure item, density item and velocity item are checked in the pre-processing.

[0033] (3) Based on the spherical bubble radiation theory, when performing numerical calculation of the noise generated by the bubble volume fluctuation, the bubble is regarded as a simple pulsation source, if the bubble diameter is much smaller than the characteristic wavelength, the bubble can be regarded as a point pulsation source, and the sound pressure of the monitoring point at a distance r from the center of the sphere is:

[0034]

[0035] (Formula (1) parameter definition: r represents the distance of the monitoring point from the center of the sphere, t represents time, Q is the mass fluctuation rate of the unit volume of liquid, q is the volume fluctuation rate of the unit volume of liquid, and p0 is the density)

[0036] The time domain sound pressure signal and the frequency domain sound pressure signal of the monitoring field point are obtained by data processing of the bubble volume of the last 10 rotation periods according to the above formula;

[0037] (4) After the convergence of the numerical calculation of the unsteady flow of the vaned pump cavitation, the flow field information obtained by the cavitation unsteady calculation in the last 10 rotation periods is derived in the post-processing of the computational fluid dynamics software, and the flow field information includes the pressure term, the velocity term and the density term. The above flow field information is the sound source information for the subsequent acoustic finite element calculation. The acoustic mesh of the acoustic calculation domain of the vaned pump is divided. In order to ensure the accuracy of the acoustic finite element calculation, the maximum size of the acoustic mesh should be less than 1 / 6 of the wavelength corresponding to the maximum calculation frequency, that is: L max max , wherein L max represents the maximum size of the acoustic mesh, c represents the sound speed, and f max is the maximum frequency of acoustic calculation. In the acoustic calculation software, an acoustic analysis model is created, boundary conditions are set, the calculation domain wall and entity are renamed, and test conditions are used. A shell with a thickness of 0.003m is added to the calculation domain wall, in order to make the cavitation-induced noise propagate from inside to outside without reflection, a pipe mode is added to the inlet and outlet of the hydrofoil calculation domain. The time domain information of the sound source is extracted by using iCFD function, the "HANGNING" window is selected to perform fast Fourier transform on the time domain information, and then the saved iCFD file is executed for calculation.

[0038] (5) Set the field point and the output of the sound pressure cloud map, save the frequency domain file in the format of.dat, and execute the sound propagation calculation by using the solver;

[0039] (6) In the post-processing, the sound pressure level and the sound source cloud map distribution and the sound pressure level spectrum curve of the monitoring point are viewed respectively.

[0040] For (4)-(6), the flow-induced noise except the monopole noise generated by the pulsation of the cavity volume is solved based on the Lighthill acoustic analogy theory. In this solving process, the sound source term is divided into surface sound source and volume sound source. The sound source information generated in the impeller rotation area is all equivalent to the surface sound source on the interface between the impeller and the volute, and the sound source generated in the pump cavity area is treated as a volume sound source. That is, the sound generation phenomenon of fluid motion is analogous to the acoustic problem of equivalent sound source existing in static medium, and the solving theory is shown in equations (2) and (3):

[0041]

[0042]

[0043] In equation (2), the first term on the right side represents the volume sound source, and the second term on the right side represents the surface sound source.

[0044] ​The above shows and describes the basic principles and main features of the present application and the advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A numerical prediction method for cavitation-induced noise in a vane pump, characterized in that: This method numerically calculates the cavitation-induced noise of a bladed pump based on experiments. It solves for the cavitation flow-induced noise of the bladed pump using large eddy simulation and the ZGB cavitation model. The numerical prediction method for cavitation-induced noise of the bladed pump includes the following steps: Step 1: Conduct external characteristic tests of the model pump under non-air-condition and air-condition conditions; Step 2: Establish a three-dimensional model of the water body of the impeller pump and perform structured mesh generation; perform numerical calculations of non-cavitation steady, cavitation steady, and cavitation unsteady, and compare the results with experimental results. Step 3: Based on the spherical cavitation radiation theory, when performing numerical calculation of the noise generated by the cavitation volume pulsation, the cavitation volume in the last 10 rotation cycles is processed according to the formula (1) using data processing software to obtain the time-domain sound pressure and frequency-domain sound pressure of the monopole noise generated by the cavitation volume pulsation. The expression of formula (1) is as follows. Equation (1) parameter definition: r represents the distance of the monitoring point from the center of the ball, t represents time, Q is the mass pulsation rate per unit volume of liquid, q is the volume pulsation rate per unit volume of liquid, and ρ0 is the liquid density; Step 4: Extract the flow field information from the computational fluid dynamics results as the sound source information for the acoustic finite element method (FED) calculation. Divide the acoustic calculation domain of the impeller pump into an acoustic mesh. To ensure the accuracy of the acoustic finite element method, the maximum size of the acoustic mesh should be less than 1 / 6 of the wavelength corresponding to the maximum calculation frequency, i.e., L. max <c / 6f max In the formula L max The maximum size of the acoustic grid is represented by c, where c represents the speed of sound, and f represents the maximum size of the acoustic grid. max To calculate the maximum frequency in acoustics, a computational acoustics method is used to create an acoustic analysis model, set boundary conditions, and save the volumetric sound source information in .nff format as a result file. The time-domain information of the sound source is extracted using the iCFD function, a fast Fourier transform is performed on the time-domain information, and then the saved iCFD file is used to perform calculations. Step 5: Set the field point and sound pressure contour map output in the preprocessing, save the frequency domain file in .edat format, and use the solver to perform sound propagation calculations; Step 6: In post-processing, check the sound pressure level, sound source cloud map distribution, and sound pressure level spectrum curve of the monitoring point.

2. The numerical prediction method for cavitation-induced noise of a bladed pump according to claim 1, characterized in that: The specific steps of the second step include: (2-1) Based on the specific geometric parameters, perform three-dimensional modeling and structured mesh generation of the water body for the impeller pump, establish the control equations, set the boundary conditions according to the external characteristic test, first perform numerical simulation of non-airflow conditions, select the SST k-ω model for turbulence model, and compare the numerical calculation results with the head, efficiency and shaft power measured by the test. (2-2) If the error between the two is greater than 5%, it indicates that the noncavitation numerical calculation is inaccurate. Go back to (2-1) and modify the relevant content. (2-3) If the error between the two is less than 5%, it indicates that the non-cavitation numerical calculation is reliable. Then, using the non-cavitation numerical calculation result as the initial condition, cavitation steady-state numerical calculations are performed under different cavitation numbers. The SST k-ω model is selected as the turbulence model, and the ZGB model is selected as the cavitation model. The obtained cavitation external characteristics are compared with the head, efficiency, and shaft power measured by the experiment. (2-4) If the error between the two is greater than 5%, it indicates that the calculation of the cavitation steady constant value is inaccurate. Go back to (2-3) and modify the relevant content. (2-5) If the error between the two is less than 5%, then perform unsteady cavitation numerical calculations under different cavitation numbers. The turbulence model is selected as Large Eddy Simulation (LES), and the cavitation model is selected as ZGB model. The unsteady cavitation numerical calculations are used to solve for the cavitation volume generated by the impeller pump in multiple rotation cycles, and the pressure, density, and velocity terms in .ensight format are usually selected from the data of the last 10 rotation cycles for acoustic calculation. To solve for the noise generated by the cavitation volume pulsation, it is first necessary to obtain the data of the cavitation volume changing with time. The cavitation volume at each moment is obtained by writing the solution language for the cavitation volume at each moment in the user-defined function, while the transient pressure, density, and velocity terms are selected in the preprocessing.

3. The numerical prediction method for cavitation-induced noise of a bladed pump according to claim 2, characterized in that: In (2-5), the rotation period is set to 30 rotation periods.

Images