Method and system for noise optimization of waterjet propulsor based on three-dimensional counter-design
By optimizing the flow components of the waterjet propulsion pump through three-dimensional inverse design and the intelligent optimization algorithm NSGA-II, the problem of insufficient noise optimization of the waterjet propulsion pump was solved, and the optimal matching design within a short cycle was achieved, thereby reducing the noise level of the waterjet propulsion pump.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI JIAOTONG UNIV
- Filing Date
- 2022-09-06
- Publication Date
- 2026-06-23
Smart Images
Figure CN115455854B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of noise optimization technology, and more specifically, to a method and system for noise optimization of a waterjet propulsion pump based on three-dimensional inverse design. Background Technology
[0002] In the field of anti-submarine warfare, using sonar to detect specific components in ocean noise has become an important means of anti-submarine warfare. On the other hand, noise levels also affect the working conditions of personnel on board ships. Therefore, active control of ship noise is particularly important. The noise of the ship's propulsion system is a significant source of ship noise. Compared to other propulsion methods, waterjet propulsion can effectively reduce cavitation and lower vibration noise levels. At the same time, waterjet propulsion offers good maneuverability, enabling functions such as turning and reversing; therefore, waterjet propulsion has become an important method of ship propulsion. The noise of the propulsion system is a significant source of ship noise. The waterjet propulsion pump is the most important component of the waterjet propulsion unit and also the most significant source of noise.
[0003] Patent document CN105117564B (application number: CN201510599246.X) discloses a hydraulic model and design method of a pump-jet propulsion device with a front stator arranged circumferentially asymmetrically. The model includes stator blades and a stator hub, impeller blades and an impeller hub, and a duct. The stator blades are arranged circumferentially asymmetrically on the stator hub, and the tips of the stator blades are fixed to the inner wall of the duct. The axes of the stator hub, the impeller hub, and the duct coincide. The impeller blades have large sideslip and tail tilt characteristics. The stator blade pitch angle changes sinusoidally based on the reference pump-jet stator blade pitch angle. The change in pitch angle is related to the circumferential angle position of the stator blade, the impeller rotation direction, and the amplitude coefficient.
[0004] Previous research on the design methods of waterjet propulsion pumps focused on optimizing hydraulic performance by modifying blade geometry parameters, heavily relying on the designer's experience. Recently, intelligent optimization methods have been applied to optimization design, but they still target geometric parameters. Although optimization algorithms can achieve better selection of geometric parameters, the computational load is enormous. Current research on the noise of waterjet propulsion pumps focuses on their hydraulic noise characteristics; however, research on how to reduce noise through optimization design is scarce. Summary of the Invention
[0005] To address the shortcomings of existing technologies, the purpose of this invention is to provide a method and system for optimizing the noise of a waterjet propulsion pump based on three-dimensional inverse design.
[0006] The noise optimization method for waterjet propulsion pumps based on three-dimensional inverse design provided by the present invention includes:
[0007] Step 1: Construct a hydraulic model for noise optimization of a waterjet propulsion pump based on three-dimensional inverse design, including an impeller, a runner chamber, a guide vane hub, guide vane blades, and a guide vane chamber. The impeller blades are fixed to the impeller hub, and the tip face of the impeller blades and the inner wall of the runner chamber have an equal distance between the blade tips. The root of the guide vane blades is fixed to the guide vane hub, and the tip of the guide vane blades is fixed to the guide vane chamber.
[0008] Step 2: Use four-point cubic Bézier curves to fit the rim curves and hub curves of the runner chamber and guide vane chamber respectively, and obtain the coordinates of the control points P0, P1, P2, P3 of each curve on the circumferential XY plane and the meridional flow channel model fitted by the Bézier curves.
[0009] Step 3: Determine the inlet load values of the impeller blades and the sample space of the inlet and outlet loads of the impeller, and establish a parametric design model for the loads of the impeller blades and guide vanes.
[0010] Step 4: Use a linear angle stacking method, and use the stacking angle θ as a design parameter variable;
[0011] Step 5: Perform steady and unsteady calculations of the single-channel flow field sequentially using computational fluid dynamics to save calculation time for each optimization point. Predict flow noise using an indirect method that couples computational fluid dynamics and computational acoustics.
[0012] Step 6: Determine the value range of each design variable parameter, conduct sensitivity analysis on multiple pump parameters, and obtain the pattern of the impact of factor changes on the optimization target by changing the values of relevant design variables one by one;
[0013] Step 7: The NSGA-II intelligent optimization algorithm is used to achieve fully automatic performance optimization. After multiple iterations of calculation, the optimal combination of parameters of the current-carrying components is obtained in the global scope, realizing the performance matching design of the current-carrying components, and finally obtaining the model with the minimum noise level of the objective function.
[0014] Preferably, control point P0 is the starting point, control point P3 is the ending point, and P1 and P2 are intermediate points. The coordinates of P0 and P3 are fixed, while the coordinates of P1 and P2 can vary. Let:
[0015]
[0016] The coordinates of points P1 and P2 satisfy: P1 = P0 + c1(P3 - P0), P2 = P3 - c2(P3 - P0), where c1 ∈ [0, 1] and c2 ∈ [0, 1]. Taking c1 and c2 of each curve as parameter variables, there are a total of eight parameter variables for the four curves, denoted as c0 to c7. The range of values for c0 to c7 is [0, 1].
[0017] Preferably, step 3 includes: determining the impeller blade inlet load value based on the flow rate, head and speed of the target water jet propulsion pump; determining the range of the impeller inlet and outlet circulation difference according to the requirements; thereby determining the sample space of the impeller inlet and outlet loads; and limiting the load at the blade tip during load distribution to achieve blade tip vortex cavitation control, thereby reducing cavitation noise.
[0018] A three-stage blade load control method is adopted to control the load distribution on the blade. The load distribution control variable is used as the design variable to realize the parameterization of the blade and establish a parameterized design model for the load of the impeller blade and guide vane blade.
[0019] Preferably, step 5 includes: considering the generation and propagation of flow noise separately based on the Lighthill equation: calculating the pump body vibration coupled during sound propagation using computational acoustics; after the acoustic mesh is divided, performing acoustic calculations by combining the sound source information file and the structural modal file of the pump body to obtain the flow noise at the outlet of the guide vane chamber.
[0020] Preferably, step 6 includes: designing multiple test schemes using response surface methodology, establishing a quadratic function between the optimization objective and the design variables, performing quadratic regression analysis to obtain the design variables that significantly affect pump performance, and removing design parameters that have a smaller impact on performance.
[0021] The waterjet propulsion pump noise optimization system based on three-dimensional inverse design provided by the present invention includes:
[0022] Module M1: Construct a hydraulic model for noise optimization of a waterjet propulsion pump based on three-dimensional inverse design, including an impeller, a runner chamber, a guide vane hub, guide vane blades, and a guide vane chamber. The impeller blades are fixed to the impeller hub, and the tip face of the impeller blades and the inner wall of the runner chamber have an equal distance between the blade tips. The root of the guide vane blades is fixed to the guide vane hub, and the tip of the guide vane blades is fixed to the guide vane chamber.
[0023] Module M2: Four-point cubic Bézier curves are used to fit the rim curves and hub curves of the runner chamber and guide vane chamber respectively, to obtain the coordinates of the control points P0, P1, P2, and P3 of each curve in the circumferential XY plane and the meridional flow channel model fitted by the Bézier curves.
[0024] Module M3: Determines the inlet load values of the impeller blades and the sample space of the inlet and outlet loads of the impeller, and establishes a parametric design model for the loads of the impeller blades and guide vanes.
[0025] Module M4: Uses a linear angle stacking method, with the stacking angle θ as a design parameter variable;
[0026] Module M5: It uses computational fluid dynamics to perform steady and unsteady calculations of the single-channel flow field in sequence to save the calculation time of a single optimization point. It uses an indirect method that couples computational fluid dynamics and computational acoustics to predict flow noise.
[0027] Module M6: Determine the value range of each design variable parameter, perform sensitivity analysis on multiple pump parameters, and obtain the pattern of the impact of factor changes on the optimization target by changing the values of relevant design variables one by one;
[0028] Module M7: It adopts the intelligent optimization algorithm NSGA-II to achieve fully automatic performance optimization. After multiple iterations of calculation, it obtains the optimal combination of parameters of the current flow components in the global scope, realizes the performance matching design of the current flow components, and finally obtains the model with the minimum noise level of the objective function.
[0029] Preferably, control point P0 is the starting point, control point P3 is the ending point, and P1 and P2 are intermediate points. The coordinates of P0 and P3 are fixed, while the coordinates of P1 and P2 can vary. Let:
[0030]
[0031] The coordinates of points P1 and P2 satisfy: P1 = P0 + c1(P3 - P0), P2 = P3 - c2(P3 - P0), where c1 ∈ [0, 1] and c2 ∈ [0, 1]. Taking c1 and c2 of each curve as parameter variables, there are a total of eight parameter variables for the four curves, denoted as c0 to c7. The range of values for c0 to c7 is [0, 1].
[0032] Preferably, the module M3 includes: determining the inlet load value of the impeller blades based on the flow rate, head and speed of the target water jet propulsion pump; determining the range of the circulation difference between the impeller inlet and outlet according to the requirements; thereby determining the sample space of the impeller inlet and outlet loads; and limiting the load at the blade tip during load distribution to achieve blade tip vortex cavitation control, thereby reducing cavitation noise.
[0033] A three-stage blade load control method is adopted to control the load distribution on the blade. The load distribution control variable is used as the design variable to realize the parameterization of the blade and establish a parameterized design model for the load of the impeller blade and guide vane blade.
[0034] Preferably, the module M5 includes: considering the generation and propagation of flow noise separately based on the Lighthill equation; calculating the pump body vibration coupled during sound propagation using computational acoustics; and performing acoustic calculations by combining the sound source information file and the structural modal file of the pump body after the acoustic mesh is divided, to obtain the flow noise at the outlet of the guide vane chamber.
[0035] Preferably, module M6 includes: designing multiple test schemes using response surface methodology, establishing a quadratic function between the optimization objective and design variables, performing quadratic regression analysis to obtain design variables that significantly affect pump performance, and removing design parameters with minor impact on performance.
[0036] Compared with the prior art, the present invention has the following beneficial effects:
[0037] (1) The present invention can complete the optimized design of the flow-through components of the water jet propulsion pump in a relatively short period of time, and obtain the best matching of the flow-through component parameters and the water jet propulsion pump hydraulic model with the least noise.
[0038] (2) This invention takes into account the performance and noise requirements of water jet propulsion pumps. This design method is also applicable to the design of pump spray. After promotion, it can be widely applied to the design of water jet propulsion systems. Attached Figure Description
[0039] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:
[0040] Figure 1 A flowchart of a noise optimization design method for a waterjet propulsion pump based on a three-dimensional inverse problem;
[0041] Figure 2 is a schematic diagram of the fitting of the parametric curve of the meridional plane; Figure 2a This is a schematic diagram of Bézier curve fitting. Figure 2b This is a schematic diagram of the meridional flow channel;
[0042] Figure 3 Schematic diagram of blade load control;
[0043] Figure 4 Schematic diagram of blade stacking angle control;
[0044] Figure 5 Hydraulic model diagram of a waterjet propulsion pump. Detailed Implementation
[0045] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.
[0046] Example:
[0047] This invention proposes a high-performance, low-noise optimized hydraulic model for a waterjet propulsion pump based on a three-dimensional inverse problem design. The model includes an impeller, a runner chamber, and guide vanes. The impeller blades are fixed to the impeller hub, with an equidistant tip clearance between the blade tip and the inner wall of the runner chamber. The guide vane body includes a guide vane hub, guide vane blades, and a guide vane chamber. The guide vane blade roots are fixed to the guide vane hub, and the guide vane blade tips are fixed to the guide vane chamber. The contours of both the runner chamber and the guide vane chamber are parametrically fitted using four-point cubic Bézier curves to control the axial flow path. The airfoils of both the impeller and guide vane blades are obtained using a parametric three-dimensional inverse problem design method. The cross-sectional load distribution of both the impeller and guide vane blades adopts a three-segment control method, and the blades are stacked at a fixed angle.
[0048] The impeller has 5 blades and the guide vane has 9 blades.
[0049] There is a gap between the tip of the impeller blade and the runner chamber, and the gap size is 0.2 to 0.5 mm.
[0050] This invention proposes a noise optimization design method for waterjet propulsion pumps based on a combination of three-dimensional inverse problem design and multi-objective multi-parameter optimization. The method employs a three-dimensional inverse problem design to control the spatial airfoil in the blade design, and uses Bézier curve fitting to achieve the meridional design. In the optimization design, the coordinates of the control points of the Bézier curves constituting the meridional flow channel and the control parameters of the blade load distribution of the airfoil are used as design parameter variables. The efficiency and head of the waterjet propulsion pump are used as constraints, and the average flow noise level of the first five blade frequencies of the waterjet propulsion pump is used as the optimization objective.
[0051] like Figure 1 The optimized design method for the flow-through components of the high-pressure pump for seawater desalination proposed in this invention mainly includes:
[0052] Step 1: Fit the rim curves and hub curves of the impeller chamber and guide vane chamber using four-point cubic Bézier curves respectively, to obtain the coordinates of the control points P0, P1, P2, and P3 on the circumferential XY plane. Control point P0 is the starting point, control point P3 is the ending point, and P1 and P2 are intermediate points. Due to actual installation limitations, the coordinates of P0 and P3 are often fixed, while the coordinates of P1 and P2 can vary. Let: The coordinates of points P1 and P2 satisfy P1 = P0 + c1(P3 - P0) and P2 = P3 - c2(P3 - P0), where c1 ∈ [0, 1] and c2 ∈ [0, 1]. c1 and c2 of each curve are used as parameter variables. There are eight parameter variables in total for the four curves, denoted as c0 to c7, where the values of c0 to c7 are all within the range [0, 1]. The meridional flow channel model fitted by the Bézier curve is obtained, as follows: Figure 2a and Figure 2b As shown.
[0053] Step 2: Determine the impeller blade inlet load value based on the target waterjet propulsion pump's flow rate, head, and speed. The blade load reflects the strength of the work done by the blades, and the load distribution is essentially the distribution of blade circulation. The head of the waterjet propulsion pump is determined by the circulation difference between the pump inlet and outlet, and the head is the external manifestation of the waterjet propulsion pump's work capacity. Determine the approximate range of the impeller inlet and outlet circulation difference as needed, thereby defining the sample space for the impeller inlet and outlet loads. Simultaneously, considering that the linear velocity is highest at the blade tip, making cavitation more likely, limit the load (circulation) at the blade tip during load distribution to achieve blade tip vortex cavitation control, thereby reducing cavitation noise. Calculate the impeller outlet profile circulation using Γ = 2πRvu as the guide vane inlet circulation for guide vane design.
[0054] Step 3: Use a three-stage blade load control method to control the load distribution on the blades, such as... Figure 3 As shown, the load distribution control variable is used as the design variable to realize the parameterization of the blade, and a parameterized design model of the load of the impeller blade and guide vane blade is established.
[0055] Step 4: The stacking method uses linear angular stacking, and the stacking angle θ is used as a design parameter variable, such as... Figure 4 As shown.
[0056] Step 5: Computational fluid dynamics (CFD) is used to perform both steady and unsteady calculations of the single-channel flow field sequentially to save computation time per optimization point. Unsteady calculations employ separated eddy simulation (DES). Flow noise prediction uses an indirect method combining CFD and computational acoustics (CA), considering the generation and propagation of flow noise separately based on the Lighthill equations. When calculating sound propagation using computational acoustics, pump vibration is coupled in between. After acoustic mesh generation, acoustic calculations are performed using a specific procedure, combining the sound source information file and the pump's structural modal file, to obtain the flow noise at the guide vane chamber outlet.
[0057] Step 6: Determine the value range of each design variable parameter, and then perform sensitivity analysis on multiple pump parameters. Explain the influence of these factors on the optimization objective by changing the values of the relevant design variables one by one. Simultaneously, design multiple experimental schemes using response surface methodology, establish a quadratic function between the optimization objective and the design variables, and perform quadratic regression analysis to obtain the design variables that significantly affect pump performance, and remove design parameters with less impact on performance.
[0058] Step 7: Employ the NSGA-II intelligent optimization algorithm to achieve fully automated performance optimization. Through multiple iterative calculations, the optimal combination of parameters for the current-carrying components can be obtained globally, achieving performance-matched design of the current-carrying components and ultimately obtaining the model with the minimum noise level of the objective function, such as... Figure 5 As shown.
[0059] The waterjet propulsion pump noise optimization system based on three-dimensional inverse design provided by the present invention includes: Module M1: Constructing a hydraulic model for noise optimization of the waterjet propulsion pump based on three-dimensional inverse design, including an impeller, a runner chamber, a guide vane hub, guide vane blades, and a guide vane chamber. The impeller blades are fixed to the impeller hub, and the blade tip faces of the impeller blades and the inner wall of the runner chamber have equidistant blade tip gaps. The root of the guide vane blades is fixed to the guide vane hub, and the blade tip of the guide vane blades is fixed to the guide vane chamber; Module M2: Fitting the rim curves and hub curves of the runner chamber and guide vane chamber respectively using four-point cubic Bézier curves to obtain the coordinates of the control points P0, P1, P2, and P3 of each curve on the circumferential XY plane and the meridional flow channel model fitted by the Bézier curves; Module M3: Determining the sample space of the impeller blade inlet load value and the impeller inlet and outlet loads, and establishing the impeller blades and guide vane blades. The load parameterization design model includes: Module M4: using a linear angle stacking method, with the stacking angle θ as a design parameter variable; Module M5: using computational fluid dynamics to perform steady and unsteady calculations of the single-channel flow field sequentially to save calculation time for individual optimization points, and using an indirect method that couples computational fluid dynamics and computational acoustics to predict flow noise; Module M6: determining the value range of each design variable parameter, performing sensitivity analysis on multiple pump parameters, and obtaining the law of the magnitude of the impact of factor changes on the optimization objective by changing the values of relevant design variables one by one; Module M7: using the intelligent optimization algorithm NSGA-II to achieve fully automatic performance optimization, obtaining the optimal combination of flow component parameters globally after multiple iterative calculations, realizing the performance matching design of flow components, and finally obtaining the model with the minimum noise level of the objective function.
[0060] Control point P0 is the starting point, control point P3 is the ending point, and P1 and P2 are intermediate points. The coordinates of P0 and P3 are fixed, while the coordinates of P1 and P2 are variable. Let: The coordinates of points P1 and P2 satisfy: P1 = P0 + c1(P3 - P0), P2 = P3 - c2(P3 - P0), where c1 ∈ [0, 1] and c2 ∈ [0, 1]. Taking c1 and c2 of each curve as parameter variables, there are a total of eight parameter variables for the four curves, denoted as c0 to c7. The range of values for c0 to c7 is [0, 1].
[0061] Module M3 includes: determining the impeller blade inlet load value based on the flow rate, head, and speed of the target waterjet propulsion pump; determining the range of impeller inlet and outlet circulation difference according to requirements, thereby determining the sample space of impeller inlet and outlet loads; limiting the load at the blade tip during load distribution to achieve blade tip vortex cavitation control, thereby reducing cavitation noise; using a three-segment blade load control method to control the load distribution on the blades; using the load distribution control variable as the design variable to achieve blade parameterization; and establishing a parameterized design model for the impeller blade and guide vane blade loads. Module M5 includes: considering the generation and propagation of flow noise separately based on the Lighthill equation; using computational acoustics to calculate the pump body vibration coupled during sound propagation; after acoustic mesh generation, combining the sound source information file and the pump body structural modal file to perform acoustic calculations to obtain the flow noise at the guide vane chamber outlet. Module M6 includes: designing multiple test schemes using response surface methodology; establishing a function of quadratic terms between the optimization objective and design variables; performing quadratic regression analysis to obtain design variables that significantly affect pump performance; and removing design parameters with less impact on performance.
[0062] This invention can complete the optimized design of the flow-through components of a water jet propulsion pump in a relatively short period of time, achieving optimal matching of flow-through component parameters and minimizing pump noise.
[0063] Those skilled in the art will understand that, in addition to implementing the system, apparatus, and their modules provided by this invention in purely computer-readable program code, the same program can be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system, apparatus, and their modules provided by this invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; alternatively, modules for implementing various functions can be considered both software programs implementing the method and structures within the hardware component.
[0064] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.
Claims
1. A noise optimization method for a waterjet propulsion pump based on three-dimensional inverse design, characterized in that, include: Step 1: Construct a hydraulic model for noise optimization of a waterjet propulsion pump based on three-dimensional inverse design, including an impeller, a runner chamber, a guide vane hub, guide vane blades, and a guide vane chamber. The impeller blades are fixed to the impeller hub, and the tip face of the impeller blades and the inner wall of the runner chamber have an equal distance between the blade tips. The root of the guide vane blades is fixed to the guide vane hub, and the tip of the guide vane blades is fixed to the guide vane chamber. Step 2: Use four-point cubic Bézier curves to fit the rim curves and hub curves of the runner chamber and guide vane chamber respectively, and obtain the coordinates of the control points P0, P1, P2, P3 of each curve on the circumferential XY plane and the meridional flow channel model fitted by the Bézier curves. Step 3: Determine the inlet load values of the impeller blades and the sample space of the inlet and outlet loads of the impeller, and establish a parametric design model for the loads of the impeller blades and guide vanes. Step 4: Use a linear angle stacking method and use the stacking angle θ as a design parameter variable; Step 5: Use computational fluid dynamics to perform steady and unsteady calculations of the single-channel flow field in sequence to save the calculation time of a single optimization point, and use an indirect method that couples computational fluid dynamics and computational acoustics to predict flow noise; Step 6: Determine the value range of each design variable parameter, conduct sensitivity analysis on multiple pump parameters, and obtain the pattern of the influence of factor changes on the optimization target by changing the values of relevant design variables one by one; design multiple test schemes using response surface methodology, and establish a quadratic term function between the optimization target and design variables, and conduct quadratic regression analysis to obtain the design variables that significantly affect pump performance and remove design parameters with less impact on performance. Step 7: The NSGA-II intelligent optimization algorithm is used to achieve fully automatic performance optimization. After multiple iterations of calculation, the optimal combination of parameters of the current-carrying components is obtained in the global scope, realizing the performance matching design of the current-carrying components, and finally obtaining the model with the minimum noise level of the objective function.
2. The noise optimization method for a waterjet propulsion pump based on three-dimensional inverse design according to claim 1, characterized in that, Control point P0 is the starting point, control point P3 is the ending point, and P1 and P2 are intermediate points. The coordinates of P0 and P3 are fixed, while the coordinates of P1 and P2 are variable. Let: ， Then the coordinates of points P1 and P2 satisfy: , ,in , With c1 and c2 of each curve as parameter variables, there are a total of eight parameter variables for the four curves, denoted as c0~c7, and the value range of c0~c7 is [0, 1].
3. The noise optimization method for a waterjet propulsion pump based on three-dimensional inverse design according to claim 1, characterized in that, Step 3 includes: determining the inlet load value of the impeller blades based on the flow rate, head and speed of the target water jet propulsion pump; determining the range of the circulation difference between the impeller inlet and outlet according to the requirements; thereby determining the sample space of the impeller inlet and outlet loads; and limiting the load at the blade tip during load distribution to achieve blade tip vortex cavitation control, thereby reducing cavitation noise. A three-stage blade load control method is adopted to control the load distribution on the blade. The load distribution control variable is used as the design variable to realize the parameterization of the blade and establish a parameterized design model for the load of the impeller blade and guide vane blade.
4. The noise optimization method for a waterjet propulsion pump based on three-dimensional inverse design according to claim 1, characterized in that, Step 5 includes: considering the generation and propagation of flow noise separately based on the Lighthill equation; calculating the pump body vibration coupled during sound propagation using computational acoustics; after the acoustic mesh is divided, performing acoustic calculations by combining the sound source information file and the structural modal file of the pump body to obtain the flow noise at the outlet of the guide vane chamber.
5. A noise optimization system for a waterjet propulsion pump based on three-dimensional inverse design, characterized in that, include: Module M1: Construct a hydraulic model for noise optimization of a waterjet propulsion pump based on three-dimensional inverse design, including an impeller, a runner chamber, a guide vane hub, guide vane blades, and a guide vane chamber. The impeller blades are fixed to the impeller hub, and the tip face of the impeller blades and the inner wall of the runner chamber have an equal distance between the blade tips. The root of the guide vane blades is fixed to the guide vane hub, and the tip of the guide vane blades is fixed to the guide vane chamber. Module M2: Four-point cubic Bézier curves are used to fit the rim curves and hub curves of the runner chamber and guide vane chamber respectively, to obtain the coordinates of the control points P0, P1, P2, and P3 of each curve in the circumferential XY plane and the meridional flow channel model fitted by the Bézier curves. Module M3: Determines the inlet load values of the impeller blades and the sample space of the inlet and outlet loads of the impeller, and establishes a parametric design model for the loads of the impeller blades and guide vanes. Module M4: Uses a linear angle stacking method, with the stacking angle θ as a design parameter variable; Module M5: It uses computational fluid dynamics to perform steady and unsteady calculations of the single-channel flow field in sequence to save the calculation time of a single optimization point. It uses an indirect method that couples computational fluid dynamics and computational acoustics to predict flow noise. Module M6: Determine the value range of each design variable parameter, perform sensitivity analysis on multiple pump parameters, and obtain the pattern of the impact of factor changes on the optimization target by changing the values of relevant design variables one by one; Multiple experimental schemes were designed using response surface methodology, and a quadratic term function between the optimization objective and design variables was established. Quadratic regression analysis was then performed to obtain the design variables that significantly affect pump performance and remove design parameters with less impact on performance. Module M7: It adopts the intelligent optimization algorithm NSGA-II to achieve fully automatic performance optimization. After multiple iterations of calculation, it obtains the optimal combination of parameters of the current flow components in the global scope, realizes the performance matching design of the current flow components, and finally obtains the model with the minimum noise level of the objective function.
6. The noise optimization system for a waterjet propulsion pump based on three-dimensional inverse design according to claim 5, characterized in that, Control point P0 is the starting point, control point P3 is the ending point, and P1 and P2 are intermediate points. The coordinates of P0 and P3 are fixed, while the coordinates of P1 and P2 are variable. Let: ， Then the coordinates of points P1 and P2 satisfy: , ,in , With c1 and c2 of each curve as parameter variables, there are a total of eight parameter variables for the four curves, denoted as c0~c7, and the value range of c0~c7 is [0, 1].
7. The noise optimization system for a waterjet propulsion pump based on three-dimensional inverse design according to claim 5, characterized in that, The module M3 includes: determining the inlet load value of the impeller blades based on the flow rate, head and speed of the target water jet propulsion pump; determining the range of the circulation difference between the impeller inlet and outlet according to the requirements; thereby determining the sample space of the impeller inlet and outlet loads; and limiting the load at the blade tip during load distribution to achieve blade tip vortex cavitation control, thereby reducing cavitation noise. A three-stage blade load control method is adopted to control the load distribution on the blade. The load distribution control variable is used as the design variable to realize the parameterization of the blade and establish a parameterized design model for the load of the impeller blade and guide vane blade.
8. The noise optimization system for a waterjet propulsion pump based on three-dimensional inverse design according to claim 5, characterized in that, The module M5 includes: considering the generation and propagation of flow noise separately based on the Lighthill equation; using computational acoustics to calculate the pump body vibration coupled during sound propagation; after the acoustic mesh is divided, combining the sound source information file and the structural modal file of the pump body to perform acoustic calculations to obtain the flow noise at the outlet of the guide vane chamber.