CN112487563B - Underwater vehicle propeller damage detection method based on wavelet analysis - Google Patents

Underwater vehicle propeller damage detection method based on wavelet analysis Download PDF

Info

Publication number
CN112487563B
CN112487563B CN202011481561.XA CN202011481561A CN112487563B CN 112487563 B CN112487563 B CN 112487563B CN 202011481561 A CN202011481561 A CN 202011481561A CN 112487563 B CN112487563 B CN 112487563B
Authority
CN
China
Prior art keywords
propeller
wavelet
software
basin
setting
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202011481561.XA
Other languages
Chinese (zh)
Other versions
CN112487563A (en
Inventor
刘宗凯
郭正阳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Science and Technology
Original Assignee
Nanjing University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Science and Technology filed Critical Nanjing University of Science and Technology
Priority to CN202011481561.XA priority Critical patent/CN112487563B/en
Publication of CN112487563A publication Critical patent/CN112487563A/en
Application granted granted Critical
Publication of CN112487563B publication Critical patent/CN112487563B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/15—Vehicle, aircraft or watercraft design
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/10—Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Computing Systems (AREA)
  • Fluid Mechanics (AREA)
  • Mathematical Physics (AREA)
  • Algebra (AREA)
  • Computer Graphics (AREA)
  • Software Systems (AREA)
  • Automation & Control Theory (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Computational Mathematics (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)

Abstract

The invention provides a CFD and wavelet analysis-based propeller damage detection method for an underwater vehicle, which is characterized in that three-dimensional computer aided design software SOLIDWORKS is used as platform software to establish geometric models of a submarine body, a surrounding shell, a tail vane, a propeller wall, a propeller blade and a flow field around the underwater vehicle; ANSYSICEMCFD is used as platform software to carry out geometric body repair, calculation domain division and grid generation; opening a CFX-Pre module of ANSYSCFX software, and carrying out numerical calculation pretreatment; starting CFD simulation, and acquiring the force of the propeller in the y direction, the force of the propeller in the z direction and the monitoring moment in the x direction by taking the head end point of the submersible vehicle as a monitoring point; and (3) calling a wavelet tool box by using MATLAB as platform software, performing 6-layer wavelet decomposition by using the wavelet tool box and adopting db5 wavelet, and comparing complete blade propeller warp data with broken blade propeller data to establish a fault database for subsequent fault diagnosis. The invention can avoid the interference of the environment to the visual detection.

Description

Underwater vehicle propeller damage detection method based on wavelet analysis
Technical Field
The invention relates to the field of computational fluid mechanics, in particular to a CFD and wavelet analysis-based propeller damage detection method for an underwater vehicle.
Background
The underwater vehicle in the ocean strategy is an indispensable role, the operational capability of the underwater vehicle plays an extremely important role, and the viability of the underwater vehicle is one of important indexes for measuring the operational capability of the submarine. The underwater vehicle works in a complex and crisis-four-volt marine environment, and the propeller of the underwater vehicle has the possibility of being broken down in many emergency situations. The propeller consists of an engine, a shaft, a propeller hub and blades, and can cause faults, wherein the blades are easy to damage, the blades of the propeller are radially inward in the working process, resistance of fluid to the blades of the propeller and force generated by collision of impurity particles in water on the blades of the propeller are relatively high in risk of blade damage, the blades of the propeller can cause periodic unbalance of radial stress of the propeller, if the blades of the propeller are not detected in time, a main shaft and the engine can be damaged, serious consequences can be generated, and therefore how to detect the damage of the blades is a very important part in propeller fault detection.
A vibration method fault monitoring system based on numerical simulation fault signals is established in Europe and the West, hydrodynamic performance is forecasted by a CFD method, response of shafting rotary vibration is calculated, and finally data are processed and analyzed, so that reference and guidance are provided for propeller detection and diagnosis. In addition, ANSYS Fluent software is also utilized to carry out three-dimensional abnormal numerical calculation on the ducted propeller, pressure pulsation monitoring points are set, extracted pressure signals are subjected to Fast Fourier Transform (FFT), and the frequency spectrum characteristics of the ducted propeller in normal working and blade breakage fault states are analyzed, so that the CFD numerical simulation method for predicting the blade breakage fault of the ducted propeller is provided. However, in an actual working environment, noise signals, such as underwater background noise, mechanical operation noise, and the like, are necessarily mixed in the acquired signals, and the noise signals exist in the frequency spectrum in the form of different frequency components, so that if only the frequency spectrum analysis is performed, the obtained frequency spectrum information is necessarily interfered by the noise, and further the fault detection effect is affected.
Disclosure of Invention
The invention aims to provide a CFD and wavelet analysis-based propeller damage detection method for an underwater vehicle.
The technical solution for realizing the purpose of the invention is as follows: a CFD and wavelet analysis-based propeller damage detection method for an underwater vehicle comprises the following steps:
the method comprises the following steps: establishing geometric models of a submarine body, a surrounding shell, a tail rudder, a propeller hub, a propeller blade and a peripheral flow field of the underwater vehicle by using three-dimensional computer aided design software SOLIDWORKS as platform software, and outputting geometric model files;
step two: ANSYS ICEM CFD is used as platform software, the geometric model established in the step one is led into the platform software for geometric body repair, calculation domain division and grid generation, and grid files are output;
step three: carrying out numerical calculation pretreatment, opening a CFX-Pre module of ANSYS CFX software, importing a grid file, establishing a basin, a boundary and an interface, setting a calculation type, calculating total time, time step length, a turbulence model, a convergence threshold, a convergence iteration step number, a domain parameter, a boundary condition and an output physical quantity;
step four: starting CFD simulation, and acquiring the force of the propeller in the y direction, the force of the propeller in the z direction and the monitoring moment in the x direction by taking the head end point of the submersible vehicle as a monitoring point;
step five: and (3) taking MATLAB as platform software to call a wavelet tool box, importing the data obtained in the step four into the MATLAB, performing 6-layer wavelet decomposition by using the wavelet tool box and adopting db5 wavelet, and comparing the data of the complete blade propeller channel with the data of the broken blade propeller to establish a fault database for subsequent fault diagnosis.
Further, in the first step, three-dimensional computer aided design software SOLIDWORKS is used as platform software to establish geometric models of the submarine body, the surrounding shell, the tail vane, the propeller blades and the surrounding flow field of the underwater vehicle and output geometric model files, and the specific method comprises the following steps:
the SOLIDWORKS software is turned on, and various parts of the submersible vehicle are loaded into the software: the submarine comprises a submarine body, a surrounding shell, a tail rudder, a propeller, and part models corresponding to propeller blades, wherein the parts are assembled according to a reference coordinate system, and a basin geometric model is drawn, and the basin is divided into two parts, namely a static basin and a dynamic basin, wherein the static basin is a cuboid, and the submersible vehicle is positioned in the cuboid; the dynamic watershed is a cylinder area divided from a cuboid area, and the propeller part formed by the propeller blades and the propeller driver is surrounded by the dynamic watershed; determining the boundary of the drainage basin, including an inlet boundary, an outlet boundary, an upper boundary, a lower boundary, a left boundary and a right boundary, respectively corresponding to the front and back, the upper and lower surfaces, the left and right surfaces of the cuboid, and the surface boundary of the submersible vehicle, corresponding to the surface of the geometric body of the submersible vehicle, and outputting a geometric model file.
Further, in step two, ANSYS ICEM CFD is used as platform software, the geometric model established in step one is imported into the platform software for geometric body repair, calculation domain division and mesh generation, and a mesh file is output, and the specific method is as follows:
opening ICEM CFD software, importing the geometric file output in the step one into the ICEM CFD software, and repairing the geometric body by utilizing a geometric body repairing function in the software to achieve the purpose of removing the flaw. The static drainage basin is structurally divided, an initial calculation domain is established, the initial calculation domain corresponds to the whole cuboid drainage basin, the initial calculation domain is divided into a plurality of calculation domains according to geometric characteristics of a geometric body, mapping is established between each calculation domain and a corresponding geometric body part, so that mapping from a physical domain to the calculation domains is established, each edge of each calculation domain is subjected to node setting, the structural division of the static drainage basin is completed, and a structural grid is generated; generating an unstructured surface grid and an unstructured body grid in the mobility domain; mesh merging is carried out on the overlapped meshes at the interface of the static watershed and the dynamic watershed; and finally outputting the grid file.
Further, in step three, pretreatment of numerical calculation is performed, a CFX-Pre module of ANSYS CFX software is opened, a mesh file is imported, a basin, a boundary and an interface are established, a calculation type, a total calculation time, a time step length, a turbulence model, a convergence threshold, a convergence iteration step number, a domain parameter, a boundary condition and an output physical quantity are set, and the specific method comprises the following steps:
opening a CFX-Pre module of ANSYS CFX software, creating an example file, entering an example setting interface, setting the calculation type as an unsteady calculation type, setting the total calculation time length to be 40 seconds, the time step length to be 0.01 second, setting the convergence threshold value to be 0.01 percent, setting the convergence iteration step number to be 100 steps, setting a static basin to be a stable basin, setting a dynamic basin to be a rotary basin, setting media in the static basin and the dynamic basin to be constant-temperature fluid water at 25 ℃ degrees, setting turbulence models of the static basin and the dynamic basin to be SST models, setting an inlet boundary to be a speed inlet, setting an outlet boundary to be a pressure outlet, setting upper and lower boundaries and left and right boundaries to be symmetrical surfaces, setting a surface boundary of a sneaker to be a non-slip wall surface, and setting output physical quantities to be speed and pressure.
Further, in the fourth step, CFD simulation is started to obtain a force of the propeller in the y direction, a force received by the propeller in the z direction, and a monitoring moment in the x direction with the first end point of the submersible vehicle as a monitoring point, and the specific method is as follows:
ANSYS CFX software solves a continuity equation and a Navier-Stokes equation through a finite volume method to obtain the force of the propeller in the y direction, the force of the propeller in the z direction and the monitoring torque along the x direction by taking the head end point of the sneaker as a monitoring point, and after calculation is finished, the result data is taken out from the CFX-Solver Manager in a dat format file.
Further, in the fifth step, using MATLAB as platform software to call a wavelet tool box, importing the data obtained in the fourth step into MATLAB, using the wavelet tool box to perform 6-layer wavelet decomposition by adopting db5 wavelet, comparing the complete blade screw pass data with the broken blade screw data, and establishing a fault database for subsequent fault diagnosis, wherein the specific method comprises the following steps:
and (3) opening MATLAB software, importing result data in a dat format, adding the result data into a working area, storing the result data in a mat format, calling an MATLAB wavelet tool box, inputting the result data in the mat format, selecting a db5 wavelet for 6-level wavelet analysis, comparing complete blade propeller warp data with broken blade propeller data, and establishing a fault database for subsequent fault diagnosis.
A CFD and wavelet analysis-based propeller damage detection system of an underwater vehicle is used for detecting the damage of the propeller of the underwater vehicle based on the CFD and wavelet analysis based on any one of the methods.
A computer apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing any one of the methods when executing the computer program for CFD and wavelet analysis based detection of damage to a propeller of a submersible vehicle.
A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements any of the methods for performing CFD and wavelet analysis-based damage detection of a propeller of a submersible vehicle.
Compared with the prior art, the invention has the following remarkable advantages: 1) in the signal processing, a high-frequency signal needs to be analyzed by a narrow time window method, and a low-frequency signal needs to be analyzed by a wide time window method; 2) the underwater vehicle is a closed body, in an underwater working environment, the deeper the depth, the higher the liquid density, the less light can reach, the environment is dim, and the fluid around the propeller moves at a high speed, the flow mode is complex, and the fault detection cannot be performed on the propeller of the vehicle by direct visual detection under the conditions.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Fig. 2 is a schematic view of a submersible vehicle and a watershed.
Fig. 3 is a schematic view of a quiet flow field.
Fig. 4 is a grid schematic of a full blade propeller.
Fig. 5 is a grid schematic of a damaged blade propeller.
FIG. 6 is a schematic diagram of the y-direction force of a full blade propeller and its wavelet analysis results.
FIG. 7 is a schematic diagram of the y-direction force of a propeller with a blade broken by 30% and the wavelet analysis results thereof.
FIG. 8 is a schematic diagram of z-direction stress of a complete blade propeller and wavelet analysis results thereof.
FIG. 9 is a schematic diagram of z-direction force of a propeller with a blade broken by 30% and wavelet analysis results thereof.
FIG. 10 is a schematic illustration of the x-direction monitoring moment of the head end of a full blade propeller submersible and its wavelet analysis results.
FIG. 11 is a schematic diagram of a monitoring moment in the x direction of the head end of a propeller submersible with a blade broken by 30% and a wavelet analysis result thereof.
Fig. 12 is a cross-sectional pressure profile of the entire propeller and the damaged propeller at different positions perpendicular to the axis of rotation, where (a) is the cross-sectional pressure profile at position 4.285 in the direction x of the axis of rotation of the damaged propeller; (b) the pressure distribution diagram of the section of the damaged propeller in the rotating shaft direction x is 4.300; (c) the pressure distribution diagram of the section of the damaged propeller in the rotating shaft direction x is 4.315; (d) the pressure distribution diagram of the section of the complete propeller in the rotating shaft direction x is 4.285; (e) the pressure distribution diagram of the section of the complete propeller in the rotating shaft direction x is 4.300; (f) the pressure distribution diagram of the section of the complete propeller at the position of 4.315 is shown in the rotating shaft direction x, wherein the rotating shaft direction is the x axis.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The invention relates to a CFD and wavelet analysis-based propeller damage detection method for an underwater vehicle, which utilizes computational fluid dynamics simulation to calculate navigation combination of the underwater vehicle, takes ANSYS CFX as platform software to simulate the distribution characteristics, stress conditions and overall influence on the underwater vehicle when the underwater vehicle navigates around the propeller, outputs the result in the form of a physical quantity-time curve, inputs the output result into an MATLAB wavelet analysis tool, adopts db5 wavelet function to carry out 6-level wavelet decomposition, and inputs a signal of force F of the propeller in the y direction y Force F received by the propeller in the z-direction z And a monitoring moment M along the x direction by taking the head end point of the sneaker as a monitoring point x The characteristics of the decomposed signals are combined with the wavelet singularity theory to serve as a detection basis. The method comprises the following specific steps:
the method comprises the following steps: establishing geometric models of a submarine body, a surrounding shell, a tail rudder, a propeller hub, a propeller blade and a peripheral flow field of the underwater vehicle by using three-dimensional computer aided design software SOLIDWORKS as platform software, and outputting geometric model files;
the method for establishing the geometric model comprises the following steps of opening SOLIDWORKS software, and loading various parts of the submersible vehicle into the software: the method comprises the following steps of assembling parts according to a reference coordinate system by using part models corresponding to a hull, a surrounding shell, a tail vane, a propeller hub and a propeller blade, and drawing a watershed geometric model, wherein the watershed is divided into two parts, namely a static watershed and a dynamic watershed; the static drainage basin is a cuboid, and the submersible vehicle is positioned in the cuboid; the dynamic watershed is a cylinder area divided from a cuboid area, the propeller part is surrounded by the dynamic watershed and consists of propeller blades and propeller jacks, and the boundary comprises an inlet boundary, an outlet boundary, an upper boundary, a lower boundary, a left boundary and a right boundary which respectively correspond to the front, the rear, the upper and the lower sides, the left and the right sides of the cuboid; the sneaker surface boundary corresponds to the sneaker geometry surface. And outputting the geometric model file.
Step two: and ANSYS ICEM CFD, taking platform software, importing the geometric model established in the step one into the platform software for geometric body repair, calculation domain division and grid generation, and outputting a grid file.
Opening ICEM CFD software, importing the geometric file output in the step I into the ICEM CFD software, and repairing a geometric body by utilizing a geometric body repairing function in the software to achieve the purpose of removing flaws. The static drainage basin is structurally divided, an initial calculation domain is established, the initial calculation domain corresponds to the whole cuboid drainage basin, the initial calculation domain is divided into a plurality of calculation domains according to geometric characteristics of a geometric body, mapping is established between each calculation domain and a corresponding geometric body part, so that mapping from a physical domain to the calculation domains is established, each edge of each calculation domain is subjected to node setting, the structural division of the static drainage basin is completed, and a structural grid is generated; generating an unstructured surface grid and an unstructured body grid in the mobility domain; mesh merging is carried out on the overlapped meshes at the interface of the static drainage basin and the dynamic drainage basin; and finally outputting the grid file.
Step three: after outputting the grid file, preprocessing the numerical value calculation, opening a CFX-Pre module of ANSYS CFX software, importing the grid file, and establishing a basin, a boundary and an interface by using the functions of the software module after importing the grid file. Setting by using a software control tree on the left side of the software interface: setting a calculation type, a total calculation time, a time step length, a turbulence model, a convergence threshold value and a convergence iteration step number; setting a domain parameter; boundary conditions are set, and output physical quantities are set.
The method specifically comprises the steps of opening a CFX-Pre module of ANSYS CFX software, enabling a newly-built calculation case file to enter a calculation case setting interface, setting a calculation type to be an unsteady calculation type by means of a control tree on the left side of the interface, setting the total calculation time to be 40 seconds, the time step length to be 0.01 second, setting a convergence threshold to be 0.01%, setting the number of convergence iteration steps to be 100 steps, setting a static basin to be a stable basin, setting a dynamic basin to be a rotary basin, setting media in the static basin and the dynamic basin to be constant-temperature fluid water at 25 ℃ degrees, setting turbulence models of the static basin and the dynamic basin to be SST models, setting an inlet boundary to be a speed inlet, setting an outlet boundary to be a pressure outlet, setting an upper boundary, a lower boundary, a left boundary and a right boundary to be symmetrical surfaces, setting a surface boundary of a sneaker to be a non-slip wall surface, and setting output physical quantities to be speed and pressure.
Step four: and starting CFD simulation, and taking out data after the calculation is finished.
Starting CFD numerical simulation, wherein a simulation tool is a CFX-Solver Manager module of ANSYS CFX software, and the specific principle is as follows: the numerical simulation of the propeller by CFD was performed by solving continuity equations and Navier-Stokes equations (N-S equations) by finite volume method, which were established by french scientist c.l.h.m Navier and english scientist G.G Stokes in 1821 and 1845, respectively, in vector form:
Figure BDA0002838262370000061
in the rectangular coordinate system, the following can be expressed:
Figure BDA0002838262370000071
where ρ is the fluid density; v is a velocity vector; p is pressure; u, v, w are velocity components at the point (x, y, z) in the flow field, f is an external force applied to the unit three-dimensional fluid, and if only gravity is considered, f is ρ g, and g is a gravitational acceleration; μ is the kinematic viscosity coefficient.
The finite volume method, also called control volume method, is to divide the calculation area into grids, the grids mentioned here are provided by step two, so that there is a control volume around each grid point which is not repeated, the differential equation to be solved is integrated for each control volume, thereby obtaining a set of discrete equations, and the unknowns in the equations are dependent variables on the grid nodes. The subfield method plus the discretization is the basic idea of the finite volume method. The basic idea of the finite volume method is easy to understand and can directly derive a physical explanation. The physical meaning of a discrete equation is the principle of conservation of a dependent variable in a control volume of finite size, as a differential equation represents the principle of conservation in a control volume where the dependent variable is infinitesimally small. The discrete equation obtained by the finite volume method requires that the integral conservation of the dependent variable is satisfied for any group of control sets, and naturally satisfies for the real calculation region.
The solution method of the finite volume method of the invention adopts the PISO algorithm, which is an algorithm based on the high approximate relation between pressure and velocity corrections. To improve computational efficiency, the PISO algorithm performs two additional corrections, an adjacent correction and a skew correction. The main idea of the PISO algorithm is that after one or more PISO cycles, the corrected velocity tends to more closely satisfy the continuity equation and the momentum equation, and after the solution of the pressure correction equation is initialized, the pressure correction gradient needs to be recalculated, and then the mass flow correction is updated with the recalculated value. The long time required for each cycle and the fact that many iterations are not required is an advantage.
The process of solving the continuity equation and the N-S equation comprises the steps of solving the continuity equation and the N-S equation by using a finite volume method aiming at one time step, comparing and checking obtained numerical solutions to obtain a residual error, calculating after iteration if the residual error is larger than a convergence threshold value, judging the residual error to be convergence if the residual error is smaller than the convergence threshold value, and continuing to carry out numerical calculation aiming at the next time step by using the same method, wherein a turbulence model is added in the process of comparing the residual error with the convergence threshold value to be used as compensation, so that the calculation process is accelerated to converge.
And after the calculation is finished, the result data is taken out from the CFX-Solver Manager in a dat format file.
Step five: and (3) calling a wavelet tool box by taking MATLAB as platform software, importing data into the MATLAB to change the format of the data, inputting the data with the changed format into the wavelet tool box to perform wavelet analysis, performing 6-layer wavelet decomposition on the wavelet analysis by adopting db5 wavelet, comparing complete blade propeller warp data with broken blade propeller data, and establishing a fault database for subsequent fault diagnosis.
The specific method is that MATLAB software is opened, result data in a dat format is imported, needed partial data is selected and added into a working area, and the partial data is stored in a mat format. And (4) calling an MATLAB wavelet tool box, inputting result data in a mat format, and selecting a db5 wavelet to perform 6-level wavelet analysis.
The wavelet analysis principle is as follows:
let F (t) be L 2 (R) is Fourier transformed into F (omega) and F (omega) satisfies the tolerance condition
Figure BDA0002838262370000081
Wherein R is a real number domain; l is 2 (R) represents a function space of a real number domain.
In this case, f (t) is referred to as a base wavelet or mother wavelet. On the basis, a series of wavelets called as a wavelet sequence can be obtained by performing scale transformation and time-varying transformation on the base wavelet, and can be expressed mathematically as:
Figure BDA0002838262370000082
wherein a, b belongs to R; a ≠ 0, a is a scale transform factor, and b is a time-shift transform factor. Wavelet series F a,b (t) acts as an observation window in the signal analysis, so F a,b (t) must satisfy the condition that the function must satisfy in the usual case, i.e.
Figure BDA0002838262370000083
Furthermore, F (ω) must be at the originIs 0. The transformation is a continuous wavelet transform, denoted as W f (a, b) of the form:
Figure BDA0002838262370000084
in practical application, the wavelet needs to be discretized, so that the purpose of saving time can be achieved by realizing wavelet analysis through a computer. To F a,b (t) and Wf ( a, b) are discretized, in effect a and b are discretized, not for time t. Order to
Figure BDA0002838262370000085
j belongs to Z, and Z is an integer set; a is 0 Not equal to 1 and is constant, and a discretized wavelet function can be derived and is marked as W j，k (t) in the form of:
Figure BDA0002838262370000086
wavelet singularity theory:
the singularity is a concept derived from mathematics, and is defined as that if a function is discontinuous at a certain position or a derivative of a certain order is discontinuous, the function has singularity, and a point with the singularity is called a singularity point. The singularities can be characterized by a Lipschitz index, which is defined as: let n be an integer and n<α<n +1, signal x (t) at t 0 Is Lipschitz α if and only if the constants A and h are present 0 >Polynomial P of degree 0 and n n (h) So that for all h<h 0 Are all provided with
|x(t 0 +h)-P n (h)|≤A|h| α (8)
Wherein, P n (h) Is x (t) at t 0 The first n terms taylor series. A is called Lipschitz index and describes the mutation degree of the function at the corresponding point, and the function is smoother when alpha is larger; the smaller the α, the less smooth the function and the more singular the corresponding. If the Lipschitz index of a function at a point is alpha, the Lipschitz index of the derivative of the function at the point is alphaIs alpha-1. The singularity of a function is generally shown by two phenomena, one is that the function itself jumps in function value at a certain point to cause discontinuity, such as a step function; the other is that the first or second derivative of the function jumps at some point resulting in a discontinuity, e.g. an impulse function. In practical situations, a jump component signal exists in a signal function, and the signal is often associated with faults, so that the detection of the jump of the signal function is one of the practical effects of wavelet singularity theory.
Selection of wavelet function:
the selection of the wavelet functions affects the detection effect, because different wavelet functions have different time-frequency characteristics, the detection effect of the same signal by adopting different wavelet functions is different. The selection of the wavelet function should be considered in terms of four aspects of compactness, vanishing moment, regularization and symmetry. When the wavelet function is only in a non-zero limited interval, tight support is the primary consideration, and the larger the tight support interval is, the stronger the capability of reflecting local dynamics of a frequency domain is; the smaller the tight support interval is, the stronger the ability to reflect the local dynamics of the time domain is. If F (t) is k ≦ 0<n, is provided with
Figure BDA0002838262370000091
If yes, it is called F (t) having n-order vanishing rectangles. When a certain signal function x (t) exists, the nth derivative thereof is at t 0 If there is discontinuity, a wavelet function with vanishing moments of order n +1 should be used, and the vanishing moments are related to the clearness of the detection result. In general, the requirement for the regularity of the wavelet function is satisfied if it has sufficient vanishing moments. The symmetry of the wavelet function affects the distortion in the wavelet transform and the inverse transform. In general, when selecting wavelet functions, it is mainly satisfied that there is a certain tight support interval and sufficient vanishing moments. Among wavelet functions, Daubechies wavelet function, also called db wavelet, is a kind of wavelet function with good tight support. The Daubechies wavelet family is constructed by the Belgium scientist Ingrid Daubechies and is generally designated db N Wherein N is the serial number of the wavelet, and takes the value of N as 2,3,4, …, 10. Db with number 1 in the wavelet series 1 Besides wavelets, other wavelet functions have no specific analytical expression, but the square modulus of the transfer function h is unambiguous. The effective set length of the wavelet function and the scale function of the wavelet system is 2N-1, the vanishing moment of the wavelet function is N, and the wavelet function of the wavelet system has orthogonality, compactness and good smoothness. The vanishing rectangle is another important property of the wavelet function, and the Daubechies wavelet has high vanishing moment, which brings about good smoothness. In the sneak fault detection studied in the present invention, Daubechies wavelet functions with tight support in the time domain and high vanishing moments can be the preferred choice.
Examples
To verify the validity of the inventive scheme, the following simulation experiment was performed.
The embodiment follows the flow chart shown in fig. 1 and the steps described above. The calculation type is non-constant calculation, the calculation time is 40 seconds, the time step is 0.01 second, the convergence threshold value is 1e-4, and the convergence step number is 100 steps. FIG. 2 shows a geometric model of a submersible vehicle and a dead basin, wherein the length of the submersible vehicle is 4.35m, and the height of the submersible vehicle is 0.5 m; the static flow field is a cylindrical flow field with the length of 0.1m and the diameter of 0.2 m. The rectangular parallelepiped flow field shown in fig. 3 is a static flow field, and has a length of 20m, a width of 10m, and a height of 10 m. The static watershed is set as a stable watershed, the incoming flow speed of the speed inlet boundary is 1m/s, the reference pressure of the outlet pressure boundary is 0Pa, the upper boundary, the lower boundary, the left boundary and the right boundary are set as interfaces, and the turbulence model is set as an SST model; the flowing basin is set as a rotating basin, the rotating speed is 250 revolutions per minute, the rotating shaft is an x axis, and the turbulence model is set as an SST model. The monitored physical quantity in the calculation is the moment taking the head end of the sneaker as a reference point and the stress of each part. Selecting the force F of the propeller in the y direction from the calculation result y Force F received by the propeller in the z-direction z And a monitoring moment M along the x direction by taking the head end point of the submersible vehicle as a monitoring point x As a wavelet analysis input signal.
Fig. 6 and 7 are comparative group (a), fig. 8 and 9 are comparative group (b), fig. 10 and 11 are comparative group (c), and the comparative db5 is the result of the wavelet 6 level analysis. From d in the results plot 6 To d 1 Corresponding to the decomposed signals from low frequency to high frequency, propeller failure signal oneGenerally of low frequency, so from d 6 And d 5 Can be seen obviously in the decomposed signals, and the analysis result graph of the fault paddle is seen relative to the analysis result graph of the complete paddle at d 6 And d 5 Obvious abrupt changes, namely singular points, appear in the decomposed signals, which accords with the fact that structural damage can correspond to the appearance of the singular points in the wavelet singularity theory.
FIG. 12 is a cross-sectional pressure profile of a 30% broken blade at 30 seconds and a complete propeller at different positions perpendicular to the rotation axis, FIG. 12(d) (e) (f) is a cross-sectional pressure profile of a complete propeller at different positions perpendicular to the rotation axis, and it can be seen from FIG. 12(a) (d) that the pressure profile of the inflow region in front of the propeller is symmetrical when the propeller is not damaged and in normal operation, and a small positive pressure region is arranged in front of the root of each blade, and the distribution shapes of the positive pressure regions are approximately the same; and in the coming flow area in front of the damaged propeller, the area of positive pressure is enlarged in front of the roots of the broken blades and the adjacent blades, and meanwhile, the area of positive pressure in front of the blade roots of other complete blades is reduced, so that the phenomenon that the center of a rotating shaft in the positive pressure area deflects towards the direction of the damaged blade is integrally shown. As can be seen from fig. 12(b) (e), the pressure distribution characteristics of the blade surface and the blade back of each blade of the complete propeller are approximately the same, and a symmetrical characteristic is presented; whereas a broken blade is more concentrated in its immediate area than a perfect blade. From fig. 12(c) (f) it can be seen that the negative pressure area in the wake of the propeller is offset towards the wake behind the damaged blade. From the perspective of the propeller, when one blade is broken, the center of mass of the propeller is shifted, and the moment of inertia of the propeller is changed according to the parallel axis theorem of rigid body rotation, so that the moment in the x direction fluctuates after one blade is broken; from the view of the flow field, the front flow field and the rear flow field of the propeller have positive pressure regions and negative pressure regions which deviate along the direction from the center of the rotating shaft to the damaged blade after one blade is broken, and the pressure distribution deviates. In the process of propeller rotation, the distribution areas of positive pressure and negative pressure correspondingly rotate around the rotating shaft of the propeller, so that the stressed fluctuation of the propeller with one blade broken in the y direction and the z direction is caused, and the rotating frequency of the propeller belongs to low frequency, so that the fluctuation is obviously expressed in the form of singular points in the low frequency component in the wavelet analysis result. The examples demonstrate that the detection method of the present invention is effective.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent application shall be subject to the appended claims.

Claims (7)

1. The underwater vehicle propeller damage detection method based on CFD and wavelet analysis is characterized by comprising the following steps:
the method comprises the following steps: establishing geometric models of a submarine body, a surrounding shell, a tail rudder, a propeller hub, a propeller blade and a peripheral flow field of the underwater vehicle by using three-dimensional computer aided design software SOLIDWORKS as platform software, and outputting geometric model files;
step two: ANSYS ICEM CFD is used as platform software, the geometric model established in the step one is led into the platform software for geometric body repair, calculation domain division and grid generation, and grid files are output;
step three: carrying out numerical calculation pretreatment, opening a CFX-Pre module of ANSYS CFX software, importing a grid file, establishing a basin, a boundary and an interface, setting a calculation type, calculating total time, time step length, a turbulence model, a convergence threshold, a convergence iteration step number, a domain parameter, a boundary condition and an output physical quantity;
step four: starting CFD simulation, and acquiring the force of the propeller in the y direction, the force of the propeller in the z direction and the monitoring moment in the x direction by taking the head end point of the submersible vehicle as a monitoring point;
step five: using MATLAB as platform software to call a wavelet tool box, importing the data obtained in the fourth step into the MATLAB, performing 6-layer wavelet decomposition by using the wavelet tool box and adopting db5 wavelet, comparing the complete blade propeller meridian data with the broken blade propeller data, and establishing a fault database for subsequent fault diagnosis;
in the second step, ANSYS ICEM CFD is used as platform software, the geometric model established in the first step is imported into the platform software for geometric body repair, calculation domain division and grid generation, and a grid file is output, and the specific method comprises the following steps:
opening ICEM CFD software, importing the geometric file output in the step one into the ICEM CFD software, repairing a geometric body by utilizing a geometric body repairing function in the software, structurally dividing a static current domain, establishing an initial calculation domain, wherein the initial calculation domain corresponds to the whole cuboid current domain, dividing the initial calculation domain into a plurality of calculation domains according to geometric characteristics of the geometric body, establishing mapping between each calculation domain and the corresponding geometric body part, thereby establishing mapping from the physical domain to the calculation domain, performing node setting on each edge of the calculation domain, completing structural division of the static current domain, and generating a structural grid; generating an unstructured surface grid and an unstructured body grid in the mobility domain; mesh merging is carried out on the overlapped meshes at the interface of the static drainage basin and the dynamic drainage basin; finally, outputting the grid file;
in the fourth step, CFD simulation is started, and the force of the propeller in the y direction, the force of the propeller in the z direction and the monitoring moment in the x direction with the head end point of the submersible as a monitoring point are obtained, wherein the specific method comprises the following steps:
ANSYS CFX software solves a continuity equation and a Navier-Stokes equation through a finite volume method to obtain the force of the propeller in the y direction, the force of the propeller in the z direction and the monitoring torque in the x direction by taking the head end point of the sneaker as a monitoring point, and after the calculation is finished, the result data is taken out from the CFX-Solver Manager in a dat format file.
2. The method for detecting propeller damage of an underwater vehicle based on CFD and wavelet analysis as claimed in claim 1, wherein in step one, three-dimensional computer aided design software SOLIDWORKS is used as platform software to establish geometric models of a submarine body, a surrounding shell, a tail rudder, a propeller driver, a propeller blade and a flow field around the vehicle, and a geometric model file is output, and the specific method is as follows:
opening SOLIDWORKS software, and loading various parts of the submarine vehicle into the software: the submarine comprises a submarine body, a surrounding shell, a tail rudder, a propeller, and part models corresponding to propeller blades, wherein the parts are assembled according to a reference coordinate system, and a basin geometric model is drawn, and the basin is divided into two parts, namely a static basin and a dynamic basin, wherein the static basin is a cuboid, and the submersible vehicle is positioned in the cuboid; the dynamic watershed is a cylinder area divided from a cuboid area, and the propeller part formed by the propeller blades and the propeller driver is surrounded by the dynamic watershed; determining the boundaries of the drainage basin, including an inlet boundary, an outlet boundary, an upper boundary, a lower boundary, a left boundary and a right boundary, respectively corresponding to the front and back, the upper and lower surfaces, the left and right surfaces of the cuboid, the surface boundary of the submersible vehicle, and the surface of the submersible vehicle geometry, and outputting a geometry model file.
3. The method for detecting propeller damage of an underwater vehicle based on CFD and wavelet analysis of claim 1, wherein in the third step, numerical calculation preprocessing is performed, a CFX-Pre module of ANSYS CFX software is opened, a grid file is imported, a basin, a boundary and an interface are established, calculation types are set, total time, time step length, a turbulence model, a convergence threshold, convergence iteration steps, domain parameters, boundary conditions and output physical quantities are calculated, and the method comprises the following specific steps:
opening a CFX-Pre module of ANSYS CFX software, creating an example file, entering an example setting interface, setting the calculation type to be an unsteady calculation type, setting the total calculation time length to be 40 seconds, setting the time step length to be 0.01 second, setting the convergence threshold value to be 0.01%, setting the convergence iteration step number to be 100 steps, setting a static basin to be a stable basin, setting a dynamic basin to be a rotary basin, setting media in the static basin and the dynamic basin to be constant-temperature fluid water at 25 ℃, setting turbulence models of the static basin and the dynamic basin to be SST models, setting an inlet boundary to be a speed inlet, setting an outlet boundary to be a pressure outlet, setting upper and lower boundaries and left and right boundaries to be symmetrical surfaces, setting a surface boundary of a sneak device to be a non-slip wall surface, and setting output physical quantities to be speed and pressure.
4. The CFD and wavelet analysis-based propeller damage detection method for the underwater vehicle of claim 1, wherein in the fifth step, MATLAB is used as platform software to call a wavelet tool box, the data obtained in the fourth step is imported into the MATLAB, db5 wavelet is adopted by the wavelet tool box to carry out 6-layer wavelet decomposition, and the data of the complete blade screw channel and the data of the broken blade propeller are compared to establish a fault database for subsequent fault diagnosis, and the specific method is as follows:
and (3) opening MATLAB software, importing result data in a dat format, adding the result data into a working area, storing the result data in a mat format, calling an MATLAB wavelet tool box, inputting the result data in the mat format, selecting a db5 wavelet for 6-level wavelet analysis, comparing complete blade propeller warp data with broken blade propeller data, and establishing a fault database for subsequent fault diagnosis.
5. The system for detecting damage to a propeller of an underwater vehicle based on CFD and wavelet analysis is characterized in that the system for detecting damage to the propeller of the underwater vehicle based on CFD and wavelet analysis is performed based on the method of any one of claims 1 to 4.
6. A computer apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1-4 when executing the computer program for performing CFD and wavelet analysis based detection of damage to a propeller of a submersible vehicle.
7. A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any one of claims 1-4 for performing CFD and wavelet analysis based detection of damage to a propeller of an underwater vehicle.
CN202011481561.XA 2020-12-16 2020-12-16 Underwater vehicle propeller damage detection method based on wavelet analysis Active CN112487563B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011481561.XA CN112487563B (en) 2020-12-16 2020-12-16 Underwater vehicle propeller damage detection method based on wavelet analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011481561.XA CN112487563B (en) 2020-12-16 2020-12-16 Underwater vehicle propeller damage detection method based on wavelet analysis

Publications (2)

Publication Number Publication Date
CN112487563A CN112487563A (en) 2021-03-12
CN112487563B true CN112487563B (en) 2022-09-13

Family

ID=74917241

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011481561.XA Active CN112487563B (en) 2020-12-16 2020-12-16 Underwater vehicle propeller damage detection method based on wavelet analysis

Country Status (1)

Country Link
CN (1) CN112487563B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117349970B (en) * 2023-12-05 2024-03-26 招商局金陵船舶(威海)有限公司 Automatic rationality verification method for ship assembly arrangement

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111177903A (en) * 2019-12-18 2020-05-19 智慧航海（青岛）科技有限公司 Propeller thrust performance test method based on simulation technology

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150094975A1 (en) * 2013-10-01 2015-04-02 King Fahd University Of Petroleum And Minerals Wavelet transform system and method for voltage events detection and classification

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111177903A (en) * 2019-12-18 2020-05-19 智慧航海（青岛）科技有限公司 Propeller thrust performance test method based on simulation technology

Also Published As

Publication number Publication date
CN112487563A (en) 2021-03-12

Similar Documents

Publication Publication Date Title
Venditti et al. Grid adaptation for functional outputs: application to two-dimensional inviscid flows
Ku et al. Numerical investigation of tip-vortex cavitation noise of submarine propellers using hybrid computational hydro-acoustic approach
US20130218536A1 (en) Time and Space Scaled S-Model for Turbulent Fluid Flow Simulations
CN111859530A (en) Iterative propulsion disturbance domain updating method for aircraft dynamic aerodynamic characteristic simulation
Kersken et al. Time-Linearized and Time-Accurate 3D RANS Methods for Aeroelastic Analysis in Turbomachinery
CN112487563B (en) Underwater vehicle propeller damage detection method based on wavelet analysis
CN111859529A (en) Multi-grid disturbance domain updating acceleration method for aircraft streaming numerical simulation
Dönmez et al. The effect of inlet blade angle variation on cavitation performance of a centrifugal pump: A parametric study
Nadery et al. Numerical investigation of the hydrodynamic performance of the propeller behind the ship with and without WED
Mehdipour Simulating propeller and propeller-hull interaction in openfoam
CN113378440A (en) Fluid-solid coupling numerical simulation calculation method, device and equipment
Dubas Robust automated computational fluid dynamics analysis and design optimisation of rim driven thrusters
Kersken et al. Flutter analysis of an embedded blade row with a harmonic balance solver
Park et al. Validation of an output-adaptive, tetrahedral cut-cell method for sonic boom prediction
CN111177903A (en) Propeller thrust performance test method based on simulation technology
Petrie-Repar Three-dimensional non-reflecting boundary condition for linearized flow solvers
EP2816332B1 (en) Method and computer product for modeling the sound emission and propagation of systems over a wide frequency range
CN112632726B (en) Flow field reconstruction method for aeroelastic simulation of impeller mechanical blade
CN115292844A (en) Impeller mechanical blade flutter boundary prediction method based on depth map learning
Bruzzone et al. Rudder-propeller interaction: analysis of different approximation techniques
Wang et al. Large-scale parallel simulation with implicit conservative overlapped grid technique for compressor flow
Król et al. Experimental and numerical validation of the improved vortex method applied to CP745 marine propeller model
Heners et al. Evaluating the aerodynamic damping at shock wave boundary layer interacting flow conditions with harmonic balance
Sumi et al. Generalized characteristic interface conditions for high-order multi-block computation
Henninger et al. Time-domain implementation of higher-order non-reflecting boundary conditions for turbomachinery applications

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant