CN116029131B - Magnetic signal simulation and processing method for underwater metal structure of offshore wind farm - Google Patents
Magnetic signal simulation and processing method for underwater metal structure of offshore wind farm Download PDFInfo
- Publication number
- CN116029131B CN116029131B CN202310027871.1A CN202310027871A CN116029131B CN 116029131 B CN116029131 B CN 116029131B CN 202310027871 A CN202310027871 A CN 202310027871A CN 116029131 B CN116029131 B CN 116029131B
- Authority
- CN
- China
- Prior art keywords
- signal
- wavelet
- magnetic
- metal structure
- signals
- 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
Links
- 239000002184 metal Substances 0.000 title claims abstract description 57
- 238000003672 processing method Methods 0.000 title description 4
- 238000004088 simulation Methods 0.000 title description 4
- 230000006698 induction Effects 0.000 claims abstract description 61
- 238000000034 method Methods 0.000 claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 17
- 238000004364 calculation method Methods 0.000 claims abstract description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 24
- 230000000694 effects Effects 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000009826 distribution Methods 0.000 claims description 6
- 238000009827 uniform distribution Methods 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 238000013139 quantization Methods 0.000 claims description 3
- 230000005358 geomagnetic field Effects 0.000 abstract description 9
- 238000001914 filtration Methods 0.000 abstract description 6
- 230000006870 function Effects 0.000 description 34
- 238000005070 sampling Methods 0.000 description 10
- 239000013535 sea water Substances 0.000 description 9
- 230000008569 process Effects 0.000 description 7
- 108010076504 Protein Sorting Signals Proteins 0.000 description 5
- 230000008859 change Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 230000005672 electromagnetic field Effects 0.000 description 4
- 239000000523 sample Substances 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000005684 electric field Effects 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 230000010355 oscillation Effects 0.000 description 2
- 230000035699 permeability Effects 0.000 description 2
- 238000011158 quantitative evaluation Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 101150004636 SYM1 gene Proteins 0.000 description 1
- 230000008033 biological extinction Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 150000002500 ions Chemical class 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000009659 non-destructive testing Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000009991 scouring Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 101150065184 sym-2 gene Proteins 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Abstract
The invention discloses a method for simulating and processing a magnetic signal of an underwater metal structure of an offshore wind farm, which can be used for simulating a detected magnetic signal in the ocean, filtering an interference induction magnetic field signal caused by the main current speed of ocean current by using a ocean current induction magnetic field calculation model, and filtering a noise induction magnetic field signal caused by the disturbance speed by combining a wavelet threshold denoising method so as to obtain the magnetic signal of the underwater metal structure in a geomagnetic field environment with high signal to noise ratio.
Description
Technical Field
The invention belongs to the technical field of underwater metal structure magnetic field models of offshore wind power, and relates to a method for simulating and processing an underwater metal structure magnetic signal of an offshore wind power plant.
Background
In recent years, with the development of industrial technology, the demand for energy is increasing, and the development of offshore energy is an effective solution. The existing coastal offshore wind power generation facility effectively relieves the pressure of power resources while maintaining the green development concept. The offshore wind power pile foundation is divided into an offshore wind power pile foundation and an underwater wind power pile foundation, wherein the underwater wind power pile foundation is divided into an offshore wind power pile foundation and the underwater wind power pile foundation, the underwater wind power pile foundation is soaked in highly corrosive sea water for a long time, and in addition, corrosive pits and cracks are easily generated due to ocean current scouring or impact, so that great potential safety hazards are generated. The buried cable and the buried oil pipeline on land can be subjected to health monitoring by using a magnetic detection technology, and the pipeline can be ensured to run safely by regular detection. The offshore wind pile foundation also needs to be detected regularly, so that the safety of facilities is ensured.
The earth is a natural magnet, and the sea water in a continuous motion state cuts the geomagnetic field to generate induced current, so that an induced magnetic field is generated, and a marine electromagnetic field environment is formed. The research of marine electromagnetics plays an important role in the development of marine resources and the detection of underwater targets, and the earliest related research can be traced to the 70 th century. With the development of socioeconomic technology and the technological innovation in the related fields, marine electromagnetic technology has been rapidly developed in recent decades. The law of sea water movement cutting geomagnetic field to generate an induction electric field and an induction magnetic field is always a research hot spot, a plurality of scholars construct a sea electromagnetic field model, an analytical equation of the sea water generation induction electromagnetic field in different movement forms is obtained, influences of sea water movement speed, period, water depth and the like on the size of the induction field and the time-frequency characteristic of the sea induction magnetic field are analyzed, and research direction and theoretical basis are provided for denoising of the magnetic detection signal of the underwater metal structure. However, in the research process of the existing ocean electromagnetic field model, the ocean magnetic signal acquisition difficulty is high, and the acquisition equipment and process cost is high.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a method for simulating and processing a magnetic signal of an underwater metal structure of an offshore wind farm, so as to solve the problems of high acquisition difficulty of the ocean magnetic signal and high process cost of acquisition equipment in the prior art.
In order to achieve the purpose, the invention is realized by adopting the following technical scheme:
a method for simulating and processing a magnetic signal of an underwater metal structure of an offshore wind farm comprises the following steps:
collecting original noise-containing magnetic signals of a multi-channel metal structure, wherein the original noise-containing magnetic signals of the metal structure are collected in a still water layer; the original noise-containing signals of the metal structure comprise pure signals, main flow induction magnetic field signals and disturbance induction magnetic field signals;
the main flow induction magnetic field signals are subtracted through vector phase, the disturbance induction magnetic field signals are removed through a wavelet threshold denoising method, and pure signals are obtained and are metal structure magnetic signals.
The invention further improves that:
preferably, the main stream induced magnetic field signals are identical at the same water depth, and the disturbance induced signals are uniformly distributed in compliance with U (-c, c).
Preferably, the calculation formula of the original noise-containing magnetic signal of the metal structure is as follows:
wherein s is 1 (t)、s 2 (t) and s 3 (t) Metal structure original noise-containing magnetic signals acquired by three channels respectively, x 1 (t)、x 2 (t) and x 3 (t) pure signals in three channels, n 1 (t) is a current induced magnetic field signal in three channels; the ocean current induced magnetic field signal is composed of a main current induced magnetic field signal and a disturbance induced magnetic field signal.
Preferably, the calculation formulas of the main flow induction magnetic field signal and the disturbance induction signal are as follows:
wherein B is x1 、B x2 And B x3 The magnetic induction intensities in the x direction of the three channels are respectively A x1 、A x2 And A x3 Coefficients in x direction of three channels, U x1 (-c，c)、U x2 (-c, c) and U x3 (-c, c) respectivelyIs evenly distributed in the x direction of the three channels; b (B) y1 、B y2 And B y3 The magnetic induction intensities in the y directions of the three channels are respectively A y1 、A y2 And A y3 Coefficients in y directions of three channels, U y1 (-c，c)、U y2 (-c, c) and U y3 (-c, c) are respectively an even distribution in the y-direction of the three channels; b (B) z1 、B z2 And B z3 The magnetic induction intensities in the z direction of the three channels are respectively A z1 、A z2 And A z3 Coefficients in the z direction of three channels, U z1 (-c，c)、U z2 (-c, c) and U z3 (-c, c) are respectively a uniform distribution in the z direction of the three channels.
Preferably, the wavelet function threshold denoising method comprises the following steps:
(1) Performing wavelet decomposition on the original metal structure noise-containing magnetic signal through a wavelet base to obtain a layered metal structure original noise-containing magnetic signal;
(2) Based on a set threshold value, carrying out quantization processing on each coefficient of the coefficient through a threshold function to obtain a wavelet coefficient;
(3) And reconstructing the original noise-containing magnetic signals of the metal structures of the layers through wavelet coefficients to obtain final denoising signals.
Preferably, in step (1), the wavelet basis includes db wavelet, bior wavelet, coif wavelet and sym wavelet.
Preferably, a wavelet base with a support length of 5 to 9 is selected.
Preferably, in the step (2), a set threshold value is obtained through a heursure threshold algorithm; the threshold function is either a hard threshold function or a soft threshold function.
Preferably, the method is characterized in that the denoising effect of the original noise-containing magnetic signal of the metal structure is evaluated through signal-to-noise ratio and root mean square error.
Compared with the prior art, the invention has the following beneficial effects:
the invention discloses a method for simulating and processing a magnetic signal of an underwater metal structure of an offshore wind farm, which can be used for simulating a detected magnetic signal in the ocean, filtering an interference induction magnetic field signal caused by the main current speed of ocean current by using a ocean current induction magnetic field calculation model, and filtering a noise induction magnetic field signal caused by the disturbance speed by combining a wavelet threshold denoising method so as to obtain the magnetic signal of the underwater metal structure in a geomagnetic field environment with high signal to noise ratio. Before a magnetic nondestructive testing technology based on weak magnetic signal measurement is implemented on an ocean underwater metal structure, a magnetic field fine change model under the ocean underwater disturbance flow velocity state is constructed, corresponding weak magnetic field fine change is obtained according to the method, and effective processing of ocean underwater metal component magnetic detection signals is further obtained.
Drawings
FIG. 1 is a ground condition model;
FIG. 2 is a main flow velocity model;
FIG. 3 is a model of current induced current;
FIG. 4 is a model of a magnetic signal of an underwater metal structure;
FIG. 5 is a wavelet threshold denoising algorithm flow chart;
FIG. 6 is a schematic diagram of the collection of magnetic signals of a metallic structure in still water;
FIG. 7 is a model of main flow velocity for a particular sea area;
FIG. 8 is a channel one analog marine magnetic signal;
FIG. 9 is a channel two analog marine magnetic signal;
FIG. 10 is a channel three analog marine magnetic signal;
FIG. 11 is a graph showing a comparison of the effects of one of four wavelet functions on a channel;
FIG. 12 is a comparative analysis of the effects of two four wavelet functions of a channel;
fig. 13 is a comparison analysis of the effects of three four wavelet functions on a channel.
Detailed Description
The invention is described in further detail below with reference to the attached drawing figures:
in the description of the present invention, it should be noted that, directions or positional relationships indicated by terms such as "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., are based on directions or positional relationships shown in the drawings, are merely for convenience of description and simplification of description, and do not indicate or imply that the apparatus or element to be referred to must have a specific direction, be constructed and operated in the specific direction, and thus should not be construed as limiting the present invention; the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance; furthermore, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixed or removable, for example; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
Step 1, a magnetic field model under local ocean disturbance flow velocity
According to marine hydrologic data, an interference magnetic field of the ocean in an offshore area mainly comes from a ocean current cutting geomagnetic field, and a ocean current induction magnetic field model is built. The actual ocean current speed is always disturbed near a certain value, the ocean current speed is decomposed into two components of a main current speed and a disturbance speed aiming at the characteristic, the main current speed is usually a 'fixed value', and the disturbance speed is a difference value of a real-time speed deviating from the main current speed; the two velocity components are consistent in direction and are overlapped to form the actual ocean current velocity. In the xyz three-dimensional coordinate system, the positive direction of the z axis points to the seabed ground, the xoy plane with z=0 is the sea level, and the three coordinate axes are orthogonal to each other. The ocean current length is very long, and at any point, ocean current can be regarded as one-dimensional motion, the ocean current length is set along the x axis, and the main current ocean current speed direction is correspondingly set along the positive direction of the x axis. According to the hydrodynamic rules, the ocean current velocity varies continuously in all directions. The ocean current length is far greater than the ocean current thickness, and at any point, the main current speed can be considered to be unchanged along with the x-axis; the ocean current only exists on the ocean surface, namely the main flow speed is 0 above a certain depth, so that the main flow speed has a gradual change process from the maximum value to zero in the z-axis (the water depth direction). The ground condition model is shown in figure 1, the main current speed of the ocean current is represented by v, and the ocean current is from the sea level to the water depth h 1 The main flow speed is a constant value v=v 0 Depth of water h 1 To h 2 The main flow velocity v decreases linearly with increasing water depth, k being the slope of the change and b being the intercept. Depth of water h 2 The main flow velocity v is reduced to 0, i.e. the water depth h 2 Main flow velocity to the seabed ground is 0, h 1 Is usually smaller, h 2 Up to hundreds of meters. Mu is the actual permeability of the seawater, and sigma is the conductivity of the seawater. The main flow velocity model is shown in fig. 2. The disturbance speed is generally distributed in a tiny variation interval, a group of data which is compliant with uniform distribution of U (-c, c) is introduced to simulate the disturbance speed, and U (-c, c) refers to interval [ -c, c)]Any number within etc. may occur, c being on the order of 10 -1 The unit is the same as the main flow speed and is m/s. The disturbance speed may be in either direction xyz, as opposed to the main flow speed. The expression of the actual ocean current velocity V in the x-axis direction as a function of depth is as follows:
wherein U is x (-c, c) is the disturbance velocity in the x-axis direction.
Let the vertical component and horizontal component of geomagnetic field magnetic induction intensity be F z 、F y The current moves in the positive x-axis direction and its main flow velocity v assumes the lorentz force on ions of charge +q:
fz is the geomagnetic field induction intensity horizontal component F y The induced lorentz force is directed in the positive y-axis direction; f (f) y Is the vertical component F of the geomagnetic field magnetic induction intensity z The resulting lorentz force is directed in the positive z-axis direction.
Geomagnetic field horizontal component F y Generating an induced electric field E in the z-direction z Vertical component F z Generating an induced electric field E in the y-direction y As can be obtained from coulomb's law,
because the air medium is not conductive, a current loop cannot be formed in the z direction, and in the y direction, seawater and the ground form a closed loop to form induced current, and the current density J:
J＝σE y ＝σvF z (4)
an xyz three-dimensional rectangular coordinate system is established, the positive direction of the z axis points to the ocean floor, the main current flow direction of ocean current is parallel to the x axis, and J is induction current, as shown in figure 2.
The existence of the induced current must generate an induced magnetic field, and the induced magnetic field at any point from the current source r can be solved according to the law of biot-savory:
wherein μ is the actual permeability of seawater, I is the source current, dl is the tiny line element of the source current, i=js, S is the current cross-sectional area, e r Is the direction vector of the current source pointing to the point to be measured.
Because the ocean current width is very wide, usually more than tens of kilometers, the magnetic induction intensity of a certain point can be abstracted into superposition of the magnetic induction intensity of a plurality of direct current wires with infinite length in the y direction at the point, and a xoz plane is an integration plane. Assuming that the distance between the to-be-measured point and the current source is r, the magnetic induction intensity of the infinitely long straight wire at the distance r is:
the ocean current length is far greater than the ocean current thickness, so that the influence of the ocean current length direction (x-axis direction) on the magnetic induction intensity is equal everywhere, the magnetic induction intensity contribution of the induction current in the limited length interval to the to-be-measured point is calculated, the distance between the limited length interval is set to be 10m, and the magnetic induction intensity contribution of the induction current to the magnetic induction intensity is almost 0 beyond the distance. In the model, it is assumed that the to-be-measured point is positioned in yx 0 A certain point in the z plane is taken as x 0 Each 10m on the left and right is taken as an x-direction integration interval, and the integration interval in the z direction is from the sea level (z=0) to the water depth h where the main current velocity of the ocean becomes 0 2 . The magnetic induction intensity generated by the main flow speed at a certain point in the ocean current can be obtained by combining formulas (4) - (6):
B y ＝0 (8)
B z ＝0 (9)
among them, since the main flow velocity varies with the water depth, the sea current main flow velocity is collectively denoted as v (z).
The formulae (7) to (9) show that the induced magnetic field generated by the ocean current with constant speed is a certain value, namely the induced magnetic field caused by the same ocean current main flow speed in the model is the same. The sea current induction magnetic field is divided into a main current induction magnetic field generated by the main current speed and a disturbance induction magnetic field generated by the disturbance speed corresponding to the main current speed and the disturbance speed, and the disturbance induction magnetic field introduces noise components to the underwater metal structure magnetic signal. The induced magnetic field generated by the perturbation velocity following the uniform distribution is also subject to the uniform distribution by equations (7) - (9), the specific value of which is the perturbation velocity multiplied by a real coefficient a. The ocean current induced magnetic field D is expressed in each direction.
In the scheme, a water area with the water flow speed of O is defined as still water, a multichannel high-precision fluxgate sensor is used for collecting magnetic signals of a metal structure in the still water, and a sampling sequence is expressed by x (t).
The magnetic signal model of the underwater metal structure is shown in fig. 4, and the ocean current induced magnetic field signals collected by each channel are shown in the following formula (11):
wherein B is x1 、B x2 And B x3 The magnetic induction intensities in the x direction of the three channels are respectively A x1 、A x2 And A x3 Coefficients in x direction of three channels, U x1 (-c，c)、U x2 (-c, c) and U x3 (-c, c) are respectively an even distribution in the x-direction of the three channels; b (B) y1 、B y2 And B y3 The magnetic induction intensities in the y directions of the three channels are respectively A y1 、A y2 And A y3 Coefficients in y directions of three channels, U y1 (-c，c)、U y2 (-c, c) and U y3 (-c, c) are respectively an even distribution in the y-direction of the three channels; b (B) z1 、B z2 And B z3 The magnetic induction intensities in the z direction of the three channels are respectively A z1 、A z2 And A z3 Coefficients in the z direction of three channels, U z1 (-c，c)、U z2 (-c, c) and U z3 (-c, c) are respectively a uniform distribution in the z direction of the three channels.
Based on the formula (11), the finally obtained total magnetic field signal sequence containing noise is shown in the following formula (12), and consists of a metal structure magnetic signal sequence in still water and a ocean current induced magnetic field signal.
Subscripts 1, 2, 3 respectively represent three sampling channels, n (t) is a current induced magnetic field signal, x (t) is a metal structure magnetic signal sequence in still water, i.e. a pure signal, and s (t) is a total magnetic field signal sequence, i.e. a current induced magnetic field signal.
2. Array magnetic signal processing method under local ocean disturbance flow velocity
To obtain a magnetic signal of an undisturbed underwater metal structure, the induced magnetic signal generated by ocean currents must be filtered out. The magnitude of the induced magnetic field caused by the main current velocity of the ocean current at any depth is obtained by the formulas (7) - (9), and the magnitude of the induced magnetic field caused by the main current velocity of the ocean current at the same depth is the same. Therefore, aiming at the interference induced magnetic field generated by the main flow speed, the noise component is removed by adopting a method of subtracting the original noise-containing magnetic signal of the underwater metal structure from the vector of the induced magnetic field signal generated by the main flow speed. The induced magnetic field generated by the disturbance speed has uncertainty and cannot be filtered by a vector or algebraic cancellation method, so that the induced magnetic field is filtered by adopting a wavelet threshold denoising algorithm.
(1) The steps of the wavelet threshold denoising method are shown in fig. 5, and the specific steps are as follows:
(1) and selecting a wavelet base according to the characteristics of the sample signal and determining the decomposition layer number to carry out wavelet decomposition on the noise-containing signal.
Let wavelet functionIf a, b continuously change, a family of functions ψ can be obtained from ψ (t) a，b (t). For the signal s (t), x (t) ∈L 2 (R), wavelet transform of signal s (t):
where a is the scale factor, b is the time shift factor,the factor being a normalization constant to ensure conservation of energy of the transformation, i.e
Psi (t) is the basic wavelet, psi a，b (t) is a family of functions generated by the translation and expansion of the basic wavelet, called wavelet basis functions. The time shift b determines the time position of the analysis of s (t) and the scale factor a scales the basic wavelet ψ (t). a is more than 1, the signal waveform contracts, a is less than 1, and the waveform extends.
Examples of the wavelet base that is commonly used include db wavelet, bior wavelet, coif wavelet, sym wavelet, etc., and some of the properties are shown in Table 1. The larger the vanishing moment of the wavelet basis function, the longer the corresponding filter length, the flatter the obtained filter spectrum response passband characteristic, the steeper the transition band, and the more ideal the stopband characteristic. The larger vanishing moment makes the wavelet coefficient of the high-frequency subband smaller, which is beneficial to compressing and eliminating noise. However, the higher the vanishing moment is, the longer the support length of the wavelet basis function is, and the boundary effect is liable to occur. A compromise must be made in terms of moment of extinction and support length, generally selecting wavelets between 5 and 9 support lengths. The orthogonality of the tight wavelet makes the analysis simpler and more convenient, which is beneficial to the accurate reconstruction of the signal.
(2) And setting a proper threshold value, and carrying out quantization treatment on the wavelet coefficients of each layer by using a threshold function to obtain quantized wavelet coefficients. In the wavelet domain, the wavelet coefficients of the real signal tend to be larger than the wavelet coefficients of the noise signal.
Threshold algorithm:
heursure thresholding algorithm: when the signal to noise ratio is small, determining a threshold value by using a Stein-based unbiased risk estimation theory; when the signal-to-noise ratio is large, a fixed threshold algorithm is adopted.
Threshold function:
the hard threshold function is easy to cause the denoising signal to generate additional oscillation, generate jumping points and reduce smoothness; the soft threshold function does not generate additional oscillations but compresses the signal, thereby affecting the approximation of the reconstructed signal domain real signal. In the application process, the hard threshold function or the soft threshold function is selected through a specific effect.
(3) And carrying out wavelet signal reconstruction by using the wavelet coefficient subjected to threshold processing to obtain a denoising signal.
(4) Denoising effect evaluation method
In the signal denoising effect evaluation, the signal-to-noise ratio (SNR) and the Root Mean Square Error (RMSE) are important quantitative evaluation criteria, but the smooth continuity of a reconstructed signal curve is also required to be evaluated, and a reconstructed signal with a high signal-to-noise ratio but poor smoothness is not practical. Therefore, it is necessary to consider the quantitative index and the visual presentation effect in combination.
The signal-to-noise ratio (SNR) and Root Mean Square Error (RMSE) are calculated as follows:
s (t) is an original noise-containing signal, s' (t) is a reconstructed signal (denoised signal), and N is the number of sampling points of the signal.
The invention discloses a magnetic signal simulation processing system of an underwater metal structure of an offshore wind farm, which comprises the following components:
the acquisition module is used for acquiring the original metal structure noise-containing magnetic signals of the multiple channels, and the original metal structure noise-containing magnetic signals are acquired in the still water layer; the original noise-containing signals of the metal structure comprise pure signals, main flow induction magnetic field signals and disturbance induction magnetic field signals;
the denoising module is used for subtracting the main stream induction magnetic field signal through a vector phase and removing the disturbance induction magnetic field signal through a wavelet threshold denoising method to obtain a pure signal which is a metal structure magnetic signal.
(2) Experimental protocol
Simulating three groups of original noise-containing signals, selecting four wavelet functions of db wavelet, bior wavelet, coif wavelet and sym wavelet, and changing wavelet function forms (such as sym1, sym2, … and symN) to perform experiments. Recording the signal-to-noise ratio and root mean square error of the reconstructed signals in different forms of the four wavelet functions, evaluating the smoothness of the reconstructed signals, and comprehensively selecting the wavelet function with the best effect and the form thereof as a fixed algorithm program for denoising the wavelet threshold of the ocean magnetic signal.
A marine magnetic signal simulation and processing method comprises the following specific steps:
s1: simulating an underwater magnetic field signal;
(1) In this example, a multichannel high-precision magnetic sensor is used for collecting magnetic signals of a metal structure in still water, the collection scheme is shown in fig. 6, the scanning is performed at a constant speed along the surface of a metal member, and the scanning path of the example is parallel to the sea level, namely, the whole scanning path is formedThe paths are at the same depth, and the magnetic induction intensity of the surface of the metal structure collected by the probe is set to be parallel to the direction of the magnetic field generated by the main current speed of the ocean current. Collecting a metal structure magnetic field signal x (t) in still water, and collecting a total of three channel signal sequences x 1 (t)，x 2 (t)，x 3 (t) the sampling frequency is 200Hz.
(2) F in this example z ＝3.9×10 4 nT, the depth of water at the scanning path is 22m, and the main current speed model of the sea area is shown in fig. 7. The induced magnetic field B caused by the main current velocity of the ocean current on the path is calculated by the formula 7 x The size was 0.9303nT.
(3) The magnitude of the induced magnetic field caused by the disturbance speed of the sea area is set to be changed within the range from minus 0.3nT to 0.3nT and is compliant with U (-0.3, 0.3) distribution, and then the disturbance induced magnetic field in the three channels can be used as U 1 (-0.3，0.3)，U 2 (-0.3，0.3)，U 3 (-0.3, 0.3) to obtain three original noisy signals s 1 (t)，s 2 (t)，s 3 (t). As shown in fig. 8, 9 and 10.
The original noise-containing signal is constructed by experiments, and in the practical application process, the acquired metal structure original noise-containing magnetic signal is obtained.
S2: denoising the ocean magnetic signals;
interference induction magnetic field signal filtering generated by main flow speed:
in this embodiment, the scanning paths are at the same depth, and the magnetic induction intensity of the metal structure surface collected by the probe is parallel to the direction of the magnetic field generated by the main current flow speed, so that the magnitude direction of the magnetic field generated by the main current flow speed collected by the probe in the whole process is consistent. Removing the disturbing magnetic field generated by the main current velocity of the ocean current according to the principle of vector subtraction, and the formula (17) becomes:
the induced magnetic field signal U (-0.3, 0.3) generated by the disturbance speed is filtered:
(1) Filtering by adopting a wavelet threshold denoising method, and the steps are as follows.
The sampling frequency of the signal is 200Hz, so the number of wavelet decomposition layers is 3.
Order theThe square integrable signal x (t), then the wavelet of s (t) is transformed into:
wherein a is a scale factor, b is a time shift factor, ψ (t) is a basic wavelet, ψ a，b (t) is a family of functions generated by shifting and scaling the basic wavelet.
The support length of the wavelet function is controlled between 5 and 9. The experimental example changes the wavelet function form to perform a plurality of experiments. The mapping relation between the wavelet function form and the support length is shown in table 1.
TABLE 1 several wavelet function Properties
(2) Thresholding
Adopting a heursure threshold algorithm rule with better adaptability: when the signal-to-noise ratio is large, a fixed threshold algorithm is used, and when the signal-to-noise ratio is small, a Stein unbiased risk estimation algorithm is used.
heursure threshold expression:
wherein:s is all wavelet coefficientsN is the number of samples of the signal.
(3) Threshold function
The soft threshold function can effectively avoid interruption, so that the reconstructed signal is smoother and better in continuity. When the absolute value of the wavelet coefficient is larger than a given threshold value, the wavelet coefficient is made to subtract the threshold value; and when the value is smaller than the threshold value, the wavelet coefficient is set to zero.
(4) Signal reconstruction
And (3) performing wavelet inverse transformation on the wavelet coefficient to obtain a reconstruction signal s' (t).
(5) Evaluation of denoising Effect
Signal-to-noise ratio (SNR) and Root Mean Square Error (RMSE) are important criteria for quantitative evaluation of denoising effects.
The calculation formulas of SNR and RMSE can be expressed as:
wherein: s (t) is the original noise-containing signal, s' (t) is the denoised signal (reconstructed signal), and N is the number of sampling points of the signal.
And when the signal ratio and the root mean square error are equal, comparing the smoothness degree of the reconstructed signal, and determining the most suitable algorithm program by combining the signal-to-noise ratio and the smoothness degree of the reconstructed signal curve. The signal ratio and root mean square error of each reconstructed signal for channel one are shown in table 2. The signal ratio and root mean square error of each reconstructed signal for channel 2 are shown in table 3. The signal ratio and root mean square error of each reconstructed signal for channel three are shown in table 4.
TABLE 2 SNR and root mean square error of a reconstructed Signal for channel one
As shown in Table 2, the four wavelet functions in channel one have the highest signal-to-noise ratios of sym3 wavelet 84.51dB, bior3.1 wavelet 88.74dB, coif1 wavelet 85.26dB and dB5 wavelet 95.14dB, respectively. In order to comprehensively measure the noise reduction effect, a group of data with the highest wavelet signal to noise ratio is drawn and analyzed to analyze the smoothness of a noise reduction signal curve. Since the total sampling points are more, the smoothness of the signal curve cannot be judged through the overall view of the signal curve, and 1000 th to 1100 th sampling points are selected for observation. Channel a set of data partial images with the highest four wavelet signal-to-noise ratios is shown in fig. 11.
As shown in Table 3, the four wavelet functions in the second channel have the highest signal-to-noise ratios of 86.71dB for sym3 wavelet, 94.71dB for bior3.1 wavelet, 87.68dB for coif1 wavelet and 99.10dB for dB4 wavelet. And the method is similar to the channel, and 1000 to 1100 sampling points are selected for observation. The partial image of the data set with the highest signal-to-noise ratio of the two four wavelets of the channel is shown in fig. 12.
As shown in Table 4, the four wavelet functions in the third channel have the highest signal-to-noise ratios of sym3 wavelet 87.76dB, bior3.3 wavelet 92.66dB, coif1 wavelet 88.55dB and db4 wavelet 101.53B, respectively. And the method is similar to the channel, and 1000 to 1100 sampling points are selected for observation. The set of data partial images with the highest signal-to-noise ratios of the three four wavelets of the channel is shown in fig. 13.
TABLE 3 channel reconstruction signal SNR and root mean square error
TABLE 4 channel triplets signal to noise ratio and root mean square error channel three
/>
The partial images of the three-channel reconstructed signals are synthesized, the smoothness of the coif wavelet reconstructed signals is poor (marked by red circles), the smoothness of the reconstructed signals of sym wavelets, bd wavelets and bin wavelets is good, and the signal to noise ratio of the reconstructed signals of db wavelets is highest.
In summary, in the ocean interference magnetic field wavelet threshold noise reduction processing, db4 wavelet performance is optimal.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (6)
1. The method for simulating and processing the magnetic signal of the underwater metal structure of the offshore wind farm is characterized by comprising the following steps of:
collecting original noise-containing magnetic signals of a multi-channel metal structure, wherein the original noise-containing magnetic signals of the metal structure are collected in a still water layer; the original noise-containing signals of the metal structure comprise pure signals, main flow induction magnetic field signals and disturbance induction magnetic field signals;
removing the disturbance induction magnetic field signal by subtracting the main stream induction magnetic field signal from the vector phase and removing the disturbance induction magnetic field signal by a wavelet threshold denoising method to obtain a pure signal which is a metal structure magnetic signal;
the calculation formula of the original noise-containing magnetic signal of the metal structure is as follows:
wherein s is 1 (t)、s 2 (t) and s 3 (t) Metal structure original noise-containing magnetic signals acquired by three channels respectively, x 1 (t)、x 2 (t) and x 3 (t) pure signals in three channels, n 1 (t) is a current induced magnetic field signal in three channels; the ocean current induced magnetic field signal is composed of a main current induced magnetic field signal and a disturbance induced magnetic field signalComposition;
the calculation formulas of the main stream induction magnetic field signal and the disturbance induction signal are as follows:
wherein B is x1 、B x2 And B x3 The magnetic induction intensities in the x direction of the three channels are respectively A x1 、A x2 And A x3 Coefficients in x direction of three channels, U x1 (-c,c)、U x2 (-c, c) and U x3 (-c, c) are respectively an even distribution in the x-direction of the three channels; b (B) y1 、B y2 And B y3 The magnetic induction intensities in the y directions of the three channels are respectively A y1 、A y2 And A y3 Coefficients in y directions of three channels, U y1 (-c,c)、U y2 (-c, c) and U y3 (-c, c) are respectively an even distribution in the y-direction of the three channels; b (B) z1 、B z2 And B z3 The magnetic induction intensities in the z direction of the three channels are respectively A z1 、A z2 And A z3 Coefficients in the z direction of three channels, U z1 (-c,c)、U z2 (-c, c) and U z3 (-c, c) are respectively a uniform distribution in the z-direction of the three channels;
the wavelet function threshold denoising method comprises the following steps:
(1) Performing wavelet decomposition on the original metal structure noise-containing magnetic signal through a wavelet base to obtain a layered metal structure original noise-containing magnetic signal;
(2) Based on a set threshold value, carrying out quantization processing on each coefficient of the coefficient through a threshold function to obtain a wavelet coefficient;
(3) And reconstructing the original noise-containing magnetic signals of the metal structures of the layers through wavelet coefficients to obtain final denoising signals.
2. The method for simulating and processing the magnetic signals of the underwater metal structure of the offshore wind farm according to claim 1, wherein the signals of the main flow induction magnetic fields are identical at the same water depth, and the disturbance induction signals are uniformly distributed in U (-c, c).
3. The method for simulating and processing the magnetic signals of the underwater metal structure of the offshore wind farm according to claim 1, wherein in the step (1), the wavelet basis comprises db wavelet, bior wavelet, coif wavelet and sym wavelet.
4. A method for simulating and processing magnetic signals of an underwater metallic structure of a marine wind farm according to claim 3, wherein a wavelet base with a support length of 5-9 is selected.
5. The method for simulating and processing the magnetic signals of the underwater metal structure of the offshore wind farm according to claim 1, wherein in the step (2), a set threshold value is obtained through a heursure threshold algorithm; the threshold function is either a hard threshold function or a soft threshold function.
6. The method for simulating and processing the magnetic signals of the underwater metal structure of the offshore wind farm according to any one of claims 1 to 5, wherein the denoising effect of the original noise-containing magnetic signals of the metal structure is evaluated through signal-to-noise ratio and root mean square error.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310027871.1A CN116029131B (en) | 2023-01-09 | 2023-01-09 | Magnetic signal simulation and processing method for underwater metal structure of offshore wind farm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310027871.1A CN116029131B (en) | 2023-01-09 | 2023-01-09 | Magnetic signal simulation and processing method for underwater metal structure of offshore wind farm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116029131A CN116029131A (en) | 2023-04-28 |
| CN116029131B true CN116029131B (en) | 2024-02-23 |
Family
ID=86073664
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310027871.1A Active CN116029131B (en) | 2023-01-09 | 2023-01-09 | Magnetic signal simulation and processing method for underwater metal structure of offshore wind farm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116029131B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105182429A (en) * | 2015-09-29 | 2015-12-23 | 吉林大学 | Marine controlled-source electromagnetic data seawater disturbance noise correction method |
| CN108490494A (en) * | 2018-03-12 | 2018-09-04 | 中国科学院电子学研究所 | Marine magnetic survey noise suppressing method based on spectrum-subtraction and wavelet analysis |
| CN113253166A (en) * | 2021-04-21 | 2021-08-13 | 西北工业大学 | Magnetic anomaly detection method based on underwater swing platform |
-
2023
- 2023-01-09 CN CN202310027871.1A patent/CN116029131B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105182429A (en) * | 2015-09-29 | 2015-12-23 | 吉林大学 | Marine controlled-source electromagnetic data seawater disturbance noise correction method |
| CN108490494A (en) * | 2018-03-12 | 2018-09-04 | 中国科学院电子学研究所 | Marine magnetic survey noise suppressing method based on spectrum-subtraction and wavelet analysis |
| CN113253166A (en) * | 2021-04-21 | 2021-08-13 | 西北工业大学 | Magnetic anomaly detection method based on underwater swing platform |
Non-Patent Citations (4)
| Title |
|---|
| NDE/SHM of Underwater Structures;P. Rizzo;《Advancesin Science and Technology》;20120930;208-216 * |
| 一种海洋磁异常检测噪声抑制算法;费春娇;张群英;吴佩霖;方广有;朱万华;许鑫;;电子与信息学报(11);全文 * |
| 基于小波阈值的海洋磁力测量数据消噪方法;于波;翟国君;刘雁春;肖付民;;测绘科学(05);第89-91页 * |
| 海洋可控源电磁数据中海水扰动噪声的小波校正方法研究;李肃义;蒋善庆;王跃洋;于生宝;;石油物探(05);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116029131A (en) | 2023-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Song et al. | Second generation wavelet transform for data denoising in PD measurement | |
| CA2593030C (en) | Estimating noise at one frequency by sampling noise at other frequencies | |
| Trichtchenko et al. | Modeling geomagnetically induced currents using geomagnetic indices and data | |
| CN109581516B (en) | Denoising method and system for data of curvelet domain statistic adaptive threshold value ground penetrating radar | |
| Abd Rahman et al. | Autonomous localization of partial discharge sources within large transformer windings | |
| Li et al. | Period-assisted adaptive parameterized wavelet dictionary and its sparse representation for periodic transient features of rolling bearing faults | |
| CN108983287B (en) | Curvelet transform anti-aliasing seismic data reconstruction method based on convex set projection algorithm | |
| CN110908001B (en) | Reconstruction method and system of magnetotelluric sounding signal | |
| CN107356843A (en) | The partial discharge of transformer method for diagnosing faults of small echo is synchronously extruded based on gradient threshold | |
| Lan et al. | Partial discharge location of power cables based on an improved phase difference method | |
| CN116029131B (en) | Magnetic signal simulation and processing method for underwater metal structure of offshore wind farm | |
| US7991570B2 (en) | Poynting-vector filter | |
| Kumar et al. | Variations in wave direction estimated using first and second order Fourier coefficients | |
| Zhang et al. | Suppression of UHF partial discharge signals buried in white-noise interference based on block thresholding spatial correlation combinative de-noising method | |
| Güzel et al. | Application of wavelet-based denoising techniques to remote sensing very low frequency signals | |
| Kolobov et al. | The KVVN-7 multifunction digital measuring station for electromagnetic monitoring of seismoactive zones | |
| CN109725276B (en) | Optical fiber current transformer random error suppression method based on wavelet analysis | |
| CN115826068A (en) | MRS signal envelope extraction method based on self-adaptive Gaussian filtering | |
| CN113556169A (en) | Wavelet threshold denoising method and optical time domain reflectometer based on same | |
| Zhiyu et al. | A time–frequency ridge extraction diagnostic method for composite faults of bearing gears in wind turbine gearboxes | |
| Hussein et al. | Acoustic partial discharge signal denoising using power spectral subtraction | |
| Jiang et al. | Integration of Extreme-point Symmetric Mode Decomposition and Singular Value Decomposition for Dynamic Deflection Denoising of Urban Bridges Obtained by Ground-based Microwave Interferometry. | |
| Liu et al. | Research on a matching detection method for magnetic anomaly of underwater targets | |
| CN117148432B (en) | Shallow profile data space interpolation method based on multi-scale component extraction | |
| Jiaqi et al. | Magnetic anomaly detection based on magnetic gradient orthonormal basis function and wavelet packet |
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 |