CN110389168A - It is a kind of that detection method is considered to be worth doing based on the engine metal of magnetic detecting principle and inductance method - Google Patents
It is a kind of that detection method is considered to be worth doing based on the engine metal of magnetic detecting principle and inductance method Download PDFInfo
- Publication number
- CN110389168A CN110389168A CN201910665514.1A CN201910665514A CN110389168A CN 110389168 A CN110389168 A CN 110389168A CN 201910665514 A CN201910665514 A CN 201910665514A CN 110389168 A CN110389168 A CN 110389168A
- Authority
- CN
- China
- Prior art keywords
- signal
- detection
- metal
- matrix
- metal fillings
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/72—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables
- G01N27/82—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws
- G01N27/90—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws using eddy currents
- G01N27/9046—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws using eddy currents by analysing electrical signals
Landscapes
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
Detection method is considered to be worth doing based on the engine metal of magnetic detecting principle and inductance method the invention discloses a kind of, it is matched by differential vortex array detection with magnetic field modulation, according to the modulation deviation of metal fillings, the response signal of regularity can be formed in differential vortex array detection coil；Due to the non-intellectual of metal fillings material, granular size etc., complete motion model can not be obtained in advance, therefore, first integrated use Stochastic signal processing, modulation /demodulation, Blind Signal Separation scheduling algorithm, metal fillings response signal is separated and extracted from a series of coils, and reconstructs motion profile, and the size and material of metal fillings are then estimated according to each motion profile, and counted, realize the metal fillings quantitative detection and positioning of engine oil system.
Description
Technical field
The invention belongs to defect detecting technique fields, more specifically, are related to a kind of based on magnetic detecting principle and electricity
The online metal fillings detection method of the engine of sense method.
Background technique
Aero-Engine Lubrication System detection is divided into offline inspection and on-line checking, and common off-line checking method has iron spectrum
Analysis, spectrum analysis etc..But largely work is carried out in laboratory to these offline inspection means, and real-time is low, and right
The severity of failure and development more depend on the subjective judgement of people, big so as to cause its random error, easily report by mistake and leak
Report the metal filings in oil system.Therefore offline inspection means are gradually eliminated, and on-line checking means become research emphasis.
Foreign study person has just carried out engine metal from mid-term the 1980s and has considered grinding for last on-line monitoring technique to be worth doing
Study carefully, emphasis focuses mostly in the online survey sensor exploitation of metal filings.It can be divided into according to principle: (1) magnetic detecting method
(2) online x-ray spectrometry (7) the electrostatic charge detection of inductance method (3) capacitance method (4) supersonic testing method (5) optoelectronic induction method (6)
Method.
(1) magnetic detecting method
Magnetic detecting method principle is that magnetic element is installed in engine oil system, ferromagnetic in absorption or interception lubricating oil
Property metal filings, metal filings accumulation can be such that warning circuit is connected to a certain extent, and the inductive signal generated is strong and weak and metal
Bits amount is directly proportional.Due to magnetic force size to apart from related, its different, oil also different to the reaction force of coercive force of oil viscosity
The content of ferromagnetic material will affect testing result in fluid viscosity and metal fillings.
No. 5 " the MEASURING of volume 37 of 1985 " Industrial Lubrication and Tribology "
A THE DEBRIS " text describes the online metal fillings detector of CDM (continuous Debris Monitor).The sensor
By 1 permanent magnet, the magnetoresistance device composition of 1 bipolar boots and 2 heat pairings.2 temperature sensing are mounted with against mistor
Device compensates zero point and heat drift of sensibility, and entire component is encapsulated in the sleeve of 1 nonferromagnetic with epoxy resin.This
The detectable iron particle greater than 100um of method, but the detection sensitivity of this method and the ferromagnetic material content of metal fillings, magnetism and
Shape is related and leads to poor dimensional precision, and effectively detection zone is influenced by oil viscosity, the metal fillings amount being collected into and lubrication
In oil the relationship with metal fillings amount be difficult to determine, qualitative auxiliary detection means can only be used as.
(2) inductance method
Inductance method principle is that coiling or use planar technology constitute inductance on lubrication duct, is generated just using inductance coil
Beginning magnetic field, when having metal fillings to pass through, magnetic field is disturbed, and the variation in magnetic field is converted into the change of voltage, variable quantity and metal fillings
Size, shape and metal material are closely related.
Volume 3 the 7th " the An online of 1992 " Measurement Science and Technology "
A wear debris monitor " text discloses the resonant inductance method using coil as oscillator, and phaselocked loop is arranged can be certainly
Any slow drift in dynamic compensated oscillator frequency, while can detecte and amplify the temporary disturbance of metallic particles generation.Frequently
The direction of shifting can distinguish ferromagnetism and non-ferromagnetic metal particle, and signal amplitude can measure the size and concentration of fragment.But
Due to the interference by abrasive grain shape, this method can not quantitative detection grit size, press 100 μ m in size range segment processings more,
The quantitative judge that cannot achieve abrasive grain quantity, when abrasive grain generation speed is higher, multiple while abrasive grains by sensor can quilt
Erroneous detection is single abrasive grain, and verification experimental verification abrasive grain distance is less than 5mm and there is erroneous detection.
(3) capacitance method
Capacitance method principle can be changed between grid capacitor two boards when having metal filings to pass through in oil system
Dielectric constant causes capacitor to change, and detects change in dielectric constant caused by metal fillings using impedance analyzer.The party
The advantages of method is at low cost, easy to process, and the dielectric constant of metal material and bubble, water, oil liquid and other non-metallic materials
It differs greatly, therefore capacitive method is insensitive to material property interference, oil liquid color nor affects on detection effect.But pole plate ruler
Very little and distance will affect its detection accuracy, i.e. runner increase will lead to the reduction of its effective resolution, therefore detection accuracy and sampling
There are contradictions for representativeness.
Doctoral thesis " Modeling a microfluidic capacitive sensor for metal in 2009
It is proposed in a wear debris detection in lubrication oil " text a kind of flat based on Al Kut counting principle
Andante MEMS capacitor, the tiny wear particle being able to detect in sub-micrometer range.Experiments verify that the capacitance type transducers can
Detect the metal fillings of 20 μm~6mm." sensor and micro-system " volume 35 the 10th in 2016 is " winged based on capacitance sensor
A machine lubricating oil wear particle detection system design " text devises an arcuation polar-plate-type capacitance sensor, using alternating current bridge-type capacitor
Detection method.The system applies pumping signal to AC bridge with DDS signal generator, through C/V conversion, differential amplification, phase sensitivity
Real-time voltage variation is obtained after the processing such as demodulation and low-pass filtering, is realized the accurate detection of weak signal, is able to detect size and exists
The metal fillings of 100um-500um, and can effectively overcome the influence of stray capacitance and parasitic capacitance.But both methods is in oil circuit
Abrasive grain generate speed it is lower when, since detection interval runner is small, certainly exist false dismissal probability, and abrasive grain generate speed it is higher when,
It to be single abrasive grain by erroneous detection that multiple abrasive grains pass through pole plate simultaneously
(4) ultrasonic detecting technology
Ultrasonic detecting technology principle is that the metal fillings in oil liquid can generate scattering/reflex to ultrasonic wave, so that super
Acoustic attenuation and echo amplitude and metal fillings size it is cube directly proportional, while also with the shape of metal fillings and the close phase of material
It closes.Scattering sensor uses two face paste formula ultrasonic transducers, and one is used to emit, another is for receiving.Due to ultrasonic wave
It can only be scattered by solids, so bubble cannot be identified with this principle.Reflective sensor is made only with a sensor
For transmitting and receiving unit.One sound emission pulse will obtain two echo impulse responses, and one comes from suspended particles, another
From wall surface, the abrasive grain information in engine is identified using the significant difference of two kinds of echo strengths.Scattering principle is relative to anti-
The advantages of penetrating principle is that, due to the natural acoustic impedance of bubble, it can distinguish bubble by the relative phase of bubble.
" Chinese journal of scientific instrument " volume 38 the 7th " aircraft engine oil abrasive grain on-line monitoring " Wen Zhongti in 2017
To a kind of comprehensive transmission case wear particle detection equipment, sensor structure design is removed using the double outlets of oil liquid using gravity principle
Bubble, adjusting screw can detect as low as 45 μm in oil of abrasive grain, by the way that different gains is arranged and gathers for adjusting focal spot parameters
Burnt parameter can survey 3~2000 μm of abrasive grains.But this method is only effective when there is a small amount of foam.In 10mm diameter oil pipe, 1mm
The focal spot of diameter will lead to 99% false dismissal probability.And sensor mounting location, ultrasonic frequency select and vibrate and make an uproar
Influence of sound etc. also can all seriously affect detection performance.
5) optoelectronic induction method
The light transmission of lubricating oil is influenced by metal fillings.Therefore, metal fillings is monitored by measuring the transmission of light.Pass through
Light is measured by the optical density decaying after candidate oil, can quantify, qualitatively detect wear debris.
United States Patent (USP) " On-line measurement of contaminant level in 2000
A lubricating oil " text discloses a kind of optoelectronic induction method, and this method keeps ferromagnetism and nonferromagnetic broken by solenoid
Piece moves respectively in lubricating oil, and is surveyed before and after coil working using LED by the optical system of light source and light perception diode
Light intensity reflect the content of ferromagnetic particle and non-ferric magnetic particle.It is shown based on optical sensor and accurately identifies single grain
The advantage of son.Therefore, they are counted suitable for metal fillings, are widely used in some off-line instruments.However,
For on-line monitoring, optical means may be influenced by many unpredictable factors, such as bubble, oil liquid color, and metal fillings is close
Degree etc., when granule density is high, precision may be cannot keep.
(6) online x-ray spectrometry
Online x-ray spectrometry is excited the principle for generating dispersion spectra, detection gold using metal fillings by X-ray energy
Belong to bits metallic alloying element content.Wherein, mechanical resonant type has no concrete application；Differential pressure type and X-ray type real-time are poor
(more than tens seconds), and it is unable to quantitative detection metal fillings size and number；Microwave-type is normal to water in detection zone and oil liquid dielectric
The interference of the electrical parameters such as number is sensitive；X-ray type is not interfered by environmental factor, but only sensitive to metal element content.Differential pressure
Power type is applied to pollution detection, can overcome the defect that optical method is sensitive to bubble, be widely used in opaque oil liquid detection, but nothing
Method detects metal fillings quantity and maintenance is complicated.
" Tribology International " " Effect of the fabrication of volume 128 in 2018
A method on the wear properties of copper silicon carbide composites " text discloses
The method that embedded X Fluorescence Spectrometer (XRF) detects oil system, this method can quantitatively detect 12 kinds of elements in oil liquid, but XRF
Metal fillings total amount can only be monitored, cannot judge the features such as size, the pattern of metal fillings, and involve great expense, not answer so far
It is detected for sudden abrasion pre-warning.
(7) electrostatic charge detection method
Electrostatic charge detection method principle is that the metal filings in engine oil system have certain electricity through " triboelectrification "
Lotus, the friction conditions such as electrostatic entrained by metal fillings and load, speed, surface roughness are related.Charged metal bits end is by passing
When sensor, charge inducing is generated since electrostatic induction acts on rapier surface.The concentration of charged particle is bigger, and electrostatic signal is got over
By force, it can reflect the abrasion condition of engine gearbox and bearing by the electrostatic signal of metal fillings in monitoring oil liquid.
In master thesis " lubricant passage way wear particle electrostatic on-line monitoring and identification technology research " texts in 2014
The embedded on-line monitoring electrostatic transducer of full flow is devised, lubricating oil circuit can be reflected using the amplitude and waveform of induced voltage
Middle metal, nonmetallic charged abrasive grain feature.But the number that metal fillings is charged, by factors shadows such as temperature, oil liquid dielectric constants
It rings, charged metal fillings is suspended in when flowing in lubricating oil, and oil liquid is charged by friction and contacts charged effect and can make metal fillings charge
Amount further changes, and the process of such secondary charges distributional effects has randomness and complexity, therefore metal fillings size and electricity
Lotus is without stringent quantitative relationship, it is thus possible to which there are erroneous detection and missing inspections.
Summary of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of based on magnetic detecting principle and inductance method
The online metal fillings detection method of engine applies shift perturbation to metal fillings by intelligent magnetic field modulation, and by circumferentially
Differential type eddy current array realize the detection of metal fillings lateral displacement disturbance, realize metal fillings quantity, size, position and form
On-line checking has the characteristics that at low cost, high sensitivity, detection accuracy are high.
For achieving the above object, a kind of considered to be worth doing based on the engine metal of magnetic detecting principle and inductance method of the present invention is examined
Survey method, which comprises the following steps:
(1), low frequency modulating magnetic field is applied to the metal fillings in oil circuit
(1.1), metal fillings on-line checking section is set by a certain section in aviation engine lubricating oil system, then examined
It surveys in section, according to fixed interval arrangement multiple groups low-frequency coil one on the other；
(1.2), periodic square wave signal is applied to low-frequency coil, makes to detect the modulation magnetic for generating in section and attracting metal fillings
?；
(2), the detection coil of arrangement symmetrical above and below is arranged in detection section, to constitute differential vortex array, and examining
The coil-span for surveying section the same side is equal；
(3), according to skin effect, the vortex penetration depth that detection section internal modulation magnetic field generates is calculated；
Wherein, δ is vortex penetration depth, and σ is the magnetic conductivity of material, and μ is the magnetic conductivity of material, and f is field frequency；
(4), according to vortex penetration depth, suitable excitation signal frequency is chosen in conjunction with actual tubing diameter, thus will
Metal fillings displacive modulation information, which is transmitted in differential vortex array, carries out signal detection；
(5), calculus of differences
Differential vortex array and modulation magnetic field under the action of, by the modulation deviation of metal fillings in differential vortex array shape
At the response signal of regularity, it may be assumed that ferromagnetic metal bits are detected section upper half area when moving to by modulation magnetic field suction movement
When, the signal that superposed detection coil measures will be better than the signal that the detection coil positioned at lower part measures, conversely, when fortune
It moves to when detecting section lower half region, the signal that the detection coil positioned at lower part measures can be better than the letter that upper detection coil measures
Number；And non-ferromagnetic metal bits by modulation magnetic field influenced it is smaller, can detection section on continue according to inertia motion；
The amplitude that differential vortex array will test the signal that upper and lower half region of section measures carries out difference, obtains ferromagnetism gold
Belong to the observation signal of bits with non-ferromagnetic metal bits response；
(6), observation signal is modulated demodulation Stochastic signal processing, then signal is stored in observation square by treated
Battle array X,Wherein,Indicate i-th of moment collected observation signal of j-th of detection coil, j=1,2 ..., n, n
Indicate the total number of detection coil in differential vortex array；
(7), Blind Signal Separation is carried out using ICA algorithm；
(7.1), equalization processing is carried out to observing matrix X, obtains matrix X*；
(7.2) covariance matrix of calculating matrix X* obtains its characteristic value Λ and feature vector U；
(7.3), orthogonal matrix Z is constructed；
Z=W0*X
Wherein, W0For whitening matrix, meet:
(7.4), setting initialization weight matrixMaximum number of iterations T' and nonlinear function g (x)；
(7.5), weight matrix when the t times iteration is calculated
Wherein, g'(x) it is led for the single order of g (x), expectation is asked in E { } expression, | | | | indicate vector field homoemorphism；
(7.6), the judgement front and back weight matrix after iteration twiceWithDot product whether be 1, if it is 1,Convergence, enters step (7.7)；Otherwise, it enables the number of iterations t from adding 1, returns again to step (7.5), if working as the number of iterations
When t reaches maximum number of iterations T', then the weight matrix after T' iteration is exportedAnd enter step (7.7)；
(7.7), separation matrix is constructedWherein, the first row of separation matrix indicates ferromagnetic metal bits
Signal matrix is denoted as S1, secondary series indicate non-ferromagnetic metal consider to be worth doing signal matrix, be denoted as S2；
(8), according to S1With S2, drawn respectively using the plot order of software Matlab, wherein the abscissa of image
Indicate sample frequency, ordinate indicates voltage amplitude；Then the fluctuation number in two width figures is counted, the fluctuation number of statistics indicates
The quantity of ferromagnetic metal bits and non-ferromagnetic metal bits；
(9), according to the two images of drafting, by S1With S2Included in signal voltage amplitude and sample frequency conduct
The input of neural network, the output of neural network are the positions and dimensions of ferromagnetic metal bits and non-ferromagnetic metal bits.
Goal of the invention of the invention is achieved in that
The present invention is based on the engine metals of magnetic detecting principle and inductance method to consider detection method to be worth doing, passes through differential vortex array
Detection is matched with magnetic field modulation, and according to the modulation deviation of metal fillings, rule can be formed in differential vortex array detection coil
The response signal of property；Due to the non-intellectual of metal fillings material, granular size etc., complete motion model can not be obtained in advance, because
This, first integrated use Stochastic signal processing, modulation /demodulation, Blind Signal Separation scheduling algorithm are separated and extracted from a series of coils
Metal fillings response signal, and motion profile is reconstructed, the size and material of metal fillings are then estimated according to each motion profile,
And counted, realize the metal fillings quantitative detection and positioning of engine oil system.
Meanwhile also having the present invention is based on the engine metal of magnetic detecting principle and inductance method bits detection method and following have
Beneficial effect:
(1), the present invention applies shift perturbation to metal fillings by developing intelligent magnetic field modulation technique, overcomes conventional method
In passive type extract metal fillings weak electromagnetic signal, and be converted into specific displacive modulation information extraction.
(2), the present invention is retaining conventional three-wire circle electromagnetic induction method by developing differential type eddy current array circumferentially
While the good characteristics such as highly sensitive high reliability wide temperature range, the detection of metal fillings lateral displacement disturbance is realized.
(3), present invention fusion modulation /demodulation, Stochastic signal processing and Blind Signal Separation algorithm, building metal fillings move rail
Mark reduces more abrasive grains in conventional method while passing through the probability for generating erroneous detection missing inspection, realizes that (speed adds based on kinematic parameter
Speed etc.) metal fillings quantity, size, position and form on-line checking.
(4), oil circuit is axially symmetric structure, and by the way that axisymmetric difference coil is arranged, effective compensation temperature etc. joins coil
Several influences.
(5), the present invention can reduce metal fillings detection range using modulation magnetic field, while be superimposed additional position for metal fillings
Move disturbance information.
Detailed description of the invention
Fig. 1 is that the present invention is based on the online metal fillings detection principle diagrams of the engine of magnetic detecting principle and inductance method；
Fig. 2 is that the present invention is based on the online metal fillings detection method flow charts of the engine of magnetic detecting principle and inductance method；
Specific embodiment
A specific embodiment of the invention is described with reference to the accompanying drawing, preferably so as to those skilled in the art
Understand the present invention.Requiring particular attention is that in the following description, when known function and the detailed description of design perhaps
When can desalinate main contents of the invention, these descriptions will be ignored herein.
Embodiment
Fig. 1 is that the present invention is based on the engine metals of magnetic detecting principle and inductance method to consider detection principle diagram to be worth doing.
In the present embodiment, as shown in Figure 1, when, there are when metal filings, being utilized in tested aviation engine lubricating oil road
Intelligent magnetic field modulation technique adjusts magnetic field according to flow velocity and oil viscosity, so that metal fillings is controllably moved in oil circuit.
As shown in Figure 1, when modulating magnetic field closing, metal fillings free movement in oil pipe, when modulating magnetic field unlatching, by
It is different in the mode of action of the modulation magnetic field to ferromagnetism abrasive dust with non-ferromagnetic debris, two kinds of material metal fillings can be caused in oil circuit
In show different motion profiles.When ferromagnetism abrasive dust is after being modulated magnetic fields, meeting occurs bright towards tube wall direction
Aobvious displacement, and non-ferromagnetic debris is influenced small by modulation magnetic field, motion profile is substantially unaffected, and continues to do free movement.
Therefore modulation magnetic fields are that additional shift perturbation information is superimposed for metal fillings while reducing metal fillings detection range.
Fig. 2 is that the present invention is based on the engine metals of magnetic detecting principle and inductance method to consider detection method flow chart to be worth doing.
In the present embodiment, as shown in Fig. 2, a kind of engine metal based on magnetic detecting principle and inductance method of the present invention
Consider detection method to be worth doing, comprising the following steps:
S1, low frequency modulating magnetic field is applied to the metal fillings in oil circuit
S1.1, metal fillings on-line checking section is set by a certain section in aviation engine lubricating oil system, then examined
It surveys in section, according to fixed interval arrangement multiple groups low-frequency coil one on the other；
S1.2, periodic square wave signal is applied to low-frequency coil, makes to detect the modulation magnetic field for generating in section and attracting metal fillings；
In the present embodiment, it is excellent to retain conventional three-wire circle electromagnetic induction method high sensitivity high reliability wide temperature range etc.
While characteristic, the detection of metal fillings lateral displacement disturbance is realized.Field region can be such that ferromagnetic metal considers to be worth doing to tube wall in this way
Direction movement, and non-ferromagnetic metal bits then not by modulation magnetic field effect still according to inertia motion.Entire modulation magnetic field fortune
With Frequency Synchronization, pulse modulation technology so as to adjust magnetic field strength according to flow velocity and oil viscosity, guarantee metal fillings fortune
It will not be adsorbed and fix by magnetic field again while dynamic range maximum.
S2, apply differential vortex array
The probe configuration design of eddy current array is carried out according to the size of detected oil circuit caliber and type face, is popped one's head in by multiple
Independent coil arranges, and arranges the detection coil of arrangement symmetrical above and below in detection section, so that differential vortex array is constituted,
And the coil-span in detection section the same side is equal；This structure is more sensitive to metal fillings motion information, by with modulation magnetic
Field is used in combination, and according to the modulation deviation of metal fillings, the response of regularity can be formed in differential vortex array detection coil
Signal.
In the present embodiment, the number of detection coil at least meets 12 in differential vortex array, equidistant symmetrical respectively
It is placed at left and right sides of the 2nd group of low-frequency coil, i.e., has 6 groups of detection coils between the 1st group and the 2nd group of low-frequency coil, at the 2nd group
There are other 6 groups of detection coils between the 3rd group of low-frequency coil.
S3, the vortex penetration depth generated according to skin effect, calculating detection section internal modulation magnetic field；
Wherein, δ is vortex penetration depth, and σ is the magnetic conductivity of material, and μ is the magnetic conductivity of material, and f is field frequency；
S4, displacive modulation is converted into signal detection
According to vortex penetration depth, suitable excitation signal frequency is chosen in conjunction with actual tubing diameter, due to needs
It is deeper to be vortexed penetration depth, it is therefore desirable to low frequency signal is used, excitation signal frequency can be chosen in the range of 30Hz-150Hz,
This modulation technique is by actively applying shift perturbation to metal fillings, so that metal fillings displacive modulation information is transmitted to difference whirlpool
Signal detection is carried out in stream array.
S5, calculus of differences
Differential vortex array and modulation magnetic field under the action of, by the modulation deviation of metal fillings in differential vortex array shape
At the response signal of regularity, it may be assumed that ferromagnetic metal bits are detected section upper half area when moving to by modulation magnetic field suction movement
When, the signal that superposed detection coil measures will be better than the signal that the detection coil positioned at lower part measures, conversely, when fortune
It moves to when detecting section lower half region, the signal that the detection coil positioned at lower part measures can be better than the letter that upper detection coil measures
Number；And non-ferromagnetic metal bits by modulation magnetic field influenced it is smaller, can detection section on continue according to inertia motion；
The amplitude that differential vortex array will test the signal that upper and lower half region of section measures carries out difference, obtains ferromagnetism gold
Belong to the observation signal of bits with non-ferromagnetic metal bits response；
S6, the reconstruct of metal fillings motion profile
Observation signal is modulated demodulation Stochastic signal processing, then signal is stored in observing matrix X by treated,
Observing matrix X is transmitted to PC machine again；Wherein,Indicate the collected sight of j-th of detection coil of i-th of moment
Signal, j=1,2 ..., n are surveyed, n indicates the total number of detection coil in differential vortex array；
S7, PC machine carry out Blind Signal Separation using ICA algorithm；
S7.1, equalization processing is carried out to observing matrix X, goes mean value i.e. centralization, is substantially to keep mixed signal equal
Value is zero, to obtain matrix X*；
The covariance matrix of S7.2, calculating matrix X* obtain its characteristic value Λ and feature vector U；
S7.3, building orthogonal matrix Z；
Z=W0*X
Wherein, W0For whitening matrix, meet:
S7.4, setting initialization weight matrixMaximum number of iterations T' and nonlinear function g (x)；
In the present embodiment, nonlinear function g (x) is specifically desirable: g (x)=tanh (x) or g (x)=x*exp (- x2/2)
Or g (x)=x3。
Weight matrix when the t times S7.5, calculating iteration
Wherein, g'(x) it is led for the single order of g (x), expectation is asked in E { } expression, | | | | indicate vector field homoemorphism；
S7.6, judgement the front and back weight matrix after iteration twiceWithDot product whether be 1, if it is 1,Convergence, enters step S7.7；Otherwise, it enables the number of iterations t from adding 1, returns again to step S7.5, if working as the number of iterations t
When reaching maximum number of iterations T', then the weight matrix after T' iteration is exportedAnd enter step S7.7；
S7.7, building separation matrixWherein, the first row of separation matrix indicates ferromagnetic metal bits letter
Number matrix, is denoted as S1, secondary series indicate non-ferromagnetic metal consider to be worth doing signal matrix, be denoted as S2；
The quantity of S8, statistics ferromagnetic metal bits and non-ferromagnetic metal bits
According to S1With S2, drawn respectively using the plot order of software Matlab, wherein the abscissa of image indicates
Sample frequency, ordinate indicate voltage amplitude；Then the fluctuation number in two width figures is counted, the fluctuation number of statistics indicates ferromagnetic
Property metal fillings and non-ferromagnetic metal bits quantity；
S9, the positions and dimensions for determining ferromagnetic metal bits and non-ferromagnetic metal bits
In the present embodiment, it is necessary first to neural network is trained and examined, keep the output of neural network different
The metal fillings size of position, to realize accurate estimation, details are not described herein for the process for being trained and examining to neural network；
According to the two images of drafting, by S1With S2Included in signal voltage amplitude and sample frequency as nerve
The input of network, the output of neural network are the positions and dimensions of ferromagnetic metal bits and non-ferromagnetic metal bits.
In this way according to the above method, ferromagnetic metal in aviation engine lubricating oil system is can be monitored in real time in staff
The size of the number and ferromagnetic metal bits and non-ferromagnetic metal bits at different location of bits and non-ferromagnetic metal bits, then
Judge whether to meet aviation engine lubricating oil system context needs, to carry out aviation engine lubricating oil system real
Shi Qingli.
Although the illustrative specific embodiment of the present invention is described above, in order to the technology of the art
Personnel understand the present invention, it should be apparent that the present invention is not limited to the range of specific embodiment, to the common skill of the art
For art personnel, if various change the attached claims limit and determine the spirit and scope of the present invention in, these
Variation is it will be apparent that all utilize the innovation and creation of present inventive concept in the column of protection.
Claims (3)
1. a kind of consider detection method to be worth doing based on the engine metal of magnetic detecting principle and inductance method, which is characterized in that including following
Step:
(1), low frequency modulating magnetic field is applied to the metal fillings in oil circuit
(1.1), metal fillings on-line checking section is set by a certain section in aviation engine lubricating oil system, then in detection section
On, according to fixed interval arrangement multiple groups low-frequency coil one on the other；
(1.2), periodic square wave is applied to low-frequency coil and believes signal, make to detect the modulation magnetic field for generating in section and attracting metal fillings；
(2), the detection coil of arrangement symmetrical above and below is arranged in detection section, to constitute differential vortex array, and in detection section
The coil-span of the same side is equal；
(3), according to skin effect, the vortex penetration depth that detection section internal modulation magnetic field generates is calculated；
Wherein, δ is vortex penetration depth, and σ is the magnetic conductivity of material, and μ is the magnetic conductivity of material；
(4), according to vortex penetration depth, suitable excitation signal frequency is chosen in conjunction with actual tubing diameter, thus by metal
Bits displacive modulation information, which is transmitted in differential vortex array, carries out signal detection；
(5), calculus of differences
Under the action of differential vortex array and modulation magnetic field, the modulation deviation of metal fillings is formed to rule in differential vortex array
The response signal of rule property, it may be assumed that ferromagnetic metal bits are by modulation magnetic field suction movement, when moving to detection section upper half area, position
The signal that detection coil in top measures will be better than the signal that the detection coil positioned at lower part measures, conversely, when moving to
When detecting section lower half region, the signal that the detection coil positioned at lower part measures can be better than the signal that upper detection coil measures；And
Non-ferromagnetic metal bits by modulation magnetic field influenced it is smaller, can detection section on continue according to inertia motion；
The amplitude that differential vortex array will test the signal that upper and lower half region of section measures carries out difference, obtains ferromagnetic metal bits
With the observation signal of non-ferromagnetic metal bits response；
(6), observation signal is modulated demodulation Stochastic signal processing, then signal is stored in observing matrix X by treated,Wherein,Indicate that i-th of moment collected observation signal of j-th of detection coil, j=1,2 ..., n, n indicate
The total number of detection coil in differential vortex array；
(7), Blind Signal Separation is carried out using ICA algorithm；
(7.1), equalization processing is carried out to observing matrix X, obtains matrix X*；
(7.2) calculating matrix X*Covariance matrix, obtain its characteristic value Λ and feature vector U；
(7.3), orthogonal matrix Z is constructed；
Z=W0*X
Wherein, W0For whitening matrix, meet:
(7.4), setting initialization weight matrixMaximum number of iterations T' and nonlinear function g (x)；
(7.5), weight matrix when the t times iteration is calculated
Wherein, g'(x) it is led for the single order of g (x), expectation is asked in E { } expression, | | | | indicate vector field homoemorphism；
(7.6), the judgement front and back weight matrix after iteration twiceWithDot product whether be 1, if it is 1,It receives
It holds back, enters step (7.7)；Otherwise, it enables the number of iterations t from adding 1, returns again to step (7.5), if when the number of iterations t reaches
When maximum number of iterations T', then the weight matrix after T' iteration is exportedAnd enter step (7.7)；
(7.7), separation matrix is constructedWherein, the first row of separation matrix indicates that ferromagnetic metal considers signal square to be worth doing
Battle array, is denoted as S1, secondary series indicate non-ferromagnetic metal consider to be worth doing signal matrix, be denoted as S2；
(8), according to S1With S2, drawn respectively using the plot order of software Matlab, wherein the abscissa of image indicates
Sample frequency, ordinate indicate voltage amplitude；Then the fluctuation number in two width figures is counted, the fluctuation number of statistics indicates ferromagnetic
Property metal fillings and non-ferromagnetic metal bits quantity；
(9), according to the two images of drafting, by S1With S2Included in signal voltage amplitude and sample frequency as nerve
The input of network, the output of neural network are the positions and dimensions of ferromagnetic metal bits and non-ferromagnetic metal bits.
2. according to claim 1 consider detection method to be worth doing based on the engine metal of magnetic detecting principle and inductance method, special
Sign is that the nonlinear function g (x) is specifically desirable: g (x)=tanh (x) or g (x)=x*exp (- x2/ 2) or g (x)=x3。
3. according to claim 1 consider detection method to be worth doing based on the engine metal of magnetic detecting principle and inductance method, special
Sign is, the excitation signal frequency selection purposes are as follows: in the range of 30Hz-150Hz.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910665514.1A CN110389168B (en) | 2019-07-23 | 2019-07-23 | Engine metal scrap detection method based on magnetic detection principle and inductance method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910665514.1A CN110389168B (en) | 2019-07-23 | 2019-07-23 | Engine metal scrap detection method based on magnetic detection principle and inductance method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110389168A true CN110389168A (en) | 2019-10-29 |
| CN110389168B CN110389168B (en) | 2022-11-22 |
Family
ID=68287090
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910665514.1A Active CN110389168B (en) | 2019-07-23 | 2019-07-23 | Engine metal scrap detection method based on magnetic detection principle and inductance method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110389168B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112347635A (en) * | 2020-11-03 | 2021-02-09 | 中国航空工业集团公司北京长城航空测控技术研究所 | Numerical analysis method for electrostatic to weak magnetic field perturbation mechanism |
| CN112362540A (en) * | 2020-10-26 | 2021-02-12 | 重庆邮电大学 | Oil abrasive particle motion trajectory image monitoring system and detection method |
| CN112775722A (en) * | 2020-12-23 | 2021-05-11 | 江苏恒力组合机床有限公司 | Horizontal double-sided machining center |
| CN113219048A (en) * | 2021-07-09 | 2021-08-06 | 西南交通大学 | Steel bridge damage detection system and method based on eddy current and digital twinning technology |
| CN113984600A (en) * | 2021-10-27 | 2022-01-28 | 北京信息科技大学 | High-sensitivity metal wear particle online detection sensor based on magnetostatic iron |
| CN115950558A (en) * | 2022-12-22 | 2023-04-11 | 国家石油天然气管网集团有限公司 | Orthogonal alternating electromagnetism-based pipeline axial stress detection method, device and equipment |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4906925A (en) * | 1979-12-05 | 1990-03-06 | Outokumpu Oy | Apparatus for detecting conductive materials in a moving flow |
| EP1333079A1 (en) * | 2002-02-01 | 2003-08-06 | Delphi Technologies, Inc. | Conductive adhesive material with metallurgically-bonded conductive particles |
| KR20060014008A (en) * | 2004-08-09 | 2006-02-14 | 김현수 | A stick for eyesight impedment person |
| AU2006203385A1 (en) * | 1997-07-15 | 2006-09-07 | Google Inc. | Hand-Held Camera With Ink Cartridge Carrying Image Capture Integrated Circuitry |
| US20100109686A1 (en) * | 2008-10-31 | 2010-05-06 | The University Of Akron | Metal wear detection apparatus and method employing microfluidic electronic device |
| CN102331389A (en) * | 2010-11-30 | 2012-01-25 | 蒋伟平 | High-sensitivity oil abrasive grain on-line monitoring sensor |
| US8203343B1 (en) * | 2005-10-12 | 2012-06-19 | Seektech, Inc. | Reconfigurable portable locator employing multiple sensor array having flexible nested orthogonal antennas |
| CN102728458A (en) * | 2011-04-12 | 2012-10-17 | 邹吉武 | Electromagnetic metal sieving device |
| CN102818754A (en) * | 2012-09-06 | 2012-12-12 | 爱德森（厦门）电子有限公司 | Method and device of improving online monitoring accuracy of engine oil metal abrasive particles |
| CN103646153A (en) * | 2013-12-25 | 2014-03-19 | 哈尔滨工业大学 | Method for describing motion curve of metal particles in alternating magnetic field |
| AU2013234409A1 (en) * | 2012-10-05 | 2014-04-24 | Egrs Technology Pty Ltd | Apparatus and Method for the Separation of Particulates |
| WO2015148981A1 (en) * | 2014-03-28 | 2015-10-01 | The Board Of Trustees Of The University Of Illinois | Label free analyte detection by field effect transistors |
| CN105065085A (en) * | 2015-07-31 | 2015-11-18 | 吉林大学 | On-line engine oil quality monitoring system and method adopting multi-sensor information fusion |
| CN105300853A (en) * | 2015-11-24 | 2016-02-03 | 大连海事大学 | Series resonance type oil liquid metal particle measurement apparatus and series resonance type oil liquid metal particle measurement method |
| CN105628419A (en) * | 2015-12-18 | 2016-06-01 | 国网安徽省电力公司 | System and method of diagnosing GIS (Gas Insulated Switchgear) mechanical defects based on independent component analysis denoising |
| CN106018192A (en) * | 2016-05-12 | 2016-10-12 | 绍兴文理学院 | Double-coil oil monitoring method with wave suppression, separation and molding |
| CN107990820A (en) * | 2017-11-28 | 2018-05-04 | 四川元匠科技有限公司 | A kind of plate thickness information detecting method based on impulse eddy current |
| US20180243756A1 (en) * | 2015-09-10 | 2018-08-30 | University Of Utah Research Foundation | Variable frequency eddy current metal sorter |
| CN208109645U (en) * | 2018-04-04 | 2018-11-16 | 善测（天津）科技有限公司 | A kind of Large Diameter Pipeline high-precision metal bits end on-line monitoring sensor |
| CN109579894A (en) * | 2018-09-20 | 2019-04-05 | 西人马联合测控（泉州）科技有限公司 | A kind of scaling method and system of lubricating oil metal filings sensor |
| CN109759686A (en) * | 2019-03-22 | 2019-05-17 | 哈尔滨工业大学（威海） | A kind of resistance spot welding method under controllable rotating magnetic fields |
-
2019
- 2019-07-23 CN CN201910665514.1A patent/CN110389168B/en active Active
Patent Citations (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4906925A (en) * | 1979-12-05 | 1990-03-06 | Outokumpu Oy | Apparatus for detecting conductive materials in a moving flow |
| AU2006203385A1 (en) * | 1997-07-15 | 2006-09-07 | Google Inc. | Hand-Held Camera With Ink Cartridge Carrying Image Capture Integrated Circuitry |
| EP1333079A1 (en) * | 2002-02-01 | 2003-08-06 | Delphi Technologies, Inc. | Conductive adhesive material with metallurgically-bonded conductive particles |
| US20030146266A1 (en) * | 2002-02-01 | 2003-08-07 | Chaudhuri Arun K. | Conductive adhesive material with metallurgically-bonded conductive particles |
| KR20060014008A (en) * | 2004-08-09 | 2006-02-14 | 김현수 | A stick for eyesight impedment person |
| US8203343B1 (en) * | 2005-10-12 | 2012-06-19 | Seektech, Inc. | Reconfigurable portable locator employing multiple sensor array having flexible nested orthogonal antennas |
| US20100109686A1 (en) * | 2008-10-31 | 2010-05-06 | The University Of Akron | Metal wear detection apparatus and method employing microfluidic electronic device |
| CN102331389A (en) * | 2010-11-30 | 2012-01-25 | 蒋伟平 | High-sensitivity oil abrasive grain on-line monitoring sensor |
| CN102728458A (en) * | 2011-04-12 | 2012-10-17 | 邹吉武 | Electromagnetic metal sieving device |
| CN102818754A (en) * | 2012-09-06 | 2012-12-12 | 爱德森（厦门）电子有限公司 | Method and device of improving online monitoring accuracy of engine oil metal abrasive particles |
| AU2013234409A1 (en) * | 2012-10-05 | 2014-04-24 | Egrs Technology Pty Ltd | Apparatus and Method for the Separation of Particulates |
| CN103646153A (en) * | 2013-12-25 | 2014-03-19 | 哈尔滨工业大学 | Method for describing motion curve of metal particles in alternating magnetic field |
| WO2015148981A1 (en) * | 2014-03-28 | 2015-10-01 | The Board Of Trustees Of The University Of Illinois | Label free analyte detection by field effect transistors |
| US20170176379A1 (en) * | 2014-03-28 | 2017-06-22 | The Board Of Trustees Of The University Of Illinois | Label Free Analyte Detection by Electronic Desalting and Field Effect Transistors |
| CN105065085A (en) * | 2015-07-31 | 2015-11-18 | 吉林大学 | On-line engine oil quality monitoring system and method adopting multi-sensor information fusion |
| US20180243756A1 (en) * | 2015-09-10 | 2018-08-30 | University Of Utah Research Foundation | Variable frequency eddy current metal sorter |
| CN105300853A (en) * | 2015-11-24 | 2016-02-03 | 大连海事大学 | Series resonance type oil liquid metal particle measurement apparatus and series resonance type oil liquid metal particle measurement method |
| CN105628419A (en) * | 2015-12-18 | 2016-06-01 | 国网安徽省电力公司 | System and method of diagnosing GIS (Gas Insulated Switchgear) mechanical defects based on independent component analysis denoising |
| CN106018192A (en) * | 2016-05-12 | 2016-10-12 | 绍兴文理学院 | Double-coil oil monitoring method with wave suppression, separation and molding |
| CN107990820A (en) * | 2017-11-28 | 2018-05-04 | 四川元匠科技有限公司 | A kind of plate thickness information detecting method based on impulse eddy current |
| CN208109645U (en) * | 2018-04-04 | 2018-11-16 | 善测（天津）科技有限公司 | A kind of Large Diameter Pipeline high-precision metal bits end on-line monitoring sensor |
| CN109579894A (en) * | 2018-09-20 | 2019-04-05 | 西人马联合测控（泉州）科技有限公司 | A kind of scaling method and system of lubricating oil metal filings sensor |
| CN109759686A (en) * | 2019-03-22 | 2019-05-17 | 哈尔滨工业大学（威海） | A kind of resistance spot welding method under controllable rotating magnetic fields |
Non-Patent Citations (10)
| Title |
|---|
| LEE SH 等: "Nuclear magnetic resonance in a metal cylinder using an RF eddy current", 《JOURNAL OF THE KOREAN PHYSICAL SOCIETY》 * |
| RUEDIGER G 等: "Tayler instability of toroidal magnetic fields in MHD Taylor-coutte flows", 《ASTRONOMISCHE NACHRICHTEN》 * |
| RUETER DIRK 等: "Induction coil as a non-contacting ultrasound transmitter and detector: Modeling of magnetic fields for improving the performance", 《ULTRASONICS》 * |
| ZHU PEIPEI 等: "Local Sparseness and Image Fusion for Defect Inspection in Eddy Current Pulsed Thermography", 《IEEE SENSORS JOURNAL》 * |
| 冯文文: "滑油磨粒复合传感器结构优化设计", 《中国优秀硕士学位论文全文数据库》 * |
| 常祥 等: "透射式涡流渗透深度的仿真与实验", 《传感器与微系统》 * |
| 张晓菊: "仪器分析技术在变速箱齿轮油检测中的应用", 《汽车零部件》 * |
| 杨可 等: "铌钛碳氮析出物对硬面合金耐高温磨损行为的影响", 《摩擦学学报》 * |
| 杨帆 等: "不锈钢覆面焊缝脉冲涡流热成像检测技术研究", 《科技视界》 * |
| 母文省 等: "航空减速器滑油污染监控探究", 《液压与气动》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112362540A (en) * | 2020-10-26 | 2021-02-12 | 重庆邮电大学 | Oil abrasive particle motion trajectory image monitoring system and detection method |
| CN112362540B (en) * | 2020-10-26 | 2024-06-04 | 重庆矩子兴智能科技有限公司 | Oil abrasive particle motion track image monitoring system and detection method |
| CN112347635A (en) * | 2020-11-03 | 2021-02-09 | 中国航空工业集团公司北京长城航空测控技术研究所 | Numerical analysis method for electrostatic to weak magnetic field perturbation mechanism |
| CN112347635B (en) * | 2020-11-03 | 2023-09-01 | 中国航空工业集团公司北京长城航空测控技术研究所 | Numerical analysis method for weak magnetic field perturbation mechanism by static electricity |
| CN112775722A (en) * | 2020-12-23 | 2021-05-11 | 江苏恒力组合机床有限公司 | Horizontal double-sided machining center |
| CN112775722B (en) * | 2020-12-23 | 2022-02-22 | 江苏恒力组合机床有限公司 | Horizontal double-sided machining center |
| CN113219048A (en) * | 2021-07-09 | 2021-08-06 | 西南交通大学 | Steel bridge damage detection system and method based on eddy current and digital twinning technology |
| CN113219048B (en) * | 2021-07-09 | 2021-09-14 | 西南交通大学 | Steel bridge damage detection system and method based on eddy current and digital twinning technology |
| CN113984600A (en) * | 2021-10-27 | 2022-01-28 | 北京信息科技大学 | High-sensitivity metal wear particle online detection sensor based on magnetostatic iron |
| CN115950558A (en) * | 2022-12-22 | 2023-04-11 | 国家石油天然气管网集团有限公司 | Orthogonal alternating electromagnetism-based pipeline axial stress detection method, device and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110389168B (en) | 2022-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110389168A (en) | It is a kind of that detection method is considered to be worth doing based on the engine metal of magnetic detecting principle and inductance method | |
| Ren et al. | Inductive debris sensor using one energizing coil with multiple sensing coils for sensitivity improvement and high throughput | |
| Wu et al. | Progress and trend of sensor technology for on-line oil monitoring | |
| Du et al. | A high throughput inductive pulse sensor for online oil debris monitoring | |
| Vasic et al. | Pulsed eddy-current nondestructive testing of ferromagnetic tubes | |
| Du et al. | Real-time monitoring of wear debris in lubrication oil using a microfluidic inductive Coulter counting device | |
| CA2204352C (en) | Method and system for analyzing a two-phase flow | |
| Du et al. | Inductive Coulter counting: detection and differentiation of metal wear particles in lubricant | |
| Du et al. | Parallel sensing of metallic wear debris in lubricants using undersampling data processing | |
| Shi et al. | An impedance debris sensor based on a high-gradient magnetic field for high sensitivity and high throughput | |
| Zhang et al. | Research on the output characteristics of microfluidic inductive sensor | |
| Du et al. | On-line wear debris detection in lubricating oil for condition based health monitoring of rotary machinery | |
| Krabicka et al. | Finite-element modeling of electrostatic sensors for the flow measurement of particles in pneumatic pipelines | |
| Qian et al. | Ultrasensitive inductive debris sensor with a two-stage autoasymmetrical compensation circuit | |
| Jia et al. | Online wear particle detection sensors for wear monitoring of mechanical equipment—A review | |
| US11061010B2 (en) | Coaxial capacitive sensor and a method for on-line monitoring and diagnosing engine lubricating oil abrasive particles | |
| CN109187736A (en) | A kind of crude oil pipeline internal corrosion detection system | |
| CN108519268A (en) | Wear particle detection device and method under a kind of lubricating condition | |
| Flanagan et al. | Wear-debris detection and analysis techniques for lubricant-based condition monitoring | |
| Ma et al. | Investigation on the effect of debris position on the sensitivity of the inductive debris sensor | |
| Qian et al. | Interference reducing by low-voltage excitation for a debris sensor with triple-coil structure | |
| Shi et al. | An ultrasensitive debris microsensor for oil health monitoring based on resistance–inductance parameter | |
| Zhang et al. | A method for estimating the composition and size of wear debris in lubricating oil based on the joint observation of inductance and resistance signals: theoretical modeling and experimental verification | |
| Wu | Detection of foreign particles in lubrication oil with a microfluidic chip | |
| CN112345624A (en) | High-sensitivity metal wear particle detection sensor based on giant magnetoresistance effect |
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 |