CN114997252B - Vehicle-mounted detection method for wheel polygon based on inertia principle - Google Patents

Vehicle-mounted detection method for wheel polygon based on inertia principle Download PDF

Info

Publication number
CN114997252B
CN114997252B CN202210935872.1A CN202210935872A CN114997252B CN 114997252 B CN114997252 B CN 114997252B CN 202210935872 A CN202210935872 A CN 202210935872A CN 114997252 B CN114997252 B CN 114997252B
Authority
CN
China
Prior art keywords
wheel
axle box
signal
components
vertical acceleration
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210935872.1A
Other languages
Chinese (zh)
Other versions
CN114997252A (en
Inventor
陈是扦
王开云
谢博
宋沛泽
凌亮
翟婉明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southwest Jiaotong University
Original Assignee
Southwest Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Southwest Jiaotong University filed Critical Southwest Jiaotong University
Priority to CN202210935872.1A priority Critical patent/CN114997252B/en
Publication of CN114997252A publication Critical patent/CN114997252A/en
Application granted granted Critical
Publication of CN114997252B publication Critical patent/CN114997252B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M17/00—Testing of vehicles
    • G01M17/08—Railway vehicles
    • G01M17/10—Suspensions, axles or wheels
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
    • G06F17/141—Discrete Fourier transforms
    • G06F17/142—Fast Fourier transforms, e.g. using a Cooley-Tukey type algorithm
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/15—Correlation function computation including computation of convolution operations
    • G06F17/156—Correlation function computation including computation of convolution operations using a domain transform, e.g. Fourier transform, polynomial transform, number theoretic transform
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Discrete Mathematics (AREA)
  • Computing Systems (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

The invention discloses a vehicle-mounted detection method for a wheel polygon based on an inertia principle, which comprises the following steps: first, an axle box vertical acceleration signal is acquired and decomposed into a plurality of IMF components. And then, combining the IMF components and the original signals, constructing a rapid independent component analysis observation matrix, calculating to obtain mutually independent components, and screening out effective signal components related to the polygonal excitation of the wheel by adopting a correlation coefficient method. And further, performing secondary integration on the effective signal component based on an inertia principle to obtain an acceleration integration result, and performing trend term removing processing on the acceleration integration result to obtain the radial deviation displacement of the wheel. And finally, the radial deviation displacement of the wheel can be used for quantitatively identifying the order and the amplitude of the polygon of the wheel after fast Fourier transform. The invention is applied to the field of rail transit, realizes the online continuous monitoring of the wheel polygon, and has the characteristics of high efficiency and high precision.

Description

Vehicle-mounted detection method for wheel polygon based on inertia principle
Technical Field
The invention belongs to the technical field of vehicle-mounted detection of wheel polygons, and particularly relates to a vehicle-mounted detection method of wheel polygons based on an inertia principle.
Background
The railway is an important transportation mode, which not only effectively improves the traffic conditions of all areas, but also drives the local economic development and relieves the travel pressure of people. The development of the rail transit field is receiving more and more attention. With the development of railways towards high speed and heavy load, the safety and comfort of rolling stock are more and more emphasized. The wheels of the train are used as the vital components of the rail vehicle, so that the running and the guiding of the vehicle are ensured, and loads in all directions between the vehicle and the rail are borne, so that the stability and the safety of the vehicle in the running process are directly influenced. The polygonal wheel is one of main expression forms of out-of-round and out-of-order wheels, widely exists on the wheels of the rail vehicles, can cause sharp increase of wheel-rail contact force, causes severe vibration of vehicle bodies, influences riding comfort of passengers, shortens service lives of vehicle rail structural components such as steel rails, wheels and wheel shafts, and can cause derailment of trains in severe cases to endanger personal safety of the passengers. Therefore, the monitoring of the polygon of the wheel plays an important role in ensuring the safe and smooth running of the train.
The current detection methods for wheel polygons can be classified into two categories, static detection and dynamic detection. Static detection relies on the manual work to carry out when the train is static or the wheel is dismantled, and detection efficiency is low, and it is great that the detection precision receives human factor to seriously influence the operating efficiency of train. The dynamic detection method does not affect the normal running of the train, has high detection efficiency, and is divided into a trackside detection method and a vehicle-mounted detection method, wherein the trackside detection method can only detect when the speed of the train is low, and can only detect the wheel state when the train passes through. The vehicle-mounted detection method has the advantages that the acceleration sensor is arranged on the key part of the vehicle, the wheel fault diagnosis and identification are carried out based on the vibration response, the continuous on-line monitoring can be realized, the detection efficiency is high, the detection cost is low, and the method is a simple and efficient detection method.
Disclosure of Invention
In order to overcome the defects of the existing detection method, the inventor of the invention provides a vehicle-mounted detection method for the polygon of the wheel based on the inertia principle through long-term exploration and test and continuous improvement and innovation. A signal processing method of Variational Mode Decomposition (VMD) and Fast Independent Component Analysis (FastICA) is utilized to separate an axle box vertical acceleration signal caused by a wheel polygon from an axle box mixed signal, then based on the inertia principle, the separated effective axle box vertical acceleration signal Component is subjected to secondary integration, a trend term after the integration is removed, a wheel radial deviation displacement is obtained, and Fast Fourier Transform (FFT) is carried out on the wheel radial deviation displacement so as to identify the order and the amplitude of the wheel polygon. The wheel polygon detection method provided by the invention has the advantages of high detection precision, high detection efficiency and the like, can accurately and quantitatively detect the order and the amplitude of the wheel polygon, and provides a basis for turning the wheel.
In order to solve the technical problem, the invention provides a vehicle-mounted detection method of a wheel polygon based on an inertia principle, which comprises the following steps of:
1) Signal acquisition: acquiring an axle box vertical acceleration signal of a train in a stable running state, and dividing the axle box vertical acceleration signal by taking a wheel rotation period as a time window;
2) Original signal decomposition: sequentially and adaptively decomposing the axle box vertical acceleration signals in a single time window into K IMF components by using a VMD decomposition method;
3) Effective signal component separation: combining the K IMF components and the axle box vertical acceleration signal to construct a FastICA observation signal, and calculating to obtain M independent components;
4) Screening the effective signal component: screening out independent components related to the wheel polygon by adopting a correlation coefficient method, and determining the independent component with the maximum correlation value as an effective signal component;
5) And (3) calculating the radial deviation displacement of the wheel: performing secondary integration on the effective signal component to obtain an acceleration integration result, and removing a trend term from the acceleration integration result to obtain the radial deviation displacement of the wheel;
6) Order and magnitude estimation of wheel polygons: and sequentially carrying out fast Fourier transform on the radial deviation displacement of the wheel in the rotation period of the wheel to obtain the order and the amplitude of the polygon of the wheel.
Preferably, in the step 2), the VMD decomposition method adaptively decomposes the axle box vertical acceleration signal into an IMF component with sparse characteristics, and comprises the following steps:
step 2.1: calculating vertical acceleration signals of axle box through Hilbert
Figure DEST_PATH_IMAGE001
And constructing the following variation optimization problem with constraints:
Figure DEST_PATH_IMAGE002
in the formula:
Figure DEST_PATH_IMAGE003
represents the sum of the K elements, and the K elements,
Figure DEST_PATH_IMAGE004
represents
Figure DEST_PATH_IMAGE005
Is/are as followsl 2 The number of the norm is calculated,
Figure DEST_PATH_IMAGE006
represents the partial derivative with respect to time and,
Figure DEST_PATH_IMAGE007
stands for dirac
Figure DEST_PATH_IMAGE008
The function, j, represents the unit of an imaginary number,
Figure DEST_PATH_IMAGE009
which is representative of the center frequency of the signal,
Figure DEST_PATH_IMAGE010
representing the IMF components to be estimated, K representing the number of components,
Figure DEST_PATH_IMAGE011
representing a constraint condition, and t represents a time variable;
step 2.2: solving the variation optimization problem with the constraint by using an augmented Lagrange multiplier method, and introducing a secondary penalty parameter
Figure DEST_PATH_IMAGE012
And lagrange multiplier
Figure DEST_PATH_IMAGE013
Changing equation (1) into an unconstrained optimization problem, the augmented lagrange format of equation (1) is expressed as:
Figure DEST_PATH_IMAGE014
wherein the content of the first and second substances,
Figure DEST_PATH_IMAGE015
representing an inner product operation;
step 2.3: the saddle point of the formula (2) is solved by using an alternating multiplier direction method, and the optimization problem is solved by using an iterative algorithm to realize that the kth signal component
Figure DEST_PATH_IMAGE016
And its center frequency
Figure DEST_PATH_IMAGE017
The updating is as follows:
Figure DEST_PATH_IMAGE018
Figure DEST_PATH_IMAGE019
in the formula:
Figure DEST_PATH_IMAGE020
represents
Figure DEST_PATH_IMAGE021
W represents a frequency variable, n represents the number of iterations; obtaining K IMF components of the axle box vertical acceleration signal at the end of the iteration
Figure DEST_PATH_IMAGE022
。
Preferably, in the step 3), fastICA is adopted to calculate independent components, an axle box vertical acceleration signal and IMF component form a FastICA observation matrix X, and M independent components are reconstructed from the FastICA observation matrix X by constructing a separation matrix L
Figure DEST_PATH_IMAGE023
，
Figure DEST_PATH_IMAGE024
Are the individual component numbers.
Preferably, the criterion for screening the effective signal component in the step 4) of the invention is as follows, and the correlation coefficient of the independent component and the axle box vertical acceleration signal is calculated according to the formula (5):
Figure DEST_PATH_IMAGE025
in the formula (I), the compound is shown in the specification,
Figure DEST_PATH_IMAGE026
which represents the operation of the covariance,
Figure DEST_PATH_IMAGE027
it is indicated that the variance operation is performed,
Figure DEST_PATH_IMAGE028
representing the vertical acceleration signal of the axle box,
Figure DEST_PATH_IMAGE029
the independent components are reconstructed and the reconstruction is carried out,
Figure DEST_PATH_IMAGE030
are independent component numbers;
arranging the independent components in sequence from large to small according to the correlation coefficient, and determining the independent component with the maximum correlation coefficient as the effective signal component
Figure DEST_PATH_IMAGE031
。
Preferably, in step 5), the effective signal component of the vertical acceleration signal of the axle box is subjected to quadratic integral filtering, and the transfer function of the integral filter is deduced by adopting a rectangular integral method as follows:
Figure DEST_PATH_IMAGE032
in the formula (I), the compound is shown in the specification,
Figure DEST_PATH_IMAGE033
to sample
Figure DEST_PATH_IMAGE034
Point and previous sample
Figure DEST_PATH_IMAGE035
The time interval between the points is such that,
Figure DEST_PATH_IMAGE036
is the active signal component.
Preferably, the trend term elimination is carried out on the acceleration integral result based on the least square principle in the step 5) of the invention.
Preferably, the specific operation of the present invention for eliminating the trend term from the acceleration integration result is as follows: firstly, a trend term polynomial is listed by using a least square principle to solve an equation; secondly, solving a trend item fitting curve by using a matrix method; and finally, subtracting the trend term from the axle box vertical acceleration signal to eliminate the trend term of the acceleration integration result, and obtaining the wheel radial deviation displacement caused by the wheel polygon.
Preferably, in the step 6), the fast Fourier transform is performed on the radial deviation displacement of the wheel, and the order and the amplitude of the polygon of the wheel are obtained through the time-frequency domain signal conversion.
Compared with the prior art, the invention has the beneficial effects that:
1. the method realizes effective separation of axle box vertical acceleration signals caused by wheel polygons based on the VMD-FastICA, and provides a basis for accurate detection of the wheel polygons.
2. The effective signal components are subjected to secondary integration based on the inertia principle to obtain an acceleration integration result, a trend term is removed from the acceleration integration result, and the obtained wheel radial deviation displacement can be accurately calculated through FFT (fast Fourier transform algorithm) to obtain the order and the amplitude of a wheel polygon.
3. The wheel polygon detection method provided by the invention is a vehicle-mounted dynamic detection method, not only overcomes the defects of the traditional detection method, but also screens out effective signal components related to the wheel polygon by adopting a correlation coefficient method, realizes the quantitative identification of the order and the amplitude of the wheel polygon, and improves the detection precision.
Drawings
FIG. 1 is a flow chart of a vehicle-mounted detection method for wheel polygon based on inertia principle according to the invention,
figure 2 is a graph of the results of a wheel polygon test,
figure 3 shows the VMD decomposition of the vertical axle box acceleration signal,
figure 4 shows the VMD decomposition of the vertical axle box acceleration signal,
figure 5 shows the VMD decomposition of the vertical axle box acceleration signal,
figure 6 shows the VMD decomposition of the vertical axle box acceleration signal,
figure 7 is a result of the independent component calculation,
figure 8 is a result of the independent component calculation,
figure 9 is a result of the independent component calculation,
figure 10 is a result of the independent component calculation,
figure 11 is a result of the independent component calculation,
FIG. 12 shows the order and magnitude recognition results for wheel polygons.
Detailed Description
The following description of specific embodiments of the present invention is provided to facilitate understanding of the technical contents of the present invention by those skilled in the art. The following description is of the preferred embodiments of the present application and it is intended that the scope of the invention not be limited by such specific statements and examples. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification made within the spirit and principles of the present invention should also be considered within the scope of the present application.
Examples
The invention provides a vehicle-mounted detection method for a wheel polygon based on an inertia principle, which has a main working flow chart shown in figure 1 and specifically comprises the following steps:
1) Acquiring a vertical acceleration signal of an axle box of the electric locomotive in a stable running state, and dividing the vertical acceleration signal of the axle box by taking a wheel rotation period as a time window;
in this embodiment, the wheel polygon of the selected electric locomotive is mainly 17 th and 18 th orders, the corresponding acceleration amplitudes are 0.083 mm and 0.068 mm, respectively, and the wheel polygon test result is shown in fig. 2. The method comprises the steps of collecting axle box vertical acceleration signals of the electric locomotive under the condition that the running speed is 60 km/h, setting the sampling frequency to be 4096 Hz, and determining the number of acceleration sampling points in a single wheel rotation period to be 966 according to the radius of a wheel of 0.625 m.
2) The method comprises the following steps of utilizing a VMD decomposition method to adaptively decompose an acquired axle box vertical acceleration signal into a plurality of IMF components:
step 2.1: calculating vertical acceleration signals of axle box through Hilbert
Figure DEST_PATH_IMAGE037
And constructing the following variation optimization problem with constraints:
Figure DEST_PATH_IMAGE038
in the formula:
Figure DEST_PATH_IMAGE039
represents the sum of the K elements, and the K elements,
Figure DEST_PATH_IMAGE040
represents
Figure DEST_PATH_IMAGE041
Isl 2 The norm of the number of the first-order-of-arrival,
Figure DEST_PATH_IMAGE042
represents the partial derivative with respect to time and,
Figure DEST_PATH_IMAGE043
stands for dirac
Figure DEST_PATH_IMAGE044
The function, j, represents the unit of an imaginary number,
Figure DEST_PATH_IMAGE045
which represents the center frequency of the signal at the center,
Figure DEST_PATH_IMAGE046
representing the IMF components to be estimated, K representing the number of components,
Figure DEST_PATH_IMAGE047
representing a constraint condition, and t represents a time variable;
step 2.2: solving the variation optimization problem with constraints by using an augmented Lagrange multiplier method, and introducing a secondary penalty parameter
Figure DEST_PATH_IMAGE048
And lagrange multiplier
Figure DEST_PATH_IMAGE049
Changing equation (1) into an unconstrained optimization problem, the augmented lagrange format of equation (1) is expressed as:
Figure DEST_PATH_IMAGE050
wherein the content of the first and second substances,
Figure DEST_PATH_IMAGE051
representing an inner product operation;
step 2.3: solving the saddle point of the formula (2) by using an alternative multiplier direction method, and solving an optimization problem by using an iterative algorithm to realize that the kth signal component
Figure DEST_PATH_IMAGE052
And its center frequency
Figure DEST_PATH_IMAGE053
The updating is as follows:
Figure DEST_PATH_IMAGE054
Figure DEST_PATH_IMAGE055
in the formula:
Figure DEST_PATH_IMAGE056
represents
Figure DEST_PATH_IMAGE057
W represents a frequency variable, n represents the number of iterations; obtaining K IMF components of the axle box vertical acceleration signal at the end of the iteration
Figure 525143DEST_PATH_IMAGE022
。
In this embodiment, the penalty parameter of the VMD algorithm is set
Figure DEST_PATH_IMAGE058
And the number of decompositions of IMF componentsKFor example, =4, the axle box vertical acceleration signal can be decomposed into 4 IMF components at different center frequencies, and then the data can be substituted into the above formula, and the results are shown in fig. 3-6.
3) Forming a FastICA observation signal by the IMF component and the axle box vertical acceleration signal, and calculating to obtain an independent component;
4) And constructing a FastICA algorithm observation matrix X by using 4 IMF components obtained by VMD decomposition and the axle box vertical acceleration signal, and reconstructing 5 independent components from X by constructing a separation matrix L, as shown in figures 7-11.
5) Screening out effective signal components according to effective signal component screening criteria; the criterion for screening the effective signal component by adopting a correlation coefficient method is as follows:
the correlation coefficient of the independent component and the vertical acceleration signal of the axle box is calculated according to the formula (5):
Figure DEST_PATH_IMAGE059
in the formula (I), the compound is shown in the specification,
Figure DEST_PATH_IMAGE060
which represents the operation of the covariance,
Figure DEST_PATH_IMAGE061
it is shown that the operation of the variance,
Figure DEST_PATH_IMAGE062
representing the vertical acceleration signal of the axle box,
Figure DEST_PATH_IMAGE063
the independent components are reconstructed and the reconstruction is carried out,
Figure DEST_PATH_IMAGE064
are independent component numbers;
in a specific embodiment, correlation coefficients between the 5 independent components and the axle box vertical acceleration signal are calculated by using a correlation coefficient expression, wherein the correlation coefficients are respectively 0.9270, 0.9947, 0.8620, 0.9297 and 0.4926, and the 2 nd independent component with the largest correlation coefficient is used for next analysis to obtain an effective signal component.
6) And (3) calculating the radial deviation displacement of the wheel, namely performing secondary integration on the effective signal component to obtain an acceleration integration result, and removing a trend term from the acceleration integration result to obtain the radial deviation displacement of the wheel.
In a specific embodiment, the sampling frequency is calculated according to the vertical acceleration signal sampling frequency of the axle box
Figure DEST_PATH_IMAGE065
Therefore, the transfer function numerator denominator coefficient of the filter integration method is determined, and the displacement trend term after integration is removed by a least square method.
7) And performing FFT on the radial deviation displacement of the wheel in the wheel rotation period to obtain the order and the amplitude of the wheel polygon.
In a specific embodiment, the FFT is performed on the displacement signal in the wheel rotation period according to the signal sampling frequency to obtain the amplitudes corresponding to the major orders and the respective orders included in the wheel polygon.
As shown in fig. 12, the figure is an order and amplitude recognition result of the wheel polygon in this embodiment, and the recognition result shows that the wheel polygon detection method provided by the present invention can accurately recognize the primary order of the wheel polygon, the root mean square error between the amplitude and the true value of each order is 0.0027, the detection precision is high, and the test result can provide a data basis for whether the wheel needs turning repair.
The above are only preferred embodiments of the present invention, and it should be noted that the above preferred embodiments should not be considered as limiting the present invention, and the protection scope of the present invention should be subject to the scope defined by the claims. It will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the spirit and scope of the invention, and should be considered to be within the scope of the invention.

Claims (8)

1. A vehicle-mounted detection method for a wheel polygon based on an inertia principle is characterized by comprising the following steps:
1) Signal acquisition: acquiring an axle box vertical acceleration signal of a train in a stable running state, and dividing the axle box vertical acceleration signal by taking a wheel rotation period as a time window;
2) Original signal decomposition: sequentially and adaptively decomposing the axle box vertical acceleration signals in a single time window into K IMF components by using a VMD decomposition method;
3) Effective signal component separation: combining the K IMF components and the axle box vertical acceleration signal to construct a FastICA observation signal, and calculating to obtain M independent components;
4) Screening the effective signal component: screening out independent components related to the wheel polygon by adopting a correlation coefficient method, and determining the independent component with the maximum correlation value as an effective signal component;
5) And (3) calculating the radial deviation displacement of the wheel: performing secondary integration on the effective signal component to obtain an acceleration integration result, and removing a trend term from the acceleration integration result to obtain the radial deviation displacement of the wheel;
6) Order and magnitude estimation of wheel polygons: and sequentially carrying out fast Fourier transform on the radial deviation displacement of the wheel in the wheel rotation period to obtain the order and the amplitude of the wheel polygon.
2. The vehicle-mounted detection method of the wheel polygon based on the inertia principle as claimed in claim 1, wherein: in the step 2), the VMD decomposition method adaptively decomposes the axle box vertical acceleration signal into IMF components with sparse characteristics, and comprises the following steps:
1) Calculating vertical acceleration signals of axle box through Hilbert
Figure RE-161678DEST_PATH_IMAGE001
And constructing the following variation optimization problem with constraints:
Figure RE-406715DEST_PATH_IMAGE002
in the formula:
Figure RE-452031DEST_PATH_IMAGE003
represents the sum of the K elements, and the K elements,
Figure RE-50503DEST_PATH_IMAGE004
represent
Figure RE-130454DEST_PATH_IMAGE005
Is/are as followsl 2 The norm of the number of the first-order-of-arrival,
Figure RE-31546DEST_PATH_IMAGE006
represents the partial derivative with respect to time and,
Figure RE-298579DEST_PATH_IMAGE007
stands for dirac
Figure RE-763058DEST_PATH_IMAGE008
The function, j, represents the unit of an imaginary number,
Figure RE-635199DEST_PATH_IMAGE009
which represents the center frequency of the signal at the center,
Figure RE-222038DEST_PATH_IMAGE010
representing the IMF components to be estimated, K representing the number of components,
Figure RE-241947DEST_PATH_IMAGE011
representing a constraint condition, and t represents a time variable;
2) Solving the variation optimization problem with the constraint by using an augmented Lagrange multiplier method, and introducing a secondary penalty parameter
Figure RE-182221DEST_PATH_IMAGE012
And lagrange multiplier
Figure RE-705606DEST_PATH_IMAGE013
Changing equation (1) to an unconstrained optimization problem, the augmented lagrange format of equation (1) is expressed as:
Figure RE-401030DEST_PATH_IMAGE014
wherein, the first and the second end of the pipe are connected with each other,
Figure RE-219819DEST_PATH_IMAGE015
representing an inner product operation;
3) Solving the saddle point of the formula (2) by using an alternative multiplier direction method, and solving an optimization problem by using an iterative algorithm to realize that the kth signal component
Figure RE-760522DEST_PATH_IMAGE016
And its center frequency
Figure RE-607255DEST_PATH_IMAGE017
The updating is as follows:
Figure RE-208001DEST_PATH_IMAGE018
Figure RE-264818DEST_PATH_IMAGE019
in the formula:
Figure RE-343633DEST_PATH_IMAGE020
represents
Figure RE-107190DEST_PATH_IMAGE021
Fourier transform of (2)In other words, w represents a frequency variable, and n represents the number of iterations; obtaining K IMF components of the axle box vertical acceleration signal at the end of the iteration
Figure RE-82099DEST_PATH_IMAGE022
。
3. The method as claimed in claim 1, wherein FastICA is used to calculate independent components in step 3), fastICA observation matrix X is formed by axle box vertical acceleration signal and IMF component, and M independent components are reconstructed from FastICA observation matrix X by constructing separation matrix L
Figure RE-298317DEST_PATH_IMAGE023
，
Figure RE-259450DEST_PATH_IMAGE024
Are the individual component numbers.
4. The method as claimed in claim 1, wherein the criteria for selecting the effective signal components in step 4) are as follows, and the correlation coefficient between the independent components and the axle box vertical acceleration signal is calculated according to equation (5):
Figure RE-877514DEST_PATH_IMAGE025
in the formula (I), the compound is shown in the specification,
Figure RE-23324DEST_PATH_IMAGE026
which represents the operation of the covariance,
Figure RE-726838DEST_PATH_IMAGE027
it is indicated that the variance operation is performed,
Figure RE-475351DEST_PATH_IMAGE028
representing the vertical acceleration signal of the axle box,
Figure RE-213500DEST_PATH_IMAGE029
the independent components are reconstructed and the reconstruction is carried out,
Figure RE-530212DEST_PATH_IMAGE030
are independent component numbers;
arranging the independent components in sequence from large to small according to the correlation coefficient, and determining the independent component with the maximum correlation coefficient as the effective signal component
Figure RE-721022DEST_PATH_IMAGE031
。
5. The vehicle-mounted detection method for the wheel polygon based on the inertial principle as claimed in claim 1, wherein: in the step 5), effective signal components of the vertical acceleration signals of the axle box are subjected to quadratic integral filtering, and a transfer function of the integral filter is deduced by a rectangular integral method as follows:
Figure RE-945330DEST_PATH_IMAGE032
in the formula (I), the compound is shown in the specification,
Figure RE-583990DEST_PATH_IMAGE033
to sample
Figure RE-133920DEST_PATH_IMAGE034
Point and previous sampling
Figure RE-749709DEST_PATH_IMAGE035
The time interval between the points of the image,
Figure RE-43288DEST_PATH_IMAGE036
is the effective signal component.
6. The vehicle-mounted detection method of the wheel polygon based on the inertia principle as claimed in claim 5, wherein: and in the step 5), trend term elimination is carried out on the acceleration integral result based on the least square principle.
7. The vehicle-mounted inertia-principle-based wheel polygon detection method as claimed in claim 6, wherein the trend term elimination is performed on the acceleration integral result as follows: firstly, a trend term polynomial is listed by using a least square principle to solve an equation; secondly, solving a trend item fitting curve by using a matrix method; and finally, subtracting the trend term from the axle box vertical acceleration signal to eliminate the trend term of the acceleration integration result, and obtaining the wheel radial deviation displacement caused by the wheel polygon.
8. The vehicle-mounted detection method for the wheel polygon based on the inertial principle as claimed in claim 1, wherein: and 6) performing fast Fourier transform on the radial deviation displacement of the wheel, and converting according to the time-frequency domain signal to obtain the order and the amplitude of the wheel polygon.
CN202210935872.1A 2022-08-05 2022-08-05 Vehicle-mounted detection method for wheel polygon based on inertia principle Active CN114997252B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210935872.1A CN114997252B (en) 2022-08-05 2022-08-05 Vehicle-mounted detection method for wheel polygon based on inertia principle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210935872.1A CN114997252B (en) 2022-08-05 2022-08-05 Vehicle-mounted detection method for wheel polygon based on inertia principle

Publications (2)

Publication Number Publication Date
CN114997252A CN114997252A (en) 2022-09-02
CN114997252B true CN114997252B (en) 2022-10-25

Family

ID=83023144

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210935872.1A Active CN114997252B (en) 2022-08-05 2022-08-05 Vehicle-mounted detection method for wheel polygon based on inertia principle

Country Status (1)

Country Link
CN (1) CN114997252B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116252820B (en) * 2023-05-12 2023-09-05 西南交通大学 Polygonal quantitative detection method for high-speed train wheels driven by improved frequency domain integration method
CN116353660B (en) * 2023-06-01 2023-08-22 兰州交通大学 High-speed railway wheel polygon fault detection method and system based on BWO-VMD

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106250613A (en) * 2016-07-28 2016-12-21 南京理工大学 A kind of wheel service state security domain is estimated and method for diagnosing faults
CN107650945A (en) * 2017-09-19 2018-02-02 华东交通大学 A kind of recognition methods of wheel polygon and its device based on vertical wheel rail force
CN109059840A (en) * 2018-05-29 2018-12-21 南京理工大学 A kind of city rail vehicle wheel out of round is along detection method
CN112381027A (en) * 2020-11-23 2021-02-19 西南交通大学 Wheel polygon wave depth estimation method based on train axle box vertical acceleration signal
RO134895A2 (en) * 2019-10-04 2021-04-29 Albert Mihai Suvac Self-propelled electric tricycle to be combined with small agricultural equipments
CN112991577A (en) * 2021-02-25 2021-06-18 成都运达科技股份有限公司 Railway vehicle wheel polygon state diagnostic system
CN113229829A (en) * 2021-04-15 2021-08-10 广东工业大学 Quaternion electroencephalogram signal extraction method and system
CN114595719A (en) * 2022-03-07 2022-06-07 河南理工大学 Mine aftershock monitoring method based on VMD and IRCNN

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2017158768A1 (en) * 2016-03-16 2018-10-11 本田技研工業株式会社 Vehicle control system, vehicle control method, and vehicle control program
CN110084185B (en) * 2019-04-25 2021-03-16 西南交通大学 Method for rapidly extracting small-amplitude snaking operation characteristics of high-speed train
CN110595765A (en) * 2019-08-26 2019-12-20 西安理工大学 Wind turbine generator gearbox fault diagnosis method based on VMD and FA _ PNN
CN114580460A (en) * 2022-01-17 2022-06-03 西南交通大学 Railway vehicle wheel rail fault diagnosis method based on morphological filtering and HHT conversion

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106250613A (en) * 2016-07-28 2016-12-21 南京理工大学 A kind of wheel service state security domain is estimated and method for diagnosing faults
CN107650945A (en) * 2017-09-19 2018-02-02 华东交通大学 A kind of recognition methods of wheel polygon and its device based on vertical wheel rail force
CN109059840A (en) * 2018-05-29 2018-12-21 南京理工大学 A kind of city rail vehicle wheel out of round is along detection method
RO134895A2 (en) * 2019-10-04 2021-04-29 Albert Mihai Suvac Self-propelled electric tricycle to be combined with small agricultural equipments
CN112381027A (en) * 2020-11-23 2021-02-19 西南交通大学 Wheel polygon wave depth estimation method based on train axle box vertical acceleration signal
CN112991577A (en) * 2021-02-25 2021-06-18 成都运达科技股份有限公司 Railway vehicle wheel polygon state diagnostic system
CN113229829A (en) * 2021-04-15 2021-08-10 广东工业大学 Quaternion electroencephalogram signal extraction method and system
CN114595719A (en) * 2022-03-07 2022-06-07 河南理工大学 Mine aftershock monitoring method based on VMD and IRCNN

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
《Research of multi-concurrent fault diagnosis of rotating machinery based on VMD and KICA》;Hongchun Sun等;《VIBROENGINEERING PROCEDIA》;20181031;第20卷;第52-56页 *
《基于变分模态分解和支持向量机的钢轨波磨辨识》;许官儒;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》;20220515(第05期);第C033-189页 *
《车轮多边形激励下高速列车制动界面摩擦学行为分析》;王志伟 等;《https://kns.cnki.net/kcms/detail/62.1095.O4.20220727.0853.001.html》;20220727;第1-19页 *
《高速列车液压减振器故障建模与诊断方法研究》;张楷;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20180315(第03期);第C033-273页 *

Also Published As

Publication number Publication date
CN114997252A (en) 2022-09-02

Similar Documents

Publication Publication Date Title
CN114997252B (en) Vehicle-mounted detection method for wheel polygon based on inertia principle
Ye et al. Fault diagnosis of high-speed train suspension systems using multiscale permutation entropy and linear local tangent space alignment
Chen et al. A two-level adaptive chirp mode decomposition method for the railway wheel flat detection under variable-speed conditions
CN110658005B (en) Method for identifying rail corrugation diseases based on vehicle body acceleration
CN107403139B (en) Urban rail train wheel flat scar fault detection method
CN106092015B (en) A kind of raceway surface recess length detecting method
CN107423692A (en) A kind of rail corrugation fault detection method based on wavelet-packet energy entropy
Chen et al. Quantitative detection of locomotive wheel polygonization under non-stationary conditions by adaptive chirp mode decomposition
Li et al. Rail Corrugation Detection of High‐Speed Railway Using Wheel Dynamic Responses
CN110501172A (en) A kind of rail vehicle wheel condition recognition methods based on axle box vibration
CN111444574B (en) Sensor layout optimization method based on dynamics analysis
CN113281414B (en) Method and device for identifying short-wave disease types of steel rails and electronic equipment
Hayashi et al. Real time fault detection of railway vehicles and tracks
Ye et al. Multislice Time-Frequency image Entropy as a feature for railway wheel fault diagnosis
CN105501250B (en) Railway wheelset flat sliding method for diagnosing faults based on vehicle-mounted detection apparatus
CN116252820B (en) Polygonal quantitative detection method for high-speed train wheels driven by improved frequency domain integration method
CN116467570B (en) Heavy-duty locomotive coupler swing angle quantitative identification method based on data driving
Xie et al. High-speed railway wheel polygon detection framework using improved frequency domain integration
Kojima et al. Fault detection of railway track by multi-resolution analysis
CN114580460A (en) Railway vehicle wheel rail fault diagnosis method based on morphological filtering and HHT conversion
Xu et al. Detection method for polygonalization of wheel treads based on dynamic response
Huang et al. A Fault Diagnosis Method for Out‐of‐Round Faults of Metro Vehicle Wheels with Strong Noise
Kim et al. Defect Prediction of Railway Wheel Flats based on Hilbert Transform and Wavelet Packet Decomposition
He et al. Dynamic responses of the metro train’s bogie frames: Field tests and data analysis
Li et al. A Multitask Learning Method for Rail Corrugation Detection Using In-Vehicle Responses and Noise Data

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