CN116522217A - Method and system for detecting degree of unsmooth degradation of railway low joint and electronic equipment - Google Patents
Method and system for detecting degree of unsmooth degradation of railway low joint and electronic equipment Download PDFInfo
- Publication number
- CN116522217A CN116522217A CN202310304538.0A CN202310304538A CN116522217A CN 116522217 A CN116522217 A CN 116522217A CN 202310304538 A CN202310304538 A CN 202310304538A CN 116522217 A CN116522217 A CN 116522217A
- Authority
- CN
- China
- Prior art keywords
- irregularity
- low joint
- rail
- low
- railway
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 230000015556 catabolic process Effects 0.000 title claims abstract description 53
- 238000006731 degradation reaction Methods 0.000 title claims abstract description 53
- 230000004044 response Effects 0.000 claims abstract description 116
- 229910000831 Steel Inorganic materials 0.000 claims abstract description 94
- 239000010959 steel Substances 0.000 claims abstract description 94
- 238000001514 detection method Methods 0.000 claims abstract description 45
- 238000012216 screening Methods 0.000 claims abstract description 40
- 230000001133 acceleration Effects 0.000 claims abstract description 33
- 208000012659 Joint disease Diseases 0.000 claims abstract description 28
- 238000004458 analytical method Methods 0.000 claims abstract description 24
- 230000008878 coupling Effects 0.000 claims description 36
- 238000010168 coupling process Methods 0.000 claims description 36
- 238000005859 coupling reaction Methods 0.000 claims description 36
- 238000000354 decomposition reaction Methods 0.000 claims description 30
- 238000001228 spectrum Methods 0.000 claims description 26
- 238000004590 computer program Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 13
- 230000011218 segmentation Effects 0.000 claims description 13
- 238000004088 simulation Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 7
- 230000006866 deterioration Effects 0.000 claims description 6
- 239000004973 liquid crystal related substance Substances 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 230000007547 defect Effects 0.000 claims description 5
- 230000009467 reduction Effects 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 230000003862 health status Effects 0.000 claims description 4
- 230000007704 transition Effects 0.000 claims description 4
- 230000003595 spectral effect Effects 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 abstract description 3
- 238000004422 calculation algorithm Methods 0.000 description 24
- 238000010586 diagram Methods 0.000 description 7
- 238000012360 testing method Methods 0.000 description 7
- 238000011160 research Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 5
- 238000009826 distribution Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008439 repair process Effects 0.000 description 4
- 241001669679 Eleotris Species 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000003466 welding Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000007635 classification algorithm Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000009412 basement excavation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005498 polishing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011895 specific detection Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61K—AUXILIARY EQUIPMENT SPECIALLY ADAPTED FOR RAILWAYS, NOT OTHERWISE PROVIDED FOR
- B61K9/00—Railway vehicle profile gauges; Detecting or indicating overheating of components; Apparatus on locomotives or cars to indicate bad track sections; General design of track recording vehicles
- B61K9/08—Measuring installations for surveying permanent way
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Automation & Control Theory (AREA)
- Aviation & Aerospace Engineering (AREA)
- Mechanical Engineering (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention provides a method, a system and electronic equipment for detecting the degree of unsmooth degradation of a railway low joint, which belong to the technical field of railway detection operation and maintenance and calculate the vibration acceleration response of wheels when a train passes through steel rails in different states; decoupling wheel set vibration response caused by rail low joint irregularity from complex vibration response signals by utilizing WPD-EFD combined kurtosis threshold screening; performing time domain analysis on the decoupled vibration response signals, and screening out sensitive features in the vibration response signals by combining Fisher criteria; and respectively inputting the screened sensitive features and the vehicle speed into PSO-SVM as classification features for detecting the degradation degree of the low joint diseases of the steel rail, so as to realize the detection of the irregularity degradation degree of the low joint of the heavy haul railway. The method accurately identifies the irregularity of the low joint of the track; and performing time domain analysis on the decoupled characteristic signals, screening out 5 sensitive characteristics by using Fisher criteria, and identifying and classifying the characteristic signals by using PSO-SVM in combination with the speed of the vehicle as the classification characteristic for detecting the degradation degree of the low joint diseases of the steel rail.
Description
Technical Field
The invention relates to the technical field of railway detection operation and maintenance, in particular to a method, a system and electronic equipment for detecting the degree of unsmooth degradation of a railway low joint based on wheel set vibration response.
Background
The seamless track is widely applied to heavy haul railways due to excellent smoothness and stability. The welded joint and the common steel rail are subjected to discontinuous material and rigidity phenomena under the repeated actions of climate change and wheel-rail load, and the steel rail joint is deformed and collapsed to form the low joint irregularity of the steel rail. The unsmooth low joint of the heavy-duty railway rail can bring a series of outstanding wheel rail dynamic action problems under the load action of the heavy-duty train, further worsen the state of the heavy-duty railway line, further influence the running safety and stability of the heavy-duty train, and even cause serious safety accidents such as derailment, subversion and the like of the heavy-duty train when serious. Therefore, the method for detecting the unsmooth faults of the low joints of the steel rail has very important significance for ensuring the running safety of the train and making a maintenance plan.
The scholars at home and abroad develop a series of researches on the influence of the rail joint irregularity on the dynamics of the vehicle rail, pei Chengjie and the like are used for carrying out simulation analysis on the dynamics response of the high-speed vehicle passing through the rail weld irregularity, and comparing the impact effect of different rail weld irregularities on the wheel rail. Yang Yunfan and the like test and analyze the irregularity of the welded joint of the seamless rail of the linear motor subway, study the influence of the wavelength and the wave depth of the irregularity of the joint on the dynamic response of the wheel rail, and provide the safety limit value and the polishing limit value of the irregularity of the welded joint of different types. Wang et al put forward an improved excitation model to simulate the irregularity of the welded rail joints of heavy haul railways based on the measured irregularity data of the welded rail joints of heavy haul railways, and studied the influence of the irregularity of the welded rail joints of heavy haul railways on the wheel rail power index. Plateau et al analyze and calculate the influence of the irregularity of the welded joint of the heavy-duty railway in different service periods on the vibration acceleration of the wheels and the steel rail and the vertical force of the wheels and the rail from two angles of a time domain and a frequency domain by utilizing a vehicle-rail vertical coupling dynamics model. Yang Yike et al investigated the effect of the wavelength and amplitude of the irregularity of a female and male weld joint at a small radius curve of a heavy haul railway on the dynamic response of the wheel track. Niu Liu et al utilize ABAQUS software to establish a wheel rail finite element contact model, utilize model simulation to calculate wheel rail vertical forces under different magnitudes of high-speed railway welded joint irregularity, and analyze wheel rail vertical force equipotential line distribution characteristics.
For detecting the irregularity of the rail joints, maria et al propose a detection algorithm based on the frequency characteristics of the acceleration signals of the axle boxes, so that the detection of the damage of the insulated rail joints is realized, and the detection success rate is 78%. M. Oregui et al employ a vehicle-mounted monitoring system to automatically detect and evaluate rail joint conditions based on wavelet transformation. Nunez et al propose a technique for processing passenger car axle box acceleration signals using hilbert spectra to detect and evaluate weld quality, and prioritizing the repair order of the weld by establishing two objective functions of performance and repair cost, thereby providing repair support to repair personnel. Tsai-Hsin Chu et al adopts Hilbert-Huang transform to analyze acceleration signals of an axle box of a vehicle, and realizes detection of rail joint diseases and wave mill diseases. Liu Jinchao et al combine axle box acceleration and wheel rail force high frequency vehicle dynamic response, and effectively diagnose short wave diseases such as rail ripples grinds, welded joint convex-concave based on the time-frequency characteristic excavation algorithm of the short time Fourier transform of improved synchronous compression.
In summary, the research on the rail joint irregularity at home and abroad at present mainly focuses on two aspects of dynamic response research caused by the rail joint irregularity and detection of the rail joint irregularity. For detection research of rail joint irregularity, most of researches directly analyze uncoupled vibration response signals through signal processing methods such as Hilbert transform, wavelet transform and short-time Fourier transform to obtain rail joint irregularity characteristics, and the final detection result is influenced by other irrelevant response components, so that the detection accuracy is reduced. The method realizes qualitative detection of the rail low joint irregularity, but has poor detection effect on the degree of degradation of the rail low joint irregularity.
Disclosure of Invention
The invention aims to provide a method, a system and electronic equipment for detecting the degree of unsmooth degradation of a railway low joint, which are used for solving at least one technical problem in the background technology.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
in one aspect, the present invention provides a method for detecting a degree of degradation of a low joint irregularity of a railway, comprising:
establishing a vehicle-track vertical coupling model and a steel rail low joint irregularity model, analyzing and calculating vibration acceleration response of wheels when a train passes through steel rails in different states;
decoupling wheel set vibration response caused by rail low joint irregularity from complex vibration response signals by utilizing WPD-EFD combined kurtosis threshold screening;
performing time domain analysis on the decoupled vibration response signals, and screening out sensitive features in the vibration response signals by combining Fisher criteria;
and respectively inputting the screened sensitive features and the vehicle speed into PSO-SVM as classification features for detecting the degradation degree of the low joint diseases of the steel rail, so as to realize the detection of the irregularity degradation degree of the low joint of the heavy haul railway.
Preferably, the method for establishing the vehicle-rail vertical coupling model and the rail low joint irregularity model, analyzing and calculating the vibration acceleration response of the wheels when the train passes through the rails in different states comprises the following steps:
According to the vehicle-track coupling dynamics, a heavy truck-track vertical coupling model is established, a track irregularity spectrum is a yellow heavy haul railway high-low irregularity spectrum, and a high-low irregularity spectrum density function and characteristic parameters are as follows: s is S v (f)＝(af+b)/(1+bf+cf 2 +df 3 ) 2 Wherein S is v (f) A, b, c, d is a characteristic parameter for simulating the track irregularity density;
the rail low joint model is modeled by adopting a displacement input method model:
wherein x represents the longitudinal coordinate of the steel rail, d is the descending depth of the steel rail joint, and L is the length of the low joint defect of the steel rail;
superposing a rail irregularity model equivalent to the rail low joint and the high-low irregularity of the heavy haul railway to serve as input of a vehicle-rail coupling model;
and according to the previously established vehicle-track coupling dynamics model, vibration response signals of a vehicle body, a framework and wheel sets when the vehicle runs on a normal steel rail and a steel rail containing low joint diseases under the high-low irregularity spectrum of the yellow heavy haul railway are obtained through simulation.
Preferably, decoupling the wheel set vibration response caused by rail low joint irregularity from the complex vibration response signal using WPD-EFD in combination with kurtosis threshold screening comprises:
performing Wavelet Packet Decomposition (WPD) calculation on the wheel set vibration acceleration signals;
Performing empirical Fourier decomposition on the wheel set vibration response signals after noise reduction;
calculating the kurtosis value of each modal component and determining a kurtosis threshold value;
and screening and extracting the decoupled characteristic vibration response signals according to the kurtosis threshold value.
Preferably, the time domain analysis is performed on the decoupled vibration response signal, and sensitive features in the vibration response signal are screened out by combining with Fisher criteria, including:
calculating the training sample at the ith feature Z i Inter-class divergence on the first class; calculating the feature Z of the training sample belonging to the c class at the i i Upper intra-class divergence; calculation of the ith feature Z i Fisher score of (2); ith feature Z i The greater the Fisher score of (2), the i-th feature Z i The larger the inter-class divergence, the smaller the intra-class divergence, indicating that the feature is more sensitive to changes in health status; and calculating and extracting Fisher scores of various features on the training data set by using Fisher feature selection criteria, and selecting the first 5 features to form a sensitive feature subset according to the Fisher scores from high to low.
Preferably, the empirical fourier decomposition of the noise-reduced wheel set vibration response signal includes:
defining the Fourier spectrum of the signal to be decomposed in a normalized frequency range [ -pi, pi]Determining the boundary of the segmented segment by utilizing an improved segmentation technology on the obtained Fourier spectrum; wherein, in the improved segmentation technique, [0, pi ] ]Is divided into N adjacent frequency segments, each segment is divided into S n ＝[ω n-1 ,ω n ]，n∈[1,N]Representation omega n And omega N The values of (2) are determined in an adaptive ordering process; identifying and extracting local maxima of the Fourier amplitude values at omega=0 and omega=pi to be decomposed into a sequence, wherein the values in the sequence are arranged in descending order; the first N maxima in the permutation sequence correspond to the frequency applications [ omega ] 1 ,Ω 2 ,...,Ω N ]A representation; and define omega 0 =0 and Ω N+1 =pi; the boundary of each frequency bin is determined by:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing omega n ～Ω n+1 Between (a) and (b)Fourier spectrum amplitude;
constructing a zero-phase filter bank based on the frequency band obtained by the improved segmentation technology; in each frequency band, the zero-phase filter is a band-pass filter with a cut-off frequency of ω n-1 And omega n It has no transition phase; finally, the signal after the zero-phase filter is subjected to inverse Fourier transform to obtain a component decomposed in the time domain.
Preferably, calculating the kurtosis value for each modal component and determining the kurtosis threshold comprises:
the calculation formula of the kurtosis value K is as follows:
wherein x is i Vibration amplitude corresponding to discrete sequence points of the time domain waveform,is the average amplitude of the discrete sequence, and n is the number of discrete sequence points.
Preferably, the kurtosis value of each component after EFD decomposition is calculated, a signal component with the kurtosis value larger than a threshold value is extracted, the signal is reconstructed, a characteristic vibration signal is obtained, and if no component with the kurtosis value larger than the threshold value exists in the components, the characteristic vibration signal is set to be zero.
In a second aspect, the present invention provides a system for detecting a degree of deterioration of a railway low joint irregularity, comprising:
the calculation module is used for establishing a vehicle-track vertical coupling model and a steel rail low joint irregularity model, analyzing and calculating vibration acceleration response of wheels when the train passes through steel rails in different states;
the decoupling module is used for decoupling wheel set vibration response caused by the irregularity of the low joint of the steel rail from complex vibration response signals by utilizing WPD-EFD combined with kurtosis threshold screening;
the screening module is used for carrying out time domain analysis on the decoupled vibration response signals and screening sensitive characteristics in the vibration response signals by combining with Fisher criteria;
the classification module is used for respectively inputting the screened sensitive features and the vehicle speed as classification features for detecting the degradation degree of the low joint diseases of the steel rail into the PSO-SVM to realize the detection of the irregularity degradation degree of the low joint of the heavy haul railway.
In a third aspect, the present invention provides a non-transitory computer readable storage medium for storing computer instructions which, when executed by a processor, implement the railway low joint irregularity degradation level detection method as described above.
In a fourth aspect, the present invention provides a computer program product comprising a computer program for implementing the railway low joint irregularity degradation level detection method as described above when run on one or more processors.
In a fifth aspect, the present invention provides an electronic device, comprising: a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes the instructions for implementing the method for detecting the degree of degradation of the railway low joint irregularity as described above.
The invention has the beneficial effects that: calculating wheel set vibration acceleration response by establishing a vehicle-track vertical coupling model and a steel rail low joint irregularity model, and researching a heavy-load railway low joint irregularity degradation degree detection algorithm based on the wheel set vibration acceleration response; providing a wavelet packet empirical Fourier decomposition (WPD-EFD) combined kurtosis threshold sieve steel rail low joint irregularity characteristic signal decoupling algorithm, decoupling a rail low joint irregularity characteristic signal hidden in a complex vehicle vibration response signal, and accurately identifying the rail low joint irregularity; and performing time domain analysis on the decoupled characteristic signals, screening out 5 sensitive characteristics by using Fisher criteria, and identifying and classifying the characteristic signals by using PSO-SVM in combination with the speed of the vehicle as the classification characteristic for detecting the degradation degree of the low joint diseases of the steel rail.
The advantages of additional aspects of the invention will be set forth in part in the description which follows, or may be learned by practice of the invention.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic view of rail low joint irregularity.
FIG. 2 is a schematic view of a rail irregularity input curve. In the figure, the line on the left side is a case where there is no smoothness.
FIG. 3 shows the vibration response signals of the body, frame and wheels of a vehicle running on a rail containing low joint failure (depth 0.1 mm) at 80 km/h. Fig. 3 (a) shows the vibration of the vehicle running on the normal rail, and fig. 3 (b) shows the vibration of the vehicle running on the rail with low joint failure.
FIG. 4 is a wavelet packet decomposition of the vibrational response signal of the body of a vehicle at 40km/h running on a normal rail and a rail containing low joint failure (depth 0.1 mm).
FIG. 5 is an EFD decomposition after noise reduction of vibration response signals of a wheel set when the vehicle is running on a normal rail and a rail containing low joint failure (depth 0.1 mm) at 40 km/h.
FIG. 6 is a kurtosis value distribution of wheel vibration acceleration signals when a vehicle is running on a normal rail at different speeds.
FIG. 7 is a characteristic signal of the vehicle wheel vibration signal extracted by WPD-EFD at 40km/h during normal rail operation.
FIG. 8 is a characteristic signal obtained by WPD-EFD extraction of wheel vibration signals of a vehicle at 40km/h when the vehicle is running on a low joint (depth 0.1 mm) rail.
FIG. 9 is a Fisher score for 11 features extracted on a training dataset.
Figure 10 is a plot of the first 5 features at 40km/h as a function of different low joint depths.
FIG. 11 is a flow chart of rail low joint diagnostics based on WPD-EFD and temporal features.
FIG. 12 is a graph showing a comparison of accuracy in detecting the depth of a rail joint.
Fig. 13 is a schematic diagram showing classification results of the depth of the low joint of the rail detected by combining 5 time domain features of the characteristic signal obtained by WPD-EFD with PSO-SVM.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements throughout or elements having like or similar functionality. The embodiments described below by way of the drawings are exemplary only and should not be construed as limiting the invention.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless expressly stated otherwise, as understood by those skilled in the art. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, and/or groups thereof.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
In order that the invention may be readily understood, a further description of the invention will be rendered by reference to specific embodiments that are illustrated in the appended drawings and are not to be construed as limiting embodiments of the invention.
It will be appreciated by those skilled in the art that the drawings are merely schematic representations of examples and that the elements of the drawings are not necessarily required to practice the invention.
Example 1
In this embodiment 1, there is provided first a railway low joint irregularity degradation degree detection system including:
the calculation module is used for establishing a vehicle-track vertical coupling model and a steel rail low joint irregularity model, analyzing and calculating vibration acceleration response of wheels when the train passes through steel rails in different states;
the decoupling module is used for decoupling wheel set vibration response caused by the irregularity of the low joint of the steel rail from complex vibration response signals by utilizing WPD-EFD combined with kurtosis threshold screening;
the screening module is used for carrying out time domain analysis on the decoupled vibration response signals and screening sensitive characteristics in the vibration response signals by combining with Fisher criteria;
the classification module is used for respectively inputting the screened sensitive features and the vehicle speed as classification features for detecting the degradation degree of the low joint diseases of the steel rail into the PSO-SVM to realize the detection of the irregularity degradation degree of the low joint of the heavy haul railway.
In this embodiment 1, with the system described above, a method for detecting a degree of degradation of a low joint irregularity of a railway is realized, including:
utilizing a calculation module to establish a vehicle-track vertical coupling model and a steel rail low joint irregularity model, and analyzing and calculating vibration acceleration response of wheels when a train passes through steel rails in different states;
decoupling wheel set vibration response caused by rail low joint irregularity from complex vibration response signals by using a decoupling module and utilizing WPD-EFD in combination with kurtosis threshold screening;
the method comprises the steps of utilizing a screening module to conduct time domain analysis on a decoupled vibration response signal, and screening sensitive features in the vibration response signal by combining Fisher criteria;
and the screened sensitive features are respectively input into the PSO-SVM by using the classification module and combining the speed as classification features for detecting the degradation degree of the low joint diseases of the steel rail, so that the detection of the irregularity degradation degree of the low joint of the heavy haul railway is realized.
The method for establishing the vehicle-rail vertical coupling model and the rail low joint irregularity model, analyzing and calculating the vibration acceleration response of wheels when a train passes through rails in different states comprises the following steps:
according to the vehicle-track coupling dynamics, a heavy truck-track vertical coupling model is established, a track irregularity spectrum is a yellow heavy haul railway high-low irregularity spectrum, and a high-low irregularity spectrum density function and characteristic parameters are as follows: s is S v (f)＝(af+b)/(1+bf+cf 2 +df 3 ) 2 Wherein S is v (f) A, b, c, d is a characteristic parameter for simulating the track irregularity density;
the rail low joint model is modeled by adopting a displacement input method model:
wherein x represents the longitudinal coordinate of the steel rail, d is the descending depth of the steel rail joint, and L is the length of the low joint defect of the steel rail;
superposing a rail irregularity model equivalent to the rail low joint and the high-low irregularity of the heavy haul railway to serve as input of a vehicle-rail coupling model;
and according to the previously established vehicle-track coupling dynamics model, vibration response signals of a vehicle body, a framework and wheel sets when the vehicle runs on a normal steel rail and a steel rail containing low joint diseases under the high-low irregularity spectrum of the yellow heavy haul railway are obtained through simulation.
Decoupling the wheel set vibration response caused by rail low joint irregularity from the complex vibration response signal using WPD-EFD in combination with kurtosis threshold screening, comprising:
performing Wavelet Packet Decomposition (WPD) calculation on the wheel set vibration acceleration signals;
performing empirical Fourier decomposition on the wheel set vibration response signals after noise reduction;
calculating the kurtosis value of each modal component and determining a kurtosis threshold value;
and screening and extracting the decoupled characteristic vibration response signals according to the kurtosis threshold value.
Performing time domain analysis on the decoupled vibration response signals, and screening out sensitive features in the vibration response signals by combining Fisher criteria, wherein the time domain analysis comprises the following steps:
calculating the training sample at the ith feature Z i Inter-class divergence on the first class; calculating the feature Z of the training sample belonging to the c class at the i i Upper intra-class divergence; calculation of the ith feature Z i Fisher score of (2); ith feature Z i The greater the Fisher score of (2), the i-th feature Z i The larger the inter-class divergence, the smaller the intra-class divergence, indicating that the feature is more sensitive to changes in health status; and calculating and extracting Fisher scores of various features on the training data set by using Fisher feature selection criteria, and selecting the first 5 features to form a sensitive feature subset according to the Fisher scores from high to low.
Performing an empirical fourier decomposition of the noise-reduced wheel set vibration response signal, comprising:
defining the Fourier spectrum of the signal to be decomposed in a normalized frequency range [ -pi, pi]Determining the boundary of the segmented segment by utilizing an improved segmentation technology on the obtained Fourier spectrum; wherein, in the improved segmentation technique, [0, pi ]]Is divided into N adjacent frequency segments, each segment is divided into S n ＝[ω n-1 ,ω n ]，n∈[1,N]Representation omega n And omega N The values of (2) are determined in an adaptive ordering process; for fourier magnitudes at ω=0 and ω=pi Identifying and extracting the values and the local maximum value of the amplitude of the signal to be decomposed into a sequence, wherein the values in the sequence are arranged in descending order; the first N maxima in the permutation sequence correspond to the frequency applications [ omega ] 1 ,Ω 2 ,...,Ω N ]A representation; and define omega 0 =0 and Ω N+1 =pi; the boundary of each frequency bin is determined by:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing omega n ～Ω n+1 Fourier spectral magnitudes in between;
constructing a zero-phase filter bank based on the frequency band obtained by the improved segmentation technology; in each frequency band, the zero-phase filter is a band-pass filter with a cut-off frequency of ω n-1 And omega n It has no transition phase; finally, the signal after the zero-phase filter is subjected to inverse Fourier transform to obtain a component decomposed in the time domain.
Calculating the kurtosis value of each modal component and determining a kurtosis threshold value comprises:
the calculation formula of the kurtosis value K is as follows:
wherein x is i Vibration amplitude corresponding to discrete sequence points of the time domain waveform,is the average amplitude of the discrete sequence, and n is the number of discrete sequence points.
And calculating the kurtosis value of each component after EFD decomposition, extracting a signal component with the kurtosis value larger than a threshold value, reconstructing the signal to obtain a characteristic vibration signal, and if no component with the kurtosis value larger than the threshold value exists in the components, assigning the characteristic vibration signal to be zero.
Example 2
In this embodiment 2, an automatic detection algorithm for the degree of deterioration of a low joint irregularity of a heavy haul railway based on the vehicle vibration response is provided. Firstly, a vehicle-track vertical coupling model and a rail low joint irregularity model are established, and the vibration acceleration response of wheels when a train passes through rails in different states is calculated by solving the models. And decoupling the wheel set vibration response caused by the rail low joint irregularity from the complex vibration response signal by using WPD-EFD in combination with kurtosis threshold screening. And performing time domain analysis on the decoupled vibration response signals, screening out 5 sensitive features by using Fisher criteria, and respectively inputting the 5 sensitive features into the PSO-SVM by combining the vehicle speed as classification features for detecting the degradation degree of the low joint diseases of the steel rail, thereby realizing the detection of the irregularity degradation degree of the low joint of the heavy haul railway. The result shows that the proposed algorithm is superior to the recognition algorithm based on the WPD and EFD decomposition algorithm and the widely used recognition algorithm based on the combination of EMD and WPD with each component energy and kurtosis characteristic on the basis of the low joint depth recognition of the steel rail, and the average accuracy can reach 99.60%.
The automatic detection algorithm for the degree of unsmooth degradation of the low joint of the heavy haul railway based on the vehicle vibration response provided in the embodiment 2 specifically comprises the following steps:
(1) Acquiring vehicle vibration response data
(1) And establishing a heavy-duty truck-rail vertical coupling model according to the vehicle-rail coupling dynamics:
the vehicle-track coupling dynamics model in the embodiment of the invention comprises a heavy load truck vehicle model, a steel rail-sleeper-ballast bed-roadbed three-layer ballasted track model and a vehicle-track coupling relation. The vehicle parameters of the C80 type open car special for the coal mine are selected, and the axle weight is 25t. The main line of the track parameters adopts 75kg/m steel rail, and the sleeper adopts the parameters of the III type sleeper.
In the vehicle model, the vehicle is simulated into a multi-rigid-body system running on a track structure at the speed v, and the multi-rigid-body system consists of a vehicle body, front and rear bogies and wheels, and the vertical vibration and nodding of the vehicle body and the bogies and the vertical vibration of the wheels are considered, so that 10 degrees of freedom are taken into account. The track model adopts a rail-sleeper-ballast bed-roadbed three-layer structure, wherein the rail adopts an infinite length Eular beam supported by elastic points, the sleeper adopts a rigid body unit, and the ballast bed is longitudinally scattered into ballast bed block units. The vertical coupling of the vehicle system to the rail system is achieved by vertical contact of the wheel rails, in particular in terms of vertical forces between the wheel rails, the most classical effective Hertz nonlinear elastic contact model being used in this example 2.
The track irregularity spectrum adopts a yellow heavy haul railway high-low irregularity spectrum, and the high-low irregularity spectrum density function and characteristic parameters are as follows:
S v (f)＝(af+b)/(1+bf+cf 2 +df 3 ) 2 ； (1)
wherein S is v (f) To simulate track irregularity density in mm 2 /(1/m); f is the spatial frequency (1/m), a, b, c, d are characteristic parameters 3217.2, -27.9, 847.1, 413.5, respectively.
(2) Establishing a steel rail low joint irregularity model:
TB/T1632.1-2005 Steel rail welding part 1: the flatness and surface quality requirements of the welding area rail head working face within the length range of 1.0m are specified in general technical condition: when the steel rail is welded by flash welding and the vehicle operation speed is not more than 120km/h, the vertical space irregularity of the welded joint is smaller than 0.3mm, and the phenomenon of concave is not allowed to occur.
The low joint model of the steel rail adopts a displacement input method model, and because the outline of the low joint disease of the steel rail on the vertical-longitudinal plane is very similar to cosine waveform, the modeling can be carried out by the following formula:
wherein x (m) is a coordinate axis, namely a longitudinal coordinate of the steel rail; d (mm) is the descending depth of the rail joint; l (m) is the length of the low joint defect of the steel rail. A schematic of rail low joint irregularity is shown in fig. 1.
And superposing the rail irregularity model equivalent to the rail low joint and the high and low irregularity of the plastic heavy haul railway to serve as input of the vehicle-rail coupling model. Fig. 2 is a graph of the magnitude of a plastic heavy haul railway rail irregularity.
(3) Vehicle vibration acceleration response analysis under low joint irregularity:
and according to the previously established vehicle-track coupling dynamics model, vibration response signals of a vehicle body, a framework and wheel sets when the vehicle runs on a normal steel rail and a steel rail containing low joint diseases under the high-low irregularity spectrum of the yellow heavy haul railway are obtained through simulation. Fig. 3 (a) shows vibration response signals of the body, frame and wheel set when the vehicle is running on a normal rail, and fig. 3 (b) shows vibration response signals of the body, frame and wheel set when the vehicle is running on a rail containing low joint failure (depth 0.1 mm).
(2) Decoupling algorithm design of low joint characteristic signals of steel rail
(1) Wavelet Packet Decomposition (WPD) calculation is performed on the wheel set vibration acceleration signal:
the wavelet packet decomposition is a signal decomposition method, and is composed of a series of wavelet functions which are linearly combined, so that the frequency band can be divided into multiple layers. Meanwhile, the high-frequency part and the low-frequency part can be analyzed, and the accurate local analysis capability is achieved. In order to preliminarily extract the frequency range of the wheel set vibration response component caused by the low joint irregularity of the steel rail, wavelet packet decomposition is performed on the wheel set vibration response signal in the embodiment. The number of the decomposition layers is 3, and a meyer wavelet basis function is selected as a mother function in wavelet transformation to extract the fault characteristics of the signals. The vibration acceleration signal of the wheel set is decomposed and calculated through a 3-layer wavelet packet to obtain 8 modal components, as shown in fig. 4. And extracting the first node signal to reconstruct the signal to obtain a noise-reduced vibration response signal.
(2) Performing an Empirical Fourier Decomposition (EFD) on the noise-reduced wheel set vibration response signal:
EFD comprises two key steps: improved segmentation techniques and construction of zero-phase filter banks.
Firstly, a Fourier transform is utilized to obtain a Fourier spectrum of a signal f (t) to be decomposed. Defining the Fourier spectrum of the signal to be decomposed in a normalized frequency range [ -pi, pi]On top of that, then, improved utilization of the resulting Fourier spectrumThe segmentation technique determines the boundaries of the segments. In the improved segmentation technique, [0, pi ]]Is divided into N adjacent frequency segments, each segment is divided into S n ＝[ω n-1 ,ω n ]，n∈[1,N]Representation omega n And omega N The values of (2) are determined in an adaptive ordering process. The local maxima of the fourier magnitudes at ω=0 and ω=pi of the signal to be decomposed are identified and extracted as a sequence. The values in the sequence are arranged in descending order. The first N maxima in the permutation sequence correspond to the frequency applications [ omega ] 1 ,Ω 2 ,...,Ω N ]And (3) representing. And define omega 0 =0 and Ω N+1 Pi. The boundary of each frequency bin is determined by:
wherein the method comprises the steps ofRepresenting omega n ～Ω n+1 Fourier spectral amplitude values therebetween.
A zero-phase filter bank is then constructed based on the frequency bins resulting from the improved segmentation technique. In each frequency band, the zero-phase filter is a band-pass filter with a cut-off frequency of ω n-1 And omega n It has no transition phase. Finally, the signal after the zero-phase filter is subjected to inverse Fourier transform to obtain a component decomposed in the time domain.
FIG. 5 is a graph showing the signal decomposition after EFD decomposition after noise reduction of vibration response signals of a wheel set of a vehicle running on a normal rail and a rail containing low joint failure (depth 0.1 mm) at 40 km/h. When the number of the decomposition layers of the EFD is 8, the effect of extracting the low joint diseases of the steel rail is best. In the right column of fig. 5, the signal component in the small box indicated by the arrow in the second top-down box is the extracted vibration of the wheel caused by the rail low joint.
(3) Calculating the kurtosis value of each modal component and determining a kurtosis threshold value:
the calculation formula of the kurtosis value K is as follows:
wherein x is i Vibration amplitude corresponding to discrete sequence points of the time domain waveform,is the average amplitude of the discrete sequence, and n is the number of discrete sequence points.
And calculating kurtosis values of vibration acceleration signals of wheels when the vehicle runs on a normal steel rail at different speeds (40 km/h-120km/h and a step length of 5 km/h). The distribution results are shown in fig. 6. The kurtosis threshold is determined to be 30.
(4) And screening and extracting the decoupled characteristic vibration response signals according to the kurtosis threshold value.
And calculating the kurtosis value of each component after EFD decomposition, extracting a signal component with the kurtosis value larger than a threshold value, reconstructing the signal to obtain a characteristic vibration signal, and if no component with the kurtosis value larger than the threshold value exists in the components, assigning the characteristic vibration signal to be zero. According to the method, the characteristic signals of the wheel vibration acceleration signal of the vehicle in normal rail operation and the wheel vibration acceleration signal of the vehicle in low joint defect-containing rail operation under the condition of 40km/h can be extracted, and are shown in fig. 7 and 8.
(3) Detection of degree of unsmooth degradation of low joint of steel rail
(1) And carrying out time domain analysis on the decoupled characteristic signals:
in order to analyze the extracted characteristic signals under various working conditions, 11 common characteristic values of the decoupling signals are calculated, as shown in table 1.
TABLE 1
(2) Selecting sensitive features by using Fisher feature selection criteria:
the Fisher feature selection criteria are calculated as follows:
computational trainingSample at ith feature Z i Inter-class divergence above:
wherein n is c The number of samples of the c type;the mean value of the class c sample on the ith feature; mu (mu) i Is the mean of all samples over the ith feature.
Calculating the feature Z of the training sample belonging to the c class at the i i Intra-class divergence above:
in the method, in the process of the invention,for the ith sample in the c-th samples at the ith feature Z i The value of the above value.
Calculation of the ith feature Z i Fisher score of (2):
from formula (7): v (V) Fisher (Z i ) The larger the value is, the i-th feature Z i The larger the inter-class divergence, the smaller the intra-class divergence, indicating that the feature is more sensitive to changes in health status. Fisher scores for the 11 extracted features on the training dataset were calculated using Fisher feature selection criteria, as shown in FIG. 9. The top 5 features are selected to form a sensitive feature subset according to the Fisher score from high to low, and the change curve of the top 5 features with different low joint depths at 40km/h is shown in FIG. 10.
(3) The detection of the degree of degradation of the low joint irregularity of the steel rail is realized by using a PSO-SVM model:
according to the embodiment, 5 sensitive features obtained by using the decoupled characteristic signals are input into a PSO-SVM model together with the combination speed, so that the detection of the degree of unsmooth degradation of the low joint of the steel rail is realized.
A specific detection algorithm flow is shown in fig. 11.
(4) Simulation calculation and result analysis
According to the embodiment, through establishing a vehicle track vertical coupling model and a steel rail low joint disease model, under the condition that the height of a plastic yellow heavy haul railway is unsmooth, vibration response of wheels is calculated in a simulation mode when the vehicle passes through a normal steel rail and a steel rail with a low joint, 170 groups of wheel vibration response data under the normal condition of different vehicle speeds (40-120 km/h and 5km/h step length) and 850 groups of data under the condition of different steel rail low joint depths (0.1-0.5 mm and 0.1mm step length) with different speeds are respectively obtained, and each 170 groups of data are obtained. The data size distribution for the different conditions is shown in table 2.
TABLE 2
Characteristic signal decoupling is carried out on wheel vibration response data obtained through simulation under the condition that the steel rail normally contains low joints with different depths (0.1 mm-0.5 mm) and the steel rail, 5 sensitive time domain characteristics are extracted from the decoupled signals through time domain analysis, the speed is combined to be used as the detection steel rail low joint depth classification characteristic of the PSO-SVM model, 85% of data of the normal condition and the condition containing low joints with different depths are taken for training, the corresponding output is 0 and represents that the steel rail is normal, and 1, 2, 3, 4 and 5 respectively represent that the steel rail contains low joint diseases with the depths of 0.1mm, 0.2mm, 0.3mm, 0.4mm and 0.5 mm. And inputting the rest 15% data under each working condition into a PSO-SVM model for testing.
The decoupling algorithm based on the WPD-EFD combined kurtosis threshold screening provided by the invention is compared with the WPD-EFD decomposition algorithm only and the EMD decomposition algorithm which uses more at present so as to verify the superiority of the method. To verify generalization and robustness of the test results, each test algorithm randomly extracted the training set and the test set from the dataset for 6 tests, and the test accuracy of each algorithm is shown in fig. 12.
(1) Classification algorithm combining WPD and EFD only with time domain features
After the wheel vibration signals of the normal steel rail and the steel rail containing the low joint diseases are decomposed, the characteristic signals are extracted, 5 sensitive time domain features of the characteristic signals are extracted, and the obtained time domain features are combined with the speed as classification features for detecting the low joint diseases depth of the steel rail and are respectively input into a PSO-SVM.
(2) Classification algorithm for comparing WPD and EMD combined with each component energy and kurtosis characteristics
After WPD and EMD decomposition are carried out on wheel vibration signals of a normal steel rail and a steel rail with low joint diseases, 8 layers of signal components are respectively extracted to serve as research objects, energy and kurtosis characteristics of each component are extracted, the kurtosis characteristics of each component are comprehensively obtained to obtain multi-scale kurtosis characteristics which serve as classification characteristics for detecting the depth of the steel rail low joint diseases, and the multi-scale kurtosis characteristics are respectively input into PSO-SVM.
The experimental comparison results are shown in table 3, and the decoupling algorithm of the low joint irregularity characteristic signals of the steel rail based on the screening of the WPD-EFD combined kurtosis threshold value provided by the embodiment has obviously improved detection accuracy compared with the detection algorithm only using the WPD and the EFD; compared with widely used detection algorithms based on the combination of WPD and EMD with energy and kurtosis characteristics of each component, the accuracy is improved.
The algorithm provided by the embodiment has the advantages that the vibration characteristic signals caused by the low joint irregularity can be decoupled more accurately, so that the detection accuracy of the low joint irregularity with different depths is improved. The detection classification result of one of the 6 detection experiments is shown in fig. 13, and it can be seen from the graph that the method provided by the invention can be used for 100% detection of the low joint diseases of the steel rail, and the detection classification accuracy of the low joint depth is up to 99.9%. There is one sample misprediction.
TABLE 3 Table 3
In summary, in the proposed detection algorithm for the degree of degradation of the low joint irregularity of the heavy haul railway based on the vehicle vibration response in this embodiment 2, the wheel set vibration acceleration response is calculated by establishing the vehicle-track vertical coupling model and the rail low joint irregularity model, and the detection algorithm for the degree of degradation of the low joint irregularity of the heavy haul railway based on the wheel set vibration acceleration response is studied. Due to the nonlinearity of the heavy haul railway system, the diversity of operating conditions (operating speed, line conditions, etc.), the vehicle vibration response has nonlinear, non-stationarity characteristics, while the low joint irregularity characteristic signal of the track is hidden in the complex vehicle vibration response signal. How to decouple the characteristic signals is a key to accurately identify the rail low joint irregularity, and in view of this problem, a rail low joint irregularity characteristic signal decoupling algorithm of wavelet packet empirical fourier decomposition (WPD-EFD) combined with kurtosis threshold sieve is proposed in this embodiment 2. And performing time domain analysis on the decoupled characteristic signals, screening out 5 sensitive characteristics by using Fisher criteria, and identifying and classifying the characteristic signals by using PSO-SVM in combination with the speed of the vehicle as the classification characteristic for detecting the degradation degree of the low joint diseases of the steel rail.
Example 3
Embodiment 3 provides a non-transitory computer-readable storage medium for storing computer instructions which, when executed by a processor, implement the method for detecting a degree of degradation of a railway low joint irregularity as described above, the method comprising:
establishing a vehicle-track vertical coupling model and a steel rail low joint irregularity model, analyzing and calculating vibration acceleration response of wheels when a train passes through steel rails in different states;
decoupling wheel set vibration response caused by rail low joint irregularity from complex vibration response signals by utilizing WPD-EFD combined kurtosis threshold screening;
performing time domain analysis on the decoupled vibration response signals, and screening out sensitive features in the vibration response signals by combining Fisher criteria;
and respectively inputting the screened sensitive features and the vehicle speed into PSO-SVM as classification features for detecting the degradation degree of the low joint diseases of the steel rail, so as to realize the detection of the irregularity degradation degree of the low joint of the heavy haul railway.
Example 4
Embodiment 4 provides a computer program product comprising a computer program for implementing a railway low joint irregularity degradation level detection method as described above when run on one or more processors, the method comprising:
Establishing a vehicle-track vertical coupling model and a steel rail low joint irregularity model, analyzing and calculating vibration acceleration response of wheels when a train passes through steel rails in different states;
decoupling wheel set vibration response caused by rail low joint irregularity from complex vibration response signals by utilizing WPD-EFD combined kurtosis threshold screening;
performing time domain analysis on the decoupled vibration response signals, and screening out sensitive features in the vibration response signals by combining Fisher criteria;
and respectively inputting the screened sensitive features and the vehicle speed into PSO-SVM as classification features for detecting the degradation degree of the low joint diseases of the steel rail, so as to realize the detection of the irregularity degradation degree of the low joint of the heavy haul railway.
Example 5
Embodiment 5 provides an electronic apparatus including: a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes instructions for implementing the method for detecting the degree of degradation of the low joint irregularity of the railway as described above, the method comprising:
establishing a vehicle-track vertical coupling model and a steel rail low joint irregularity model, analyzing and calculating vibration acceleration response of wheels when a train passes through steel rails in different states;
Decoupling wheel set vibration response caused by rail low joint irregularity from complex vibration response signals by utilizing WPD-EFD combined kurtosis threshold screening;
performing time domain analysis on the decoupled vibration response signals, and screening out sensitive features in the vibration response signals by combining Fisher criteria;
and respectively inputting the screened sensitive features and the vehicle speed into PSO-SVM as classification features for detecting the degradation degree of the low joint diseases of the steel rail, so as to realize the detection of the irregularity degradation degree of the low joint of the heavy haul railway.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the foregoing description of the embodiments of the present invention has been presented in conjunction with the drawings, it should be understood that it is not intended to limit the scope of the invention, but rather, it should be understood that various changes and modifications could be made by one skilled in the art without the need for inventive faculty, which would fall within the scope of the invention.
Claims (10)
1. A method for detecting a degree of deterioration of a railway low joint irregularity, comprising:
establishing a vehicle-track vertical coupling model and a steel rail low joint irregularity model, analyzing and calculating vibration acceleration response of wheels when a train passes through steel rails in different states;
decoupling wheel set vibration response caused by rail low joint irregularity from complex vibration response signals by utilizing WPD-EFD combined kurtosis threshold screening;
performing time domain analysis on the decoupled vibration response signals, and screening out sensitive features in the vibration response signals by combining Fisher criteria;
and respectively inputting the screened sensitive features and the vehicle speed into PSO-SVM as classification features for detecting the degradation degree of the low joint diseases of the steel rail, so as to realize the detection of the irregularity degradation degree of the low joint of the heavy haul railway.
2. The method for detecting the degree of deterioration of a low joint of a railway according to claim 1, wherein the steps of establishing a vehicle-rail vertical coupling model and a rail low joint irregularity model, analyzing and calculating the vibration acceleration response of wheels when a train passes through rails in different states, include:
according toThe vehicle-track coupling dynamics establishes a heavy truck-track vertical coupling model, a track irregularity spectrum adopts a yellow heavy haul railway high-low irregularity spectrum, and a high-low irregularity spectrum density function and characteristic parameters are as follows: s is S v (f)＝(af+b)/(1+bf+cf 2 +df 3 ) 2 Wherein S is v (f) A, b, c, d is a characteristic parameter for simulating the track irregularity density;
the rail low joint model is modeled by adopting a displacement input method model:
wherein x represents the longitudinal coordinate of the steel rail, d is the descending depth of the steel rail joint, and L is the length of the low joint defect of the steel rail;
superposing a rail irregularity model equivalent to the rail low joint and the high-low irregularity of the heavy haul railway to serve as input of a vehicle-rail coupling model;
and according to the previously established vehicle-track coupling dynamics model, vibration response signals of a vehicle body, a framework and wheel sets when the vehicle runs on a normal steel rail and a steel rail containing low joint diseases under the high-low irregularity spectrum of the yellow heavy haul railway are obtained through simulation.
3. The method for detecting the degree of degradation of a low joint irregularity of a railway according to claim 1, wherein decoupling the wheel set vibration response caused by the low joint irregularity of the rail from the complex vibration response signal using WPD-EFD in combination with kurtosis threshold screening, comprises:
performing Wavelet Packet Decomposition (WPD) calculation on the wheel set vibration acceleration signals;
performing empirical Fourier decomposition on the wheel set vibration response signals after noise reduction;
calculating the kurtosis value of each modal component and determining a kurtosis threshold value;
And screening and extracting the decoupled characteristic vibration response signals according to the kurtosis threshold value.
4. The method for detecting the degree of degradation of the low-joint irregularity of the railway according to claim 1, wherein the step of performing time-domain analysis on the decoupled vibration response signals and screening out sensitive features in the vibration response signals by combining Fisher criteria comprises the steps of:
calculating the training sample at the ith feature Z i Inter-class divergence on the first class; calculating the feature Z of the training sample belonging to the c class at the i i Upper intra-class divergence; calculation of the ith feature Z i Fisher score of (2); ith feature Z i The greater the Fisher score of (2), the i-th feature Z i The larger the inter-class divergence, the smaller the intra-class divergence, indicating that the feature is more sensitive to changes in health status; and calculating and extracting Fisher scores of various features on the training data set by using Fisher feature selection criteria, and selecting the first 5 features to form a sensitive feature subset according to the Fisher scores from high to low.
5. The method for detecting the degree of deterioration of a railway low joint irregularity of claim 3, characterized by performing an empirical fourier decomposition of the noise-reduced wheel set vibration response signal, comprising:
defining the Fourier spectrum of the signal to be decomposed in a normalized frequency range [ -pi, pi ]Determining the boundary of the segmented segment by utilizing an improved segmentation technology on the obtained Fourier spectrum; wherein, in the improved segmentation technique, [0, pi ]]Is divided into N adjacent frequency segments, each segment is divided into S n ＝[ω n-1 ,ω n ]，n∈[1,N]Representation omega n And omega N The values of (2) are determined in an adaptive ordering process; identifying and extracting local maxima of the Fourier amplitude values at omega=0 and omega=pi to be decomposed into a sequence, wherein the values in the sequence are arranged in descending order; the first N maxima in the permutation sequence correspond to the frequency applications [ omega ] 1 ,Ω 2 ,...,Ω N ]A representation; and define omega 0 =0 and Ω N+1 =pi; the boundary of each frequency bin is determined by:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing omega n ～Ω n+1 Fourier spectral magnitudes in between;
constructing a zero-phase filter bank based on the frequency band obtained by the improved segmentation technology; in each frequency band, the zero-phase filter is a band-pass filter with a cut-off frequency of ω n-1 And omega n It has no transition phase; finally, the signal after the zero-phase filter is subjected to inverse Fourier transform to obtain a component decomposed in the time domain.
6. The method for detecting the degree of degradation of a railway low joint irregularity according to claim 5, wherein calculating the kurtosis value of each modal component and determining the kurtosis threshold value comprises:
The calculation formula of the kurtosis value K is as follows:
wherein x is i Vibration amplitude corresponding to discrete sequence points of the time domain waveform,is the average amplitude of the discrete sequence, and n is the number of discrete sequence points.
7. The method for detecting the degree of deterioration of a railway low joint according to claim 6, wherein the kurtosis value of each component after the EFD decomposition is calculated, the signal component with the kurtosis value larger than a threshold value is extracted, the signal is reconstructed to obtain the characteristic vibration signal, and if no component with the kurtosis value larger than the threshold value exists in the components, the characteristic vibration signal is set to zero.
8. A railway low joint irregularity degradation level detection system, comprising:
the calculation module is used for establishing a vehicle-track vertical coupling model and a steel rail low joint irregularity model, analyzing and calculating vibration acceleration response of wheels when the train passes through steel rails in different states;
the decoupling module is used for decoupling wheel set vibration response caused by the irregularity of the low joint of the steel rail from complex vibration response signals by utilizing WPD-EFD combined with kurtosis threshold screening;
the screening module is used for carrying out time domain analysis on the decoupled vibration response signals and screening sensitive characteristics in the vibration response signals by combining with Fisher criteria;
The classification module is used for respectively inputting the screened sensitive features and the vehicle speed as classification features for detecting the degradation degree of the low joint diseases of the steel rail into the PSO-SVM to realize the detection of the irregularity degradation degree of the low joint of the heavy haul railway.
9. A non-transitory computer readable storage medium storing computer instructions which, when executed by a processor, implement the method of detecting a level of degradation of a railway low joint irregularity of any of claims 1-7.
10. An electronic device, comprising: a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes the instructions for implementing the method for detecting the degree of degradation of the railway low joint irregularity according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310304538.0A CN116522217A (en) | 2023-03-27 | 2023-03-27 | Method and system for detecting degree of unsmooth degradation of railway low joint and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310304538.0A CN116522217A (en) | 2023-03-27 | 2023-03-27 | Method and system for detecting degree of unsmooth degradation of railway low joint and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116522217A true CN116522217A (en) | 2023-08-01 |
Family
ID=87407261
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310304538.0A Pending CN116522217A (en) | 2023-03-27 | 2023-03-27 | Method and system for detecting degree of unsmooth degradation of railway low joint and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116522217A (en) |
-
2023
- 2023-03-27 CN CN202310304538.0A patent/CN116522217A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108845028B (en) | Method and device for dynamically detecting high-speed railway rail corrugation | |
| Berggren et al. | A new approach to the analysis and presentation of vertical track geometry quality and rail roughness | |
| Mosleh et al. | Early wheel flat detection: An automatic data-driven wavelet-based approach for railways | |
| CN113705412B (en) | Track state detection method and device for multi-source data fusion based on deep learning | |
| CN114169422A (en) | Subway rail corrugation identification method and system based on vehicle vibration and noise combined test | |
| Ye et al. | Shock detection of rotating machinery based on activated time-domain images and deep learning: An application to railway wheel flat detection | |
| CN106092015A (en) | A kind of raceway surface depression length detecting method | |
| Li et al. | Rail corrugation detection of high-speed railway using wheel dynamic responses | |
| Ning et al. | Feature recognition of small amplitude hunting signals based on the MPE-LTSA in high-speed trains | |
| CN106218668A (en) | Wheel out of round degree detection method and device | |
| Kostrzewski | Analysis of selected acceleration signals measurements obtained during supervised service conditions–study of hitherto approach | |
| Ye et al. | Multislice Time-Frequency image Entropy as a feature for railway wheel fault diagnosis | |
| CN113281414B (en) | Method and device for identifying short-wave disease types of steel rails and electronic equipment | |
| CN111444574A (en) | Sensor layout optimization method based on dynamic analysis | |
| Ding et al. | Time-frequency analysis of wheel-rail shock in the presence of wheel flat | |
| CN116522217A (en) | Method and system for detecting degree of unsmooth degradation of railway low joint and electronic equipment | |
| Li et al. | Assessment of vertical track geometry quality based on simulations of dynamic track—vehicle interaction | |
| Zhang et al. | Ballasted Track Behaviour Induced by Absent Sleeper Support and its Detection Based on a Convolutional Neural Network Using Track Data | |
| CN113501028B (en) | Method and device for diagnosing poor welded joint of heavy-duty railway steel rail | |
| Lang et al. | A Rail Corrugation Detection Method Based on Wavelet Packet Energy Entropy | |
| Miao et al. | Ballastless track mortar layer void detection by high-order statistical analysis of axle box acceleration | |
| Yue et al. | WSN-based vibration characteristic research for various railway track structures for pattern classification | |
| Guo et al. | Experimental and numerical research on the bogie hunting of a high-speed train caused by the empty stroke of yaw damper | |
| Ji et al. | Research on wheel-rail local impact identification based on axle box acceleration | |
| Chen et al. | Fault diagnosis of train wheels based on empirical mode decomposition generalized energy |
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 |