CN102501886B - Method and system for detecting states of steel rail welding joints of high-speed railway - Google Patents

Method and system for detecting states of steel rail welding joints of high-speed railway Download PDF

Info

Publication number
CN102501886B
CN102501886B CN201110344671.6A CN201110344671A CN102501886B CN 102501886 B CN102501886 B CN 102501886B CN 201110344671 A CN201110344671 A CN 201110344671A CN 102501886 B CN102501886 B CN 102501886B
Authority
CN
China
Prior art keywords
axle box
vertical acceleration
box vertical
welded joint
effective value
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
CN201110344671.6A
Other languages
Chinese (zh)
Other versions
CN102501886A (en
Inventor
黎国清
刘金朝
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.)
Infrastructure Inspection Institute of CARS
Beijing IMAP Technology Co Ltd
Original Assignee
Infrastructure Inspection Institute of CARS
Beijing IMAP Technology Co Ltd
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 Infrastructure Inspection Institute of CARS, Beijing IMAP Technology Co Ltd filed Critical Infrastructure Inspection Institute of CARS
Priority to CN201110344671.6A priority Critical patent/CN102501886B/en
Publication of CN102501886A publication Critical patent/CN102501886A/en
Application granted granted Critical
Publication of CN102501886B publication Critical patent/CN102501886B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention discloses a method and system for detecting the states of steel rail welding joints of a high-speed railway. The method comprises the following steps of: sampling the time of the vertical acceleration of an axle box to obtain a time sampling signal of the vertical acceleration of the axle box; carrying out band-pass filtration on the time sampling signal of the vertical acceleration of the axle box; calculating an effective value of the filtered vertical acceleration of the axle box; determining the positions of the welding joints according to the effective value of the vertical acceleration of the axle box, and extracting the effective value of the vertical acceleration of the axle box at each welding joint; and if the effective value of the vertical acceleration of the axle box at each welding joint is larger than a threshold value, judging that the welding joint is in a poor state. The states of the welding joints on the whole railway are detected according to data of the vertical acceleration of the axle box in the technical scheme of the embodiment of the invention, therefore, the current using states of the welding joints of the high-speed railway can be effectively detected, and a powerful theoretical basis is provided for reasonably and economically guiding the maintenance and the repair of the welding joints.

Description

Method of inspection and the system of rail in high speed railway welded joint state
Technical field
The present invention relates to high speed railway safe operation technical field, particularly a kind of method of inspection and system of rail in high speed railway welded joint state.
Background technology
Steel rail weld joint is the weak link of high speed railway, has become the key equipment of restriction road speed and decision maintenance quality.The judge of steel rail weld joint is comprised to static judge and two kinds of dynamic judges, and main employing static state judge at present, dynamically passes judgment on and lacks.
Static judge operation is fairly simple, first utilizes the straight instrument of electronics to measure the flatness of steel rail weld joint, then according to certain criterion, passes judgment on its state.As, within the scope of the 1m centered by weld seam, if flatness height fluctuation difference in any 200mm section is not more than 0.2mm, thinks that state is qualified, otherwise need repairing.Fig. 1 is the static measurement result schematic diagram of the flatness of steel rail weld joint, and wherein upper and lower bound forms optimum waveform, and the ultimate range between them is called optimum peak value, conventionally value 0.2.
Static optimum waveform and the optimum peak value of passing judgment on rule of thumb determined, and lacks theoretical foundation.In addition, the impact of the flatness of welded joint on vehicle-rail system dynamic characteristics, except relevant with its optimum peak value, simultaneously with the contact condition of wheel track, the suspension parameter of vehicle is closely related.Therefore, in order dynamically to grasp the state of steel rail weld joint, instruct maintenance, be necessary very much to carry out the dynamic judge of welded joint.
According to wavelength division, steel rail weld joint irregularity belongs to Short wave irregularity, wavelength is generally between 0.1~1m, and the detection wavelength of track geometry checking system is conventionally between 3~120m, the Short wave irregularity that is difficult to pass judgment on steel rail weld joint, on the safety of high speed railway track-Vehicular system operation and the impact of traveling comfort, need to be studied other auxiliary judge means.
Axle box directly with wheel to being connected, track irregularity is by taking turns being directly delivered on axle box.Suppose wheel pairing approximation regard rigid construction as, the impact of the excitation power that axle box vertical acceleration can directly reflect the caused external world of track Short wave irregularity on vehicle dynamics.Therefore, select to adopt the auxiliary track Short wave irregularity of passing judgment on of axle box vertical acceleration on the impact of vehicle dynamics performance [list of references 1-7] both at home and abroad.Fig. 2 is the axle box vertical acceleration signal schematic representation of surveying on ripple abrasion rail, and Fig. 3 is the axle box vertical acceleration signal schematic representation on normal rail.
Although both at home and abroad carried out a large amount of research [list of references 8-10] utilizing aspect axle box vertical acceleration diagnosis rail wear, but how to utilize axle box vertical acceleration to pass judgment on welded joint state and yet there are no pertinent literature report and patent description, and the state of welded joint is significant for transporting safely of high speed railway.
Summary of the invention
For the defect of prior art, the embodiment of the present invention provides a kind of method of inspection and system of high speed railway welded joint state, and this scheme utilizes axle box vertical acceleration to detect the state of welded joint, has filled up the domestic and international technological gap in this area.
To achieve these goals, the embodiment of the present invention provides a kind of method of inspection of rail in high speed railway welded joint state, and described method comprises: axle box vertical acceleration is carried out to time sampling, obtain the time sampling signal of axle box vertical acceleration; Time sampling signal to described axle box vertical acceleration carries out bandpass filtering; The effective value of the axle box vertical acceleration after calculation of filtered; According to the effective value of described axle box vertical acceleration, determine the position of welded joint, and extract the effective value of the axle box vertical acceleration at each welded joint place; Whether the axle box vertical acceleration effective value that judges welded joint place is greater than threshold value, if be greater than threshold value, judges that this welded joint state is bad.
Before the time sampling signal to described axle box vertical acceleration carries out bandpass filtering, described method also comprises: the lower-frequency limit F that presets bandpass filtering l, and the upper cut-off frequency F that dynamically determines bandpass filtering h; Wherein, dynamically determine the upper cut-off frequency F of bandpass filtering hspecifically comprise: in frequency limit corresponding to the power spectrum of axle box vertical acceleration, axle box vertical acceleration is carried out to resonance and demodulation, obtain respectively the power envelope spectrum of the axle box vertical acceleration corresponding with each frequency in described frequency limit; Calculate the percentum that the first dominant frequency energy in each power envelope spectrum accounts for this power envelope spectrum gross energy, draw the relation curve of described percentum and its corresponding frequency; The frequency corresponding to peak value of percentum in described relation curve is defined as to the upper cut-off frequency of bandpass filtering.
Described in frequency limit corresponding to the power spectrum of axle box vertical acceleration, axle box vertical acceleration is carried out to resonance and demodulation, and the power envelope spectrum that obtains respectively the axle box vertical acceleration corresponding with each frequency in described frequency limit comprises: the energy of supposing axle box vertical acceleration is mainly distributed in [F a-F b] between Hz, to frequency from F ahz is to F bhz scans, and supposes that frequency increment is F chz, the sweep rate of the j time is F j=Fa+ (j-1) * FcHz; Repeat the following step A-step C, until complete [F a-F b] scanning of Hz frequency limit; A, axle box vertical acceleration is carried out to [F l, F j] Hz bandpass filtering, generate filtered signal x j(t); F wherein lfor lower-frequency limit; B, utilize Hilbert conversion to extract x j(t) envelope, obtains a j(t); C, to a j(t) carrying out refinement Fourier analysis obtains and F jthe power envelope spectrum of corresponding axle box vertical acceleration.
The effective value of the axle box vertical acceleration after described calculation of filtered comprises: suppose that the axle box vertical acceleration signal after bandpass filtering is { x i, i=1,2 ... N}, N represents sampling number, the window width of supposing effective value is K; The sum of squares of K point before calculating obtain the 1st effective value: to j circulation, j=2,3 ... N-K+1, calculates all the other effective values: RMS j = ( S + x j + K - 1 2 - x j - 1 2 ) / K .
According to the effective value of described axle box vertical acceleration, determine the position of welded joint, and the effective value that extracts the axle box vertical acceleration at each welded joint place comprises: a plurality of effective values are divided into one group, calculate aviation value m and the variances sigma of every group of effective value, according to described aviation value m and variances sigma, draw the reference value R of axle box vertical acceleration effective value t; Axle box vertical acceleration effective value is greater than to reference value R tthe transfinite location positioning at a place of large value be the large value welded joint that transfinites; According to welded joint periodic spacings, determine the transfinite position of all welded joints of welded joint centre of large value; Get near a plurality of large values same welded joint and transfinite the maxim of an effective value as the effective value of the axle box vertical acceleration at this welded joint place.
Before whether the axle box vertical acceleration effective value that judges welded joint place is greater than threshold value, described method also comprises: according to the effective value of the axle box vertical acceleration at described welded joint place, draw the distribution curve of the axle box vertical acceleration effective value at welded joint place; By in the distribution curve of the axle box vertical acceleration effective value at welded joint place, corresponding probability is greater than sets the effective value of threshold value as the threshold value of detection welded joint state.
To achieve these goals, the embodiment of the present invention also provides a kind of checking system of rail in high speed railway welded joint state, described system comprises: acceleration/accel sampling setup, for axle box vertical acceleration is carried out to time sampling, obtains the time sampling signal of axle box vertical acceleration; Band pass filter means, carries out bandpass filtering for the time sampling signal to described axle box vertical acceleration; Effective value computer device, for the effective value of the axle box vertical acceleration after calculation of filtered; Welded joint registration device, for determine the position of welded joint according to the effective value of described axle box vertical acceleration, and extracts the effective value of the axle box vertical acceleration at each welded joint place; Connector status detecting device, for judging whether the axle box vertical acceleration effective value at welded joint place is greater than threshold value, if be greater than threshold value, judges that this welded joint state is bad.
Described system also comprises the logical upper cut-off frequency generating apparatus of band, the logical upper cut-off frequency generating apparatus of described band comprises: resonance and demodulation unit, for in frequency limit corresponding to the power spectrum of axle box vertical acceleration, axle box vertical acceleration is carried out to resonance and demodulation, obtain respectively the power envelope spectrum of the axle box vertical acceleration corresponding with each frequency in described frequency limit; Percentum Drawing of Curve unit, accounts for the percentum of this power envelope spectrum gross energy for calculating the first dominant frequency energy of each power envelope spectrum, draw the relation curve of described percentum and its corresponding frequency; Upper cut-off frequency determining unit, for being defined as frequency corresponding to the peak value of described relation curve percentum the upper cut-off frequency of bandpass filtering.
Described welded joint registration device comprises: reference value generation unit, for a plurality of effective values are divided into one group, calculate aviation value m and the variances sigma of every group of effective value, and according to described aviation value m and variances sigma, draw the reference value R of axle box vertical acceleration effective value t; The first welded joint determining unit, for being greater than reference value R by axle box vertical acceleration effective value tthe transfinite location positioning at a place of large value be the large value welded joint that transfinites; The second welded joint determining unit, for according to welded joint periodic spacings, determines the transfinite position of all welded joints of welded joint centre of large value; Welded joint effective value determining unit, transfinites the maxim of an effective value as the effective value of the axle box vertical acceleration at this welded joint place for getting near a plurality of large value same welded joint.
Described system also comprises threshold process unit, described threshold process unit specifically comprises: distribution curve drawing unit, for according to the effective value of the axle box vertical acceleration at described welded joint place, draw the distribution curve of the axle box vertical acceleration effective value at welded joint place; Threshold value determining unit, for by the distribution curve of the axle box vertical acceleration effective value at welded joint place, corresponding probability be greater than set threshold value effective value as the threshold value that detects welded joint state.
The technical scheme of the embodiment of the present invention detects welded joint state completely according to the data of axle box vertical acceleration, can effectively detect the current use state of high speed railway welded joint, for reasonable economy instructs the maintenance of welded joint that strong theoretical foundation is provided.
Accompanying drawing explanation
Fig. 1 is the static measurement result schematic diagram of the flatness of prior art steel rail weld joint;
Fig. 2 is the axle box vertical acceleration signal schematic representation of surveying on ripple abrasion rail;
Fig. 3 is the axle box vertical acceleration signal schematic representation on normal rail;
Fig. 4 is the detailed schematic diagram of the method for inspection of embodiment of the present invention high speed railway welded joint state;
Fig. 4 a is the diagram of circuit of the high speed railway welded joint state-evaluation method of the embodiment of the present invention;
Fig. 5 is the power envelope spectrum schematic diagram of the corresponding axle box vertical acceleration of certain frequency of the embodiment of the present invention;
Fig. 6 is that the embodiment of the present invention the first dominant frequency power percentum is with frequency filtering change curve;
Fig. 7 is the axle box acceleration waveform schematic diagram of embodiment of the present invention actual measurement;
Fig. 8 a is one of effective value schematic diagram of the axle box vertical acceleration that calculates of the embodiment of the present invention;
Fig. 8 b be the axle box vertical acceleration that calculates of the embodiment of the present invention effective value schematic diagram two;
Fig. 8 c be the axle box vertical acceleration that calculates of the embodiment of the present invention effective value schematic diagram three;
Fig. 9 is the distribution curve schematic diagram of embodiment of the present invention actual measurement axle box vertical acceleration effective value;
Figure 10 is embodiment of the present invention high speed railway welded joint state final detection result schematic diagram;
Figure 11 is the schematic diagram of the checking system of embodiment of the present invention rail in high speed railway welded joint state;
Figure 11 a is the refinement functional block diagram of the logical upper cut-off frequency generating apparatus 115 of band in Figure 11;
Figure 11 b is the refinement functional block diagram of welded joint registration device 113 in Figure 11;
Figure 11 c is the refinement functional block diagram of threshold process device 116 in Figure 11.
The specific embodiment
The embodiment of the present invention provides a kind of method of inspection and system of high speed railway welded joint state.Wear away and compare with irregular wear, wave and ripple, the Short wave irregularity at welded joint place had both had high-frequency percussion characteristic, had low-frequency cycle characteristic simultaneously.Utilize axle box vertical acceleration to pass judgment on steel rail weld joint and will run into two crucial problems, i.e. the selection of bandpass filtering frequency and the location of welded joint.Different in concentrating on narrower high-frequency range from undulatory wear, the high-frequency percussion ripple that welded joint causes is distributed in very wide frequency band, and changes.
According to Fourier transform principle, steel rail weld joint momentary impulse is an infinite bandwidth signal from frequency domain, there are in theory all frequency contents, and uniform along whole frequency axis energy, thereby must contain the p-axle box system eigentone composition of wheel, also just must cause the characteristic oscillation of system, produce generalized resonance.The embodiment of the present invention, based on above-mentioned principle, utilizes resonance and demodulation to propose new high speed railway welded joint condition detection method and system.
The groundwork of axle box vertical acceleration being carried out to resonance and demodulation is as follows:
1, axle box vertical acceleration signal is carried out to bandpass filtering, remember that filtered axle box vertical acceleration continuous signal is x (t), t is sampling instant.
2, utilize Hilbert conversion to extract the envelope of x (t), obtain a (t).
3, a (t) is carried out to refinement Fourier analysis, obtain the refinement power envelope spectrum of axle box vertical acceleration.
The dynamic testing method of embodiment of the present invention steel rail weld joint mainly comprises following algorithm: the axle box vertical acceleration based on resonance and demodulation moves the fast algorithm of effective value; Axle box vertical acceleration bandpass filtering upper cut-off frequency self adaptation Algorithms of Selecting based on resonance and demodulation; And the accurate location of the welded joint based on resonance and demodulation and state-detection algorithm.
For making object, technical scheme and the advantage of the embodiment of the present invention clearer, below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is the present invention's part embodiment, rather than whole embodiment.Embodiment based in the present invention, those of ordinary skills are not making the every other embodiment obtaining under creative work prerequisite, all belong to the scope of the claims in the present invention protection.
Fig. 4 is the detailed schematic diagram of the method for inspection of embodiment of the present invention high speed railway welded joint state, as shown in Figure 4, the method comprises two steps: the 1st step (left-half of Fig. 4) utilizes a large amount of actual measurement axle box vertical acceleration data to calculate the effective value at welded joint place, and according to effective value distribution character, calculate threshold value; Particularly, Fig. 7 is the axle box acceleration waveform schematic diagram of actual measurement; Wherein left-half is axle box vertical acceleration signal, and right half part is axle box transverse acceleration signal, can find out, when transverse acceleration variable is large, also phase strain is large for vertical acceleration; Fig. 8 a-Fig. 8 c is the effective value schematic diagram of the axle box vertical acceleration that calculates; Fig. 9 is the distribution curve schematic diagram of axle box vertical acceleration effective value; The 2nd step (right half part of Fig. 4), according to vertical acceleration effective value and the threshold value at welded joint place, detects the state of welded joint online.
The key algorithm of the 1st step and the 2nd step is identical.Key algorithm comprises: axle box vertical acceleration moves fast algorithm, the axle box vertical acceleration bandpass filtering upper cut-off frequency self adaptation Algorithms of Selecting of effective value, the accurate location algorithm of welded joint.In addition, the 1st step also comprises the algorithm of the axle box vertical acceleration effective value distribution of calculating welded joint place, the embodiment of the present invention adopts histogram table to show the probability distribution function of effective value, then it is normalized to the distribution function that obtains effective value, concrete minute 3 steps complete: the 1st step calculates the axle box vertical acceleration effective value at welded joint place, the 2nd step calculates the axle box vertical acceleration effective value histogram at welded joint place, i.e. probability distribution function; To histogram, normalization method obtains its distribution function to the 3rd step.
Fig. 4 a is the diagram of circuit of the high speed railway welded joint state-evaluation method of the embodiment of the present invention, and as shown in Fig. 4 a, the method comprises:
S401, axle box vertical acceleration is carried out to time sampling, obtain the time sampling signal of axle box vertical acceleration.
The collecting device of axle box vertical acceleration is arranged on the axle box of comprehensive detection train, and train, in operational process, gathers the axle box vertical acceleration of a plurality of positions on rail, to the sampling of the vertical acceleration of axle box, is therefore a time sampling.
S402, the time sampling signal of axle box vertical acceleration is carried out to bandpass filtering.
The frequency limit upper limit of bandpass filtering will obtain by described bandpass filtering upper cut-off frequency self adaptation Algorithms of Selecting below, because axle box vertical acceleration is mainly radio-frequency component, low-frequency component proportion seldom, therefore the selection of lower limit on the result impact of filtering seldom, can preset a lower frequency limit F according to actual needs l, as 20Hz.
Suppose the upper cut-off frequency F that the embodiment of the present invention adopts bandpass filtering upper cut-off frequency self adaptation Algorithms of Selecting to obtain h450Hz, with F l=20HZ is example, and the frequency limit of bandpass filtering is [20,450] Hz.It should be noted that; this concrete frequency limit is only for explaining the embodiment of the present invention; and be not used in, the protection domain of claim is limited, what this case claim needed protection is the method that how to obtain this frequency limit, but not concrete frequency limit numerical value.
The axle box vertical acceleration bandpass filtering upper cut-off frequency self adaptation Algorithms of Selecting that the embodiment of the present invention adopts comprises: first, in frequency limit corresponding to the power spectrum of axle box vertical acceleration, axle box vertical acceleration is carried out to resonance and demodulation, obtain respectively the power envelope spectrum of the axle box vertical acceleration corresponding with each frequency in described frequency limit; Then, calculate the percentum that the first dominant frequency energy in each power envelope spectrum accounts for this power envelope spectrum gross energy, draw the relation curve of described percentum and its corresponding frequency; Finally, the frequency corresponding to peak value of percentum in described relation curve is defined as to bandpass filtering cutoff frequency.
Concrete algorithm is as follows:
(1) in frequency limit corresponding to the power spectrum of axle box vertical acceleration, axle box vertical acceleration is carried out to resonance and demodulation, obtain respectively the power envelope spectrum of the axle box vertical acceleration corresponding with each frequency in described frequency limit.
The energy of supposing axle box vertical acceleration is mainly distributed in [F a-F b] between Hz, to frequency from F ahz is to F bhz scans, and supposes that frequency increment is F chz, the sweep rate of the j time is F j=Fa+ (j-1) * FcHz; Repeat the following step A-step C, until complete [F a-F b] scanning of Hz frequency limit.
A, axle box vertical acceleration is carried out to [F l, F j] Hz bandpass filtering, generate filtered signal x j(t); F wherein lfor lower-frequency limit, as 20Hz;
B, utilize Hilbert conversion to extract x j(t) envelope, obtains a j(t);
C, to a j(t) carry out refinement Fourier analysis, obtain and F jthe power envelope spectrum of the axle box vertical acceleration that Hz is corresponding.
In the present embodiment, by known to the power spectral analysis of axle box vertical acceleration, the energy of axle box vertical acceleration is mainly distributed between [100-600] Hz, therefore frequency is scanned from 100Hz to 600Hz, suppose that frequency increment is 10Hz, the sweep rate of the j time is F j=100+ (j-1) * 10Hz; This frequency increment can also change according to realistic accuracy demand.
Steps A, B, C are referred to as the process of resonance and demodulation, by this resonance and demodulation process just can access and frequency limit [100-600] Hz of the power spectrum of axle box vertical acceleration between the corresponding a plurality of power envelopes spectrums of each frequency, Fig. 5 be that the power envelope of the corresponding axle box vertical acceleration of certain frequency is composed schematic diagram.
(2) calculate the percentum that the first dominant frequency energy in each power envelope spectrum accounts for this power envelope spectrum gross energy, draw the relation curve of described percentum and its corresponding frequency.
D, from Fig. 5, therefore the energy of front 5Hz can represent gross energy, calculates respectively the ENERGY E 1 of the first dominant frequency of envelope spectrum and the ENERGY E in front 5Hz;
E, calculating the first dominant frequency energy account for gross energy percentum
Step D, E are also the processes of a circulation, the corresponding power envelope of each frequency spectrum need to be carried out one time, draw the percentum of each power envelope spectrum, draw the relation curve of described percentum and corresponding frequency, as shown in Figure 6.
(3) frequency corresponding to peak value of percentum in described relation curve is defined as to bandpass filtering upper cut-off frequency.
Peak value in Fig. 6 is 450HZ, so obtain the upper limit of frequency filtering scope, is 450Hz.
The effective value of the axle box vertical acceleration after S403, calculation of filtered.
The embodiment of the present invention is when calculating axle box vertical acceleration moves effective value, by time sampling.Suppose that the axle box vertical acceleration signal after bandpass filtering is { x i, i=1,2 ... N}, wherein N represents sampling number, and the window width of supposing effective value is K, and the algorithm of the mobile effective value of axle box vertical acceleration is as follows:
(1) calculate the sum of squares of front K point obtain the 1st effective value:
RMS 1 = S / K - - - ( 1 )
(2) to j circulation, j=2,3 ... N-K+1, calculates all the other effective values:
RMS j = ( S + x j + K - 1 2 - x j - 1 2 ) / K - - - ( 2 )
Casehistory is as follows: suppose sampling number N=11, and the long K=7 of window, the axle box vertical acceleration signal after bandpass filtering is { x i, i=1,2 ... 11}, the 1st definition to N-K+1=5 effective value and being calculated as follows:
RMS 1 = 1 7 Σ i = 1 7 x i 2
RMS 2 = 1 7 Σ i = 2 8 x i 2
RMS 3 = 1 7 Σ i = 3 9 x i 2
RMS 4 = 1 7 Σ i = 4 10 x i 2
RMS 5 = 1 7 Σ i = 5 11 x i 2
By calculated amount, analyze knownly, be linear the computing time of the algorithm of the mobile effective value of above-mentioned calculating axle box vertical acceleration, and computation complexity is O (N).
S404, according to the effective value of axle box vertical acceleration, determine the position of welded joint, and extract the effective value of the axle box vertical acceleration at each welded joint place.
Specifically comprise: a plurality of effective values are divided into one group, calculate aviation value m and the variances sigma of every group of effective value, according to described aviation value m and variances sigma, draw the reference value R of axle box vertical acceleration effective value t; Axle box vertical acceleration effective value is greater than to reference value R tthe transfinite location positioning at a place of large value be the large value welded joint that transfinites; According to welded joint periodic spacings, determine the transfinite position of all welded joints of welded joint centre of large value; Get near a plurality of large values same welded joint and transfinite the maxim of an effective value as the effective value of the axle box vertical acceleration at this welded joint place.
Welded joint Precision Orientation Algorithm is as follows:
(1) a plurality of effective values are divided into one group, calculate aviation value m and the variances sigma of the effective value of each group, as, the reference value R of the effective value of axle box vertical acceleration can be calculated by formula (3) t, the reference value obtaining, as shown in the dotted line in Fig. 8 a-Fig. 8 c, has only demonstrated the reference value of a group in Fig. 8 a-Fig. 8 c;
R T=m+2.5σ (3)
(2) find out effective value and be greater than reference value R tpoint, being defined as large value transfinites a little, the transfinite position at a place of these large values is exactly welded joint, is designated as result is as being greater than in Fig. 8 a as shown in zero ' o ' point; Because same welded joint has a plurality of large values to transfinite a little around, so there is a plurality of ' o ' point;
(3) right carry out polymerization, the large value at same welded joint place is transfinited and to retain a maxim, with ' o ' point that guarantees that same welded joint only has 1 large value to transfinite, the welded joint that transfinites of the large value after polymerization is designated as result is as being greater than in Fig. 8 b as shown in zero ' o ' point;
(4) according to welded joint periodic spacings, be approximately 100m, find out the welded joint that large value transfinites in the middle of welded joint, finally obtain all welded joints, and axle box vertical acceleration effective value corresponding to all welded joints, as being greater than in Fig. 8 c as shown in zero ' o ' point.
If the axle box vertical acceleration effective value at S405 welded joint place is greater than threshold value, judge that this welded joint state is bad.
The mode of definite threshold is as follows: according to the effective value of the axle box vertical acceleration at described welded joint place, draw the distribution curve of the axle box vertical acceleration effective value at welded joint place; By in the distribution curve of the axle box vertical acceleration effective value at welded joint place, corresponding probability is greater than sets the effective value of threshold value as the threshold value of detection welded joint state.
Take Fig. 9 as example, and in the distribution curve of the axle box vertical acceleration effective value at the welded joint place shown in Fig. 9, hypothetical probabilities threshold value is 99.5%, and corresponding probability is that 99.5% effective value 50m/s/s is exactly the threshold value of passing judgment on welded joint state.As shown in figure 10, the effective value of the welded joint that state is bad is apparently higher than the effective value at other welded joint place for final detection result.
Corresponding to the method for Fig. 4 a, the embodiment of the present invention also provides a kind of high speed railway welded joint state-evaluation system.Figure 11 is the schematic diagram of the checking system of embodiment of the present invention rail in high speed railway welded joint state, and as shown in figure 11, this system comprises:
Acceleration/accel sampling setup 110, for axle box vertical acceleration is carried out to time sampling, obtains the time sampling signal of axle box vertical acceleration; Band pass filter means 111, carries out bandpass filtering for the time sampling signal to described axle box vertical acceleration; Effective value computer device 112, for the effective value of the axle box vertical acceleration after calculation of filtered; Welded joint registration device 113, for determine the position of welded joint according to the effective value of described axle box vertical acceleration, and extracts the effective value of the axle box vertical acceleration at each welded joint place; Connector status detecting device 114, for judging whether the axle box vertical acceleration effective value at welded joint place is greater than threshold value, if be greater than threshold value, judges that this welded joint state is bad.
As shown in figure 11, this system also comprises the logical upper cut-off frequency generating apparatus 115 of band, as shown in Figure 11 a, the logical upper cut-off frequency generating apparatus of this band comprises: resonance and demodulation unit 1150, for in frequency limit corresponding to the power spectrum of axle box vertical acceleration, axle box vertical acceleration is carried out to resonance and demodulation, obtain respectively the power envelope spectrum of the axle box vertical acceleration corresponding with each frequency in described frequency limit; Percentum Drawing of Curve unit 1151, accounts for the percentum of this power envelope spectrum gross energy for calculating the first dominant frequency energy of each power envelope spectrum, draw the relation curve of described percentum and its corresponding frequency; Upper cut-off frequency determining unit 1152, for being defined as frequency corresponding to the peak value of described relation curve percentum the upper cut-off frequency of bandpass filtering.
Figure 11 b is the refinement functional block diagram of welded joint registration device 113, as shown in Figure 11 b, welded joint registration device 113 specifically comprises: reference value generation unit 1130, for a plurality of effective values are divided into one group, calculate aviation value m and the variances sigma of every group of effective value, according to described aviation value m and variances sigma, draw the reference value R of axle box vertical acceleration effective value t; The first welded joint determining unit 1131, for being greater than reference value R by axle box vertical acceleration effective value tthe transfinite location positioning at a place of large value be the large value welded joint that transfinites; The second welded joint determining unit 1132, for according to welded joint periodic spacings, determines the transfinite position of all welded joints of welded joint centre of large value; Welded joint effective value determining unit 1133, transfinites the maxim of an effective value as the effective value of the axle box vertical acceleration at this welded joint place for getting near a plurality of large value same welded joint.
As shown in figure 11, this system also comprises threshold process device 116; As shown in Figure 11 c, threshold process device 116 specifically comprises: distribution curve drawing unit 1160, for according to the effective value of the axle box vertical acceleration at described welded joint place, draw the distribution curve of the axle box vertical acceleration effective value at welded joint place; Threshold value determining unit 1161, for by the distribution curve of the axle box vertical acceleration effective value at welded joint place, corresponding probability be greater than set threshold value effective value as the threshold value that detects welded joint state.
The beneficial effect that technical solution of the present invention is brought: by the in-depth analysis of the axle box vertical acceleration data to more than 1000 kilometer of actual measurement of certain line, binding resonant demodulation techniques calculate the distribution function curve of welded joint place effective value can be found out, the distribution curve that under identical operating mode, twice collection calculates coincide consistent.Therefore, new welded joint Precision Orientation Algorithm is stable, and the distribution curve obtaining has good repeatability.And, utilize new method of inspection to pass judgment on welded joint state completely, can effectively detect high speed railway welded joint state, for reasonable economy instructs the maintenance of welded joint that strong theoretical foundation is provided.
Above embodiment only, in order to the technical scheme of the embodiment of the present invention to be described, is not intended to limit; Although the embodiment of the present invention is had been described in detail with reference to previous embodiment, those of ordinary skill in the art is to be understood that: its technical scheme that still can record aforementioned each embodiment is modified, or part technical characterictic is wherein equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution depart from the spirit and scope of each embodiment technical scheme of the embodiment of the present invention.
List of references:
[1]C.Esweld.Modern Railway Track[M].Duisburg:MRT-productions,2001.
[2]M.Bocciolone,A. ,A.Cigada,A.Collina.A Measurement System for Quick Rail Inspection and Effective Track Maintenance Strategy[J].Mechanical Systems and Signal Processing,2007(21):1242-1254.
[3]Sunaga Y.,T.Ide.and M.Kanao.A practical use of axle-box acceleration to control the short wave track irregularities on SHINKANSEN[J].RTRI quarterly Report 1995,9(2):35-40.
[4]Sunaga Y.,Snal,I.and Ide,T..A method to control the short wave track irregularities utilizing axle-box acceleration[J].RTRI quarterly Report,1997,38(4):176-181.
[5]Yoshida,M.,Shinowaki,S.and Sunaga,Y..A study on lateral axle-box acceleration for detecting irregularities of rail joints on sharp curves[J].Jointed Railway Technology Symposium(J-Rail2004),2004,11:91-54.
[6]Hirofumi TANAKA,Atsushi FURUKAWA.Management Method for Large Wheel Loads and Lateral Forces Using Axle-box Acceleration and Track Irregularity[J].RTRI quarterly Report,2009,50(2):116-122.
[7]Eric G.Berggren,Martin X.D.Li,Jan Spannar.A new approach to the analysis and presentation of vertical track geometry quality and rail roughness[J].Wear 2008(265):1488-1496.
[8]A Bracciali and G Cascini.Detection of corrugation and wheelflats of railway wheels using energy and cepstrum analysis of rail acceleration[J].Proc Instn Mech Engrs,1997,211,(Part E):109-116.
[9]Grassie,S.L.Rail corrugation:advances in measurement,understanding and treatment[J].Wear,2005,258:1224-1234.
[10]Hempelmann,K.and Knothe,K.An extended linear model for the prediction of short pitch corrugation[J].Wear,1996,191:161-169.

Claims (9)

1. a method of inspection for rail in high speed railway welded joint state, is characterized in that, described method comprises:
Axle box vertical acceleration is carried out to time sampling, obtain the time sampling signal of axle box vertical acceleration;
Time sampling signal to described axle box vertical acceleration carries out bandpass filtering;
The effective value of the axle box vertical acceleration after calculation of filtered;
According to the effective value of described axle box vertical acceleration, determine the position of welded joint, and extract the effective value of the axle box vertical acceleration at each welded joint place;
Whether the axle box vertical acceleration effective value that judges welded joint place is greater than threshold value, if be greater than threshold value, judges that this welded joint state is bad;
The effective value of the axle box vertical acceleration after described calculation of filtered comprises:
Suppose that the axle box vertical acceleration signal after bandpass filtering is { x i, i=1,2 ... N}, N represents sampling number, the window width of supposing effective value is K; The sum of squares of K point before calculating obtain the 1st effective value: RMS 1 = S / K ;
To j circulation, j=2,3 ... N-K+1, calculates all the other effective values:
2. method according to claim 1, is characterized in that, before the time sampling signal to described axle box vertical acceleration carries out bandpass filtering, described method also comprises: the lower-frequency limit F that presets bandpass filtering l, and the upper cut-off frequency F that dynamically determines bandpass filtering h; Wherein, dynamically determine the upper cut-off frequency F of bandpass filtering hspecifically comprise:
In frequency limit corresponding to the power spectrum of axle box vertical acceleration, axle box vertical acceleration is carried out to resonance and demodulation, obtain respectively the power envelope spectrum of the axle box vertical acceleration corresponding with each frequency in described frequency limit;
Calculate the percentum that the first dominant frequency energy in each power envelope spectrum accounts for this power envelope spectrum gross energy, draw the relation curve of described percentum and its corresponding frequency;
The frequency corresponding to peak value of percentum in described relation curve is defined as to the upper cut-off frequency of bandpass filtering.
3. method according to claim 2, it is characterized in that, described in frequency limit corresponding to the power spectrum of axle box vertical acceleration, axle box vertical acceleration is carried out to resonance and demodulation, and the power envelope spectrum that obtains respectively the axle box vertical acceleration corresponding with each frequency in described frequency limit comprises:
The energy of supposing axle box vertical acceleration is mainly distributed in [F a-F b] between Hz, to frequency from F ahz is to F bhz scans, and supposes that frequency increment is F chz, the sweep rate of the j time is F j=Fa+ (j-1) * FcHz; Repeat the following step A-step C, until complete [F a-F b] scanning of Hz frequency limit;
A, axle box vertical acceleration is carried out to [FL, F j] Hz bandpass filtering, generate filtered signal x j(t); F wherein lfor lower-frequency limit;
B, utilize Hilbert conversion to extract x j(t) envelope, obtains a j(t);
C, to a j(t) carry out refinement Fourier analysis, obtain and F jthe power envelope spectrum of the axle box vertical acceleration that Hz is corresponding.
4. method according to claim 1, is characterized in that, determines the position of welded joint according to the effective value of described axle box vertical acceleration, and the effective value that extracts the axle box vertical acceleration at each welded joint place comprises:
A plurality of effective values are divided into one group, calculate aviation value m and the variances sigma of every group of effective value, according to described aviation value m and variances sigma, draw the reference value R of axle box vertical acceleration effective value t;
Axle box vertical acceleration effective value is greater than to reference value R tthe transfinite location positioning at a place of large value be the large value welded joint that transfinites;
According to welded joint periodic spacings, determine the transfinite position of all welded joints of welded joint centre of large value;
Get near a plurality of large values same welded joint and transfinite the maxim of an effective value as the effective value of the axle box vertical acceleration at this welded joint place.
5. method according to claim 1, is characterized in that, before whether the axle box vertical acceleration effective value that judges welded joint place is greater than threshold value, described method also comprises:
According to the effective value of the axle box vertical acceleration at described welded joint place, draw the distribution curve of the axle box vertical acceleration effective value at welded joint place;
By in the distribution curve of the axle box vertical acceleration effective value at welded joint place, corresponding probability is greater than sets the effective value of threshold value as the threshold value of detection welded joint state.
6. a checking system for rail in high speed railway welded joint state, is characterized in that, described system comprises:
Acceleration/accel sampling setup, for axle box vertical acceleration is carried out to time sampling, obtains the time sampling signal of axle box vertical acceleration;
Band pass filter means, carries out bandpass filtering for the time sampling signal to described axle box vertical acceleration;
Effective value computer device, the effective value for the axle box vertical acceleration after calculation of filtered, specifically comprises: suppose that the axle box vertical acceleration signal after bandpass filtering is { x i, i=1,2 ... N}, N represents sampling number, the window width of supposing effective value is K; The sum of squares of K point before calculating obtain the 1st effective value: to j circulation, j=2,3 ... N-K+1, calculates all the other effective values: RMS j = ( S + x j + K - 1 2 - x j - 1 2 ) / K ;
Welded joint registration device, for determine the position of welded joint according to the effective value of described axle box vertical acceleration, and extracts the effective value of the axle box vertical acceleration at each welded joint place;
Connector status detecting device, for judging whether the axle box vertical acceleration effective value at welded joint place is greater than threshold value, if be greater than threshold value, judges that this welded joint state is bad.
7. system according to claim 6, is characterized in that, described system also comprises the logical upper cut-off frequency generating apparatus of band, and the logical upper cut-off frequency generating apparatus of described band comprises:
Resonance and demodulation unit, in frequency limit corresponding to the power spectrum in axle box vertical acceleration, carries out resonance and demodulation to axle box vertical acceleration, obtains respectively the power envelope spectrum of the axle box vertical acceleration corresponding with each frequency in described frequency limit;
Percentum Drawing of Curve unit, accounts for the percentum of this power envelope spectrum gross energy for calculating the first dominant frequency energy of each power envelope spectrum, draw the relation curve of described percentum and its corresponding frequency;
Upper cut-off frequency determining unit, for being defined as frequency corresponding to the peak value of described relation curve percentum the upper cut-off frequency of bandpass filtering.
8. system according to claim 6, is characterized in that, described welded joint registration device comprises:
Reference value generation unit, for a plurality of effective values are divided into one group, calculates aviation value m and the variances sigma of every group of effective value, draws the reference value R of axle box vertical acceleration effective value according to described aviation value m and variances sigma t;
The first welded joint determining unit, for being greater than reference value R by axle box vertical acceleration effective value tthe transfinite location positioning at a place of large value be the large value welded joint that transfinites;
The second welded joint determining unit, for according to welded joint periodic spacings, determines the transfinite position of all welded joints of welded joint centre of large value;
Welded joint effective value determining unit, transfinites the maxim of an effective value as the effective value of the axle box vertical acceleration at this welded joint place for getting near a plurality of large value same welded joint.
9. system according to claim 6, is characterized in that, described system also comprises threshold process unit, and described threshold process unit specifically comprises:
Distribution curve drawing unit, for according to the effective value of the axle box vertical acceleration at described welded joint place, draws the distribution curve of the axle box vertical acceleration effective value at welded joint place;
Threshold value determining unit, for by the distribution curve of the axle box vertical acceleration effective value at welded joint place, corresponding probability be greater than set threshold value effective value as the threshold value that detects welded joint state.
CN201110344671.6A 2011-11-04 2011-11-04 Method and system for detecting states of steel rail welding joints of high-speed railway Active CN102501886B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201110344671.6A CN102501886B (en) 2011-11-04 2011-11-04 Method and system for detecting states of steel rail welding joints of high-speed railway

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201110344671.6A CN102501886B (en) 2011-11-04 2011-11-04 Method and system for detecting states of steel rail welding joints of high-speed railway

Publications (2)

Publication Number Publication Date
CN102501886A CN102501886A (en) 2012-06-20
CN102501886B true CN102501886B (en) 2014-08-13

Family

ID=46214042

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201110344671.6A Active CN102501886B (en) 2011-11-04 2011-11-04 Method and system for detecting states of steel rail welding joints of high-speed railway

Country Status (1)

Country Link
CN (1) CN102501886B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102890143B (en) * 2012-10-19 2015-07-15 西南交通大学 Rail local defect vehicle-mounting detection method merging with vehicle speed information and front and rear axle box acceleration information
CN105509668B (en) * 2015-11-27 2018-11-20 北京锦鸿希电信息技术股份有限公司 Track wave grinds detection system
CN105346561A (en) * 2015-12-02 2016-02-24 北京交通大学 Rail turnout disease detection system based on operating vehicle and rail turnout disease detection method based on operating vehicle
CN108318169A (en) * 2018-01-12 2018-07-24 西南交通大学 Maximum dynamic force appraisal procedure and steel rail welding line maintenance system at a kind of steel rail welding line
CN113501028B (en) * 2021-07-07 2022-08-09 中国铁道科学研究院集团有限公司 Method and device for diagnosing poor welded joint of heavy-duty railway steel rail

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1369595A (en) * 2002-03-15 2002-09-18 西南交通大学 Tester of rail state

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6763312B1 (en) * 2003-01-11 2004-07-13 Dynamic Measurement Consultants, Llc Multiple discriminate analysis and data integration of vibration in rotation machinery

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1369595A (en) * 2002-03-15 2002-09-18 西南交通大学 Tester of rail state

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
M.Molodova et al.An investigation of the possibility to use axle box acceleration for condition monitoring of welds.《international conference of noise and vibration engineering》.2008,图8、文章第4.1及第5部分. *

Also Published As

Publication number Publication date
CN102501886A (en) 2012-06-20

Similar Documents

Publication Publication Date Title
CN108845028B (en) Method and device for dynamically detecting high-speed railway rail corrugation
CN102501886B (en) Method and system for detecting states of steel rail welding joints of high-speed railway
CN108515984B (en) Wheel damage detection method and device
CN102114855B (en) Track detection method and device
CN106394606B (en) A kind of railway car wheel loses circle detection method and detection device
CN202115547U (en) Track geometry status measuring cart and measuring apparatus
CN104359444B (en) Rail outline detection method and system
CN102874278B (en) Vehicle-mounted detection method fusing vehicle speed information and axle box vertical acceleration information for wheel flats
CN105292177A (en) Method for measuring track corrugation by utilizing axle box vibration and impact information
CN110789566B (en) Track defect monitoring method and monitoring equipment based on axle box acceleration signal
CN102706572A (en) Fault diagnosis and rehabilitation center for train wheel sets
CN202644332U (en) Rail top surface irregularity detecting device
CN106018557A (en) Track surface indentation detection method based on wavelet analysis
CN102890143B (en) Rail local defect vehicle-mounting detection method merging with vehicle speed information and front and rear axle box acceleration information
CN206307052U (en) A kind of railway car wheel loses loop truss device
CN103031787A (en) Method and device for fully automatically and accurately monitoring changes of geometric parameters of track
CN106092015A (en) A kind of raceway surface depression length detecting method
CN111332338A (en) Track disease detection system
CN104792275A (en) Tunnel deformation detection unit and method
CN104949848A (en) Hunting instability detecting algorithm based on frequency domain kurtosis
CN101701976A (en) Third rail detection device and method
CN105128888B (en) Track waveform abrasion method for quick based on many midpoint string synchro measures
CN204775291U (en) Device is exerted to track geometry not smooth -going detecting system's load based on initiative excitation
CN106052606A (en) Track surface recess detection method based on scale average wavelet energy spectrum
CN108016990A (en) The detection device and detection method that a kind of portal crane trolley track fastener loosens

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant