CN114330429A - Steel rail scratch recognition method, device, system, equipment and storage medium - Google Patents

Abstract

This paper provides a rail scratch identification method, device, system, equipment and storage medium, wherein the method includes: performing energy conversion on the acquired eddy current signal of the rail to obtain an energy signal corresponding to the eddy current signal; judging whether the amplitude of the energy signal is It is greater than the adaptive energy threshold corresponding to the energy signal; when the amplitude is less than the adaptive energy threshold, it is determined that the rail does not have scratches; when the amplitude is greater than or equal to the adaptive energy threshold, it is determined that the rail has scratches. In this paper, the energy signal obtained from the eddy current signal conversion can be compared with the adaptive energy threshold, which can eliminate the influence of factors such as rail corrugation, corrosion and other factors on the difference in electrical conductivity and magnetic permeability of the rail, and realize the automatic identification of rail scratches. , While improving the scratch recognition efficiency, the accuracy of scratch recognition is greatly improved.

Description

A kind of rail scratch identification method, device, system, equipment and storage medium

technical field

The invention relates to the technical field of non-destructive testing, in particular to a method, device, system, equipment and storage medium for rail scratch identification.

Background technique

As an important component of high-speed railway, the service status of steel rail will directly affect the safety of high-speed railway transportation. Rail scratching is one of the common diseases of rails. It is due to the high temperature generated by friction with the contact surface of the wheel and rail, and at the same time, the high contact stress reduces the phase transition temperature of the material, resulting in the phase transition of the metal structure on the top surface of the rail head, which is transformed from pearlite structure. It becomes a hard and brittle white layer, and the white layer structure is easy to be broken and broken under the action of external force, and finally forms cracks or blocks. Scratches on the rails of high-speed railways will affect the smoothness of the rails, resulting in a sharp increase in the impact force of the wheel and rails, which may damage the rail structure, and in severe cases may lead to rail breakage and affect the safety of train operation.

The existing identification methods for rail scratches mainly rely on the ultrasonic flaw detection system and visual inspection system mounted on the rail flaw detection vehicle; the ultrasonic flaw detection technology can only find scratches, but cannot evaluate the area and depth of the scratches; The inspection system can assess the area of the abrasion, but it is difficult to assess the depth of the abrasion. Therefore, for the detected suspected scratches, it is necessary to conduct on-site review combined with manual work, and determine the depth of the scratches and scratches by the method of feeler gauge, but the method of feeler gauge can only be applied to the scratches that have formed blocks. Depth measurement cannot accurately assess the severity of abrasions that only form white layer tissue and do not fall off. That is to say, the existing scratch detection methods have the problems of low detection efficiency and inaccurate detection.

In view of this, this article aims to provide a rail scratch identification method, device, system, equipment and storage medium, which can improve the efficiency and accuracy of scratch identification.

SUMMARY OF THE INVENTION

In view of the above problems in the prior art, the purpose of this paper is to provide a rail scratch identification method, device, system, equipment and storage medium to solve the problems of low efficiency and inaccuracy of rail scratch identification in the prior art.

In order to solve the above technical problems, the specific technical solutions in this paper are as follows:

In the first aspect, this paper provides a method for identifying rail scratches, including:

performing energy conversion on the acquired eddy current signal of the rail to obtain an energy signal corresponding to the eddy current signal;

Judging whether the amplitude of the energy signal is greater than the adaptive energy threshold corresponding to the energy signal;

When the amplitude is less than the adaptive energy threshold, it is determined that there is no scratch on the rail;

When the amplitude is greater than or equal to the adaptive energy threshold, it is determined that the rail is scratched.

Specifically, after judging that the rail is scratched, the method further includes:

determining parameters of the scratch based on the eddy current signal, the energy signal, and an array sensor for acquiring the eddy current signal, the parameters including the length, width and depth of the scratch;

The shape and type of the abrasion are determined according to the parameters; the shape includes an elongated shape and an oval shape, and the type of abrasion includes a superficial white layer tissue, a deep layer white layer combination and a lump.

Further, determining the parameters of the scratch based on the eddy current signal, the energy signal and the array sensor for acquiring the eddy current signal includes:

Obtain the position interval corresponding to the energy signal when its amplitude satisfies the preset condition as the first characteristic value;

obtaining the number of channels of the array sensor corresponding to the scratch as a second characteristic value;

Calculate the difference between the maximum value and the minimum value of the amplitude of the eddy current signal in the position interval as a third eigenvalue;

According to the first characteristic value, the second characteristic value and the third characteristic value, the length, width and depth of the scratch are respectively determined.

Further, determining the type of the abrasion according to the parameter, including:

judging whether the third feature value is greater than the first depth threshold and less than or equal to the second depth threshold;

When the third feature value is greater than the first depth threshold and less than or equal to the second depth threshold, determining that the type of the abrasion is deep white layer tissue;

When the third characteristic value is less than or equal to the first depth threshold, determining that the type of the abrasion is a superficial white layer tissue;

When the third feature value is greater than the second depth threshold, it is determined that the type of the scratch is a block drop.

Specifically, the obtained eddy current signal of the rail is calculated by the following formula to obtain the energy signal:

Among them, E(n) is the energy signal, the value range of n is 1～M+1-N, M is the sequence length of the eddy current signal, N is the calculation window width, and x(m) is the eddy current detected at m Signal.

Further, the energy threshold is obtained by the following steps:

Calculate the average energy of the energy signal to obtain the initial energy threshold, and the calculation formula of the average energy is:

Among them, Th is the initial energy threshold, and E(i) is the energy of the ith signal;

The initial energy threshold is corrected to obtain the adaptive energy threshold, and the correction formula is:

THR=a×Th+b;

Among them, THR is the adaptive energy threshold, a is the amplification coefficient, and b is the bias coefficient.

Preferably, before calculating the acquired eddy current signal of the rail to obtain the energy signal, the method further includes:

The eddy current signal is filtered and denoised.

Further, the filtering and denoising processing of the eddy current signal is further:

Band-pass filtering the eddy current signal by the finite impulse response of the equiripple method;

The wavelet denoising is performed on the eddy current signal processed by bandpass filtering using the soft threshold method based on the minimax criterion based on db3 basis.

In a second aspect, this article provides a rail scratch identification device, comprising:

a conversion module, configured to perform short-term energy conversion on the acquired eddy current signal of the rail to obtain an energy signal corresponding to the eddy current signal;

a judging module for judging whether the amplitude of the energy signal is greater than the adaptive energy threshold of the energy signal;

a first determination module, configured to determine that there is no scratch on the rail when the amplitude is less than the energy threshold;

The second determination module is configured to determine that the rail is scratched when the amplitude is greater than or equal to the energy threshold.

In a third aspect, this paper also provides a rail scratch identification system, the system includes a probe, a probe support structure, a signal excitation unit, a signal acquisition and processing unit, and a controller;

The probe support structure is connected with the probe, and is used for fixing the probe to keep the distance between the probe and the surface of the rail to be probed stable;

The side of the probe close to the rail is provided with an array sensor, the array sensor is connected with the signal excitation unit and the signal acquisition and processing unit, and is used to generate an AC magnetic field under the AC excitation of the signal excitation unit, and to generate an AC magnetic field. The detected eddy current signal generated by the rail under the AC magnetic field is fed back to the signal acquisition and processing unit;

The signal acquisition and processing unit is connected with the signal excitation unit and the controller, and is used for regulating the parameters of the AC excitation of the signal excitation unit under the control of the controller, and conditioning the acquired eddy current signal Amplify feedback to the controller;

The controller is used to identify rail galling based on the received conditioned amplified eddy current signal.

In a fourth aspect, this document also provides a computer device, including a memory, a processor, and a computer program stored in the memory and running on the processor, the processor implementing the computer program provided by the above technical solution when the processor executes the computer program. method.

In a fifth aspect, this document also provides a storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method provided by the above technical solution is implemented.

By adopting the above technical solution, the method, device, system, equipment and storage medium for rail scratch identification described in this paper can compare the energy signal obtained by converting the eddy current signal with the adaptive energy threshold, and the adaptive energy threshold and the energy signal are one by one. Correspondingly, it can eliminate the influence of factors such as the environment where different rails are located, the differences in electrical conductivity and magnetic permeability of different rails on the scratch discrimination, realize automatic identification of rail scratches, improve the efficiency of scratch identification, and greatly improve the scratch resistance. accuracy of injury identification.

In order to make the above-mentioned and other objects, features and advantages of this paper more obvious and easy to understand, preferred embodiments are hereinafter described in detail in conjunction with the accompanying drawings.

Description of drawings

In order to more clearly illustrate the embodiments or the technical solutions in the prior art, the following briefly introduces the accompanying drawings that are used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only For some embodiments herein, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative effort.

Fig. 1 is the step schematic diagram of a kind of rail scratch identification method provided in the embodiment of this paper;

2(a) to 2(f) are schematic diagrams of several unprocessed eddy current signals;

Figures 3(a) to 3(f) are schematic diagrams of eddy current signals obtained by bandpass filtering and wavelet denoising processing of the eddy current signals shown in Figures 2(a) to 2(f);

FIG. 4 shows a schematic diagram of steps of another rail scratch identification method provided by the embodiments of this document;

FIG. 5 shows a flow chart of steps of a method for determining a scratch size parameter provided by an embodiment of this document;

6 is a schematic diagram of obtaining a first eigenvalue of an energy signal;

FIG. 7 is a schematic diagram of the principle of the array sensor for scratch detection on the rail;

8 is a schematic diagram of obtaining a third eigenvalue of the eddy current signal corresponding to the energy signal shown in FIG. 6;

FIG. 9 shows a flow chart of steps of a method for determining the type of abrasion provided by the embodiments of this document;

FIG. 10 shows a schematic structural diagram of a rail scratch identification device provided in an embodiment of this document;

FIG. 11 shows a schematic structural diagram of a rail scratch identification system provided by an embodiment of this document;

FIG. 12 shows a schematic structural diagram of a computer device provided by an embodiment of this document.

Description of the symbols in the drawings:

10. Probe;

20. Probe support structure;

30. Signal excitation unit;

40. Signal acquisition and processing unit;

50. Controller;

60. Steel rail;

101. Conversion module;

102. Judgment module;

103. A first determination module;

104. A second determination module;

1202. Computer equipment;

1204. processor;

1206. memory;

1208. Drive mechanism;

1210. Input/output module;

1212. Input device;

1214. Output device;

1216. Presentation equipment;

1218. Graphical user interface;

1220, network interface;

1222. Communication link;

1224. A communication bus.

Detailed ways

The technical solutions in the embodiments herein will be clearly and completely described below with reference to the accompanying drawings in the embodiments herein. Obviously, the described embodiments are only a part of the embodiments herein, rather than all the embodiments. Based on the embodiments herein, all other embodiments obtained by persons of ordinary skill in the art without creative efforts shall fall within the scope of protection herein.

It should be noted that the terms "first", "second" and the like in the description and claims herein and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific sequence or sequence. It is to be understood that data so used may be interchanged under appropriate circumstances such that the embodiments herein described can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having", and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, apparatus, product or device comprising a series of steps or units is not necessarily limited to those expressly listed Rather, those steps or units may include other steps or units not expressly listed or inherent to these processes, methods, products or devices.

The existing methods for identifying rail scratches mainly rely on the ultrasonic flaw detection system and the visual inspection system mounted on the rail flaw detection vehicle. However, ultrasonic flaw detection technology can only find scratches, and cannot evaluate the area and depth of scratches; the inspection system can evaluate the area of scratches, but it is difficult to evaluate the depth of scratches. Therefore, it is necessary to perform on-site review in combination with manual work. For example, the depth of scratches and scratches can be determined by the method of feeler gauge, but the method of feeler gauge can only be applied to the depth measurement of scratches that have formed blocks. There was no scraping of the white layer tissue, and its severity could not be accurately assessed. Therefore, the existing detection methods have the problems of low detection efficiency and low accuracy.

In order to solve the above problems, the embodiments herein provide a method, device, system, equipment and storage medium for rail scratch identification, which can improve the efficiency and accuracy of scratch identification. FIG. 1 is a schematic diagram of steps of a method for identifying rail scratches provided by the embodiments of this document. The present specification provides the operation steps of the method as described in the embodiments or the flowcharts, but based on conventional or non-creative work, it may include more or more Fewer steps. The sequence of steps enumerated in the embodiments is only one of the execution sequences of many steps, and does not represent the only execution sequence. When an actual system or device product is executed, the methods shown in the embodiments or the accompanying drawings may be executed sequentially or in parallel. Specifically, as shown in FIG. 1, the method may include:

S110: Perform energy conversion on the acquired eddy current signal of the rail to obtain an energy signal corresponding to the eddy current signal.

The eddy current signal in the embodiments of this specification is detected by an array sensor disposed on the probe. Exemplarily, a 1×4 array sensor is used in the embodiment of this specification. As shown in FIG. 7 , four array sensors (channel numbers are 1, 2, 3, and 4) arranged side by side correspond to the width direction of the track. It is adapted to move along the length of the rail with the rail scratch identification system, and the rail surface is detected while moving (4 sets of eddy current signals are generated for each detection position).

The detection principle is: the probe generates an AC magnetic field, and the rail is in the AC magnetic field and generates a vortex-shaped induced alternating current (ie eddy current signal). The existence of defects and the influence of factors such as defect size and shape. By analyzing the distribution, size and phase information in the eddy current signal, the defect characteristics of the detected rail can be obtained. It should be noted that the eddy current detection used between the array sensor and the rail is a non-contact detection method, but due to the skin effect, the surface and near-surface structural states of the rail can be detected.

S120: Determine whether the amplitude of the energy signal is greater than an energy threshold adaptive to the energy signal.

In the embodiment of this specification, each energy signal is compared with its own energy threshold, which can eliminate the influence of eddy current signals of different rail surface states on the scratch determination, and improve the accuracy of scratch identification.

S130: when the amplitude is less than the energy threshold, determine that there is no scratch on the rail;

S140: When the amplitude is greater than or equal to the energy threshold, determine that the rail is scratched.

A rail scratch identification method provided by the embodiments of this specification is to convert the detected eddy current signal of the rail into an energy signal, and compare the energy signal with the adaptive energy threshold corresponding to the energy signal one-to-one, thereby, The automatic identification of the existence of scratches on the rail is realized, and compared with the existing method of using a unified judgment threshold, the influence of factors such as the environment in which the rail is located, its own electrical conductivity and magnetic permeability on the scratch judgment can be excluded. Greatly improves the accuracy of scratch recognition.

In some feasible embodiments, before performing short-term energy conversion on the acquired eddy current signal of the rail in step S110, the method further includes:

The eddy current signal is filtered and denoised.

In some preferred embodiments, band-pass filtering is performed on the eddy current signal through an Equiripple finite impulse response (Finite Impulse Response, FIR) method;

The eddy current signal processed by band-pass filtering is subjected to wavelet denoising based on db3-based minimax criterion soft threshold method.

The FIR filter has an accurate linear phase, and the system is stable; the equiripple method is based on the maximum error minimization criterion, which can minimize the maximum error between the frequency domain characteristics of the FIR filter and the frequency domain characteristics of the ideal filter, and the amplitude is in the stop band. and have equal volatility in the passband, and the error is uniformly distributed over the entire frequency band.

It should be noted that the upper and lower cut-off frequencies of the filter can be set and adjusted according to the detection frequency, as well as the scratch signal frequency and impurity signal frequency obtained from historical data, so as to filter out the influence of irrelevant clutter signals to the maximum extent. Extract more accurate scratch signals.

The dbN wavelet has better regularity and can make the signal reconstruction process smoother. N represents the vanishing moment of the wavelet. The larger the number of vanishing moment, the smoother the wavelet and the better the frequency band division. Therefore, in the case of comprehensively considering the amount of calculation and the localization capability in the frequency domain, the dB3 wavelet is selected in the embodiment of this specification. Of course, in actual use, the detection personnel can choose other filtering methods and other dbN wavelets according to actual needs.

As shown in Figure 2(a) to Figure 2(f), there are several unprocessed eddy current signals, that is, the eddy current signals detected at different track positions; Figure 3(a) to Figure 3(f) Shown are the eddy current signals obtained by sequentially bandpass filtering and wavelet denoising for the signals shown in Figure 2(a) to Figure 2(f).

Referring to Figure 2(a) and Figure 3(a), it can be seen that the amplitude of the signal fluctuates very little, and after filtering and denoising, the signal becomes smooth as a whole, and it can be inferred that the detection position corresponding to the signal is smooth; Referring to Fig. 2(b) and Fig. 3(b), it can be seen that the amplitude of the signal becomes smaller after processing, but the change is not large, and the signal shape is basically preserved. There are scratches on the rail surface at the position; comparing Fig. 2(c) and Fig. 3(c), it can be seen that there are several fluctuations in the eddy current signal, and the fluctuation amplitude of the fluctuation part is not large. It is inferred that there is corrosion on the rail surface at the detection position corresponding to the signal. The magnetic permeability of the rusted rail is different from that of the non-corroded rail, and the thickness of the corrosion at different positions of the rail surface is different, so the eddy current signal fluctuates to different degrees; Comparing Figure 2(d) and Figure 3(d), the eddy current signal before and after processing is similar to the eddy current signal before and after processing shown in Figure 2(c) and Figure 3(c). There is also a corrosion problem on the rail surface, and the degree of corrosion is more serious than that in Figure 2(c) and Figure 3(c). It fluctuates periodically, and the fluctuation amplitude is large. After processing, the starting amplitude fluctuates above and below the 0mV baseline, indicating that the low-frequency part of the eddy current signal has been filtered out. It is speculated that the detection position corresponding to the signal has wave grinding Compared with Fig. 2(f) and Fig. 3(f), the signal waveform before and after processing has a small change, but the signal shape is relatively well preserved. There are scratches around 1200mm.

In combination with the above examples, in the embodiment of this specification, band-pass filtering is performed on the eddy current signal by the finite impulse response of the equiripple method, and wavelet denoising is performed by the soft threshold method based on the db3-based minimax criterion, which can filter out the impurity signal. , to eliminate the interference of the different states of the rail surface on the eddy current signal, which is beneficial to improve the accuracy of signal extraction. After the above filtering and denoising processing, there are still interference signals such as rust and corrugation that need to be filtered out in the eddy current signal, which are not the signal objects that need to be identified by the rail scratch identification method provided in the embodiments of this specification. Therefore, in the embodiment of this specification, an adaptive energy threshold for filtering these interference signals is provided.

Specifically, in step S110, performing energy conversion on the eddy current signal, and obtaining the energy signal can be performed by the following formula:

Among them, E(n) is the energy signal, the value range of n is 1～M+1-N, N is the calculation window width, M is the sequence length of the eddy current signal, and x(m) is the detection point at the detection point m. eddy current signal.

Illustratively, in the embodiment of this specification, the detection frequency of the eddy current signal is 1 (1/mm), and it can be seen from historical data that the number of fluctuation points of the scratch signal is about 4 to 10, then the value of the calculation window width N can be set. is 10, then there are:

...

Therefore, each energy signal in the embodiment of this specification corresponds to the average value of the absolute value of the amplitude of the eddy current signal measured at N test points at the corresponding detection position, which has a noise reduction effect and can avoid abnormal signal at a single detection point. This leads to a decrease in the accuracy of subsequent scratch recognition.

It should be noted that, in the actual scratch detection scene, the energy signal can also be obtained by conversion in other ways. For example, the energy signal of each detection point is obtained by taking the absolute value of the amplitude of the eddy current signal corresponding to the detection point. ; Or take the absolute value of the amplitude of the eddy current signal for the N detection points near the detection point, and then perform weighted calculation to obtain the method.

In some feasible embodiments, the energy threshold adaptive to the energy signal can be obtained by the following steps:

Calculate the average energy of the energy signal to obtain the initial energy threshold, and the calculation formula of the average energy is:

Among them, Th is the initial energy threshold, and E(i) is the energy of the ith signal;

The initial energy threshold is corrected to obtain the adaptive energy threshold, and the correction formula is:

THR=a×Th+b;

Among them, THR is the adaptive energy threshold, a is the amplification coefficient, and b is the bias coefficient. The coefficients a and b can be determined by training based on actual scratch signals. Exemplarily, the value of a may be 7, and the value of b may be 0.3. Of course, other data may also be set according to actual scratch identification needs. By amplifying and biasing the initial energy threshold, the correction of the initial energy threshold is realized, and the energy signal is compared with the modified adaptive energy threshold for scratch identification, which can eliminate the interference such as corrosion and wave wear in the eddy current signal. The influence of noise on scratch identification improves the accuracy of scratch identification.

Further, as shown in FIG. 4, in the embodiment of this specification, in step S140: when the amplitude is greater than or equal to the adaptive energy threshold, after judging that the rail is scratched, the method further includes:

S410: Based on the eddy current signal, the energy signal, and the array sensor for acquiring the eddy current signal, determine a parameter of the scratch, where the parameter includes the length, width and depth of the scratch.

It should be noted that the length refers to the size of the scratch along the length direction of the track, the width refers to the size of the scratch along the width direction of the track, and the depth refers to the size of the scratch along the vertical direction of the track. on the size. Therefore, the rail scratch identification method provided by the embodiments of this specification can not only accurately identify and judge whether there is scratch on the rail, but also can obtain the three-dimensional data of the scratch, and can not evaluate the scratch area and depth. Compared with the prior art of manual review, the identification method provided by the embodiments of this specification can greatly reduce the workload of the inspecting personnel, improve the inspection efficiency, and avoid errors caused by human factors in the inspection process.

And S420: Determine the shape and type of the abrasion according to the parameters; the shape includes a long strip and an oval, and the types of the abrasion include superficial albuginea tissue, deep albuminous layer tissue, and abscesses.

The type of the scratch can be used to analyze the severity of the scratch; the shape of the scratch can be used to analyze the cause of the scratch, for example, a long-shaped scratch may be caused by a locked train wheel, an oval Shaped scratches may be caused by wheel spin. Of course, the above are only exemplary descriptions. In fact, most of the track scratches are caused by a combination of factors. Through the analysis of the shape and type of scratches, the maintenance and remediation of the subsequent rails can be facilitated.

The rail scratch identification method provided in the embodiments of this specification can also identify the shape and type of scratches, which is beneficial to analyze the causes of scratches and evaluate the severity of scratches, so as to not only improve the accuracy of scratch identification Efficiency, accuracy, and comprehensive scratch recognition.

As shown in FIG. 5, in some specific embodiments, step S410: based on the eddy current signal, the energy signal and the array sensor used to obtain the eddy current signal, determine the size parameter of the scratch, which may be :

S510: Acquire a position interval corresponding to the energy signal whose amplitude satisfies a preset condition as a first characteristic value; record the first characteristic value as PP.

Specifically, FIG. 6 is a schematic diagram of acquiring the first characteristic value of a certain energy signal. Exemplarily, if the first preset condition is that the amplitude is greater than or equal to 0.2 mV and lasts for 10 mm, the first characteristic value PP of the energy signal shown in FIG. 6 is [990, 1100]. Certainly, the preset condition may also have other setting manners.

It should be noted that the dotted line in FIG. 6 indicates the adaptive energy threshold corresponding to the energy signal one-to-one. It can be seen that the energy signal amplitude is greater than or equal to the adaptive energy threshold near the [1020, 1080] position interval , that is, the energy signal is scratched, and then the determination of scratch parameters and the identification of scratch shape and type are performed.

S520: Acquire the number of channels of the array sensor corresponding to the scratch as a second feature value; record the first feature value as NN.

FIG. 7 is a schematic diagram of the principle of the scratch detection of the rail by the array sensor. As shown in Figure 7, the array sensor used in the embodiment of this specification has a total of 4 channels, and each channel is placed side by side on the surface of the rail tread during flaw detection. The first channel is near the rail light strip, and the fourth channel is near the gauge angle. Exemplarily, when at a certain detection position of the rail, analyzing the eddy current signals detected by the first channel and the second channel, it is found that the corresponding rail surfaces are scratched, and the eddy current signals detected by the third channel and the fourth channel are corresponding to the rail surface. If there is no scratch, the second characteristic value NN is 2, and at the detection position of the rail, the width of the scratch is the detection width of the first channel and the second channel of the array sensor. It should be noted that the second characteristic value should be the number of channels of adjacent array sensors: when the first channel and the fourth channel detect scratches, but the second channel and the third channel do not detect scratches, The parameter analysis should be performed as two scratches, and the second characteristic value corresponding to the two scratches is both 1.

S530: Calculate the difference between the maximum and minimum amplitudes of the eddy current signal in the position interval as a third eigenvalue; record the first eigenvalue as MM.

FIG. 8 is a schematic diagram of acquiring the third eigenvalue of the eddy current signal corresponding to the energy signal shown in FIG. 6 . As shown in Figure 8, in the position interval [990, 1100] indicated by PP, the difference between the maximum and minimum amplitude values of the eddy current signal obtained is MM, that is, the obtained scratches in the position interval are obtained. deepest depth. It should be noted that, in the embodiments of the present specification, both the eigenvalue MM and the eigenvalue PP are the maximum values detected by the four-channel array sensors.

S540: Determine the length, width and depth of the scratch according to the first feature value, the second feature value and the third feature value, respectively. That is, the length of the scratch is determined according to the first feature value, the width of the scratch is determined according to the second feature value, and the depth of the scratch is determined according to the third feature value, so as to realize the determination of the size parameter of the scratch, which fills the gap in the prior art. This gap for scratch size identification.

In some feasible embodiments, S420: Determine the shape and type of the abrasion according to the parameter; the shape includes a long strip and an oval, and the type of the abrasion includes a superficial albuginea tissue, a deep alba Layer combination and block drop, which can be:

The shape of the scratch is determined based on the first eigenvalue and the second eigenvalue.

Specifically, the length of the position interval determined by the first characteristic value can be calculated in combination with Table 1, and the shape of the scratch can be determined by combining the length and the second characteristic value.

Table 1

The division of the length of the position interval calculated according to the first eigenvalue in Table 1, and the relationship between the position interval division and the number of the second eigenvalue are all exemplary, and those skilled in the art can adapt according to actual scratch recognition needs Sexual regulation.

and determining the type of the scratch according to the third feature value.

Specifically, as shown in FIG. 9 , the following methods can be used to determine the type of the scratch:

S910: Determine whether the third feature value is greater than a first depth threshold and less than or equal to a second depth threshold;

Exemplarily, the first depth threshold is an equivalent amplitude of 50mV, and the second depth threshold is 100mV. Of course, the first depth threshold and the second depth threshold may also be set as other data parameters.

S920: when the third feature value is greater than the first depth threshold and less than or equal to the second depth threshold, determine that the type of the abrasion is deep white layer tissue;

S930: when the third feature value is less than or equal to the first depth threshold, determine that the type of the abrasion is a superficial white layer tissue;

S940: When the third feature value is greater than the second depth threshold, determine that the type of the scratch is block drop.

The rail scratch identification method provided in this manual can not only detect and identify scratches, but also determine the shape and type of scratches, which has a positive effect on subsequent rail maintenance and maintenance.

In addition to the above processing methods, a three-dimensional image of the scratch can also be drawn based on the third eigenvalue of the eddy current signal detected by each channel of the array sensor at each detection point to assist in the identification of the shape and severity of the scratch, thereby More accurately analyze the causes of scratches and select more effective rail maintenance measures to improve the safety of rail traffic and the comfort of passengers.

As shown in FIG. 10 , an embodiment of the present specification further provides a rail scratch identification device, including:

The conversion module 101 is configured to perform short-term energy conversion on the acquired eddy current signal of the rail to obtain an energy signal corresponding to the eddy current signal;

Judging module 102, for judging whether the amplitude of the energy signal is greater than the adaptive energy threshold of the energy signal;

a first determination module 103, configured to determine that there is no scratch on the rail when the amplitude is less than the energy threshold;

The second determination module 104 is configured to determine that the rail is scratched when the amplitude is greater than or equal to the energy threshold.

The beneficial effects obtained by the device provided by the embodiments of this specification are consistent with the beneficial effects obtained by the above method, and are not repeated here.

As shown in FIG. 11 , the embodiment of this specification further provides a rail scratch identification system, including a probe 10 , a probe support structure 20 , a signal excitation unit 30 , a signal acquisition and processing unit 40 and a controller 50 ;

The probe support structure 20 is connected with the probe 10, and is used for fixing the probe 10 to keep the distance between the probe 10 and the surface of the rail 60 to be probed stable;

The side of the probe 10 close to the rail 60 is provided with an array sensor, the array sensor is connected to the signal excitation unit 30 and the signal acquisition and processing unit 40 , and is used for generating under the AC excitation of the signal excitation unit 30 . AC magnetic field, and feedback the detected eddy current signal generated by the rail 60 under the AC magnetic field to the signal acquisition and processing unit 40;

The signal acquisition and processing unit 40 is connected to the signal excitation unit 30 and the controller 50 , and is used to regulate and control the parameters of the AC excitation of the signal excitation unit 30 under the control of the controller 50 , and to obtain the obtained data. The eddy current signal is conditioned, amplified and fed back to the controller 50;

The controller 50 is used to identify rail galling based on the received conditioned amplified eddy current signal.

The beneficial effects obtained by the system provided by the embodiments of this specification are consistent with the beneficial effects obtained by the above method, and are not repeated here.

As shown in FIG. 12 , for a computer device provided by the embodiments herein, the computer device 1202 may include one or more processors 1204 , such as one or more central processing units (CPUs), each processing unit may implement One or more hardware threads. Computer device 1202 may also include any memory 1206 for storing any kind of information such as code, settings, data, and the like. Without limitation, for example, memory 1206 may include any one or a combination of the following: any type of RAM, any type of ROM, flash memory devices, hard disks, optical disks, and the like. More generally, any memory can use any technology to store information. Further, any memory can provide volatile or non-volatile retention of information. Further, any memory may represent a fixed or removable component of computer device 1202. In one instance, when processor 1204 executes the associated instructions stored in any memory or combination of memories, computer device 1202 may perform any operation of the associated instructions. The computer device 1202 also includes one or more drive mechanisms 1208 for interacting with any memory, such as a hard disk drive mechanism, an optical disk drive mechanism, and the like.

Computer device 1202 may also include an input/output module 1210 (I/O) for receiving various inputs (via input device 1212 ) and for providing various outputs (via output device 1214 ). A specific output mechanism may include a presentation device 1216 and an associated graphical user interface (GUI) 1218. In other embodiments, the input/output module 1210 (I/O), the input device 1212 and the output device 1214 may not be included, and only serve as a computer device in the network. Computer device 1202 may also include one or more network interfaces 1220 for exchanging data with other devices via one or more communication links 1222. One or more communication buses 1224 couple together the components described above.

Communication link 1222 may be implemented in any manner, eg, through a local area network, a wide area network (eg, the Internet), a point-to-point connection, etc., or any combination thereof. Communication link 1222 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc. governed by any protocol or combination of protocols.

Corresponding to the methods in FIG. 1 , FIG. 4 to FIG. 5 and FIG. 9 , the embodiments herein also provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor. perform the steps of the above method.

Embodiments herein also provide computer-readable instructions, wherein when a processor executes the instructions, the program therein causes the processor to perform the methods shown in FIGS. 1 , 4 to 5 and 9 .

It should be understood that, in the various embodiments herein, the size of the sequence numbers of the above-mentioned processes does not mean the sequence of execution, and the execution sequence of each process should be determined by its functions and internal logic, rather than the implementation of the embodiments herein. The process constitutes any qualification.

It should also be understood that, in the embodiments herein, the term "and/or" is only an association relationship for describing associated objects, indicating that there may be three kinds of relationships. For example, A and/or B can mean that A exists alone, A and B exist at the same time, and B exists alone. In addition, the character "/" in this document generally indicates that the related objects are an "or" relationship.

Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. Interchangeability, the above description has generally described the components and steps of each example in terms of function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of this document.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above may refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

In the several embodiments provided herein, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may also be electrical, mechanical or other forms of connection.

The units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solutions in the embodiments herein.

In addition, each functional unit in each of the embodiments herein may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

The integrated unit, if implemented in the form of a software functional unit and sold or used as an independent product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solutions in this article are essentially or make contributions to the prior art, or all or part of the technical solutions can be embodied in the form of software products, and the computer software products are stored in a storage medium , including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments herein. The aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

The principles and implementations of this paper are described by using specific examples in this paper. The descriptions of the above examples are only used to help understand the methods and core ideas of this paper; , there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this article.

Claims (12)

1. a rail scratch identification method, is characterized in that, comprises: performing energy conversion on the acquired eddy current signal of the rail to obtain an energy signal corresponding to the eddy current signal; Judging whether the amplitude of the energy signal is greater than the adaptive energy threshold corresponding to the energy signal; When the amplitude is less than the adaptive energy threshold, it is determined that there is no scratch on the rail; When the amplitude is greater than or equal to the adaptive energy threshold, it is determined that the rail is scratched. 2 . The method according to claim 1 , wherein after judging that the steel rail is scratched, the method further comprises: 2 . determining parameters of the scratch based on the eddy current signal, the energy signal, and an array sensor for acquiring the eddy current signal, the parameters including the length, width and depth of the scratch; The shape and type of the abrasion are determined according to the parameters; the shape includes an elongated shape and an oval shape, and the type of abrasion includes a superficial white layer tissue, a deep layer white layer combination and a lump. 3. The method according to claim 2, wherein the determining the parameters of the scratch based on the eddy current signal, the energy signal and the array sensor for acquiring the eddy current signal comprises: Obtain the position interval corresponding to the energy signal when its amplitude satisfies the preset condition as the first characteristic value; obtaining the number of channels of the array sensor corresponding to the scratch as a second characteristic value; Calculate the difference between the maximum value and the minimum value of the amplitude of the eddy current signal in the position interval as a third eigenvalue; According to the first characteristic value, the second characteristic value and the third characteristic value, the length, width and depth of the scratch are respectively determined. 4. The method of claim 3, wherein determining the type of abrasion according to the parameter, further comprising: judging whether the third feature value is greater than the first depth threshold and less than or equal to the second depth threshold; When the third feature value is greater than the first depth threshold and less than or equal to the second depth threshold, determining that the type of the abrasion is deep white layer tissue; When the third characteristic value is less than or equal to the first depth threshold, determining that the type of the abrasion is a superficial white layer tissue; When the third feature value is greater than the second depth threshold, it is determined that the type of the scratch is a block drop. 5. The method according to claim 1, wherein the obtained eddy current signal of the rail is calculated by the following formula to obtain the energy signal:

Among them, E(n) is the energy signal, the value range of n is 1～M+1-N, M is the sequence length of the eddy current signal, N is the calculation window width, and x(m) is the eddy current detected at m Signal. 6. The method according to claim 5, wherein the energy threshold is obtained by the following steps: Calculate the average energy of the energy signal to obtain the initial energy threshold, and the calculation formula of the average energy is:

Among them, Th is the initial energy threshold, and E(i) is the energy of the ith signal; The initial energy threshold is corrected to obtain the adaptive energy threshold, and the correction formula is: THR=a×Th+b; Among them, THR is the adaptive energy threshold, a is the amplification coefficient, and b is the bias coefficient. 7. The method according to claim 1, characterized in that, before calculating the acquired eddy current signal of the rail to obtain the energy signal, the method further comprises: The eddy current signal is filtered and denoised. 8. The method according to claim 7, wherein the filtering and denoising processing of the eddy current signal further comprises: Band-pass filtering the eddy current signal by the finite impulse response of the equiripple method; The wavelet denoising is performed on the eddy current signal processed by bandpass filtering using the soft threshold method based on the minimax criterion based on db3 basis. 9. A rail scratch identification device, characterized in that, comprising: a conversion module, configured to perform short-term energy conversion on the acquired eddy current signal of the rail to obtain an energy signal corresponding to the eddy current signal; a judging module for judging whether the amplitude of the energy signal is greater than the adaptive energy threshold of the energy signal; a first determination module, configured to determine that there is no scratch on the rail when the amplitude is less than the energy threshold; The second determination module is configured to determine that the rail is scratched when the amplitude is greater than or equal to the energy threshold. 10. A rail scratch identification system, comprising a probe, a probe support structure, a signal excitation unit, a signal acquisition and processing unit and a controller; The probe support structure is connected with the probe, and is used for fixing the probe to keep the distance between the probe and the surface of the rail to be probed stable; The side of the probe close to the rail is provided with an array sensor, the array sensor is connected with the signal excitation unit and the signal acquisition and processing unit, and is used to generate an AC magnetic field under the AC excitation of the signal excitation unit, and to generate an AC magnetic field. The detected eddy current signal generated by the rail under the AC magnetic field is fed back to the signal acquisition and processing unit; The signal acquisition and processing unit is connected with the signal excitation unit and the controller, and is used for regulating the parameters of the AC excitation of the signal excitation unit under the control of the controller, and conditioning the acquired eddy current signal Amplify feedback to the controller; The controller is used to identify rail galling based on the received conditioned amplified eddy current signal. 11. A computer device comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the processor implements any of claims 1 to 8 when the processor executes the computer program. one of the methods described. 12 . A computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images