CN107650945A - A kind of recognition methods of wheel polygon and its device based on vertical wheel rail force - Google Patents

Abstract

The invention discloses a wheel polygon identification method based on the wheel-rail vertical force and a device thereof. The identification method comprises the following steps: collecting the wheel-rail vertical force signal by the shear stress method; and analyzing the collected wheel-rail vertical force signal Carry out the EEMD decomposition of the ensemble empirical mode to obtain the aggregated intrinsic mode function IMF component; calculate the marginal spectrum of the intrinsic mode function IMF containing the wheel polygonal disease feature information; according to the obtained marginal spectrum of the intrinsic mode function IMF, obtain The damage characteristic frequency f is used to identify whether the wheel has polygons and the type of polygons according to the preset judgment criteria. Its advantages are: it can carry out real-time monitoring and damage judgment on the vehicles passing through the monitoring section, and it has the characteristics of fast and accurate; The marginal spectrum of the wheel-rail vertical force damage characteristic frequency after polygonal damage of the wheel is obtained.

Description

A wheel polygon recognition method and device based on wheel-rail vertical force

technical field

The invention belongs to the technical field of railway safety monitoring, and in particular relates to a wheel polygon recognition method based on wheel-rail vertical force and a device thereof.

Background technique

In the past ten years, high-speed railway and urban rail transit have developed rapidly due to their advantages of large capacity, energy saving and environmental protection. According to the "13th Five-Year Development Plan" of railways, the period of the 13th Five-Year Plan is still the peak period of railway construction and development , will build a high-speed railway network with "eight vertical and eight horizontal" passages as the backbone and intercity railways as supplements. At the same time, the number of cities under construction and operation of urban rail transit has doubled, and the mileage has increased by about half. However, with the increase of the operation time of my country's railways, various diseases that affect the comfort of passengers and endanger the safety of trains have appeared, which seriously restrict the further development of my country's railway industry and the implementation of the strategy of going out.

Contents of the invention

The object of the present invention is to provide a wheel polygon recognition method and its device based on the wheel-rail vertical force according to the deficiencies of the above-mentioned prior art. The rail strain signal is converted, decomposed and calculated to obtain the damage characteristic frequency f, so as to quickly and accurately identify whether the wheel has polygons and the type of polygons according to the preset judgment criteria.

The object of the present invention is realized by the following technical solutions:

A wheel polygon recognition method based on the wheel-rail vertical force, characterized in that the recognition method comprises the following steps:

(1) The vertical force signal of the wheel and rail is collected by the shear stress method;

(2) Decomposing the collected wheel-rail vertical force signal into EEMD with aggregate empirical mode, so as to obtain the lumped intrinsic mode function IMF component;

(3) Calculating the marginal spectrum of the intrinsic mode function IMF including the wheel polygonal disease feature information in the step (2);

(4) Obtain the damage characteristic frequency f according to the marginal spectrum of the intrinsic mode function IMF obtained in the step (3), and identify whether the wheel has polygons and the type of polygons according to the preset judgment criteria.

In the step (1), at the mid-span position between no less than 5 groups of adjacent sleepers, the resistance strain gauges are pasted on the rail waist, and each resistance strain gauge collects the strain signal of the rail and Converted to the wheel-rail vertical force signal.

The calculation method for converting the strain signal of the rail into the wheel-rail vertical force signal is:

P=|Q _r |+Q _l

Q _r = Q _l =(Jb/S)τ

τ=Gε

In the formula:

P is the vertical force of the wheel and rail;

Q _r and Q _l are shear force;

J is the moment of inertia of the rail section to the neutral axis;

b is the section thickness of the rail at the neutral axis;

S is the static moment of the section outside the shear stress calculation point on the neutral axis;

τ is the shear stress suffered by the rail;

G is the shear modulus of the rail;

ε is the shear strain of the rail collected by the electrical resistance strain gauge.

In the step (2), the EEMD decomposition of the ensemble empirical mode refers to adding normally distributed Gaussian white noise to the collected wheel-rail vertical force signal, and then using the EMD method to convert the modified wheel-rail The rail vertical force signal is decomposed.

In the step (3), the method for calculating the marginal spectrum of the intrinsic mode function IMF is: performing Hilbert transformation on the obtained lumped intrinsic mode function IMF components, thereby obtaining the Hilbert spectrum of the energy distribution on the time-frequency plane H(ω,t), the calculation formula is as follows:

Integrate the obtained Hilbert spectrum H(ω, t) in the time domain to obtain the marginal spectrum H(ω), and the calculation formula is as follows:

In the formula:

Re means to take the real part of the imaginary number;

a(t) is a lumped intrinsic mode function IMF component;

j is the unit imaginary number ,

ω is the instantaneous frequency in the Hilbert transform;

t is the instantaneous moment.

In the step (4), the preset judgment criteria are specifically:

If the periodic wheel-rail force wavelength λ is 2πR, and the damage characteristic frequency f is v/2πR, the wheel appears a first-order polygon and wheel eccentricity;

If the periodic wheel-rail force wavelength λ is πR, and the damage characteristic frequency f is v/πR, the wheel appears a second-order polygon and the wheel is elliptical;

If the periodic wheel-rail force wavelength λ is 2πR/3, and the damage characteristic frequency f is 3v/2πR, the wheel appears a third-order polygon and a triangular wheel;

If the periodic wheel-rail force wavelength λ is πR/2, and the damage characteristic frequency f is 2v/πR, the wheel will appear as a fourth-order polygon and a wheel quadrilateral;

If the periodic wheel-rail force wavelength λ is 2πR/N, and the damage characteristic frequency f is vN/2πR, then the wheel appears N-order polygonal, and the wheel is N-polygonal;

Wherein, R is the rolling radius of the wheel; v is the running speed of the train.

The device includes a strain collection module, a train speed collection module, a data long-distance transmission module and a remote monitoring module, the strain collection module and the train speed collection module are formed by the data long-distance transmission module and the remote monitoring module communication connection.

The remote monitoring module is also sequentially connected with a wheel-rail vertical force calculation module, an EEMD decomposition module, a signal marginal spectrum calculation module, and a wheel polygon evaluation module.

The advantages of the present invention are:

(1) In view of the fact that the existing wheel polygon detection method takes a long time and the accuracy is not high, on the basis of measuring the wheel-rail vertical force based on the shear stress method, make full use of the real-time acquired wheel-rail vertical force signal and signal processing technology, a wheel polygon recognition method and device based on the wheel-rail vertical force are proposed, which can conduct real-time monitoring and damage evaluation of vehicles passing through the monitoring section, so as to timely repair the wheels with wheel polygons Handle and avoid safety accidents;

(2) Since the EEMD method has intuitive rationality, high efficiency, adaptability, and superiority in dealing with nonlinear and non-stationary local signals, it can handle time-varying wheel-rail vertical force signals well; at the same time, after decomposing The marginal spectrum of the signal is used to obtain the characteristic frequency of the wheel-rail vertical force damage after the polygonal damage occurs to the wheel.

Description of drawings

Fig. 1 is the schematic flow chart of the wheel polygon recognition method based on wheel-rail vertical force among the present invention;

Fig. 2 is the structural representation of wheel polygon recognition device among the present invention;

Fig. 3 is a schematic diagram of the pasting and setting of the resistance strain gauge on the rail waist of the rail in the present invention;

Fig. 4 is a schematic diagram of EEMD decomposition of ensemble empirical mode in the present invention;

Fig. 5 is the time course diagram of the wheel-rail vertical force under the ovalization situation of the wheel among the present invention;

Fig. 6 is the eigenmode function figure of wheel-rail vertical force signal among the present invention;

Fig. 7 is a marginal spectrum diagram of the wheel-rail vertical force signal in the present invention.

Detailed ways

The features of the present invention and other relevant features are described in further detail below in conjunction with the accompanying drawings through the embodiments, so as to facilitate the understanding of those skilled in the art:

As shown in Figure 1-7, marks 1-3 in the figure are respectively: rail 1, sleeper 2, and resistance strain gauge 3.

Embodiment 1: As shown in Figure 1-7, this embodiment specifically relates to a wheel polygon recognition method and device based on the wheel-rail vertical force, which specifically includes the following steps:

(step 1)

1.1) As shown in Figures 1, 2, and 3, the wheel polygon recognition device is installed and arranged. The wheel polygon recognition device specifically includes a strain acquisition module, a train speed acquisition module, a long-distance data transmission module and a remote monitoring module; among them, the strain The acquisition module includes several strain gauges 3 arranged at the rail waist of the rail 1. In this embodiment, 5 groups of strain gauges 3 are arranged in total, and each strain gauge 3 is located at the mid-span position between adjacent sleepers 2. The train speed acquisition module is specifically the speed measuring instrument on the train, which is used to monitor and obtain the real-time speed of the train; the data long-distance transmission module is used to form the communication connection between the remote monitoring module and the aforementioned strain acquisition module and the train speed acquisition module , so that the remote monitoring module can receive the strain signal sent by the strain acquisition module and the train speed data sent by the train speed acquisition module. Calculation module, including wheel-rail vertical force calculation module, EEMD decomposition module of ensemble empirical mode, signal marginal spectrum calculation module and wheel polygon evaluation module;

1.2) After the installation and layout of the aforementioned wheel polygon recognition device is completed, the strain signal of the rail 1 when the train passes is measured through the strain acquisition module installed on the rail 1, and the sampling frequency can reach up to 20kHz/channel, which fully meets the sampling frequency At the same time, the running speed of the train is measured by the train speed acquisition module; and the aforementioned collected rail strain signal and train running speed are uploaded to the remote monitoring module in real time through the data long-distance transmission module, and the remote monitoring module will obtain the rail strain The signal and train running speed are stored;

1.3) After that, the remote monitoring module sends the rail strain signal to the wheel-rail vertical force calculation module, and converts the rail strain signal into a wheel-rail vertical force signal, as shown in Figure 5, and the calculation formula is:

P=|Q _r |+Q _l

Q _r = Q _l =(Jb/S)τ

τ=Gε

In the formula:

P is the wheel-rail vertical force;

Q _r and Q _l are shear force;

J is the moment of inertia of the rail 1 section to the neutral axis;

b is the section thickness of rail 1 at the neutral axis;

S is the static moment of the section outside the shear stress calculation point on the neutral axis;

τ is the shear stress on rail 1;

G is the shear modulus of rail 1;

ε is the shear strain of the rail 1 collected by the resistance strain gauge 3 .

(step 2)

As shown in Fig. 4, the wheel-rail vertical force signal is sent to the aggregate empirical mode EEMD decomposition module, and the aggregate empirical mode EEMD decomposition is performed on the wheel-rail vertical force signal, that is, the wheel-rail vertical force signal is added to the normal Distributed Gaussian white noise, and then use the EMD method to decompose the modified wheel-rail vertical force signal, and use the IMF integrated mean value obtained each time as the final signal decomposition result to obtain the lumped intrinsic mode function IMF component and residual function (r5), as shown in Figure 6.

(step 3)

Send the obtained lumped intrinsic mode function IMF component to the signal marginal spectrum calculation module, and perform Hilbert transformation on the obtained lumped intrinsic mode function IMF component through the signal marginal spectrum calculation module, so as to obtain the energy distribution on the time-frequency plane Hilbert spectrum H(ω,t), as shown in Figure 7, the calculation formula is as follows:

The obtained Hilbert spectrum H(ω, t) is integrated in the time domain to obtain the marginal spectrum H(ω), and the calculation formula is as follows:

In the formula:

Re means to take the real part of the imaginary number;

a(t) is the lumped IMF component;

j is the unit imaginary number ,

ω is the instantaneous frequency in the Hilbert transform;

t is the instantaneous moment.

(step 4)

The wheel polygon evaluation module starts to work, obtains the wheel polygon damage characteristic frequency f according to the marginal spectrum H(ω) obtained in step (3), and identifies whether the wheel appears polygon and the type of polygon according to the preset judgment criteria. For the aforementioned preset judgment criteria, please refer to Table 1 below, wherein, R is the rolling radius of the train wheel; v is the running speed of the train.

Table 1 Wheel polygon damage characteristic frequency f

The beneficial effect of this embodiment is that: the wheel polygon recognition method and device based on the wheel-rail vertical force provided by this embodiment can detect the train wheel polygons passing through the detection section in real time, and at the same time, it can be very convenient. Monitoring, to a large extent, makes up for the shortcomings of existing manual detection methods and ensures the safe operation of trains.

Embodiment 2: This embodiment specifically relates to a wheel polygon recognition method based on the wheel-rail vertical force. On the basis of Embodiment 1, it will be described in conjunction with a specific case, including the following steps:

(Step 1) As shown in Figure 3, it is preliminarily estimated that a certain wheel has been ovalized, and the wheel-rail vertical force signal is measured through the strain acquisition module set on the rail 1, as shown in Figure 5; Perform EEMD decomposition on the force signal to obtain the lumped eigenmode function IMF component, as shown in Figure 6; similarly, the other three wheel polygons use the same method for signal decomposition;

(Step 2) Integrate the lumped intrinsic mode function IMF components in the time domain to obtain the signal marginal spectrum, as shown in Figure 7. It can be seen from the figure that the damage characteristic frequency f after the wheel ellipse is 24.1Hz;

(Step 3) According to the measured train running speed v=125km/h and the wheel rolling radius R=0.46m, it can be known from the calculation formula of the damage characteristic frequency corresponding to the ovalized wheel in the preset judgment criteria in Table 1, through The damage characteristic frequency obtained by this method is completely consistent with the theoretical calculation formula, so it can be determined that the wheel ovalization of the train wheel has occurred.

Claims (8)

1. a wheel polygon recognition method based on wheel-rail vertical force, it is characterized in that described recognition method comprises the following steps: (1) The vertical force signal of the wheel and rail is collected by the shear stress method; (2) Decomposing the collected wheel-rail vertical force signal into EEMD with aggregate empirical mode, so as to obtain the lumped intrinsic mode function IMF component; (3) Calculating the marginal spectrum of the intrinsic mode function IMF including the wheel polygonal disease feature information in the step (2); (4) Obtain the damage characteristic frequency f according to the marginal spectrum of the intrinsic mode function IMF obtained in the step (3), and identify whether the wheel has polygons and the type of polygons according to the preset judgment criteria. 2. A wheel polygon recognition method based on wheel-rail vertical force according to claim 1, characterized in that in the step (1), the mid-span position between no less than 5 groups of adjacent sleepers At the position, the resistance strain gauges are pasted on the rail waist, and each resistance strain gauge collects the strain signal of the rail and converts it into the wheel-rail vertical force signal. 3. a kind of wheel polygon identification method based on wheel-rail vertical force according to claim 2, it is characterized in that the strain signal of described rail is converted into the calculation method of described wheel-rail vertical force signal as: P=|Q _r |+Q _l Q _r = Q _l =(Jb/S)τ τ=Gε In the formula: P is the vertical force of the wheel and rail; Q _r and Q _l are shear force; J is the moment of inertia of the rail section to the neutral axis; b is the section thickness of the rail at the neutral axis; S is the static moment of the section outside the shear stress calculation point on the neutral axis; τ is the shear stress suffered by the rail; G is the shear modulus of the rail; ε is the shear strain of the rail collected by the electrical resistance strain gauge. 4. A wheel polygon recognition method based on wheel-rail vertical force according to claim 1, characterized in that in the step (2), the EEMD decomposition of the ensemble empirical mode means that the collected Gaussian white noise with normal distribution is added to the wheel-rail vertical force signal, and then the EMD method is used to decompose the modified wheel-rail vertical force signal. 5. A wheel polygon recognition method based on the wheel-rail vertical force according to claim 1, characterized in that in the step (3), the calculation method of the marginal spectrum of the intrinsic mode function IMF is: to obtain Hilbert transform is performed on the lumped intrinsic mode function IMF component of the total, so as to obtain the Hilbert spectrogram H(ω, t) of the energy distribution on the time-frequency plane, and the calculation formula is as follows: Integrate the obtained Hilbert spectrum H(ω, t) in the time domain to obtain the marginal spectrum H(ω), and the calculation formula is as follows: In the formula: Re means to take the real part of the imaginary number; a(t) is a lumped intrinsic mode function IMF component; j is the unit imaginary number , ω is the instantaneous frequency in the Hilbert transform; t is the instantaneous moment. 6. A wheel polygon recognition method based on wheel-rail vertical force according to claim 1, characterized in that in the step (4), the preset judgment criterion is specifically: If the periodic wheel-rail force wavelength λ is 2πR, and the damage characteristic frequency f is v/2πR, the wheel appears a first-order polygon and wheel eccentricity; If the periodic wheel-rail force wavelength λ is πR, and the damage characteristic frequency f is v/πR, the wheel appears a second-order polygon and the wheel is elliptical; If the periodic wheel-rail force wavelength λ is 2πR/3, and the damage characteristic frequency f is 3v/2πR, the wheel appears a third-order polygon and a triangular wheel; If the periodic wheel-rail force wavelength λ is πR/2, and the damage characteristic frequency f is 2v/πR, the wheel will appear as a fourth-order polygon and a wheel quadrilateral; If the periodic wheel-rail force wavelength λ is 2πR/N, and the damage characteristic frequency f is vN/2πR, then the wheel appears N-order polygonal, and the wheel is N-polygonal; Wherein, R is the rolling radius of the wheel; v is the running speed of the train. 7. according to the device of a kind of wheel polygon recognition method based on wheel-rail vertical force described in any one in claim 1-6, it is characterized in that said device comprises strain acquisition module, train speed acquisition module, data long-distance transmission module and a remote monitoring module, the strain collection module and the train speed collection module form a communication connection with the remote monitoring module via the data long-distance transmission module. 8. The device of a wheel polygon recognition method based on the wheel-rail vertical force according to claim 7, characterized in that the remote monitoring module is also connected with a wheel-rail vertical force calculation module and a set empirical model in sequence EEMD decomposition module, signal marginal spectrum calculation module and wheel polygon evaluation module.