CN104176092A - Turnout steel rail damage monitoring method and device - Google Patents

Abstract

The invention relates to the technical field of railway equipment, and provides a method and device for monitoring damage to a turnout rail. The device includes a sensor, a monitoring extension, a monitoring host, and a data center; The monitoring extension, after the monitoring extension processes the characteristic data, transmits the processing result to the monitoring host, and the monitoring host reprocesses the processing result and then distributes it as specified, the data center manages the entire device, and controls the client terminal according to the reprocessed data; It solves the technical problem that the rapid development of tiny cracks cannot be effectively identified and there is a detection blind spot when the track circuit and large-scale track inspection vehicles are used to detect the damage of the rail in the prior art, and then the tiny cracks on the rail of the turnout can be detected Carry out effective identification and overcome detection blind spots.

Description

A method and device for monitoring damage of a turnout rail

technical field

The invention relates to the technical field of railway equipment, and provides a method and a device for monitoring damage of a turnout rail.

Background technique

High-speed turnout rails are very important in high-speed railway infrastructure and key facilities that affect driving safety. High-speed turnout rails have always been the weak link in the line and the key and difficult point in line maintenance. Therefore, the high-speed turnout rail monitoring system is very necessary, and it is also urgently needed for high-speed railway maintenance and repair.

The traditional rail damage detection is mainly completed by the railway track circuit and the rail damage inspection vehicle. The rail damage detection principle of the railway track circuit is that when the rail is damaged, the characteristic parameters of the railway track circuit change. Rail damage such as breakage that may occur; rail damage inspection vehicle, its principle is to detect whether the rail is cracked, broken, etc. by means of ultrasonic guided wave rebound.

But there are following technical problems in the prior art:

Since the track circuit and the large-scale track inspection vehicle cannot effectively identify the rapidly developing tiny cracks of the rail, there is a blind spot in the detection of the prior art.

Contents of the invention

The purpose of the present invention is to provide a method and device for monitoring the damage of a turnout rail, which solves the problem that the rapid development of tiny cracks cannot be effectively identified when the rail circuit and large-scale rail inspection vehicles are used to detect the damage of the rail in the prior art , there is a technical problem in detecting blind spots.

On the one hand, the present invention provides a switch rail damage monitoring device, including a sensor, a monitoring extension, a monitoring host, and a data center;

The sensor is installed on the turnout rail, and the sensor transmits the collected characteristic data to the monitoring extension. After the monitoring extension processes the characteristic data, it transmits the processing result to the monitoring host, and the monitoring host processes the processing result and distributes it as designated. Manage the entire device and control the client terminal based on the reprocessed data.

Furthermore, the monitoring extension includes a signal preprocessing module, a data local energy analysis module and an adaptive approximation module.

Furthermore, the signal preprocessing module is used to receive the characteristic data collected by the sensor, and obtain the processing result through charge amplification, hardware filtering, analog-to-digital conversion, and software filtering.

Further, the feature data is specifically voltage data, or temperature and humidity data, or oscillation data.

Further, the data local energy analysis module is used to perform time-domain analysis and frequency-domain analysis on the processing results, and obtain initial characteristic parameters according to the analysis results.

Further, the adaptive approximation module is used to perform adaptive approximation processing on the initial characteristic parameters according to the established rail damage database, and obtain damage data according to the processing results.

Further, the monitoring host completes the receiving, classified management and storage of the data uploaded by the monitoring extension.

Further, the data center manages the business processing of the entire turnout rail damage monitoring device, and provides data support for client terminals and various alarm devices.

Further, the client terminal sends an alarm in time for rail damage events and device component failure events.

On the other hand, the present invention also provides a method for monitoring damage of a turnout rail, comprising the following content:

Preprocess the collected feature data to obtain processing results;

Perform data local energy analysis on the processing results, and obtain initial characteristic parameters according to the analysis results;

According to the established rail damage database, the initial characteristic parameters are adaptively approximated, and the damage data is obtained according to the processing results;

Classify, manage and store the damage data, and control the client terminal according to the damage data.

The present invention adopts the above technical scheme and has the following beneficial effects:

1. Due to the adoption of the turnout rail damage monitoring device composed of sensors, monitoring extensions, monitoring hosts and data centers, the sensors are installed on the turnout rails, and the collected characteristic data are transmitted to the monitoring extension, and the monitoring extension processes the characteristic data Afterwards, the processing results are sent to the monitoring host, and the monitoring host reprocesses the processing results and distributes them accordingly. The data center manages the entire device and controls the client terminal according to the reprocessed data. It solves the technical problem that the rapid development of micro cracks cannot be effectively identified and there is a detection blind spot when the track circuit and large-scale track inspection vehicle are used to detect the damage of the rail in the prior art, and then the micro cracks on the rail of the turnout can be detected Carry out effective identification and overcome detection blind spots.

2. Since the customer terminal connected to the data center damages the rails and the faults of the device components can be alarmed in time, thereby avoiding the occurrence of unsafe accidents and ensuring safety.

Description of drawings

Fig. 1 is the structural representation of the switch rail damage monitoring device in the embodiment of the present invention;

Fig. 2 is the module schematic diagram of monitoring extension in the embodiment of the present invention;

Fig. 3 is a schematic diagram of modules of a data center in an embodiment of the present invention;

Fig. 4 is a schematic diagram of modules interacting between the monitoring host, the monitoring extension and the data center in an embodiment of the present invention;

Fig. 5 is a flow chart of a method for detecting damage to a turnout rail.

Detailed ways

The invention provides a method and device for monitoring rail damage of a turnout, which solves the problem that the rapid development of tiny cracks cannot be effectively identified when the track circuit and large-scale rail inspection vehicles are used to detect the damage of the rail in the prior art. The technical problem of detecting the blind area can effectively identify the tiny cracks on the rail of the turnout and overcome the detection blind area.

In order to solve the above-mentioned technical problems of detection blind spots, the general idea is as follows:

This scheme adopts a turnout rail damage monitoring device, including sequentially connected sensors, monitoring extension, monitoring host and data center, wherein the sensor is installed on the turnout rail, and the sensor sends the collected characteristic data to the monitoring extension. The feature data is processed, and the processing results are sent to the monitoring host. The monitoring host processes the processing results uploaded by the monitoring extensions and distributes them accordingly. The data center manages the entire device and controls the client terminal according to the reprocessed data.

In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

As shown in FIG. 1 , a switch rail damage monitoring device provided by the present invention specifically includes a sensor 101 , a monitoring extension 102 , a monitoring host 103 , and a data center 104 .

The sensor 101 is installed on the turnout rail. The sensor 101 transmits the collected characteristic data to the monitoring extension 102. After the monitoring extension 102 processes the characteristic data, it transmits the processing result to the monitoring extension 102. The monitoring host 103 uploads the monitoring extension 102. The processing results are reprocessed and distributed accordingly. The data center 104 manages the entire device and controls the client terminal according to the reprocessed data. The entire monitoring device completes the connection between all levels on the data link through the data transmission channel, and realizes fast, safe and stable uplink and downlink transmission of information with specific data messages.

In a specific embodiment, the piezoelectric sensor is used to collect the excitation data of the rail, and according to the rail damage characteristic data, the possible rail damage data is extracted through a specific algorithm. Since the rail cracking signal is a kind of acoustic emission signal, which is generally relatively small, such a small sound signal can be identified through the high-frequency and high-sensitivity data acquisition module, and then the damage data of rail cracking can be quantified and analyzed to realize the identification of cracks Compared with the existing technology, this monitoring technology can cover the monitoring blind area of the traditional technology in the field of rail flaw detection. The piezoelectric sensor has good insulation performance and does not affect the track circuit.

Specifically, as shown in FIG. 2 , the monitoring extension 102 specifically includes a signal preprocessing module 201 , a data local energy analysis module 202 and an adaptive approximation module 203 .

The monitoring extension also includes a power supply module 204 for supplying power to each module of the entire monitoring extension. Of course, it also includes a data forwarding module 205. Through the power supply module 204, the data forwarding module 205 makes the feature data processed by the monitoring extension. The data forwarding module 205 forwards the processed data to the monitoring host 103 .

In a specific embodiment, the signal preprocessing module 201 receives the characteristic data collected by the sensor 101, specifically, the collected characteristic data may specifically be voltage data, or temperature and humidity data, or characteristic data such as oscillation data, and then , preprocessing these feature data through charge amplification, hardware filtering, analog-to-digital conversion, and software filtering to obtain processing results.

The piezoelectric sensor with a frequency range of 10kHz to 300kHz is selected when the sensor is used for acquisition, and the characteristics of capturing the signal when the rail damage occurs. Due to the relationship between the sensor detection distance and the detected area and the location of the damage position, 4*4 channels are selected. The signal acquisition card, the parameter configuration of the sensor is as follows:

Sampling frequency Sampling points number of channels Sensor frequency range Filter range Sampling accuracy 1MHz 1024 4*4 10kHz-300kHz 20K-300K 16 bits

Next, the processing results of the signal after the above-mentioned preprocessing are analyzed in the data local energy analysis module 202, which includes the statistical analysis of the time domain characteristics of the data, the statistical analysis of the frequency domain characteristics of the data, and according to the analysis As a result, the local signal energy characteristics in the frequency domain are extracted, and the statistical values are quantified and output to generate the initial characteristic parameters of the rail damage data. The specific method is as follows:

Time-domain analysis, the time-domain analysis of damage signals includes maximum value analysis, mean value analysis, and mean square value analysis. The specific calculations are as follows:

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; max</span>}<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; ms</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \}</span><span\; class="notranslate">\; ;</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; ms</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; ;</span>$

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; ;</span>$

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; rms</span>}}<span\; class="notranslate">\; =</span>\sqrt{\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; N</span>}}}<span\; class="notranslate">\; ;</span>$

(2) Frequency domain analysis, the frequency domain analysis of the damaged signal includes the analysis of the maximum spectral value of the power distribution, the analysis of the power spectral density, etc. The specific calculation is as follows:

${\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; \&Integral;</span>}_{<span\; class="notranslate">\; -</span><span\; class="notranslate">\; \&infin;</span>}^{<span\; class="notranslate">\; \&infin;</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;vt</span>}}\mathrm{<span\; class="notranslate">\; dt</span>}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; \&Integral;</span>}_{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; T</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;vt</span>}}\mathrm{<span\; class="notranslate">\; dt</span>}<span\; class="notranslate">\; ;</span>$

satisfy ${\&Integral;}_{-\&infin;}^{\&infin;}{u}_T^2\left(t\right)\mathrm{dt}={\&Integral;}_{-\&infin;}^{\&infin;}{\left|u_T\right|}^2\mathrm{dv};$

available ${\lim }_{T=\&infin;}\frac{1}{2T}{\&Integral;}_{-T}^T{u}^2\left(t\right)\mathrm{dt}={\&Integral;}_{-\&infin;}^{\&infin;}{\lim }_{T\&Right\; Arrow;\&infin;}\frac{1}{2T}{\left|u_T\left(v\right)\right|}^2\mathrm{dv};$

That is, the power spectral density is obtained, and the time domain characteristics and frequency domain characteristics of the rail damage data are comprehensively analyzed to obtain the local energy analysis of the damage information in the original data. Through the analysis of the obtained local energy distribution , quantify and output the statistical value, and generate the initial characteristic parameters of the damage data.

After the initial characteristic parameters are obtained, in the adaptive approximation module 203, the initial characteristic parameters are processed according to the established rail damage database, so as to obtain damage data. The specific method is as follows:

Based on the minimum criterion of the square sum of the array output of each snapshot, that is, the least squares (LS) criterion, using all the array data information obtained after the algorithm is initialized, the recursive method is used to complete the matrix inversion operation, so the convergence speed Fast, insensitive to feature value spread, and can achieve a compromise between convergence speed and computational complexity.

Let the cost function be:

$\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}^{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}}^{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}}{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; w</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ;</span>$

In the formula, d(i) and w ^H (i)x(i) become the expected response of the array and the output response of the array respectively, and 0≤γ≤1 is the forgetting factor. Depend on Get R(k)w(k)=r(k);

in, Represents the weighted autocorrelation matrix of the array accepting vectors;

(i) x(i) represents the correlation vector between the array's accept vector and the desired output vector.

According to the above two formulas, the recursive estimation formula is obtained:

$\left\{\begin{array}{c}\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&gamma;R</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>\\ \mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&gamma;r</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; *</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>\end{array}\right.<span\; class="notranslate">\; ;</span>$

The recursive formula of inverse matrix p(k)=R ^-1 (k) can be obtained again:

$\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span>$

Available:

p(k)x(k)=g(k)

w(k)=R ^-1 (k)r(k)=p(k)r(k)

Simplify and get

e(k)=d*(k)-x ^H (k)w(k-1)

Through the above-mentioned series of calculations, effective damage data is finally obtained, thereby improving the recognition accuracy and recognition operation rate of damage detection.

After the monitoring extension 102 obtains the damage data, it uploads the damage data to the monitoring host 103, and the monitoring host 103 completes the receiving, classification management and storage of the damage data, and the data center 104 can monitor the entire turnout rail damage monitoring device Business processing, data support for the client and each alarm device. According to the damage data obtained by the data center 104, the client terminal sends an alarm to the damage event and the failure event of the entire device component in time.

The monitoring host 103 completes the reception of the data uploaded by the monitoring extension 102, and classifies and manages the received damage data, alarm data, business data, etc., and also processes and forwards instructions issued by the superior.

The data center 104 completes the management of the entire turnout rail damage monitoring device, processes relevant data according to business logic, including data classification, in-depth processing, platform forwarding and database storage, and provides data support and business support to clients at all levels.

Wherein, as shown in Figure 3, in the data center 104, include the inquiry module 301 of data, user communication module 302, host computer communication module 303, data warehouse-in module 304, the monitoring host computer 103 at the station receives host computer communication module 303 from When the damaged data is obtained in the data storage module 304, the type of the damaged data is inquired according to the data query module 301, thereby triggering the user communication module 302 to send an alarm message to the user terminal, which can be alarmed in the form of a short message, or through the Internet The push information form of the network notifies the terminal to alarm.

As shown in Figure 4, it is a data interaction module diagram between the monitoring host 103, the monitoring extension 102 and the data center 104, wherein the data center 104 is specifically the railway bureau platform, and the monitoring extension 102 and the monitoring host 103 can communicate with the railway bureau platform data interaction between them. Specifically, a lower-end communication module and a data forwarding module are arranged near the monitoring host 103 and the monitoring extension 102, and an upper-end communication module and a data forwarding module are arranged near the railway bureau platform side, and an agreement is passed between the upper-end communication module and the lower-end communication module. The stack module transmits the data uploaded by the monitoring host 103 and the monitoring extension 102 to the railway bureau platform, and also transmits the instructions issued by the railway bureau platform; the forwarded data is processed by the state processing module between the two data forwarding modules, thereby This enables the railway bureau platform to obtain identifiable data.

The client terminal connected to the data center 104 by wire or wireless can display the pre-alarm information and equipment status information generated by the switch rail damage monitoring device, and at the same time, the user can intuitively see the position of the switch rail alarm through the dynamic sensor deployment diagram.

Based on the same inventive concept, the present application also provides a method for monitoring damage of a switch rail sensor, as shown in Figure 5, including the following steps:

S10. Preprocessing the collected feature data of the turnout rail to obtain a processing result;

S20. Perform data local energy analysis on the processing result, and obtain initial characteristic parameters according to the analysis result;

S30. Perform adaptive approximation processing on the initial characteristic parameters according to the established rail damage database, and obtain damage data according to the processing results;

S40. Classify, manage and store the damage data, and control the client terminal according to the damage data.

In a specific embodiment, in S10, the collected damage signal is specifically subjected to a preprocessing process of charge amplification, hardware filtering, analog-to-digital conversion, and software filtering, so as to obtain a processing result.

In S20, time-domain analysis and frequency-domain analysis are specifically performed on the processing results obtained above, so as to obtain initial characteristic parameters according to the analysis results.

In S30, according to the established rail damage database, the initial characteristic parameters are adaptively approximated, and the damage data is obtained according to the obtained processing results. Specifically, the least square (LS) criterion is used, and the initial characteristic parameters are used after the algorithm is initialized. For all the array data information obtained, the recursive method is used to complete the inverse operation of the matrix, so the convergence speed is fast, it is not sensitive to the scatter of the eigenvalues, and a compromise between the convergence speed and the computational complexity can be achieved.

After the damage data obtained by S30, S40 can transmit the damage data to the alarm client terminal for alarming.

The method for monitoring the damage of the switch rail will not be described in detail in the embodiment of the present application.

While preferred embodiments of the invention have been described, additional changes and modifications to these embodiments can be made by those skilled in the art once the basic inventive concept is appreciated. Therefore, it is intended that the appended claims be construed to cover the preferred embodiment as well as all changes and modifications which fall within the scope of the invention.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims (10)

1. A turnout rail damage monitoring device, including a sensor, a monitoring extension, a monitoring host, and a data center, characterized in that:   The sensor is installed on the turnout rail, and the sensor transmits the collected characteristic data to the monitoring extension. After the monitoring extension processes the characteristic data, it transmits the processing result to the monitoring host, and the monitoring host processes the processing result and distributes it as designated. Manage the entire device and control the client terminal based on the reprocessed data. 2. The switch rail damage detection device according to claim 1, characterized in that the monitoring extension includes a signal preprocessing module, a data local energy analysis module and an adaptive approximation module. 3. The switch rail damage monitoring device according to claim 2, characterized in that: the signal preprocessing module is used to receive the characteristic data collected by the sensor, and through charge amplification, hardware filtering, analog-to-digital conversion, software filtering, Get the processing result. 4. The switch rail damage monitoring device according to claim 1 or 3, characterized in that the characteristic data is specifically voltage data, or temperature and humidity data, or vibration data. 5 . The turnout rail damage monitoring device according to claim 3 , wherein the data local energy analysis module is used to perform time-domain analysis and frequency-domain analysis on the processing results, and obtain initial characteristic parameters according to the analysis results. 6. The turnout rail damage monitoring device according to claim 5, wherein the adaptive approximation module is used to perform adaptive approximation processing on the initial characteristic parameters according to the established rail damage database, and obtain damage according to the processing results data. 7. The crack monitoring device for turnout rails according to claim 1, characterized in that: the monitoring host completes the reception, classification management and storage of the data uploaded by the monitoring extension. 8. The turnout rail damage monitoring device according to claim 1, characterized in that: the data center manages the business processing of the entire turnout rail damage monitoring device, and provides data support for client terminals and various alarm devices. 9 . The track damage monitoring device for turnouts according to claim 1 , characterized in that: the client terminal issues alarms in time for rail damage events and device component failure events. 10 . 10. A method for detecting damage to a turnout rail, applied to a detection device for damage to a turnout rail, characterized in that it includes the following content: Preprocess the collected feature data to obtain processing results; Perform data local energy analysis on the processing results, and obtain initial characteristic parameters according to the analysis results; According to the established rail damage database, the initial characteristic parameters are adaptively approximated, and the damage data is obtained according to the processing results; Classify, manage and store the damage data, and control the client terminal according to the damage data.