CN113946913B - Railway running fault detection method, device, electronic equipment and storage medium - Google Patents

Abstract

The invention provides a method and a device for detecting railway running faults, electronic equipment and a storage medium. The method comprises the following steps: acquiring historical driving fault detection data of fault detection stations distributed along a railway line; and determining the fault detection result of the railway vehicle according to the historical vehicle fault detection data of each fault detection station. The invention utilizes the detection data of the detection station of the full railway line range to carry out comprehensive fault detection on the railway running, improves the comprehensive detection capability of the automatic railway running safety detection, and can meet the requirements of integrated, automatic and intelligent railway running safety monitoring.

Description

Railway running fault detection method, device, electronic equipment and storage medium

technical field

The invention relates to the technical field of railway passenger transport safety, in particular to a railway running fault detection method, device, electronic equipment and storage medium.

Background technique

The safety of railway operation is the top priority of railway transportation work. Through the installation of different types of operation safety detection equipment (hereinafter referred to as the fault detection station) on each line of the whole railway, the relevant data of the passing railway operation are sampled, and according to the sampling The data uses the algorithm model to obtain the fault detection results of railway running, so as to realize the fault detection of railway running.

However, the fault detection algorithm models corresponding to different types of fault detection stations are different, and the fault detection capabilities are uneven. Accumulation, when the accumulated times reaches the preset threshold, it is considered to reach the preset fault judgment level. This method can only carry out a relatively primary cumulative evaluation of the number of faults, and cannot break through the limitation of the fault detection algorithm model calculation capability of the fault detection station itself. The existing patent similar to the technical solution of the present invention is: KR101769588B1.

SUMMARY OF THE INVENTION

The present invention provides a railway running fault detection method, device, electronic equipment and storage medium, which are used to solve the problems that the prior art does not have comprehensive detection capability for railway running safety detection and the fault detection accuracy is low, and improve the automatic railway running safety. Comprehensive detection capability of detection.

The present invention provides a railway running fault detection method, comprising:

Obtain the historical traffic fault detection data of each fault detection station distributed along the railway line;

The fault detection result of the railway running is determined according to the historical running fault detection data of each fault detection station.

Optionally, the fault detection result of the railway running is determined according to the historical running fault detection data of each fault detection station, including:

According to the historical driving fault detection data of each fault detection station, determine the fault detection result of the single-point detection station, the fault detection result of the same type of detection station and the fault detection result of all the detection stations;

According to the fault detection result of the single-point detection station, the fault detection result of the same type of detection station, and the fault detection results of all the detection stations, the comprehensive fault detection result of the railway running is determined.

Optionally, according to the fault detection results of the single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all the detection stations, determine the comprehensive fault detection results of the railway operation, including:

Input the fault detection results of the single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all detection stations into the weighted comprehensive fault detection model to obtain the comprehensive fault detection results of the railway running. ;

The weighted comprehensive fault detection model is used to assign different weight coefficients to the input fault detection results of the single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all detection stations.

Optionally, according to the historical driving fault detection data of each fault detection station, determine the fault detection result of the single-point detection station, including:

，所述第一公式为： Calculate the fault detection result of the single-point detection station according to the following first formula

, the first formula is:

or,

为常数，

₁为单点检测站的历史行车故障检测次数，

为根据单点检测站 的历史行车故障检测数据计算得到的第一故障概率，

为预设权重值。 in,

is a constant,

₁ is the number of historical driving fault detections of the single-point detection station,

is the first failure probability calculated according to the historical driving failure detection data of the single-point detection station,

is the default weight value.

Optionally, according to the historical driving fault detection data of each fault detection station, determine the fault detection result of the same type of detection station, including:

，所述第二公式为： Calculate the fault detection result of the same type of detection station according to the second formula below

, the second formula is:

or,

为常数，

₂为同类型检测站的历史行车故障检测次数，

为根据同类型检 测站的历史行车故障检测数据计算得到的第二故障概率，

为预设权重值。 in,

is a constant,

₂ is the number of historical driving fault detections of the same type of detection station,

is the second fault probability calculated according to the historical driving fault detection data of the same type of detection station,

is the default weight value.

Optionally, according to the historical driving fault detection data of each fault detection station, determine the fault detection results of all the detection stations, including:

，所述第三公式为： Calculate the fault detection results of all detection stations according to the third formula below

, the third formula is:

or,

为常数，l ₃为全部检测站的历史行车故障检测次数，P ₃为根据全部检测站 的历史行车故障检测数据计算得到的第三故障概率，

为预设权重值。 in,

is a constant, l3 is the number of historical driving fault detections of all detection stations, P3 _is the _third failure probability calculated according to the historical driving fault detection data of all detection stations,

is the default weight value.

Optionally, the weighted comprehensive fault detection model S is:

为所述单点检测站故障检测结果

的权重系数，

为所述同类型检测站 故障检测结果

的权重系数，

为所述全部检测站故障检测结果

的权重系数； in,

Fault detection results for the single point detection station

The weight coefficient of ,

Fault detection results for the same type of detection station

The weight coefficient of ,

Fault detection results for all the detection stations

The weight coefficient of ;

or,

为权重系数，

为常数。 in,

is the weight coefficient,

is a constant.

Optionally, after obtaining the comprehensive fault detection result of the railway vehicle, the method further includes:

When the comprehensive fault detection result of the railway operation is greater than the preset fault level determination threshold, determine that the comprehensive fault detection result of the railway operation is a fault;

When the comprehensive fault detection result of the railway operation is less than the preset failure level determination threshold, it is determined that the comprehensive fault detection result of the railway operation is a fault false alarm.

Optionally, after determining that the comprehensive fault detection result of the railway vehicle is a fault or a fault false alarm, the method further includes:

According to the comprehensive fault detection result of the railway operation, the weight coefficient of the weighted comprehensive fault detection model is adjusted as follows:

，

为所述单点检测站 故障检测结果

的权重系数，

为所述同类型检测站故障检测结果

的权重系数，

为所 述全部检测站故障检测结果

的权重系数，

为常数； in,

,

Fault detection results for the single point detection station

The weight coefficient of ,

Fault detection results for the same type of detection station

The weight coefficient of ,

Fault detection results for all the detection stations

The weight coefficient of ,

is a constant;

，

为所述带权重的综 合故障检测模型为了综合平衡单点检测站故障检测结果、同类型检测站故障检测结果和全 部检测站故障检测结果在求得综合故障检测结果时的不同权重而设置的系数，

为常数。 or,

,

Coefficient set for the weighted comprehensive fault detection model in order to comprehensively balance the different weights of the single-point detection station fault detection results, the fault detection results of the same type of detection stations and the fault detection results of all detection stations in obtaining the comprehensive fault detection results ,

is a constant.

The present invention also provides a railway running fault detection device, comprising:

The acquisition module is used to acquire the historical driving fault detection data of each fault detection station distributed along the railway line;

The processing module is configured to determine the fault detection result of the railway running according to the historical running fault detection data of each fault detection station.

The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, when the processor executes the program, the above-mentioned railway running fault can be realized The steps of the detection method.

The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, implements the steps of any one of the above-mentioned methods for detecting a railway running fault.

The railway running fault detection method, device, electronic equipment and storage medium provided by the present invention first obtain the historical running fault detection data of each fault detection station distributed along the railway line, and then according to the historical running fault detection data of each fault detection station, Determine the fault detection results of railway running. It can be seen from this that the present invention uses the detection data of the detection stations in the entire railway line to perform comprehensive fault detection on railway driving, improves the comprehensive detection capability of automatic railway driving safety detection and the accuracy of fault detection, reduces the fault false alarm rate, and enhances the It improves the ability and level of railway running safety monitoring, and effectively guarantees the running safety of railway vehicles.

Description of drawings

In order to explain the present invention or the technical solutions in the prior art more clearly, the following will briefly introduce the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are the For some embodiments of the invention, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.

Fig. 1 is the schematic flow chart of the railway running fault detection method provided by the present invention;

Fig. 2 is the structural representation of the railway running safety monitoring system provided by the present invention;

3 is a schematic structural diagram of a railway running fault detection device provided by the present invention;

FIG. 4 is a schematic structural diagram of an electronic device provided by the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention. , not all examples. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

As shown in Figure 1, the railway running fault detection method provided by the present invention includes:

Step 101: Acquire historical driving fault detection data of each fault detection station distributed along the railway line;

In this step, first obtain the historical driving fault detection data of each fault detection station distributed along the railway line, the historical multiple detection data provided by the present invention can set the upper limit of the detection times, and the detection data can be determined according to the performance of the fault detection station For example, the sound data of the axle of the railway forming vehicle can be obtained, the pressure data of the axle of the vehicle can also be obtained, and the image data of the railway running can also be obtained, which is not specifically limited here.

Step 102: Determine the fault detection result of the railway running according to the historical running fault detection data of each fault detection station.

In this step, when the railway traffic passes a certain fault detection station, based on the historical traffic fault detection data of each fault detection station acquired in step 101, the fault detection data of the currently passing fault detection station and the fault detection data of the current fault detection station and the difference between the fault detection stations in the whole railway line are respectively determined. The fault detection data collection of all fault detection stations of the same type as the currently passing fault detection station, and then determine the single-point fault detection results output by the currently passed fault detection station, and the results from all fault detection stations of the same type as the currently passed fault detection station. The multi-point fault detection results of the same type obtained from the fault detection data set of 1 and the multi-type fault detection results obtained from the historical driving fault detection data set of each fault detection station.

In this step, the single-point detection station fault detection results, the same-type multi-point fault detection results and the multi-type fault detection results obtained above are input into the weighted comprehensive fault detection model to calculate the comprehensive fault detection results of railway running. Among them, the weighted comprehensive fault detection model is used to assign different weight coefficients to the input single-point detection station fault detection results, the same type of multi-point fault detection results and the multi-type fault detection results. After the comprehensive fault detection results are obtained, the weight coefficients of the model are automatically optimized according to the comprehensive fault detection results, thereby effectively improving the fault detection capability of the model.

The railway running fault detection method provided by the present invention first obtains the historical running fault detection data of each fault detection station distributed along the railway line, and then determines the fault detection result of the railway running according to the historical running fault detection data of each fault detection station. It can be seen from this that the present invention uses the detection data of the detection stations in the entire railway line to perform comprehensive fault detection on railway driving, improves the comprehensive detection capability of automatic railway driving safety detection and the accuracy of fault detection, reduces the fault false alarm rate, and enhances the It improves the ability and level of railway running safety monitoring, and effectively guarantees the running safety of railway vehicles.

Based on the content of the above embodiment, in this embodiment, according to the historical driving fault detection data of each fault detection station, the fault detection result of the railway driving is determined, including:

According to the historical driving fault detection data of each fault detection station, the single-point detection station fault detection result obtained based on the historical driving fault detection data of each fault detection station, and the historical driving fault detection data set based on the same type of fault detection station are determined. The fault detection results of the same type of detection stations and the fault detection results of all detection stations based on the historical driving fault detection data set of all fault detection stations;

According to the fault detection result of the single-point detection station, the fault detection result of the same type of detection station, and the fault detection results of all the detection stations, the comprehensive fault detection result of the railway running is determined.

In this embodiment, it should be noted that when determining the fault detection results of a single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all detection stations, the algorithm model is used to calculate the probability of multiple failures, and then the The fault detection results of the railway running through the current fault detection station are obtained by comprehensive calculation of the probability of multiple faults.

Based on the content of the above-mentioned embodiment, in this embodiment, according to the fault detection result of the single-point detection station, the fault detection result of the same type of detection station, and the fault detection results of all the detection stations, the comprehensive railway operation is determined. Fault detection results, including:

Input the fault detection results of the single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all detection stations into the weighted comprehensive fault detection model to obtain the comprehensive fault detection results of the railway running. ;

The weighted comprehensive fault detection model is used to assign different weight coefficients to the input fault detection results of the single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all detection stations.

Based on the content of the above embodiment, in this embodiment, according to the historical driving fault detection data of each fault detection station, determine the single-point detection station fault detection result obtained based on the historical driving fault detection data of each fault detection station, based on The fault detection results of the same type of detection stations obtained from the historical driving fault detection data sets of the same type of fault detection stations and the fault detection results of all detection stations based on the historical driving fault detection data sets of all fault detection stations, including:

According to the historical driving fault detection data of each fault detection station, determine the first fault probability when the historical driving passes through the current fault detection station, and determine the fault detection result of the single-point detection station according to the first fault probability;

And, according to the historical driving fault detection data set of the same type of fault detection station, determine the second fault probability when the historical driving passes the current fault detection station, and determine the same type of detection station according to the second fault probability fault detection results;

And, according to the historical driving fault detection data set of all the fault detection stations, determine the third fault probability when the historical driving passes the current fault detection station, and determine the fault detection of all the detection stations according to the third fault probability result.

Based on the content of the above embodiment, in this embodiment, according to the historical driving fault detection data of each fault detection station, the fault detection result of the single-point detection station is determined, including:

，所述第一公式为： Calculate the fault detection result of the single-point detection station according to the following first formula

, the first formula is:

or,

为常数，

₁为单点检测站的历史行车故障检测数据，

为根据单点检测站 的历史行车故障检测数据计算得到的第一故障概率，

为预设权重值。 in,

is a constant,

₁ is the historical driving fault detection data of the single-point detection station,

is the first failure probability calculated according to the historical driving failure detection data of the single-point detection station,

is the default weight value.

可以理解为通过算法模型计算出的铁路行车通过故障检 测站A时当前的故障概率和此前每一次通过故障检测站A时的故障概率的集合。单点检测站 的历史行车故障检测次数

可以理解为铁路行车通过故障检测站A的历史次数，历史次数 包括当前本次通过，且每一次铁路行车通过故障检测站A就计算一次故障概率。 In this embodiment, the single-point detection station is equivalent to each individual fault detection station. For example, when a railway train passes through the fault detection station A, the fault detection station A may be referred to as the single-point detection station A. The historical traffic fault detection data of the single-point detection station can be understood as the collection of the current traffic data collected by the fault detection station A and the previous multiple traffic data when the railway traffic passes the fault detection station A. first failure probability

It can be understood as the set of the current failure probability when the railway train passes through the fault detection station A and the failure probability when it passes through the fault detection station A each time previously calculated by the algorithm model. The number of historical driving fault detections at a single-point detection station

It can be understood as the historical number of railway traffic passing through the fault detection station A, the historical number includes the current passing, and a failure probability is calculated every time the railway traffic passes the fault detection station A.

Based on the content of the above embodiment, in this embodiment, according to the historical driving fault detection data of each fault detection station, the fault detection result of the same type of detection station is determined, including:

，所述第二公式为： Calculate the fault detection result of the same type of detection station according to the second formula below

, the second formula is:

or,

为常数，

₂为同类型检测站的历史行车故障检测数据，

为根据同类型检 测站的历史行车故障检测数据计算得到的第二故障概率，

为预设权重值。 in,

is a constant,

₂ is the historical driving fault detection data of the same type of detection station,

is the second fault probability calculated according to the historical driving fault detection data of the same type of detection station,

is the default weight value.

可以理解为第一故障概率

加上与故障检测站A 型号相同的其他故障检测站通过算法模型计算出的铁路行车此前每一次通过自身故障检 测站时的故障概率。同类型检测站的历史行车故障检测次数

可以理解为铁路行车通过故 障检测站A的历史次数加上通过与故障检测站A相同型号的各个故障检测站的历史次数，历 史次数包括铁路行车通过故障检测站A的次数加上通过与故障检测站A相同型号的各个故 障检测站的次数，且每一次铁路行车通过故障检测站就计算一次故障概率。 In this embodiment, the detection stations of the same type are fault detection stations of the same model on the railway line. For example, when the railway travels through the fault detection station A, other fault detection stations of the same model as the fault detection station A together constitute the same type of detection station . The historical driving fault detection data of the same type of detection station can be understood as the current driving data collected by the fault detection station A when the railway traffic passes through the fault detection station A, and other fault detection stations of the same model as the fault detection station A have collected many times before. collection of driving data. Second Failure Probability

It can be understood as the first failure probability

Add the failure probability calculated by the algorithm model of other fault detection stations with the same model as the fault detection station A when the railway train passes through its own fault detection station each time before. The number of historical driving fault detections of the same type of detection station

It can be understood as the historical number of railway traffic passing through fault detection station A plus the historical number of each fault detection station of the same model as fault detection station A, the historical number of times includes the number of railway traffic passing through fault detection station A plus the number of passing and fault detection. The number of each fault detection station of the same model of station A, and the probability of a fault is calculated every time the railway travels through the fault detection station.

Based on the content of the above embodiment, in this embodiment, according to the historical driving fault detection data of each fault detection station, the fault detection results of all the detection stations are determined, including:

，所述第三公式为： Calculate the fault detection results of all detection stations according to the third formula below

, the third formula is:

or,

为常数，l ₃为全部检测站的历史行车故障检测数据，P ₃为根据全部检测站 的历史行车故障检测数据计算得到的第三故障概率，

为预设权重值。 in,

is a constant, l3 is the historical driving fault detection data of all the detection stations, P3 _is the _third failure probability calculated according to the historical driving fault detection data of all the detection stations,

is the default weight value.

可以理解为铁路行车 每次通过各个故障检测站时计算得到的故障概率。全部检测站的历史行车故障检测次数

可以理解为铁路行车通过各个故障检测站的总次数。 In this embodiment, all detection stations are all fault detection stations set on the railway line. The historical traffic fault detection data of all the detection stations can be understood as the collection of the current traffic data collected by each fault detection station and the traffic data collected many times before when the railway travels through all the detection stations. Third Failure Probability

It can be understood as the failure probability calculated every time a railway vehicle passes through each failure detection station. The number of historical driving fault detections of all detection stations

It can be understood as the total number of times the railway traffic passes through each fault detection station.

Based on the content of the above embodiment, in this embodiment, the weighted comprehensive fault detection model S is:

为所述单点检测站故障检测结果

的权重系数，

为所述同类型检测站 故障检测结果

的权重系数，

为所述全部检测站故障检测结果

的权重系数；in,

Fault detection results for the single point detection station

The weight coefficient of ,

Fault detection results for the same type of detection station

The weight coefficient of ,

Fault detection results for all the detection stations

The weight coefficient of ;

or,

为所述带权重的综合故障检测模型为了综合平衡单点检测站故障检测结 果、同类型检测站故障检测结果和全部检测站故障检测结果在求得综合故障检测结果时的 不同权重而设置的系数，

为常数。 in,

Coefficient set for the weighted comprehensive fault detection model in order to comprehensively balance the different weights of the single-point detection station fault detection results, the fault detection results of the same type of detection stations and the fault detection results of all detection stations in obtaining the comprehensive fault detection results ,

is a constant.

Based on the content of the foregoing embodiment, in this embodiment, after obtaining the comprehensive fault detection result of the railway operation, the method further includes:

When the comprehensive fault detection result of the railway operation is greater than the preset fault level determination threshold, determine that the comprehensive fault detection result of the railway operation is a fault;

When the comprehensive fault detection result of the railway operation is less than the preset failure level determination threshold, it is determined that the comprehensive fault detection result of the railway operation is a fault false alarm.

In this embodiment, it should be noted that the corresponding fault level determination threshold is set, and when the comprehensive fault detection result S reaches the corresponding fault level determination threshold, the fault level D of the railway operation is determined.

及其判定阈值

可以为 Among them, the failure level of railway traffic

and its decision threshold

can be

Based on the content of the foregoing embodiment, in this embodiment, after determining that the comprehensive fault detection result of the railway train is a fault or a fault false alarm, the method further includes:

According to the comprehensive fault detection result of the railway operation, the weight coefficient of the weighted comprehensive fault detection model is adjusted as follows:

，

为所述单点检测站 故障检测结果

的权重系数，

为所述同类型检测站故障检测结果

的权重系数，

为所 述全部检测站故障检测结果

的权重系数，

为常数； in,

,

Fault detection results for the single point detection station

The weight coefficient of ,

Fault detection results for the same type of detection station

The weight coefficient of ,

Fault detection results for all the detection stations

The weight coefficient of ,

is a constant;

，

为所述带权重的综 合故障检测模型为了综合平衡单点检测站故障检测结果、同类型检测站故障检测结果和全 部检测站故障检测结果在求得综合故障检测结果时的不同权重而设置的系数，

为常数。 or,

,

Coefficient set for the weighted comprehensive fault detection model in order to comprehensively balance the different weights of the single-point detection station fault detection results, the fault detection results of the same type of detection stations and the fault detection results of all detection stations in obtaining the comprehensive fault detection results ,

is a constant.

In this embodiment, it should be noted that because the models of each fault detection station distributed on a railway line are not uniform, the detection capabilities of each fault detection station are uneven. If the inspection data collected by the inspection station is used for fault detection, the accuracy of the inspection results cannot be guaranteed. In order to solve the problem that the detection accuracy of the prior art is not high, the present invention provides a brand-new railway traffic fault detection method. When the railway traffic passes through the A fault detection station at the time B, firstly obtain the current data collected by the A fault detection station. The railway driving detection data and the historically collected detection data, so as to determine the single-point fault detection result of the fault detection station A according to the above data, and assign the corresponding weight coefficient to the result. The weight coefficient can be determined according to the historical detection results. For example, when When the historical performance of the fault detection station A is excellent, a high weight coefficient can be assigned to it. When the historical performance of the fault detection station A is not good (the number of errors reported), it can be assigned a lower weight coefficient. When the railway travels through the A fault detection station, the historical detection data of each fault detection station of the same type as the A fault detection station is obtained at the same time, and comprehensive calculation is performed to obtain the same type of fault detection results when the railway travels through the same type of detection station, which is It assigns corresponding weight coefficients. When the railway traffic passes through the fault detection station A, the historical detection data of the railway traffic passing through each fault detection station on the whole road are obtained at the same time, and the fault detection results of all the detection stations are obtained through comprehensive calculation, and the corresponding weight coefficients are assigned to them. Finally, comprehensive evaluation of faults is carried out. According to the final actual inspection results of the faults determined each time by a fault detection station in the later inspection, the weights calculated by the comprehensive evaluation method S of fault detection are automatically adjusted. When the history of fault detection station A is When the performance is excellent, a higher weight coefficient is automatically assigned to the fault detection results of the single-point detection station of the A fault detection station and the fault detection results of the same type of detection station. When the historical performance of the A fault detection station is not good (more errors are reported) , which can be automatically assigned a lower weight factor. After a long period of operation, each fault detection in the whole channel can obtain a comprehensive fault evaluation calculation method with different weights that is closer to the actual failure probability, which can effectively improve the fault evaluation ability. Described below by specific embodiments:

Example 1:

In this embodiment, the railway traffic safety monitoring system is represented by the trackside acoustic diagnosis and monitoring system TADS (Trackside Acoustic Detection System) for rolling bearings of the EMU, and the trackside dynamic monitoring system TPDS (Truck Performance Detection System) for the running quality of the EMU. The above-mentioned railway traffic safety monitoring system adopts the architecture design of two-level network deployment and multi-level monitoring application. The network platform is deployed in the monitoring center of the National Railway Group and the vehicle (high-speed train) section to realize the realization of the National Railway Group, the Railway Bureau, and the train (high-speed train) section. . The multi-level application system function of the train station will centrally connect the different types of driving safety detection equipment (hereinafter referred to as the fault detection station) installed on each railway line on the whole road to realize comprehensive automatic monitoring. Among them, TADS mainly uses the principle of acoustics to sample the sound of the axle of the vehicle in operation, and calculates the probability of possible failure of the axle through the algorithm model, so as to monitor the running state of the axle; The pressure is sampled, and the probability of possible failure is calculated through the algorithm model to realize the wheel condition monitoring.

Affected by technological development capabilities and equipment operating conditions, different models of fault detection stations constructed by different manufacturers currently have different algorithm models for fault judgment, and the diagnostic capabilities are uneven, and the judgment difference rate can reach more than 40% (1. One type of fault detection station is judged as faulty and another type of fault detection station is judged as no fault). At the same time, with the continuous construction and improvement of the above-mentioned monitoring system for networked operation, the networked monitoring capability of the network structure has been gradually formed in the whole road. Therefore, using the detection data of multiple fault detection stations of different models and different locations, comprehensive fault judgment is carried out on the running state of the vehicle axle, and the fault judgment method is continuously improved and optimized in combination with the actual inspection results, which is helpful for improving the safety monitoring of automatic driving on the whole road. level is important.

At present, TADS and TPDS fault detection station equipment can use a single fault detection station to perform model calculation on multiple detection data of the same wheel or axle at different times, so as to obtain fault judgment results. This method utilizes multiple detection data, effectively shielding the operation error caused by abnormal detection, but cannot break through the limitation of the fault detection station's own model computing capability. For this reason, in the field of passenger vehicle monitoring, TADS tries to use networked equipment to conduct comprehensive fault evaluation. The current method is to accumulate the number of faults reported by each fault detection station's own fault evaluation. When the accumulated number reaches a certain value, it is considered that Reach the level of network failure judgment. This method only uses the evaluation results of each fault detection station, the comprehensive application level of monitoring data is insufficient, the fault evaluation ability is low, and the time of fault discovery is obviously delayed.

It can be seen that the existing system either does not have the ability to comprehensively judge fault networking, or can only carry out relatively preliminary judgment on the cumulative number of networking faults, and cannot effectively use the detection data of fault detection stations of different models and locations to conduct comprehensive fault judgment. Its monitoring capability is difficult to meet the needs of integrated, automated and intelligent railway traffic safety monitoring.

To this end, the present invention provides a railway traffic fault detection method based on a railway traffic safety monitoring system. The fault evaluation results of multiple detections of type fault detection stations and multiple detections of all fault detection stations, using the comprehensive evaluation model to realize the network comprehensive evaluation of wheel and axle faults, and can automatically optimize the comprehensive evaluation model according to the actual inspection results of the axles, effectively realize the railway Efficient monitoring of vehicle axle operating safety.

As shown in FIG. 2 , the present invention is based on the existing railway traffic safety monitoring system, makes full use of the networked architecture of two-level deployment, and obtains the wheel and axle detection data of each fault detection station on the whole road at the national railway group level. The railway traffic fault detection method of the traffic safety monitoring system, through the operation of the method, effectively realizes the comprehensive evaluation of the vehicle axle fault network and the automatic optimization of the evaluation model.

Among them, the two-level system of the National Railway Group and the monitoring center of each vehicle (motor train) section can synchronize data in real time. The national railway group-level system is connected to the wheel and axle detection data uploaded by the monitoring center-level system of each vehicle (high-speed train) section, and the comprehensive evaluation model is used to conduct a networked comprehensive evaluation of wheel and axle faults, and the evaluation results are sent to the corresponding vehicle (high-speed train) section The monitoring center-level system is used by the corresponding users. The monitoring center-level system of each vehicle (motor train) section is centrally connected to different locations and different types of fault detection stations on the railway line within the jurisdiction, and the wheel and axle detection data of each fault detection station is summarized and uploaded to the national railway group-level system. The basic detection data (including passing information, vehicle information, axle detection information, equipment status information, etc.) generated by the fault detection station, as well as the comprehensive evaluation results of the axle fault network, are inquired, disposed of, and the application system functions such as backfilling of the inspection results.

In order to describe the railway traffic fault detection method based on the railway traffic safety monitoring system provided by the present invention more clearly, it is assumed that the sets of fault detection station equipment of different manufacturers and models (or different identification principles) are:

Then the set of fault detection stations installed in the whole range can be recorded as:

Therefore, when an EMU passes through a fault detection station, each axle of the EMU will generate a detection data, and the collection of historical monitoring values of all axles of the EMU can be recorded as:

表示利用算法模型计算得出的故障发生概率，则同一列 车经过全路各故障检测站时，每个轮轴每一次检测的故障概率可表示为： Assume

Represents the probability of fault occurrence calculated by the algorithm model, then when the same train passes through each fault detection station on the whole road, the fault probability of each axle detection can be expressed as:

Therefore, the comprehensive evaluation method for vehicle wheel and axle faults of a railway driving safety monitoring system according to the present invention can be described as:

Step 1. Calculate the single-point multiple fault judgment results when the axle passes through a fault detection station;

Step 2. Calculate the multi-point and multiple fault judgment results of the same type when the axle passes through a fault detection station of the same type of a fault detection station;

Step 3. Calculate the multi-type, multi-point and multiple fault judgment results when the axle passes through all the fault detection stations;

Step 4. Calculate the comprehensive evaluation result of the axle fault;

Step 5. Use the actual inspection results to automatically adjust the parameters of the comprehensive evaluation method for wheel and axle faults.

时，查找该列车某一轮轴

在 该故障检测站（所谓单点）历史多次的检测结果，并计算所有的多次检测通过算法模型

计算出的多次故障概率，将算出的多次故障概率进行综合计算得到该轮轴

通过某一故障 检测站时的单点多次故障评判结果，记为

。 Among them, in step 1, when the train runs through a certain fault detection station

When , find an axle of the train

At this fault detection station (so-called single point) history multiple detection results, and calculate all multiple detections through the algorithm model

Calculate the probability of multiple failures, and comprehensively calculate the probability of multiple failures to obtain the axle

The single-point multiple fault evaluation results when passing through a fault detection station are recorded as

.

Among them, the comprehensive calculation method can be

，其中，

为常数，

为权重值。权重值

可为

。

,in,

is a constant,

is the weight value. Weights

can be

.

Can also be:

，其中，

为常数，

为权重值。

,in,

is a constant,

is the weight value.

时，同时查找全路范围内，该列车某一 轮轴

在通过相同型号故障检测站时的历史多次检测结果，并计算所有多次检测通过算法 模型计算出的故障概率，并将算出的多次故障概率进行综合计算得到该轮轴

通过某一故 障检测站时的同型多点多次故障评判结果，记为

。 In step 2, the train passes the fault detection station

At the same time, search for a certain axle of the train within the whole road range.

The historical multiple detection results when passing through the same type of fault detection station, and calculate the failure probability calculated by the algorithm model for all multiple detections, and comprehensively calculate the calculated multiple failure probability to obtain the axle.

The multi-point and multiple fault judgment results of the same type when passing through a fault detection station are recorded as

.

Among them, the comprehensive calculation method can be

，其中，

为常数，

为权重值。权重值

可为

。

,in,

is a constant,

is the weight value. Weights

can be

.

Can also be:

，其中，

为常数，

为权重值。

,in,

is a constant,

is the weight value.

时，同时查找全路范围内，该列车某一 轮轴

在通过全路所有故障检测站时的历史多次检测结果，并计算所有多次检测通过算法 模型计算出的故障概率，并将算出的多次故障概率进行综合计算得到该轮轴

通过某一故 障检测站时的多型多点多次故障评判结果，记为

。 In step 3, the train passes through the fault detection station

At the same time, search for a certain axle of the train within the whole road range.

The historical multiple detection results when passing through all fault detection stations on the whole road, and calculate the fault probability calculated by the algorithm model for all multiple detections, and comprehensively calculate the calculated multiple fault probability to obtain the axle.

The multi-type, multi-point and multiple fault judgment results when passing through a fault detection station are recorded as

.

Among them, the comprehensive calculation method can be

，其中，

为常数，

为权重值。为加快运算效率，可取

为 1，

为

。

,in,

is a constant,

is the weight value. In order to speed up the operation efficiency, it is desirable to

is 1,

for

.

Can also be:

，其中，

为常数，

为权重值。

,in,

is a constant,

is the weight value.

多次重复检测后的单点多次故障评判结 果

、该故障检测站所有同型号设备对轮轴

多次重复检测后的同型多点多次故障评判结 果

，以及全路各类型所有故障检测站对轮轴

的多次重复检测后的多型多点多次故障评 判结果

，进行故障综合评判，得到该轮轴的故障发生概率，记为

。 In step 4, it will reflect the fault detection station on the axle

Single-point multiple fault judgment results after multiple repeated detections

, All the equipment of the same type of the fault detection station are opposite to the axle

Evaluation results of multiple faults of the same type and multiple points after repeated detection

, and all types of fault detection stations on the whole road to the axles

The multi-type, multi-point and multi-fault judgment results after multiple repeated detections

, carry out a comprehensive evaluation of faults, and obtain the probability of fault occurrence of the axle, which is recorded as

.

Among them, the method of comprehensive fault evaluation can be as follows:

、

、

为权重值。权重值可以取

。 ,in,

,

,

is the weight value. The weight value can be taken

.

Can also be:

，其中，

为常数，

为权重系 数。

,in,

is a constant,

is the weight coefficient.

，当

达到相应阈值时，判定出轮轴

的故 障等级

。At the same time, set the corresponding fault level judgment threshold

,when

When the corresponding threshold is reached, the axle is determined

level of failure

.

及其判定阈值

可以为： Among them, the failure level

and its decision threshold

Can be:

In step 5, the weights calculated by the comprehensive fault evaluation method S of the fault detection station are automatically adjusted according to the final actual inspection results of the faulty axles determined each time by a fault detection station in the later inspection. After a long period of operation, each fault detection station on the whole road can obtain a comprehensive fault evaluation calculation method with different weights that is closer to the actual fault occurrence probability.

Among them, the weights that are automatically adjusted can be:

，

为所述单点检测站 故障检测结果

的权重系数，

为所述同类型检测站故障检测结果

的权重系数，

为所 述全部检测站故障检测结果

的权重系数，

为常数； in,

,

Fault detection results for the single point detection station

The weight coefficient of ,

Fault detection results for the same type of detection station

The weight coefficient of ,

Fault detection results for all the detection stations

The weight coefficient of ,

is a constant;

，

为带权重的综合故 障检测模型为了综合平衡单点检测站故障检测结果、同类型检测站故障检测结果和全部检 测站故障检测结果在求得综合故障检测结果时的不同权重而设置的系数，

为常数。 or,

,

It is the coefficient set for the weighted comprehensive fault detection model to comprehensively balance the different weights of the fault detection results of single-point detection stations, the fault detection results of the same type of detection stations and the fault detection results of all detection stations in obtaining the comprehensive fault detection results,

is a constant.

In this embodiment, it should be noted that the historical multiple detections in the present invention can set an upper limit of the number of detections. The same type of fault detection station and all fault detection stations deployed on the same line of the fault detection station are respectively set. It can be seen that the present invention comprehensively utilizes the networked monitoring data to realize fault judgment, and establishes a unique comprehensive judgment model for different fault detection stations through the design of the method. The method can automatically adjust and optimize the judgment parameters of the fault judgment model, and can effectively improve the The ability to judge faults. Specifically, the present invention proposes a comprehensive evaluation method for vehicle wheel and axle faults of a railway driving safety monitoring system, which uses the wheel and axle detection data of the fault detection stations in the whole road to calculate the multiple detections of single-point equipment, the multiple detections of the same type of equipment, Based on the fault evaluation results of multiple inspections of all equipment, the comprehensive evaluation model is used to realize the networked comprehensive evaluation of wheel and axle faults, and the comprehensive evaluation model can be automatically optimized in combination with the actual inspection results of the wheel and axle, which effectively realizes the efficient monitoring of the running status of the railway vehicle wheel and axle. It greatly improves the accuracy of wheel and axle fault judgment, reduces the false alarm rate, enhances the ability and level of driving safety monitoring, and effectively guarantees the safety of railway vehicles.

The following describes the railway running fault detection device provided by the present invention. The railway running fault detection device described below and the railway running fault detection method described above can be referred to each other correspondingly.

As shown in Figure 3, a railway running fault detection device provided by the present invention includes:

The acquisition module 1 is used to acquire the historical driving fault detection data of each fault detection station distributed along the railway line;

The processing module 2 is configured to determine the fault detection result of the railway running according to the historical running fault detection data of each fault detection station.

In this embodiment, the historical driving fault detection data of each fault detection station distributed along the railway line is obtained first. The historical multiple detection data provided by the present invention can set an upper limit of the number of detections, and the detection data can be based on the fault detection station. The performance is determined, for example, the sound data of the axle of the railway forming vehicle can be obtained, the pressure data of the axle of the vehicle can also be obtained, and the image data of the railway running can also be obtained, which is not specifically limited here.

In this embodiment, when the railway traffic passes through a certain fault detection station, based on the acquired historical traffic fault detection data of each fault detection station, the fault detection data of the currently passing fault detection station and the fault detection data of the current fault detection station and the difference between the current and the whole railway line are determined respectively. The fault detection data set of all fault detection stations of the same type as the passing fault detection station, and then determine the single-point fault detection results output by the currently passing fault detection station, and the results from all fault detection stations of the same type as the currently passing fault detection station. The same type of multi-point fault detection results obtained from the fault detection data set and the multi-type fault detection results obtained from the historical driving fault detection data set of each fault detection station.

In this embodiment, the single-point detection station fault detection results, the same-type multi-point fault detection results, and the multi-type fault detection results obtained above are input into a weighted comprehensive fault detection model to calculate the comprehensive fault detection results of railway traffic. . Among them, the weighted comprehensive fault detection model is used to assign different weight coefficients to the input single-point detection station fault detection results, the same type of multi-point fault detection results and the multi-type fault detection results. After the comprehensive fault detection results are obtained, the weight coefficients of the model are automatically optimized according to the comprehensive fault detection results, thereby effectively improving the fault detection capability of the model.

The railway running fault detection device provided by the present invention first obtains the historical running fault detection data of each fault detection station distributed along the railway line, and then determines the fault detection result of the railway running according to the historical running fault detection data of each fault detection station. It can be seen from this that the present invention uses the detection data of the detection stations in the entire railway line to perform comprehensive fault detection on railway driving, improves the comprehensive detection capability of automatic railway driving safety detection and the accuracy of fault detection, reduces the fault false alarm rate, and enhances the It improves the ability and level of railway running safety monitoring, and effectively guarantees the running safety of railway vehicles.

FIG. 4 illustrates a schematic diagram of the physical structure of an electronic device. As shown in FIG. 4 , the electronic device may include: a processor (processor) 410, a communication interface (Communications Interface) 420, a memory (memory) 430, and a communication bus 440, The processor 410 , the communication interface 420 , and the memory 430 communicate with each other through the communication bus 440 . The processor 410 can call the logic instructions in the memory 430 to execute the railway running fault detection method, the method includes: acquiring the historical running fault detection data of each fault detection station distributed along the railway line; The historical traffic fault detection data is used to determine the fault detection result of the railway traffic.

In addition, the above-mentioned logic instructions in the memory 430 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including Several instructions are used to cause 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 of the present invention. 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 .

In another aspect, the present invention also provides a computer program product, the computer program product includes a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can Execute the railway running fault detection method provided by the above methods, the method includes: acquiring historical running fault detection data of each fault detection station distributed along the railway line; according to the historical running fault detection data of each fault detection station, determining The fault detection result of the railway running.

In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, it is implemented to execute the method for detecting a railway running fault provided by the above methods. The method includes: acquiring the historical running fault detection data of each fault detection station distributed along the railway line; and determining the fault detection result of the railway running according to the historical running fault detection data of each fault detection station.

The device embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed over multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

From the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on this understanding, the above-mentioned technical solutions can be embodied in the form of software products in essence or the parts that make contributions to the prior art, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic A disc, an optical disc, etc., includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the methods described in various embodiments or some parts of the embodiments.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that it can still be The technical solutions described in the foregoing embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (6)

1. a railway running fault detection method, is characterized in that, comprises: Obtain the historical traffic fault detection data of each fault detection station distributed along the railway line; Determine the fault detection result of the railway running according to the historical running fault detection data of each fault detection station; Wherein, according to the historical running fault detection data of each fault detection station, the fault detection result of the railway running is determined, including: According to the historical driving fault detection data of each fault detection station, determine the fault detection result of the single-point detection station, the fault detection result of the same type of detection station and the fault detection result of all the detection stations; According to the fault detection result of the single-point detection station, the fault detection result of the same type of detection station and the fault detection result of all the detection stations, determine the comprehensive fault detection result of the railway running; Wherein, according to the fault detection results of the single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all the detection stations, the comprehensive fault detection results of the railway running are determined, including: Input the fault detection results of the single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all detection stations into the weighted comprehensive fault detection model to obtain the comprehensive fault detection results of the railway running. ; The weighted comprehensive fault detection model is used to assign different weight coefficients to the input fault detection results of the single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all the detection stations; Wherein, after obtaining the comprehensive fault detection result of the railway vehicle, it also includes: When the comprehensive fault detection result of the railway operation is greater than the preset fault level determination threshold, determine that the comprehensive fault detection result of the railway operation is a fault; When the comprehensive fault detection result of the railway operation is less than the preset fault level determination threshold, determine that the comprehensive fault detection result of the railway operation is a fault false alarm; Wherein, after it is determined that the comprehensive fault detection result of the railway running vehicle is a fault or a fault false alarm, the method further includes: According to the comprehensive fault detection result of the railway operation, the weight coefficient of the weighted comprehensive fault detection model is adjusted as follows:

in,

,

,

Fault detection results for the single point detection station

The weight coefficient of ,

Fault detection results for the same type of detection station

The weight coefficient of ,

Fault detection results for all the detection stations

The weight coefficient of ,

is a constant; or,

,

,

Coefficient set for the weighted comprehensive fault detection model in order to comprehensively balance the different weights of the single-point detection station fault detection results, the fault detection results of the same type of detection stations and the fault detection results of all detection stations in obtaining the comprehensive fault detection results ,

is a constant. 2. The railway running fault detection method according to claim 1, wherein, according to the historical running fault detection data of each fault detection station, determining the fault detection result of a single-point detection station, comprising: Calculate the fault detection result of the single-point detection station according to the following first formula

, the first formula is:

or,

in,

is a constant,

₁ is the number of historical driving fault detections of the single-point detection station,

is the first failure probability calculated according to the historical driving failure detection data of the single-point detection station,

is the default weight value. 3. The railway running fault detection method according to claim 1, wherein, according to the historical running fault detection data of each fault detection station, determine the fault detection result of the same type of detection station, comprising: Calculate the fault detection result of the same type of detection station according to the second formula below

, the second formula is:

or,

in,

is a constant,

₂ is the number of historical driving fault detections of the same type of detection station,

is the second fault probability calculated according to the historical driving fault detection data of the same type of detection station,

is the default weight value. 4. The railway running fault detection method according to claim 1, wherein, according to the historical running fault detection data of each fault detection station, the fault detection results of all detection stations are determined, comprising: Calculate the fault detection results of all detection stations according to the third formula below

, the third formula is:

or,

in,

is a constant,

₃ is the number of historical driving fault detections of all detection stations,

is the third failure probability calculated according to the historical driving failure detection data of all detection stations,

is the default weight value. 5. The railway running fault detection method according to claim 1, wherein the weighted comprehensive fault detection model S is:

in,

Fault detection results for the single point detection station

The weight coefficient of ,

Fault detection results for the same type of detection station

The weight coefficient of ,

Fault detection results for all the detection stations

The weight coefficient of ; or,

in,

is the weight coefficient,

is a constant. 6. A railway running fault detection device, characterized in that, comprising: The acquisition module is used to acquire the historical driving fault detection data of each fault detection station distributed along the railway line; a processing module, configured to determine the fault detection result of the railway running according to the historical running fault detection data of each fault detection station; Wherein, the processing module is specifically used for: According to the historical driving fault detection data of each fault detection station, determine the fault detection result of the single-point detection station, the fault detection result of the same type of detection station and the fault detection result of all the detection stations; According to the fault detection result of the single-point detection station, the fault detection result of the same type of detection station, and the fault detection results of all the detection stations, determine the comprehensive fault detection result of the railway running; Wherein, the processing module is also specifically used for: Input the fault detection results of the single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all detection stations into the weighted comprehensive fault detection model to obtain the comprehensive fault detection results of the railway running. ; The weighted comprehensive fault detection model is used to assign different weight coefficients to the input fault detection results of the single-point detection station, the fault detection results of the same type of detection stations, and the fault detection results of all the detection stations; Wherein, after obtaining the comprehensive fault detection result of the railway running, the processing module is also specifically used for: When the comprehensive fault detection result of the railway operation is greater than the preset fault level determination threshold, determine that the comprehensive fault detection result of the railway operation is a fault; When the comprehensive fault detection result of the railway operation is less than the preset fault level determination threshold, determine that the comprehensive fault detection result of the railway operation is a fault false alarm; Wherein, after determining that the comprehensive fault detection result of the railway train is a fault or a fault false alarm, the processing module is further specifically used for: According to the comprehensive fault detection result of the railway operation, the weight coefficient of the weighted comprehensive fault detection model is adjusted as follows:

in,

,

,

Fault detection results for the single point detection station

The weight coefficient of ,

Fault detection results for the same type of detection station

The weight coefficient of ,

Fault detection results for all the detection stations

The weight coefficient of ,

is a constant; or,

,

,

Coefficient set for the weighted comprehensive fault detection model in order to comprehensively balance the different weights of the single-point detection station fault detection results, the fault detection results of the same type of detection stations and the fault detection results of all detection stations in obtaining the comprehensive fault detection results ,

is a constant.

Images