A method, system, apparatus and medium for detecting a state of a rolling stock

Abstract

The application discloses a locomotive vehicle state detection method, system, device and medium, and relates to the field of online monitoring of rail transit locomotive vehicles. The method comprises the following steps: acquiring first data and second data collected by a vehicle-mounted terminal; the first data is data of a locomotive vehicle running state, and the second data is environmental data; inputting the first data and the second data into a pre-set joint evaluation model for representing a locomotive running state; and outputting a detection result of the locomotive vehicle state by the joint evaluation model. In the method, the locomotive vehicle data and the environmental data are subjected to joint evaluation by the joint evaluation model to realize detection of the locomotive vehicle state, automatic analysis of data and automatic information flow, and thus the state evaluation efficiency and accuracy are improved; the collected data comprises not only the locomotive vehicle data but also the environmental data, that is, the data used for detecting the locomotive vehicle state is relatively comprehensive; and the accuracy of the locomotive vehicle state detection is improved through joint judgment of multiple evaluation models.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of online monitoring of rail transit locomotives and vehicles, and in particular to a locomotive and vehicle state detection method, system, device and medium. BACKGROUND

[0002] With the continuous development of sensor technology and network communication technology, the perception and data transmission network of the locomotive has been greatly improved, and the data analysis capability of the on-board and ground computers has made great progress. On this basis, real-time data acquisition, data transmission and data analysis can be used to evaluate the running state of the locomotive in real time, make decisions and handle them in advance, prevent faults and accidents, notify relevant personnel in a timely manner after the fault occurs, realize the linkage of on-board and off-board, complete the rapid handling of the accident, and improve the operation and fault handling efficiency of the locomotive and vehicle to prevent further expansion of the fault.

[0003] Currently, the detection of the state of the locomotive and vehicle is mainly determined by the driver or the dispatcher according to experience, resulting in low accuracy of the detection of the state of the vehicle. When the detection of the state of the vehicle is misjudged, such as failure to identify serious problems such as locomotive breakdown, it is easy to cause traffic accidents.

[0004] Therefore, how to accurately detect the state of the locomotive and vehicle is a technical problem that needs to be solved by those skilled in the art. SUMMARY

[0005] The purpose of the present application is to provide a locomotive and vehicle state detection method, system, device and medium for accurately detecting the state of the locomotive and vehicle.

[0006] To solve the above technical problems, the present application provides a locomotive and vehicle state detection method applied to the ground end, which comprises:

[0007] acquiring first data and second data collected by the on-board end; wherein the first data is data for representing the running state of the locomotive and vehicle, and the second data is data for representing the environment in which the locomotive and vehicle is located;

[0008] inputting the first data and the second data into a pre-set joint evaluation model for representing the running state of the locomotive; wherein the joint evaluation model at least includes a correlation rule model, a Bayesian classification model and a classification model other than the Bayesian classification model;

[0009] outputting the detection result of the state of the locomotive and vehicle by the joint evaluation model.

[0010] Preferably, the inputting of the first data and the second data into the pre-set joint evaluation model for representing the running state of the locomotive comprises:

[0011] inputting the first data and the second data into the association rule model;

[0012] outputting a first detection result by the association rule model;

[0013] adding the first detection result into a variable set of the Bayesian classification model;

[0014] outputting a second detection result by the Bayesian classification model;

[0015] determining whether to output the second detection result by the classification model.

[0016] Preferably, the outputting a first detection result by the association rule model comprises:

[0017] matching the first data and the second data with a pre-set association rule library; the association rule library contains a corresponding relationship between the first data, the second data and the rolling stock state;

[0018] determining the first detection result of the rolling stock corresponding to the first data and the second data according to the association rule library;

[0019] Correspondingly, the outputting a second detection result by the Bayesian classification model comprises:

[0020] obtaining mutual information of the first detection result, the first data and the second data;

[0021] obtaining the first data and the second data less than or equal to a threshold value of the mutual information;

[0022] obtaining a probability of a state classification result by the Bayesian classification model and performing normalization processing;

[0023] outputting the state classification result with the maximum probability as the second detection result.

[0024] Preferably, after the outputting a detection result of the rolling stock state by the joint evaluation model, the method further comprises:

[0025] determining a fault level of the rolling stock state according to the detection result; the fault level is directly proportional to a fault handling speed;

[0026] starting from determining that the fault level is greater than or equal to a threshold value, pre-warning and / or outputting a first prompt information for the detection result corresponding to the fault level greater than or equal to the threshold value within a first pre-set time;

[0027] A second prompt information corresponding to the detection result when the fault level is less than the threshold is output within a second preset time since the fault level is determined to be less than the threshold.

[0028] Preferably, after the detection result of the locomotive vehicle state is output by the joint evaluation model, the method further comprises:

[0029] feeding back the detection result to a user so as to facilitate the user to detect and dispose the locomotive vehicle;

[0030] acquiring an actual detection result fed back by the user;

[0031] updating the joint evaluation model according to the actual detection result.

[0032] Preferably, the method comprises:

[0033] collecting first data and second data; wherein the first data is data for representing the running state of the locomotive vehicle, and the second data is data for representing the environment in which the locomotive vehicle is located;

[0034] inputting the first data and the second data into a pre-set joint evaluation model for representing the running state of the locomotive; wherein the joint evaluation model at least comprises an association rule type model, a Bayesian classification model, and a classification model other than the Bayesian classification model;

[0035] outputting a detection result of the locomotive vehicle state by the joint evaluation model.

[0036] To solve the above technical problems, the application also provides a detection system for a locomotive vehicle state, which comprises: a vehicle-mounted evaluation system, a ground evaluation system, a vehicle-mounted human-computer interaction interface, and a ground human-computer interaction interface.

[0037] The vehicle-mounted evaluation system is connected with the ground evaluation system.

[0038] The vehicle-mounted evaluation system is used to collect first data and second data; wherein the first data is data for representing the running state of the locomotive vehicle, and the second data is data for representing the environment in which the locomotive vehicle is located; the first data and the second data are inputted into a pre-set joint evaluation model for representing the running state of the locomotive; wherein the joint evaluation model at least comprises an association rule type model, a Bayesian classification model, and a classification model other than the Bayesian classification model; and a detection result of the locomotive vehicle state is output by the joint evaluation model.

[0039] The vehicle-mounted human-computer interaction interface is connected with the vehicle-mounted evaluation system and is used to receive the detection result.

[0040] The ground evaluation system is configured to acquire the first data and the second data collected by the vehicle-mounted terminal, input the first data and the second data into the joint evaluation model pre-set for representing the running state of the locomotive, and output the detection result of the state of the locomotive vehicle through the evaluation model.

[0041] To solve the above technical problems, the application further provides a detection device for the state of a locomotive vehicle, which is applied to a ground terminal and comprises:

[0042] An acquisition module is configured to acquire first data and second data collected by a vehicle-mounted terminal, wherein the first data is data for representing the running state of the locomotive vehicle, and the second data is data for representing the environment in which the locomotive vehicle is located;

[0043] An input module is configured to input the first data and the second data into a joint evaluation model pre-set for representing the running state of the locomotive, wherein the joint evaluation model at least comprises a correlation rule model, a Bayesian classification model, and a classification model other than the Bayesian classification model.

[0044] An output module is configured to output a detection result of the state of the locomotive vehicle through the joint evaluation model.

[0045] To solve the above technical problems, the application further provides a detection device for the state of a locomotive vehicle, which comprises:

[0046] A memory is configured to store a computer program.

[0047] A processor is configured to execute the computer program to realize the steps of the detection method for the state of a locomotive vehicle.

[0048] To solve the above technical problems, the application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the steps of the detection method for the state of a locomotive vehicle.

[0049] The method for detecting the state of a locomotive vehicle provided in the application is applied to a ground terminal, and the method comprises the following steps: acquiring first data and second data collected by a vehicle terminal; the first data is data for representing the running state of the locomotive vehicle, and the second data is data for representing the environment in which the locomotive vehicle is located; inputting the first data and the second data into a pre-set joint evaluation model for representing the running state of the locomotive; the joint evaluation model at least comprises a correlation rule model, a Bayesian classification model and a classification model other than the Bayesian classification model; and outputting the detection result of the state of the locomotive vehicle through the joint evaluation model. Compared with the previous method for judging the state of the locomotive vehicle according to the experience of a driver or a dispatcher, in the method of the application, the locomotive vehicle data and the environmental data are subjected to the joint evaluation model to realize the detection of the state of the locomotive vehicle, the automatic analysis of the data and the automatic flow of information are realized, the state evaluation efficiency and the accuracy are improved, the collected data comprises not only the locomotive vehicle data but also the environmental data, that is, the data for detecting the state of the locomotive vehicle is relatively comprehensive, the joint judgment of the multiple evaluation models is realized, and the accuracy of the detection of the state of the locomotive vehicle is improved.

[0050] In addition, the application further provides a method for detecting the state of a locomotive vehicle applied to a vehicle terminal, a system for detecting the state of a locomotive vehicle, a device and a computer readable storage medium, which have the same or corresponding technical features and effects as the above-mentioned method for detecting the state of a locomotive vehicle. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the application, the drawings required in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0052] Figure 1 A flow chart of a method for detecting the state of a locomotive vehicle applied to a ground terminal is provided for the embodiments of the application.

[0053] Figure 2 A flow chart of a joint evaluation model running method is provided for the embodiments of the application.

[0054] Figure 3 A schematic diagram of a system for detecting the state of a locomotive vehicle is provided for the embodiments of the application.

[0055] Figure 4 A structure diagram of a device for detecting the state of a locomotive vehicle applied to a ground terminal is provided for an embodiment of the application.

[0056] Figure 5A structural diagram of a detection device of a locomotive vehicle state provided for another embodiment of the present application is shown in the figure;

[0057] Figure 6 A flowchart of a method for detecting a locomotive vehicle state provided for an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, any other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0059] The core of the present application is to provide a detection method, system, device and medium for a locomotive vehicle state, which is used to detect the locomotive vehicle state more accurately.

[0060] The vehicle-mounted terminal is usually provided with sensors to collect data of the locomotive vehicle running, such as collecting the running speed of the vehicle through a speed sensor, and collecting road conditions and other information through an image collection device. The locomotive vehicle is detected according to the data collected by the vehicle-mounted terminal. The decision and disposal are made in advance to prevent faults and accidents, and relevant personnel are notified in time after the fault occurs, so as to realize the linkage of on-board and off-board, complete the rapid disposal of the accident, improve the operation and fault handling efficiency of the locomotive vehicle, and prevent the further expansion of the fault.

[0061] In order for the person skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Figure 1 A flowchart of a detection method for a locomotive vehicle state applied to a ground terminal provided for an embodiment of the present application is shown in the figure, which includes the following steps. Figure 1

[0062] S10: acquiring first data and second data collected by the vehicle-mounted terminal.

[0063] The first data is data for representing the running state of the locomotive vehicle, such as the speed of the vehicle, the driving time of the vehicle, etc. The second data is data for representing the environment where the locomotive vehicle is located, such as the road condition where the vehicle is located, the display of the signal light in the current scene where the vehicle is located, etc. In the embodiments of the present application, the set of the first data and the second data, i.e., the locomotive vehicle state and the environment state, is represented as X i ​The first data and the second data are input into a pre-set joint evaluation model for representing the running state of the locomotive.

[0064] S11: inputting the first data and the second data into a pre-set joint evaluation model for representing the running state of the locomotive.

[0065] S12: outputting the detection result of the state of the locomotive vehicle through the joint evaluation model.

[0066] In order to improve the accuracy of the detection of the locomotive vehicle, the first data and the second data are input into the joint evaluation model, wherein the joint evaluation model at least includes an association rule model, a Bayesian classification model, and a classification model other than the Bayesian classification model. The classification model other than the Bayesian classification model is, for example, a neural network model. The order of the joint evaluation model is not limited.

[0067] The prepared data is taken as input data, and the evaluation model is run in a set order to generate an evaluation result. The evaluation result is transmitted in an XML or JSON format and stored in a database.

[0068] The method for detecting the state of the locomotive vehicle provided in the embodiment is applied to the ground end, and the method comprises the following steps: acquiring first data and second data collected by a vehicle-mounted end; the first data is data for representing the running state of the locomotive vehicle, and the second data is data for representing the environment in which the locomotive vehicle is located; inputting the first data and the second data into a pre-set joint evaluation model for representing the running state of the locomotive; the joint evaluation model at least includes an association rule model, a Bayesian classification model, and a classification model other than the Bayesian classification model; and outputting the detection result of the state of the locomotive vehicle through the joint evaluation model. Compared with the previous method for judging the state of the locomotive vehicle according to the experience of the driver or the dispatcher, in the method of the embodiment, the data of the locomotive vehicle and the environmental data are subjected to the detection of the state of the locomotive vehicle through the joint evaluation model, the automatic analysis of the data and the automatic flow of information are realized, the efficiency and the accuracy of the state evaluation are improved, the collected data includes not only the data of the locomotive vehicle but also the environmental data, that is, the data for detecting the state of the locomotive vehicle is relatively comprehensive, the accuracy of the detection of the state of the locomotive vehicle is improved through the joint judgment of multiple evaluation models.

[0069] In order to detect the state of the locomotive vehicle, the evaluation model is run in a set order in the embodiment, and the preferred implementation manner is that the first data and the second data are input into a pre-set joint evaluation model for representing the running state of the locomotive.

[0070] inputting the first data and the second data into the association rule model;

[0071] outputting the first detection result through the association rule model;

[0072] adding the first detection result into the variable set of the Bayesian classification model;

[0073] outputting the second detection result through the Bayesian classification model;

[0074] determining whether to output the second detection result through the classification model.

[0075] Figure 2 A flowchart of a joint evaluation model running method is provided for the embodiments of the present application. As shown in the flowchart, Figure 2 the method comprises:

[0076] S13: data preparation;

[0077] S14: running the association rule model; entering step S15 and entering step S17;

[0078] S15: running the Bayesian classifier model; entering step S16 and entering step S17;

[0079] S16: running other classifier models;

[0080] S17: recording the evaluation result;

[0081] S18: comprehensive evaluation result;

[0082] S19: outputting the evaluation result.

[0083] Specifically, the joint evaluation model running method is as follows:

[0084] a. After the data preparation is completed, the association rule model is preferentially run, including the association rules extracted according to the expert knowledge and the association rules generated by applying the association rule algorithm. For the association rules extracted according to the expert knowledge, the confidence of the rules is given according to the expert knowledge, and the confidence is adjusted according to the evaluation result feedback in the application process. The support degree limit is ignored when the association rules are generated. For the association rules generated by the algorithm, the support degree and the confidence of the rules are recalculated according to the evaluation result feedback in the application process, and the rules are updated.

[0085] By writing the expert knowledge in the form of association rules, the problem of not being able to identify useful patterns due to unbalanced data samples is solved. Through the iteration of the association rules, the problem of not being able to dynamically update the expert knowledge is solved.

[0086] The support degree Support(X, Y) calculation formula is as follows:

[0087]

[0088] Wherein, X represents the collected data, Y represents the detection result.

[0089] The above confidence formula is:

[0090]

[0091] b. Run the Bayesian classification model. For the rule Y→X in the association rule base, add Y to the Bayesian classification model variable set and delete variable X.

[0092] Calculate the mutual information between the result state and the variable, set the threshold of mutual information, for the variable whose mutual information does not exceed the threshold, query whether the Bayesian classification model is applied in the association rule base, calculate the probability of state classification result, normalize, output the state classification result with the largest probability, and record the result of each state classification.

[0093] The above mutual information I(X;Y) calculation formula is:

[0094]

[0095] The above Bayesian classification model (taking the naive Bayesian model as an example) state classification probability calculation formula is:

[0096]

[0097] The normalized probability is:

[0098]

[0099] The final classification result is:

[0100]

[0101] In the system use process, the verified evaluation result is recorded as a data sample to correct the above probability parameter calculation. When the computing resources are sufficient, the above calculation process can be calculated in real time after the sample is updated. When the computing resources are limited, the calculation can be updated regularly after enough samples are updated.

[0102] For the evaluation result evaluation continuous performance good model parameter, can through the form of expert review, the evaluation model is included in the association rule model based on expert knowledge.

[0103] c. Run other classification models, such as neural network model, etc.

[0104] The evaluation model provided by the embodiment is run through multiple evaluation models and in a predetermined order, in which the association rule type model is run first, then the Bayesian classification model is run, and finally the other classification model is run. The results of the association rule type model are used to replace the variables in the Bayesian classification model, which can reduce the noise when using the Bayesian classification model. Since the reliability of the Bayesian classification model is higher than that of other classification models, the Bayesian classification model is run first, and then the other classification model is run, which improves the accuracy of the detection of the state of the locomotive vehicle.

[0105] In practice, when using the evaluation model for state detection, the preferred embodiment is that the first detection result output by the association rule type model includes:

[0106] The first data and the second data are matched with a pre-set association rule library; the association rule library contains the corresponding relationship between the first data, the second data and the state of the locomotive vehicle;

[0107] According to the association rule library, the first detection result of the locomotive vehicle corresponding to the first data and the second data is determined;

[0108] Correspondingly, the second detection result output by the Bayesian classification model includes:

[0109] The mutual information of the first detection result, the first data and the second data is obtained;

[0110] The first data and the second data less than or equal to the threshold of the mutual information are obtained;

[0111] The probability of the state classification result obtained by the Bayesian classification model is normalized;

[0112] The state classification result with the maximum probability is output as the second detection result.

[0113] Mechanical breakdown is a typical fault state of a locomotive, which refers to a non-human accident caused by mechanical or electrical failure of the locomotive during operation after leaving the depot. The following takes the mechanical breakdown state evaluation as an example to illustrate the specific embodiment.

[0114] (1) Association rule library

[0115] a. Determine the association rule library about the mechanical breakdown state through expert review;

[0116] {X1→y 机破 ,..., X n →y 机破}

[0117] Where X i is a set of states of the locomotive vehicle and environmental states, such as X i= {speed = 0, signal light = green light, traction state = traction lock,...}. For such association rules, only the confidence is calculated after updating the sample according to the feedback data, and when the confidence is below the threshold, the rule is not enabled, and when the confidence is above the threshold, the rule is re-enabled.

[0118] b. Determine the association rules generated by the algorithm through calculation on the sample data;

[0119] When Support(X, y 机破 )>a and Confidence(X, y 机破 )>b, X→y 机破 is added to the association rule library. In the example, a>0.5 and b>0.7.

[0120] (2) Evaluation process

[0121] a. Data preparation, obtain the current vehicle and environmental variable state.

[0122] b. Match the association rule library, if successful, output the machine breakage warning.

[0123] c. Calculate the mutual information of the variable and the machine breakage state through the calculation on the sample data, when I(X i ; y 机破 )>theta, X i is added to the variable set X, and in the example, theta is the average of all mutual information.

[0124] Calculate

[0125]

[0126] Then normalize the probability, and the denominator calculation can be reduced and does not need to be calculated.

[0127] According to the current variable value, when P(y 机破 |x1,..., x n )>P(y 未机破 |x1,..., x n ), output the machine breakage warning.

[0128] d. Run other models (such as neural network models) to determine whether to output the machine breakage warning.

[0129] The specific implementation method of the evaluation model provided in the embodiment realizes the detection of the state of the locomotive vehicle.

[0130] On the basis of the detection result of the locomotive vehicle obtained in the above embodiment, in order to facilitate the user to timely handle the fault of the locomotive vehicle, the preferred embodiment is that after the detection result of the state of the locomotive vehicle is output by the joint evaluation model, the detection method of the state of the locomotive vehicle further comprises:

[0131] determining the fault level of the state of the locomotive vehicle according to the detection result; the fault level is directly proportional to the fault to be handled speed;

[0132] starting from determining that the fault level is greater than or equal to the threshold value, prewarning and / or outputting the first prompt information for the detection result corresponding to the fault level greater than or equal to the threshold value within the first preset time;

[0133] starting from determining that the fault level is less than the threshold value, outputting the second prompt information for the detection result corresponding to the fault level less than the threshold value within the second preset time.

[0134] The fault level corresponding to the detection result is not limited, such as the machine breaking, which belongs to the fault with a relatively high fault level, and therefore the machine breaking fault needs to be handled in time. The threshold value, the first preset time, the second preset time, the first prompt information, the second prompt information and the like are not limited and are determined according to the actual situation. It should be noted that in order to facilitate the user to timely understand the detection result of the locomotive, the first preset time and the second preset time cannot be too long; in order to facilitate the user to understand the severity of the fault according to the prompt information, the first prompt information and the second prompt information need to adopt different prompt methods, contents and the like. For example, the first prompt information can be prompted by alarm, and the second prompt information can be output on the display screen only.

[0135] After step d of the above embodiment, the following steps are further included:

[0136] e. If there is a machine breaking warning, enter the machine breaking disposal plan program, push the message to the relevant personnel, track the processing progress. And record the machine breaking evaluation verification result this time, and add the result to the sample set.

[0137] f. If there is no machine breaking warning, enter the next determination.

[0138] g. For the situation that the machine breaking occurs, but the evaluation system does not generate the machine breaking warning, record the data and add it to the sample set.

[0139] The embodiment provided in the present embodiment provides prompt through the detection result, so that the user can timely understand the detection result and remind the user to handle the abnormal detection result in time.

[0140] In order to improve the accuracy of the evaluation model, the preferred embodiment is that after the detection result of the state of the locomotive vehicle is output by the joint evaluation model, the detection method of the state of the locomotive vehicle further comprises:

[0141] feedback the detection result to the user so as to facilitate the user to detect and dispose the rolling stock;

[0142] acquire the actual detection result of the user feedback;

[0143] update the joint evaluation model according to the actual detection result.

[0144] In the method provided by the embodiment, the evaluation model is updated in combination with the evaluation result of the evaluation model and the actual detection result of the user, so that the evaluation model adopted has stronger robustness.

[0145] The above describes the detection method of the rolling stock state applied to the ground end, and the embodiment further provides a detection method of the rolling stock state applied to the vehicle-mounted end, which comprises:

[0146] acquiring the first data and the second data; wherein the first data is data for representing the running state of the rolling stock, and the second data is data for representing the environment in which the rolling stock is located;

[0147] inputting the first data and the second data into a pre-set joint evaluation model for representing the rolling state; wherein the joint evaluation model at least comprises an association rule model, a Bayesian classification model and a classification model other than the Bayesian classification model;

[0148] outputting the detection result of the rolling stock state by the joint evaluation model.

[0149] The detection method of the rolling stock state applied to the vehicle-mounted end provided by the embodiment has the same technical features as the detection method of the rolling stock state applied to the ground end mentioned above, and the embodiment of the detection method of the rolling stock state applied to the ground end has been described in detail above. Therefore, the embodiment of the detection method of the rolling stock state applied to the vehicle-mounted end will not be described herein again, and has the same beneficial effects as the detection method of the rolling stock state applied to the ground end mentioned above.

[0150] The embodiment further provides a detection system of a rolling stock state. Figure 3 A schematic diagram of a detection system of a rolling stock state provided by the embodiment of the application is shown in FIG. 1, which comprises a vehicle-mounted evaluation system 1, a ground evaluation system 2, a vehicle-mounted human-computer interaction interface 3 and a ground human-computer interaction interface 4. Figure 3

[0151] The vehicle-mounted evaluation system 1 is connected with the ground evaluation system 2.

[0152] ​The vehicle-mounted evaluation system 1 is configured to collect first data and second data, wherein the first data is data for representing the running state of the locomotive vehicle, and the second data is data for representing the environment in which the locomotive vehicle is located; the first data and the second data are input into a pre-set joint evaluation model for representing the running state of the locomotive; wherein the joint evaluation model at least includes a correlation rule model, a Bayesian classification model, and a classification model other than the Bayesian classification model; and the joint evaluation model outputs a detection result of the running state of the locomotive vehicle.

[0153] The vehicle-mounted human-computer interaction interface 3 is connected with the vehicle-mounted evaluation system 1 and is configured to receive the detection result.

[0154] The ground evaluation system 2 is configured to acquire the first data and the second data collected by the vehicle-mounted end; input the first data and the second data into a pre-set joint evaluation model for representing the running state of the locomotive; and output a detection result of the running state of the locomotive vehicle through the evaluation model. The ground human-computer interaction interface 4 is connected with the ground evaluation system 2 and is configured to receive the detection result.

[0155] The system provided in the embodiment is composed of a vehicle-mounted part and a ground part. The vehicle-mounted part is mainly configured to complete the collection and processing of vehicle-mounted data, and the ground part is responsible for summarizing all related data and running a model to automatically evaluate the locomotive vehicle. The vehicle-mounted part and the ground part both provide human-computer interaction interfaces, which are configured to show the evaluation result to a user and receive feedback on the evaluation result.

[0156] The system has the following characteristics:

[0157] a. The vehicle-mounted evaluation system is configured to complete the collection of vehicle-mounted data and run an evaluation model with low resource consumption and high certainty to output an evaluation result.

[0158] b. The ground evaluation system is configured to complete the collection of vehicle-mounted data and environmental data and run all evaluation models to output an evaluation result. The ground evaluation system can synchronize a verified model to the vehicle-mounted evaluation system for distributed running.

[0159] c. The vehicle-mounted human-computer interaction interface is configured to face vehicle-mounted personnel and complete the display and feedback of the evaluation result of the vehicle-mounted evaluation system and the ground evaluation system.

[0160] d. The ground human-computer interaction interface is configured to face ground personnel and complete the display and feedback of the evaluation result of the vehicle-mounted evaluation system and the ground evaluation system.

[0161] The locomotive vehicle state detection system provided in the embodiment has the same beneficial effects as the locomotive vehicle state detection method described above.

[0162] In the above embodiments, the detection method of the rolling stock state is described in detail, and the application also provides corresponding embodiments of the detection device of the rolling stock state. It should be noted that the embodiments of the device part are described from two angles, one is based on the functional module angle, and the other is based on the hardware angle.

[0163] Figure 4 A structural diagram of a detection device of a rolling stock state applied to a ground end is provided for an embodiment of the application. The embodiment is based on the functional module angle and includes:

[0164] The acquisition module 10 is configured to acquire first data and second data collected by the on-board end, wherein the first data is data for representing the running state of the rolling stock, and the second data is data for representing the environment in which the rolling stock is located;

[0165] The input module 11 is configured to input the first data and the second data into a pre-set joint evaluation model for representing the running state of the rolling stock, wherein the joint evaluation model at least includes an association rule model, a Bayesian classification model, and a classification model other than the Bayesian classification model;

[0166] The output module 12 is configured to output the detection result of the rolling stock state through the joint evaluation model.

[0167] Since the embodiments of the device part correspond to the embodiments of the method part, the embodiments of the device part are described in the description of the embodiments of the method part, which will not be described here. And has the same beneficial effects as the above-mentioned detection method of the rolling stock state applied to the ground end.

[0168] Figure 5 A structural diagram of a detection device of a rolling stock state is provided for another embodiment of the application. The embodiment is based on the hardware angle, as shown in Figure 5 The detection device of the rolling stock state includes:

[0169] The memory 20 is configured to store a computer program;

[0170] The processor 21 is configured to execute the computer program to realize the steps of the detection method of the rolling stock state mentioned in the above embodiments.

[0171] The detection device of the rolling stock state provided in the embodiment can include but is not limited to a smart phone, a tablet computer, a notebook computer, or a desktop computer, etc.

[0172] The processor 21 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 21 may be implemented using at least one of the following hardware forms: Digital Signal Processor (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 21 may also include a main processor and a coprocessor. The main processor, also known as the Central Processing Unit (CPU), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 21 may integrate a Graphics Processing Unit (GPU), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 21 may also include an Artificial Intelligence (AI) processor, which is used to handle computational operations related to machine learning.

[0173] The memory 20 may include one or more computer-readable storage media, which may be non-transitory. The memory 20 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 20 is used to store at least the following computer program 201, which, after being loaded and executed by the processor 21, is capable of implementing the relevant steps of the locomotive and rolling stock status detection method disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, and the storage method may be temporary or permanent storage. The operating system 202 may include Windows, Unix, Linux, etc. The data 203 may include, but is not limited to, the data involved in the locomotive and rolling stock status detection method mentioned above.

[0174] In some embodiments, the vehicle status detection device may further include a display screen 22, an input / output interface 23, a communication interface 24, a power supply 25, and a communication bus 26.

[0175] Those skilled in the art will understand that Figure 5 The structure shown does not constitute a limitation on the detection device for the condition of locomotives and rolling stock, and may include more or fewer components than shown.

[0176] The locomotive vehicle state detection device provided by the embodiment of the present application comprises a memory and a processor, and the processor can realize the following method when executing the program stored in the memory: the locomotive vehicle state detection method, and the effects are the same as above.

[0177] The present application also provides an embodiment corresponding to a computer readable storage medium. The computer readable storage medium stores a computer program, and the computer program is executed by a processor to realize the steps recorded in the above method embodiment.

[0178] It can be understood that if the method in the above embodiment is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and executes all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0179] The computer readable storage medium provided by the present application includes the above-mentioned locomotive vehicle state detection method, and the effects are the same as above.

[0180] In order to enable personnel in the technical field to better understand the present application scheme, the present application will be further described in detail in combination with the accompanying drawings and specific embodiments. Figure 6 and specific embodiments. Figure 6 The flowchart of the method for detecting the state of the locomotive vehicle provided by the embodiment of the present application is shown in FIG. 1, which comprises the following steps. Figure 6

[0181] S20: Locomotive data and environment data preparation;

[0182] S21: Running evaluation model library model;

[0183] S22: Generating evaluation results;

[0184] S23: Judging whether it is a serious alarm such as a broken locomotive that needs to be immediately processed; if yes, entering step S24; if no, entering step S26;

[0185] S24: Simultaneously alarming the vehicle-mounted and ground man-machine interaction interface, and matching and executing the emergency plan;

[0186] ​S25: determining whether the treatment is completed; if not, returning to step S24; if yes, entering step S27;

[0187] S26: outputting prompt information;

[0188] S27: verifying the evaluation result;

[0189] S28: recording and returning to step S21.

[0190] Specifically, a. complete data preparation. Organize the vehicle-mounted data and environmental data to form a data set available for the model.

[0191] b. Running the evaluation model in a set order with the prepared data as input data to generate an evaluation result. The evaluation result is delivered in XML or JSON format and stored in a database.

[0192] c. For serious alarms such as machine breaking in the evaluation result, which need to be handled immediately, matching the handling plan in the system and pushing to the relevant personnel through the human-computer interaction interface to enter the handling guidance process.

[0193] d. For other evaluation results, output prompt information and push to the relevant personnel through the human-computer interaction interface.

[0194] e. The user checks and handles the locomotive vehicle according to the evaluation result and feeds back the system evaluation result and handling situation through the human-computer interaction interface. The system records the feedback result.

[0195] f. According to the recorded actual feedback result, the system regularly evaluates and corrects the association rule model and the Bayesian classification model and updates the self-learning ability of the neural network type model.

[0196] In this embodiment, through automatic data collection, transmission and analysis, automatic evaluation of the state is realized, the evaluation accuracy is improved through comprehensive analysis of the data; the state evaluation efficiency and decision-making processing efficiency are improved through automatic analysis of the data and automatic flow of information, abnormal state expansion is prevented, and thus the locomotive vehicle operation efficiency is improved; the evaluation result can be sent to the relevant positions in real time to realize efficient linkage of multiple positions.

[0197] The above introduces in detail a locomotive vehicle state detection method, system, device and medium provided by the present application. The embodiments in the specification are described in a progressive manner, and each embodiment mainly explains the difference from other embodiments. The same or similar parts of each embodiment can be understood by referring to each other. For the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the related parts can be understood by referring to the method part. It should be pointed out that for ordinary technical personnel in the technical field, some improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

[0198] It should also be noted that in the present specification, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the existence of other identical elements in the process, method, article or device including the element.

Claims

1. A method of detecting a state of a rolling stock, characterized by, Applied to the ground end, the method comprises: Obtaining first data and second data collected by a vehicle-mounted end; wherein the first data is data for representing a running state of the locomotive vehicle, and the second data is data for representing an environment in which the locomotive vehicle is located; Inputting the first data and the second data into a pre-set joint evaluation model for representing a running state of a locomotive; wherein the joint evaluation model at least comprises an association rule type model, a Bayesian classification model, and a classification model other than the Bayesian classification model; Outputting a detection result of the locomotive vehicle state by the joint evaluation model; The inputting the first data and the second data into the pre-set joint evaluation model for representing a running state of a locomotive comprises: Inputting the first data and the second data into the association rule type model; Outputting a first detection result by the association rule type model; Adding the first detection result into a variable set of the Bayesian classification model; Outputting a second detection result by the Bayesian classification model; Determining whether to output the second detection result by the classification model other than the Bayesian classification model; The outputting a first detection result by the association rule type model comprises: Matching the first data, the second data with a pre-set association rule library; the association rule library contains a corresponding relationship between the first data, the second data and the locomotive vehicle state; Determining the first detection result of the locomotive vehicle corresponding to the first data and the second data according to the association rule library; Correspondingly, the outputting a second detection result by the Bayesian classification model comprises: Obtaining mutual information of the first detection result, the first data and the second data; Setting a threshold of mutual information, for variables whose mutual information does not exceed the threshold of mutual information, querying whether there is an application of the Bayesian classification model in the association rule library, calculating a probability of a state classification result, normalizing, and outputting a state classification result with the largest probability as the second detection result; the variables are the first data and the second data whose mutual information does not exceed the threshold of mutual information.

2. The method of detecting a state of a rolling stock according to claim 1, characterized by, After the outputting a detection result of the locomotive vehicle state by the joint evaluation model, the method further comprises: Determining a fault level of the locomotive vehicle state according to the detection result; the fault level is directly proportional to a fault handling speed; Starting from determining that the fault level is greater than or equal to a threshold, pre-warning and / or outputting a first prompt information for the detection result corresponding to the fault level greater than or equal to the threshold within a first pre-set time; Starting from determining that the fault level is less than the threshold, outputting a second prompt information for the detection result corresponding to the fault level less than the threshold within a second pre-set time.

3. The method of detecting a state of a rolling stock according to claim 1 or 2, characterized by, After the outputting a detection result of the locomotive vehicle state by the joint evaluation model, the method further comprises: Feeding back the detection result to a user, so as to facilitate the user to detect and dispose the locomotive vehicle; Obtaining an actual detection result fed back by the user; The joint evaluation model is updated based on the actual test results.

4. A method of detecting a state of a rolling stock, characterized by, Applied to vehicle-mounted devices, the method includes: Collect first data and second data; wherein, the first data is data used to characterize the operating status of the locomotive and rolling stock, and the second data is data used to characterize the environment in which the locomotive and rolling stock are located; The first data and the second data are input into a pre-set joint evaluation model for characterizing the locomotive's operating status; wherein, the joint evaluation model includes at least an association rule class model, a Bayesian classification model, and a classification model other than the Bayesian classification model; The joint evaluation model outputs the detection results of the locomotive and rolling stock status; Inputting the first data and the second data into a pre-set joint evaluation model for characterizing the locomotive's operating status includes: Input the first data and the second data into the association rule class model; The first detection result is output through the association rule class model; The first detection result is added to the variable set of the Bayesian classification model; The second detection result is output through the Bayesian classification model; Whether to output the second detection result is determined by a classification model other than the Bayesian classification model; The first detection result output by the association rule class model includes: The first data and the second data are matched with a pre-set association rule base; the association rule base contains the correspondence between the first data, the second data and the locomotive and rolling stock status; The first detection result of the locomotive and vehicle corresponding to the first data and the second data is determined according to the association rule base; Correspondingly, the output of the second detection result through the Bayesian classification model includes: Obtain the mutual information between the first detection result and the first data and the second data; A threshold for mutual information is set. For variables whose mutual information does not exceed the threshold, it is queried whether a Bayesian classification model exists in the association rule base. The probability of the state classification result is calculated and normalized. The state classification result with the highest probability is output as the second detection result. The variables are the first data and the second data whose mutual information does not exceed the threshold.

5. A system for detecting a state of a rolling stock, characterized by comprising: The system includes: an on-board assessment system, a ground assessment system, an on-board human-machine interface, and a ground human-machine interface; The vehicle-mounted assessment system is connected to the ground-based assessment system; The on-board assessment system is used to collect first data and second data; wherein, the first data is data used to characterize the operating status of the locomotive and rolling stock, and the second data is data used to characterize the environment in which the locomotive and rolling stock are located; the first data and the second data are input into a pre-set joint assessment model used to characterize the operating status of the locomotive; wherein, the joint assessment model includes at least an association rule class model, a Bayesian classification model, and a classification model other than the Bayesian classification model; the joint assessment model outputs the detection result of the locomotive and rolling stock status; The vehicle-mounted human-machine interface is connected to the vehicle-mounted evaluation system and is used to receive the detection results; The ground evaluation system is configured to acquire the first data and the second data collected by the vehicle-mounted terminal, input the first data and the second data into the pre-set joint evaluation model for representing the running state of the locomotive, and output the detection result of the state of the locomotive vehicle through the evaluation model. The inputting of the first data and the second data into the pre-set joint evaluation model for representing the running state of the locomotive comprises: inputting the first data and the second data into the association rule type model; outputting a first detection result through the association rule type model; adding the first detection result into a variable set of the Bayesian classification model; outputting a second detection result through the Bayesian classification model; determining whether to output the second detection result through a classification model other than the Bayesian classification model. The outputting of the first detection result through the association rule type model comprises: matching the first data and the second data with a pre-set association rule library; the association rule library comprises a corresponding relationship between the first data, the second data and the state of the locomotive vehicle; determining the first detection result of the locomotive vehicle corresponding to the first data and the second data according to the association rule library. Correspondingly, the outputting of the second detection result through the Bayesian classification model comprises: acquiring mutual information of the first detection result, the first data and the second data; setting a threshold of mutual information, querying whether the Bayesian classification model exists in the association rule library for a variable whose mutual information does not exceed the threshold of mutual information, calculating a probability of a state classification result, normalizing, and outputting a state classification result with the largest probability as the second detection result; the variable is the first data and the second data whose mutual information does not exceed the threshold of mutual information.

6. A device for detecting a state of a rolling stock, characterized by comprising: The device is applied to the ground terminal, and comprises: an acquisition module configured to acquire first data and second data collected by a vehicle-mounted terminal; the first data is used for representing a running state of the locomotive vehicle, and the second data is used for representing an environment in which the locomotive vehicle is located; an input module configured to input the first data and the second data into a pre-set joint evaluation model for representing the running state of the locomotive; the joint evaluation model at least comprises an association rule type model, a Bayesian classification model, and a classification model other than the Bayesian classification model; an output module configured to output a detection result of the state of the locomotive vehicle through the joint evaluation model. The input module is specifically configured to: input the first data and the second data into the association rule type model; output a first detection result through the association rule type model; add the first detection result into a variable set of the Bayesian classification model; output a second detection result through the Bayesian classification model; determine whether to output the second detection result through a classification model other than the Bayesian classification model. The first detection result output by the association rule model comprises: matching the first data and the second data with a pre-set association rule library; the association rule library contains a corresponding relationship between the first data, the second data and the locomotive vehicle state; determining the first detection result of the locomotive vehicle corresponding to the first data and the second data according to the association rule library; correspondingly, the second detection result output by the Bayesian classification model comprises: obtaining mutual information of the first detection result, the first data and the second data; setting a threshold of mutual information, querying whether the Bayesian classification model exists in the association rule library for variables whose mutual information does not exceed the threshold of mutual information, calculating a probability of a state classification result, normalizing, and outputting a state classification result with the largest probability as the second detection result; the variables are the first data and the second data whose mutual information does not exceed the threshold of mutual information.

7. A device for detecting a state of a rolling stock, characterized by comprising: comprise: a memory for storing a computer program; a processor for executing the computer program to realize the steps of the locomotive vehicle state detection method according to any one of claims 1 to 4.

8. A computer-readable storage medium, characterized in that, The computer program is stored on the computer readable storage medium and is executed by the processor to realize the steps of the locomotive vehicle state detection method according to any one of claims 1 to 4.

Images