Track circuit fault diagnosis model training method and device, fault diagnosis method and device, equipment and medium

Abstract

The invention discloses a training method of a fault diagnosis model of a track circuit, a fault diagnosis method and device, equipment and a medium, and the method comprises the steps: constructing a circuit fault tree according to a system structure diagram corresponding to a target track circuit; determining potential fault types included in the target track circuit and importance information of each potential fault type in the circuit fault tree according to the circuit fault tree; determining a high-risk fault type from the potential fault types according to the importance information, and collecting real-time working parameters of electrical equipment in the target track circuit when a high-risk fault event occurs in the target track circuit; and constructing model training data according to the real-time working parameters, and training a to-be-trained model by adopting the model training data to generate a fault diagnosis model. According to the invention, the high-risk fault type in the track circuit is automatically diagnosed by using the trained fault diagnosis model without manual diagnosis, so that the accuracy and timeliness of track circuit fault diagnosis are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a training method, a fault diagnosis method, a device, a equipment and a medium for a track circuit fault diagnosis model. Background Art

[0002] Railway transportation is related to my country's economic and social development, has very important strategic significance, and plays a pivotal role in my country's transportation industry. The railway signal system is an indispensable and important part of the modern railway system, responsible for the information transmission and dispatching instructions of various driving equipment. The track circuit is a key equipment in the entire railway signal system, which ensures the efficient and safe operation of the train. However, since the outdoor equipment of the track circuit is placed outdoors all year round, it is easily affected by bad weather and environment, which leads to the occurrence of fault problems.

[0003] Due to the complex structure of track circuits and the harsh outdoor working environment, the current diagnosis of track circuit faults is still based on the experience of staff. However, relying solely on manual diagnosis cannot accurately and timely diagnose track circuit faults. Summary of the invention

[0004] The present invention provides a training method for a fault diagnosis model of a track circuit, a fault diagnosis method, a device, equipment and a medium to solve the problem that the fault diagnosis of the track circuit cannot be accurately and timely performed by simply relying on manual diagnosis.

[0005] According to one aspect of the present invention, a method for training a fault diagnosis model for a track circuit is provided, comprising:

[0006] According to the system structure diagram corresponding to the target track circuit, construct a circuit fault tree corresponding to the target track circuit;

[0007] Determine at least one type of potential fault type included in the target track circuit according to the circuit fault tree, and importance information corresponding to each potential fault type in the circuit fault tree;

[0008] Determine a high-risk fault type from the potential fault types according to the importance information, and collect real-time operating parameters of each electrical device in the target track circuit when a high-risk fault event occurs in the target track circuit; wherein the fault type of the high-risk fault event is the high-risk fault type;

[0009] Model training data is constructed according to the real-time working parameters, and the model training data is used to train the model to be trained to generate a fault diagnosis model.

[0010] According to another aspect of the present invention, there is provided a fault diagnosis method for a track circuit, including:

[0011] Collecting the to-be-detected working parameters of each electrical device in the target track circuit, and determining whether there is an abnormality in the to-be-detected working parameters;

[0012] If it is determined that there is an abnormality in the to-be-detected working parameters, inputting the to-be-detected working parameters into a fault diagnosis model to determine the fault type of the target track circuit; wherein, the fault diagnosis model is trained by using the training method of the fault diagnosis model of the track circuit according to any one of the present invention.

[0013] According to another aspect of the present invention, there is provided a training device for a fault diagnosis model of a track circuit, including:

[0014] A fault tree construction module for constructing a circuit fault tree corresponding to the target track circuit according to the system structure diagram corresponding to the target track circuit;

[0015] An importance information determination module for determining at least one type of potential fault type included in the target track circuit according to the circuit fault tree, and importance information corresponding to each of the potential fault types in the circuit fault tree;

[0016] A real-time working parameter collection module for determining a high-risk fault type from the potential fault types according to the importance information, and collecting the real-time working parameters of each electrical device in the target track circuit when a high-risk fault event occurs in the target track circuit; wherein, the fault type of the high-risk fault event is the high-risk fault type;

[0017] A model training module for constructing model training data according to the real-time working parameters, and training a to-be-trained model by using the model training data to generate a fault diagnosis model.

[0018] According to another aspect of the present invention, there is provided a fault diagnosis device for a track circuit, including:

[0019] A to-be-detected working parameter collection module for collecting the to-be-detected working parameters of each electrical device in the target track circuit, and determining whether there is an abnormality in the to-be-detected working parameters;

[0020] A fault diagnosis module for, if it is determined that there is an abnormality in the to-be-detected working parameters, inputting the to-be-detected working parameters into a fault diagnosis model to determine the fault type of the target track circuit; wherein, the fault diagnosis model is trained by using the training method of the fault diagnosis model of the track circuit according to any one of the present invention.

[0021] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0022] at least one processor; and

[0023] a memory communicatively connected to the at least one processor; wherein,

[0024] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform any method described in the present invention.

[0025] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement any method of the present invention when executed.

[0026] The technical solution of the embodiment of the present invention constructs a circuit fault tree corresponding to the target track circuit according to the system structure diagram corresponding to the target track circuit; determines at least one type of potential fault type included in the target track circuit and the importance information corresponding to each potential fault type in the circuit fault tree according to the circuit fault tree; determines a high-risk fault type from the potential fault type according to the importance information, and collects real-time working parameters of each electrical equipment in the target track circuit when a high-risk fault occurs in the target track circuit; wherein the fault type of the high-risk fault is a high-risk fault type; constructs model training data according to the real-time working parameters, and uses the model training data to train the model to be trained to generate a fault diagnosis model, so that the trained fault diagnosis model can be used to automatically diagnose the high-risk fault types in the track circuit without relying on manual diagnosis, thereby improving the accuracy and timeliness of track circuit fault diagnosis.

[0027] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0029] Figure 1A A flowchart of a method for training a track circuit fault diagnosis model provided in Embodiment 1 of the present invention;

[0030] Figure 1B A schematic diagram of a target track circuit system structure diagram provided in Embodiment 1 of the present invention;

[0031] Figure 2 A flowchart of a method for training a track circuit fault diagnosis model provided in the second embodiment of the present invention;

[0032] Figure 3 A flow chart of a track circuit fault diagnosis method provided in Embodiment 3 of the present invention;

[0033] Figure 4 A schematic diagram of the structure of a training device for a track circuit fault diagnosis model provided in a fourth embodiment of the present invention;

[0034] Figure 5 A schematic diagram of the structure of a track circuit fault diagnosis device provided in Embodiment 5 of the present invention;

[0035] Figure 6 It is a structural schematic diagram of an electronic device for implementing a training method for a track circuit fault diagnosis model and / or a track circuit fault diagnosis method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0037] It should be noted that the terms "potential", "high risk", "real-time", "to be detected", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0038] Embodiment 1

[0039] Figure 1AThis is a flow chart of a method for training a fault diagnosis model for a track circuit provided in the first embodiment of the present invention. This embodiment is applicable to the case of training a fault diagnosis model for automatic fault diagnosis of a track circuit. The method can be executed by a training device for a fault diagnosis model for a track circuit. The training device for a fault diagnosis model for a track circuit can be implemented in the form of hardware and / or software, for example, by a server. Figure 1A As shown, the method includes:

[0040] S101. Construct a circuit fault tree corresponding to the target track circuit according to a system structure diagram corresponding to the target track circuit.

[0041] The track circuit is a circuit composed of the steel rails of a section of railway line as conductors, which is used to automatically and continuously detect whether the section of the line is occupied by a locomotive vehicle, and is also used to control signal devices or switch devices to ensure driving safety. It can be understood that the railway signal system includes many track circuits, and the so-called target track circuit in this embodiment can be any selected track circuit in the railway signal system. The system structure diagram is a structure diagram used to describe the structure of the target track circuit, such as Figure 1B As shown, Figure 1B A schematic diagram of a target track circuit system structure diagram provided in Example 1 of the present invention.

[0042] A fault tree is a logical tree diagram that deductively represents the causes of accidents or fault events and their logical relationships. The shape of a fault tree is like an inverted tree, and the events in it are generally fault events. It can be understood that a circuit fault tree is a logical tree diagram used to represent the causes of fault events in the target track circuit and their logical relationships.

[0043] In one embodiment, a server for processing a model training process of a fault diagnosis model (hereinafter referred to as the server) obtains a system structure diagram corresponding to a target track circuit, and uses a fault tree drawing algorithm to draw a fault tree based on the system structure diagram corresponding to the target track circuit to construct a circuit fault tree corresponding to the target track circuit.

[0044] S102: Determine at least one type of potential fault type included in the target track circuit according to the circuit fault tree, and importance information corresponding to each potential fault type in the circuit fault tree.

[0045] Among them, the potential fault type refers to the fault type corresponding to all fault events that may occur in the target track circuit. The importance information reflects the influence of each potential fault type in the circuit fault tree, that is, the greater the influence of any potential fault type in the circuit fault tree, the greater the importance information of the potential fault type in the circuit fault tree; the smaller the influence of any potential fault type in the circuit fault tree, the smaller the importance information of the potential fault type in the circuit fault tree.

[0046] In one embodiment, the server uses a fault tree analysis method (FTA) to analyze the constructed circuit fault tree, and determines at least one type of potential fault type included in the target track circuit and the importance information corresponding to each type of potential fault type in the circuit fault tree based on the analysis results.

[0047] Optionally, the importance information includes at least one of structural importance, probability importance and critical importance.

[0048] Among them, 1) Structural importance: without considering the probability of occurrence of the basic event itself, or assuming that the probability of occurrence of each basic event is equal, only the impact of the occurrence of each basic event on the occurrence of the top event is analyzed from a structural perspective. Generally, I φ (i) indicates that the structural importance is calculated by:

[0049]

[0050] 2) Probability importance: The degree of change in the probability of the occurrence of the bottom event leads to the degree of change in the probability of the top event failure. Since the probability function of the top event is a multilinear function of the probability of the occurrence of n basic events, the probability importance coefficient I of the basic event can be obtained by taking a partial derivative of the independent variable. g (i) is:

[0051]

[0052] Among them, q i is the probability of occurrence of the ith bottom-level event, and g(q) is the probability of occurrence of the top-level event.

[0053] 3) Critical importance: the ratio of the rate of change of the probability of the i-th bottom event to the rate of change of the probability of the top event caused by it.

[0054] The mathematical definition of critical importance is:

[0055] S103. Determine a high-risk fault type from potential fault types according to the importance information, and collect real-time operating parameters of each electrical device in the target track circuit when a high-risk fault event occurs in the target track circuit.

[0056] Among them, the high-risk fault type is obtained by screening various potential fault types according to the importance information. Since the importance information reflects the degree of influence of the potential fault type in the circuit fault tree, when the target track circuit has a fault event corresponding to a potential fault type with greater importance information, the harm to the target track circuit will be greater. Therefore, the potential fault type with greater importance information, that is, the greater harm to the target track circuit, is regarded as a high-risk fault type. It can be understood that the fault type of a high-risk fault event is a high-risk fault type. The real-time working parameters indicate the working parameters generated by each electrical equipment in the target track circuit during operation, such as voltage data, current data, temperature data, etc.

[0057] In one implementation, the server sorts each potential fault type according to the importance information of each potential fault type, and selects a number of potential fault types with greater importance information as high-risk fault types according to the sorting result, such as selecting 5 potential fault types with greater importance information as high-risk fault types. It is understandable that the number of high-risk fault types can be one or more.

[0058] After determining at least one high-risk fault type, if the server determines that a high-risk fault event corresponding to any high-risk fault type occurs in the target track circuit, such as when the target track circuit can be manually determined based on experience, the high-risk fault event occurs, and feedback is sent to the server. The server collects the real-time operating parameters of each electrical device in the target track circuit to obtain the real-time operating parameters of each electrical device.

[0059] For example, assuming that "poor matching voltage" is a high-risk fault type, when a high-risk fault event corresponding to "poor matching voltage" occurs in the target track circuit, the real-time operating parameters of each electrical device in the target track circuit are collected to obtain the real-time operating parameters of each electrical device.

[0060] For example, assuming that "attenuator failure" is a high-risk fault type, when a high-risk fault event corresponding to "attenuator failure" occurs in the target track circuit, the real-time operating parameters of each electrical device in the target track circuit are collected to obtain the real-time operating parameters of each electrical device.

[0061] Optionally, the real-time operating parameters include voltage data and / or current data.

[0062] S104: construct model training data according to the real-time working parameters, and use the model training data to train the model to be trained to generate a fault diagnosis model.

[0063] In one embodiment, the real-time operating parameters of each electrical device in the target track circuit when any high-risk fault event occurs in the target track circuit are obtained as a set of model training data, wherein the high-risk fault type corresponding to the high-risk fault event is used as label data, and the real-time operating parameters are used as feature data. The model training data is input into the model to be trained, so that the model to be trained can predict the fault type based on the real-time operating parameters, output the predicted fault type, and then calculate the loss value based on the predicted fault type and the high-risk fault type, and adjust the hyperparameters in the model to be trained based on the calculated loss value to achieve a higher prediction accuracy, thereby generating a fault diagnosis model.

[0064] It can be understood that after the working parameters of each electrical equipment are input into the trained fault diagnosis model, the trained fault diagnosis model can predict high-risk fault types based on the working parameters, and output the high-risk fault types currently existing in the target track circuit, thereby realizing automatic diagnosis of high-risk fault types in the target track circuit.

[0065] The technical solution of the embodiment of the present invention constructs a circuit fault tree corresponding to the target track circuit according to the system structure diagram corresponding to the target track circuit; determines at least one type of potential fault type included in the target track circuit and the importance information corresponding to each potential fault type in the circuit fault tree according to the circuit fault tree; determines a high-risk fault type from the potential fault type according to the importance information, and collects real-time working parameters of each electrical equipment in the target track circuit when a high-risk fault occurs in the target track circuit; wherein the fault type of the high-risk fault is a high-risk fault type; constructs model training data according to the real-time working parameters, and uses the model training data to train the model to be trained to generate a fault diagnosis model, so that the trained fault diagnosis model can be used to automatically diagnose the high-risk fault types in the track circuit without relying on manual diagnosis, thereby improving the accuracy and timeliness of track circuit fault diagnosis.

[0066] Optionally, the type of the fault diagnosis model is a long short-term memory neural network model.

[0067] Among them, the long short-term memory neural network model is also called the LSTM model. The long short-term memory neural network model is:

[0068] f t =σ(W f ·[h t-1 ,]x t +b f )

[0069] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0070]

[0071] o t =σ(W o [h t-1 ,x t ]+b o )

[0072] h t =o t *tanh(C t )

[0073]

[0074] g(x)=σ(ωh t +b)

[0075] Among them, x t h is the real-time operating parameters of each electrical equipment in the target track circuit under different high-risk fault types at the current moment; t is the output value of the hidden layer of the real-time working parameter; W f , W i , W c , W o , ω is the weight matrix; b i , b f , b c , b o is the bias of the real-time working parameters; tanh is the activation function (hyperbolic tangent function); o t is the output gate of the real-time working parameters, f t is the forget gate of real-time working parameters, C t is the unit activation vector of the real-time working parameters, σ(x) is the sigmoid function; g(x) is the predicted value of the long short-term memory neural network model.

[0076] Considering different types of input data, when processing time series data, RNN is the main method because CNN and DNN are not suitable for processing time series information. However, typical RNN has the problem of long-term dependence. In other words, during the training process of RNN, the network gradient either explodes or disappears. To solve this problem, by setting the type of fault diagnosis model to long short-term memory neural network model, the above problem can be solved and the model training results can be better.

[0077] Embodiment 2

[0078] Figure 2 This is a flow chart of a method for training a track circuit fault diagnosis model provided by the second embodiment of the present invention. This embodiment further optimizes and expands the above embodiment and can be combined with the above optional implementations. Figure 2 As shown, the method includes:

[0079] S201. Construct a circuit fault tree corresponding to the target track circuit according to a system structure diagram corresponding to the target track circuit.

[0080] S202: Determine at least one type of potential fault type included in the target track circuit according to the circuit fault tree, and importance information corresponding to each potential fault type in the circuit fault tree.

[0081] S203 , sorting each potential fault type according to importance information of each potential fault type; and selecting a preset number of potential fault types as high-risk fault types according to the sorting result.

[0082] In one implementation, the server sorts each potential fault type in descending order according to the importance information of each potential fault type, and selects a preset number of potential fault types from large to small as high-risk fault types according to the descending sorting result. For example, assuming that the descending sorting result of the importance information of each potential fault type is: potential fault type A-potential fault type B-potential fault type C-potential fault type D, and the preset number is three, then potential fault type A, potential fault type B, and potential fault type C are selected as high-risk fault types.

[0083] By sorting each potential fault type according to its importance information, and selecting a preset number of potential fault types as high-risk fault types according to the sorting results, it is possible to ensure that high-risk fault types have corresponding greater importance information, so that subsequent model training data can be constructed only based on high-risk fault types, without relying on non-high-risk fault types to construct model training data. Under the premise of ensuring that the fault diagnosis model has the ability to predict high-risk fault types, the amount of data for model training is further reduced and the training efficiency is improved.

[0084] Optionally, each potential fault type is sorted according to importance information of each potential fault type, including:

[0085] A. Obtain a first weight value associated with the structural importance, a second weight value associated with the probability importance, and a third weight value associated with the critical importance.

[0086] The first weight value, the second weight value and the third weight value can be set according to the influence of the structural importance, the probability importance and the critical importance on the target track circuit, that is, the greater the influence of any importance information on the target track circuit, the greater the corresponding weight value.

[0087] Optionally, structural importance reflects the importance of events from the fault tree structure, probability importance reflects the impact of changes in the probability of occurrence of bottom events on the probability of occurrence of top events, and critical importance measures the importance standards of each bottom event from the dual perspectives of sensitivity and probability. Therefore, according to the analysis results, the third weight value associated with the critical importance is set to be larger.

[0088] B. Determine a first weighted score based on the structural importance and the first weight value of each potential fault type, determine a second weighted score based on the probability importance and the second weight value of each potential fault type, and determine a third weighted score based on the critical importance and the third weight value of each potential fault type.

[0089] For example, assuming that the structural importance of any potential fault type is A1, the probability importance is A2, the critical importance is A3, and the first weight value, the second weight value, and the third weight value are x1, x2, and x3, respectively, then the first weighted score of the potential fault type is A1*x1, the second weighted score is A2*x2, and the third weighted score is A3*x3.

[0090] C. Determine a weighted total score of each potential fault type according to the first weighted score, the second weighted score, and the third weighted score of each potential fault type, and sort each potential fault type according to the weighted total score of each potential fault type.

[0091] Exemplarily, assuming that the first weighted score of any potential fault type is A1*x1, the second weighted score is A2*x2, and the third weighted score is A3*x3, the total weighted score of the potential fault type is A1*x1+A2*x2+A3*x3.

[0092] In one implementation, the server sorts each potential fault type in descending order according to the weighted total score of each potential fault type, and selects a preset number of potential fault types from large to small as high-risk fault types according to the descending sorting result. For example, assuming that the descending sorting result of the weighted total score of each potential fault type is: potential fault type D-potential fault type C-potential fault type B-potential fault type A, and the preset number is three, then potential fault type B, potential fault type C, and potential fault type B are selected as high-risk fault types.

[0093] By obtaining a first weight value associated with the structural importance, a second weight value associated with the probability importance, and a third weight value associated with the critical importance; determining a first weighted score according to the structural importance and the first weight value of each potential fault type, determining a second weighted score according to the probability importance and the second weight value of each potential fault type, and determining a third weighted score according to the critical importance and the third weight value of each potential fault type; determining a weighted total score for each potential fault type according to the first weighted score, the second weighted score, and the third weighted score of each potential fault type, and sorting each potential fault type according to the weighted total score of each potential fault type. Since the weighted total score is obtained by weighted summing the structural importance, probability importance, and critical importance of each potential fault type, the weighted total score can more accurately and comprehensively reflect the degree of influence of each potential fault type on the target track circuit, so that each potential fault type is sorted according to the weighted total score, which can ensure the accuracy and reliability of the sorting result.

[0094] S204. When any high-risk fault event occurs in the target track circuit, voltage data and current data of each electrical device in the target track circuit are collected to obtain at least one set of characteristic data.

[0095] In one embodiment, when any high-risk fault event occurs in the target track circuit, the server collects voltage data and current data of each electrical device in the target track circuit at the current moment, and constructs at least one set of feature data based on the collected voltage data and current data.

[0096] S205: Use the high-risk fault type corresponding to the high-risk fault event as a training label, perform a labeling operation on the feature data according to the training label, and generate training data according to the feature data after the labeling operation.

[0097] In one embodiment, the server uses the high-risk fault type corresponding to the high-risk fault event as a training label, and performs a labeling operation on the feature data according to the training label, and then uses the feature data with the training label as training data. For example, assuming that the electrical equipment included in the target track circuit is electrical equipment 1, electrical equipment 2, and electrical equipment 3, assuming that a high-risk fault event of "main transmitter failure" occurs, the collected voltage data of electrical equipment 1 is "V1", and the current data is "A1", the collected voltage data of electrical equipment 2 is "V2", and the current data is "A2", and the collected voltage data of electrical equipment 3 is "V3", and the current data is "A3". Then "V1, V2, V3, A1, A2, and A3" are used as a group of feature data, "main transmitter failure" is used as a training label to label the feature data, and training data is generated based on the feature data after the labeling operation.

[0098] When any high-risk fault event occurs in the target track circuit, the voltage data and current data of each electrical device in the target track circuit are collected to obtain at least one set of feature data. The high-risk fault type corresponding to the high-risk fault event is used as a training label, and a labeling operation is performed on the feature data according to the training label. Training data is generated based on the feature data after the labeling operation, thereby providing a specific training data generation method.

[0099] S206: Use model training data to train the model to be trained to generate a fault diagnosis model.

[0100] Embodiment 3

[0101] Figure 3 This is a flow chart of a track circuit fault diagnosis method provided in the third embodiment of the present invention. This embodiment is applicable to the case where a track circuit fault diagnosis model is used to automatically diagnose a track circuit fault. The method can be executed by a track circuit fault diagnosis device, which can be implemented in the form of hardware and / or software, for example, by a server. Figure 3 As shown, the method includes:

[0102] S301, collecting the working parameters to be detected of each electrical device in the target track circuit, and determining whether there is any abnormality in the working parameters to be detected.

[0103] In one embodiment, a server for processing the model usage process of the fault diagnosis model (hereinafter referred to as the server) collects the working parameters to be detected of each electrical equipment in the target track circuit in real time, and performs abnormal judgment on the working parameters to be detected to determine whether there is an abnormality in the working parameters to be detected.

[0104] Optionally, determining whether the working parameter to be detected is abnormal includes:

[0105] Obtain the normal range of working parameters corresponding to each electrical device, and match the working parameters to be detected of each electrical device with the normal range of working parameters of each electrical device respectively; if there is at least one electrical device whose working parameter to be detected does not belong to the normal range of working parameters of the electrical device, it is determined that the working parameter to be detected of the electrical device is abnormal.

[0106] The normal range of working parameters refers to the parameter range that the working parameters of each electrical device should be in when the device is in a normal state. The normal range of working parameters can be set based on experience.

[0107] In one embodiment, the server matches the collected working parameters to be detected of each electrical device with the normal range of the working parameters of each electrical device. If there is at least one working parameter to be detected of the electrical device that does not belong to the normal range of the working parameters of the electrical device, it is determined that the working parameters to be detected of the electrical device are abnormal; if the working parameters to be detected of all electrical devices belong to their corresponding normal ranges of working parameters, it is determined that the working parameters to be detected of all electrical devices are normal.

[0108] By obtaining the normal range of working parameters corresponding to each electrical device, and matching the working parameters to be detected of each electrical device with the normal range of working parameters of each electrical device; if there is at least one working parameter to be detected of an electrical device that does not belong to the normal range of working parameters of the electrical device, it is determined that the working parameter to be detected of the electrical device is abnormal, thereby achieving the effect of abnormal judgment on the working parameters to be detected, and laying the foundation for further determination of the fault type.

[0109] S302: If it is determined that the working parameter to be detected is abnormal, the working parameter to be detected is input into a fault diagnosis model to determine the fault type of the target track circuit.

[0110] The fault diagnosis model is trained by using a training method for a track circuit fault diagnosis model in any of the embodiments of the present invention.

[0111] In one embodiment, if it is determined that there is an abnormality in the working parameters to be detected, the server inputs the working parameters to be detected into the fault diagnosis model, so that the fault diagnosis model can predict the fault type based on the working parameters to be detected and output the predicted fault type as the fault type of the target track circuit.

[0112] The working parameters to be detected of each electrical equipment in the target track circuit are collected, and it is determined whether the working parameters to be detected are abnormal; if it is determined that the working parameters to be detected are abnormal, the working parameters to be detected are input into the fault diagnosis model to determine the fault type of the target track circuit; wherein the fault diagnosis model is trained using the training method of the fault diagnosis model of the track circuit as claimed in any one of claims 1 to 5, so that the trained fault diagnosis model can be used to automatically diagnose high-risk fault types in the track circuit without relying on manual diagnosis, thereby improving the accuracy and timeliness of track circuit fault diagnosis.

[0113] Embodiment 4

[0114] Figure 4 This is a schematic diagram of the structure of a training device for a track circuit fault diagnosis model provided by the fourth embodiment of the present invention. Figure 4 As shown, the device comprises:

[0115] A fault tree construction module 41, configured to construct a circuit fault tree corresponding to the target track circuit according to the system structure diagram corresponding to the target track circuit;

[0116] An importance information determination module 42, configured to determine at least one type of potential fault type included in the target track circuit according to the circuit fault tree, and the importance information corresponding to each potential fault type in the circuit fault tree;

[0117] A real-time working parameter acquisition module 43, configured to determine a high-risk fault type from the potential fault types according to the importance information, and collect the real-time working parameters of each electrical device in the target track circuit when a high-risk fault event occurs in the target track circuit; wherein, the fault type of the high-risk fault event is the high-risk fault type;

[0118] A model training module 44, configured to construct model training data according to the real-time working parameters, and train a model to be trained with the model training data to generate a fault diagnosis model.

[0119] Optionally, the real-time working parameter acquisition module 43 is specifically configured to:

[0120] Sort each potential fault type according to the importance information of each potential fault type;

[0121] Select a preset number of potential fault types as high-risk fault types according to the sorting result.

[0122] Optionally, the importance information includes at least one of structural importance, probability importance, and critical importance;

[0123] The real-time working parameter acquisition module 43 is further specifically configured to:

[0124] Obtain a first weight value associated with the structural importance, a second weight value associated with the probability importance, and a third weight value associated with the critical importance;

[0125] Determine a first weighted score according to the structural importance and the first weight value of each potential fault type, a second weighted score according to the probability importance and the second weight value of each potential fault type, and a third weighted score according to the critical importance and the third weight value of each potential fault type;

[0126] Determine the weighted total score of each potential fault type according to the first weighted score, the second weighted score, and the third weighted score of each potential fault type, and sort each potential fault type according to the weighted total score of each potential fault type.

[0127] Optionally, the real-time working parameters include voltage data and / or current data;

[0128] The real-time working parameter acquisition module 43 is also specifically used for:

[0129] When any high-risk fault event occurs in the target track circuit, voltage data and current data of each electrical device in the target track circuit are collected to obtain at least one set of characteristic data;

[0130] The model training module 44 is specifically used for:

[0131] The high-risk fault type corresponding to the high-risk fault event is used as a training label, and a labeling operation is performed on the feature data according to the training label, and training data is generated according to the feature data after the labeling operation.

[0132] Optionally, the type of the fault diagnosis model is a long short-term memory neural network model.

[0133] The training device for the fault diagnosis model of the track circuit provided in the embodiment of the present invention can execute the training method for the fault diagnosis model of the track circuit provided in any embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method.

[0134] Embodiment 5

[0135] Figure 5 This is a schematic diagram of the structure of a track circuit fault diagnosis device provided in Embodiment 5 of the present invention. Figure 5 As shown, the device comprises:

[0136] The to-be-detected working parameter acquisition module 51 is used to acquire the to-be-detected working parameters of each electrical device in the target track circuit and determine whether there is an abnormality in the to-be-detected working parameters;

[0137] The fault diagnosis module 52 is used to input the working parameters to be detected into the fault diagnosis model if it is determined that there is an abnormality in the working parameters to be detected, so as to determine the fault type of the target track circuit; wherein the fault diagnosis model is trained using the training method of the fault diagnosis model of the track circuit as claimed in any one of claims 1 to 5.

[0138] Optionally, the fault diagnosis module 52 is specifically used for:

[0139] Obtaining the normal range of the working parameters corresponding to each electrical device, and matching the working parameters to be detected of each electrical device with the normal range of the working parameters of each electrical device;

[0140] If there is at least one operating parameter to be detected of the electrical device that does not belong to the normal range of the operating parameters of the electrical device, it is determined that the operating parameter to be detected of the electrical device is abnormal.

[0141] Embodiment 6

[0142] Figure 6 A schematic diagram of the structure of an electronic device 60 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0143] like Figure 6 As shown, the electronic device 60 includes at least one processor 61, and a memory connected to the at least one processor 61, such as a read-only memory (ROM) 62, a random access memory (RAM) 63, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 61 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 62 or the computer program loaded from the storage unit 68 to the random access memory (RAM) 63. In the RAM 63, various programs and data required for the operation of the electronic device 60 can also be stored. The processor 61, the ROM 62, and the RAM 63 are connected to each other via a bus 64. An input / output (I / O) interface 65 is also connected to the bus 64.

[0144] A number of components in the electronic device 60 are connected to the I / O interface 65, including: an input unit 66, such as a keyboard, a mouse, etc.; an output unit 67, such as various types of displays, speakers, etc.; a storage unit 68, such as a disk, an optical disk, etc.; and a communication unit 69, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 69 allows the electronic device 60 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0145] The processor 61 may be a variety of general and / or dedicated processing components with processing and computing capabilities. Some examples of the processor 61 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 61 executes the various methods and processes described above, such as a training method for a fault diagnosis model of a track circuit and / or a fault diagnosis method for a track circuit.

[0146] In some embodiments, the training method of the fault diagnosis model of the track circuit and / or the fault diagnosis method of the track circuit may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 68. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 60 via the ROM 62 and / or the communication unit 69. When the computer program is loaded into the RAM 63 and executed by the processor 61, one or more steps of the training method of the fault diagnosis model of the track circuit and / or the fault diagnosis method of the track circuit described above may be performed. Alternatively, in other embodiments, the processor 61 may be configured to execute the training method of the fault diagnosis model of the track circuit and / or the fault diagnosis method of the track circuit in any other appropriate manner (e.g., by means of firmware).

[0147] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0148] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0149] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0150] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0151] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0152] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0153] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0154] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for training a fault diagnosis model for a track circuit, characterized in that: include: According to the system structure diagram corresponding to the target track circuit, construct a circuit fault tree corresponding to the target track circuit; Determine at least one type of potential fault type included in the target track circuit according to the circuit fault tree, and importance information corresponding to each potential fault type in the circuit fault tree; Determine a high-risk fault type from the potential fault types according to the importance information, and collect real-time operating parameters of each electrical device in the target track circuit when a high-risk fault event occurs in the target track circuit; wherein the fault type of the high-risk fault event is the high-risk fault type; Model training data is constructed according to the real-time working parameters, and the model training data is used to train the model to be trained to generate a fault diagnosis model.

2. The method according to claim 1, characterized in that Determining a high-risk fault type from the potential fault types according to the importance information includes: sorting each of the potential fault types according to the importance information of each of the potential fault types; A preset number of the potential fault types are selected as the high-risk fault types according to the sorting result.

3. The method according to claim 2, characterized in that The importance information includes at least one of structural importance, probability importance and critical importance; The sorting of the potential fault types according to the importance information of the potential fault types includes: Obtaining a first weight value associated with the structural importance, a second weight value associated with the probability importance, and a third weight value associated with the critical importance; Determine a first weighted score according to the structural importance of each potential fault type and the first weight value, determine a second weighted score according to the probability importance of each potential fault type and the second weight value, and determine a third weighted score according to the critical importance of each potential fault type and the third weight value; A weighted total score of each of the potential fault types is determined according to the first weighted score, the second weighted score and the third weighted score of each of the potential fault types, and each of the potential fault types is sorted according to the weighted total score of each of the potential fault types.

4. The method according to claim 1, characterized in that: The real-time operating parameters include voltage data and / or current data; When a high-risk fault event occurs in the target track circuit, collecting real-time operating parameters of each electrical device in the target track circuit includes: When any of the high-risk fault events occurs in the target track circuit, voltage data and current data of each electrical device in the target track circuit are collected to obtain at least one set of characteristic data; The constructing model training data according to the real-time working parameters includes: The high-risk fault type corresponding to the high-risk fault event is used as a training label, and a labeling operation is performed on the feature data according to the training label, and the training data is generated according to the feature data after the labeling operation.

5. The method according to claim 1, characterized in that The type of the fault diagnosis model is a long short-term memory neural network model.

6. A method for diagnosing faults in a track circuit, characterized in that: include: Collecting the working parameters to be detected of each electrical device in the target track circuit, and determining whether the working parameters to be detected are abnormal; If it is determined that there is an abnormality in the working parameter to be detected, the working parameter to be detected is input into the fault diagnosis model to determine the fault type of the target track circuit; wherein the fault diagnosis model is trained using the training method of the fault diagnosis model of the track circuit as described in any one of claims 1-5.

7. The method according to claim 6, characterized in that The determining whether the working parameter to be detected is abnormal includes: Acquire the normal range of the working parameters corresponding to each of the electrical devices, and match the working parameters to be detected of each of the electrical devices with the normal range of the working parameters of each of the electrical devices; If at least one of the to-be-detected operating parameters of the electrical equipment does not belong to the normal range of the operating parameters of the electrical equipment, it is determined that the to-be-detected operating parameter of the electrical equipment is abnormal.

8. A training device for a track circuit fault diagnosis model, characterized in that: include: A fault tree construction module, used to construct a circuit fault tree corresponding to the target track circuit according to a system structure diagram corresponding to the target track circuit; An importance information determination module, used to determine at least one type of potential fault type included in the target track circuit according to the circuit fault tree, and importance information corresponding to each potential fault type in the circuit fault tree; A real-time operating parameter acquisition module, used to determine a high-risk fault type from the potential fault types according to the importance information, and to acquire real-time operating parameters of each electrical device in the target track circuit when a high-risk fault event occurs in the target track circuit; wherein the fault type of the high-risk fault event is the high-risk fault type; The model training module is used to construct model training data according to the real-time working parameters, and use the model training data to train the model to be trained to generate a fault diagnosis model.

9. A fault diagnosis device for a track circuit, characterized in that: include: A detection working parameter acquisition module is used to collect the detection working parameters of each electrical device in the target track circuit and determine whether the detection working parameters are abnormal; A fault diagnosis module is used for inputting the working parameters to be detected into a fault diagnosis model if it is determined that the working parameters to be detected are abnormal, so as to determine the fault type of the target track circuit; wherein the fault diagnosis model is trained using the training method of the fault diagnosis model of the track circuit as described in any one of claims 1 to 5.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the method of any one of claims 1-5 and / or claims 6-7.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method of any one of claims 1-5 and / or claims 6-7 when executed.

Images