Roadway fault identification method and device

Abstract

The invention provides a roadway fault identification method and device. The method comprises the following steps: acquiring multiple groups of fault data of a target roadway; wherein each group of fault data is composed of a roadway identifier of a selected faulty roadway branch and pressure difference data of a ventilation facility selected after the faulty roadway branch breaks down; preprocessing the pressure difference data in each group of fault data to obtain multiple groups of target fault data; according to the multiple groups of target fault data, training a roadway fault recognition model to obtain a trained target roadway fault recognition model; the real-time pressure difference data of the ventilation facilities in the target roadway are obtained, the target roadway fault recognition model is adopted to predict the position of the faulty roadway according to the real-time pressure difference data, and therefore the roadway fault recognition model is trained through the pressure difference data of the ventilation facilities, and the faulty roadway can be quickly recognized based on the pressure difference data. And compared with air volume, air pressure data and pressure difference data of ventilation facilities, accurate monitoring is easier, and the accuracy of roadway fault diagnosis can be effectively guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of mine ventilation network fault diagnosis, and in particular to a tunnel fault identification method and device. Background Art

[0002] In the related technologies, the research on mine ventilation network fault diagnosis is mainly through monitoring the roadway air volume, roadway air pressure and roadway air volume-air pressure composite data. However, it is difficult to accurately monitor the roadway air volume and air pressure in coal mines, which leads to certain errors or uncertainties in the mine ventilation network fault diagnosis. Summary of the invention

[0003] The present invention provides a method and device for identifying a lane fault, so as to solve one of the technical problems in the related art at least to a certain extent. The technical solution of the present invention is as follows:

[0004] According to a first aspect of an embodiment of the present invention, a tunnel fault identification method is provided, comprising: obtaining multiple groups of fault data of a target tunnel; wherein each group of fault data consists of a tunnel identification of a selected fault tunnel branch and pressure difference data of a ventilation facility selected after the fault of the fault tunnel branch; preprocessing the pressure difference data in each group of the fault data to obtain multiple groups of target fault data; wherein the pressure difference data in each group of the target fault data are at the same order of magnitude; training a tunnel fault identification model based on the multiple groups of target fault data to obtain a trained target tunnel fault identification model; wherein the target tunnel fault identification model has learned the corresponding relationship between the pressure difference data and the fault tunnel branch; obtaining real-time pressure difference data of the ventilation facility in the target tunnel, and predicting the position of the fault tunnel based on the real-time pressure difference data using the target tunnel fault identification model.

[0005] According to a second aspect of an embodiment of the present invention, a tunnel fault identification device is provided, comprising: an acquisition module, used to acquire multiple groups of fault data of a target tunnel; wherein each group of fault data consists of a tunnel identifier of a selected fault tunnel branch and pressure difference data of a target tunnel branch corresponding to the fault of the fault tunnel branch; a preprocessing module, used to preprocess the pressure difference data in the multiple groups of fault data to obtain multiple groups of target fault data; wherein the pressure difference data in the multiple groups of target fault data are at the same order of magnitude; a training module, used to train a tunnel fault identification model according to the multiple groups of target fault data to obtain a trained target tunnel fault identification model; wherein the target tunnel fault identification model has learned the correspondence between the ventilation facility monitoring pressure difference data and the fault tunnel branch; a processing module, used to acquire real-time pressure difference data of the target tunnel, and, based on the real-time pressure difference data, use the target tunnel fault identification model to predict the position of the fault tunnel.

[0006] According to a third aspect of an embodiment of the present invention, there is provided an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement the lane fault identification method as described in the embodiment of the first aspect of the present invention.

[0007] According to a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the lane fault identification method as described in the embodiment of the first aspect of the present invention.

[0008] According to a fifth aspect of an embodiment of the present invention, there is provided a computer program product, comprising: a computer program, wherein when the computer program is executed by a processor, the lane fault identification method as described in the embodiment of the first aspect of the present invention is implemented.

[0009] The technical solution provided by the embodiments of the present invention brings at least the following beneficial effects:

[0010] In the technical scheme, multiple groups of fault data of the target lane are obtained; wherein each group of fault data is composed of the lane identification of the selected fault lane branch and the pressure difference data of the ventilation facilities selected after the fault lane branch fails; the pressure difference data in each group of fault data is preprocessed to obtain multiple groups of target fault data; wherein the pressure difference data in each group of target fault data is at the same order of magnitude; according to the multiple groups of target fault data, the lane fault identification model is trained to obtain the trained target lane fault identification model; wherein the target lane fault identification model has learned the corresponding relationship between the pressure difference data and the fault lane branch; the real-time pressure difference data of the ventilation facilities in the target lane is obtained, and according to the real-time pressure difference data, the target lane fault identification model is used to predict the position of the fault lane, thereby, by using the pressure difference data of the ventilation facilities to train the lane fault identification model, and using the trained target lane fault identification model to perform lane fault diagnosis, the fault lane can be quickly identified based on the pressure difference data. And compared with the air volume and air pressure data, the pressure difference data of the ventilation facilities is easier to monitor accurately, which can effectively ensure the accuracy of lane fault diagnosis. In addition, by preprocessing the collected pressure difference data, the error caused by the difference in the values ​​of pressure difference data at different positions can be reduced, and the accuracy of tunnel fault diagnosis can be improved.

[0011] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present invention, and together with the description are used to explain the principles of the present invention, and do not constitute improper limitations of the present invention.

[0013] Figure 1 is a schematic flow chart of a lane fault identification method according to a first embodiment of the present invention;

[0014] Figure 2 is a simulation experimental ventilation network diagram of a target laneway shown in the second embodiment of the present invention;

[0015] Figure 3 is a schematic flow chart of a lane fault identification method shown in the third embodiment of the present invention;

[0016] Figure 4 is a schematic diagram of the principle of a lane fault identification method shown in a fourth embodiment of the present invention;

[0017] Figure 5 is a schematic structural diagram of a lane fault identification device shown in a fifth embodiment of the present invention;

[0018] Figure 6 It is a schematic structural diagram of an electronic device 600 according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to enable ordinary persons in the art to better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings.

[0020] It should be noted that the terms "first", "second", 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 interchanged 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. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0021] It should be noted that in the technical solution of the present invention, the collection, storage, use, processing, transmission, provision and disclosure of user personal information are all carried out with the user's consent, and comply with the relevant laws and regulations and do not violate public order and good morals.

[0022] Changes in the cross-sectional area of ​​the mine ventilation system caused by the damper switch or damage, tunnel collapse or deformation, mine car operation, cage lifting, etc. will cause changes in the wind resistance of the mine ventilation system. Among them, the situation of persistent changes in wind resistance caused by the damper switch or damage, tunnel collapse or deformation, etc. is called the mine ventilation system resistance change type fault, the location of the fault is called the fault location, and the magnitude of the wind resistance change is called the fault amount.

[0023] In the related technologies, the research on mine ventilation network fault diagnosis is mainly through monitoring the tunnel air volume, tunnel air pressure and tunnel air volume-air pressure composite data. However, the existing technical means cannot guarantee the accurate testing and monitoring of the air volume and air pressure in the underground coal mine tunnels.

[0024] In view of the above problems, the present invention proposes a tunnel fault identification method, which trains a tunnel fault identification model by using the pressure difference data of ventilation facilities, and uses the trained target tunnel fault identification model to perform tunnel fault diagnosis. This can not only effectively ensure the accuracy of tunnel fault diagnosis, but also realize rapid identification of faulty tunnels based on pressure difference data.

[0025] The tunnel fault identification method and device according to the embodiments of the present invention are described below with reference to the accompanying drawings.

[0026] Figure 1 It is a flow chart of the lane fault identification method shown in the first embodiment of the present invention.

[0027] The embodiment of the present invention takes the lane fault identification method configured in a lane fault identification device as an example. The lane fault identification device can be applied to any electronic device so that the electronic device can perform the lane fault identification function.

[0028] Among them, the electronic device can be any device with computing capabilities, such as a personal computer, a mobile terminal, a server (or cloud), etc. The mobile terminal can be, for example, a vehicle-mounted device, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, and other hardware devices with various operating systems, touch screens and / or display screens.

[0029] like Figure 1 As shown, the lane fault identification method includes the following steps:

[0030] Step 101, obtaining multiple groups of fault data of a target lane.

[0031] Each set of fault data consists of the selected lane identification of the fault lane branch and the pressure difference data of the ventilation facilities selected after the fault lane branch fails.

[0032] In the embodiment of the present invention, the lane fault identification device can obtain multiple groups of fault data of the target lane in various public, legal and compliant ways. For example, each group of fault data of the target lane can be recorded in real time during the simulated fault experiment of the target lane.

[0033] As a possible implementation method, a simulated experimental ventilation network diagram of the target tunnel can be obtained; based on the simulated experimental ventilation network diagram, at least one faulty tunnel branch and at least one ventilation facility are selected from the target tunnel; the pressure difference data of at least one ventilation facility after any faulty tunnel branch fails is collected, and the tunnel identification of the faulty tunnel branch and the collected pressure difference data of at least one ventilation facility are used as a set of fault data.

[0034] As an example, assume that the simulation experimental ventilation network diagram of the target tunnel is as follows: Figure 2 As shown, Figure 2 In, e1,e2,e3,…,e 32 is the lane branch of the target lane, and the numbers in the circle are the nodes of the lane branch. If e6, e8, e9, e 11 、e 19 、e 22 6 lane branches are selected as fault lane branches, e7 and e 10 、e 16 、e 17 、e 18 、e 24 、e 25 、e 31 Eight ventilation facilities on the tunnel branches are used to monitor the pressure difference data. The locations of the eight ventilation facilities are as follows: Figure 2 As shown in the figure, 600 sets of fault data of the target tunnel are obtained through simulated experimental ventilation, among which e6, e8, e9, e 11 、e 19 、e 22 There are 100 sets of fault data for each of the 6 lane branches. The acquisition process is to obtain e6, e8, e9, e 11 、e 19 、e 22 After any one lane branch is set to fault, collect e7, e 10 、e 16 、e 17 、e 18 、e 24 、e 25 、e 31 The pressure difference data of 8 ventilation facilities on the lane branch, thus a set of fault data is obtained based on the faulty lane branch and the pressure difference data of the 8 ventilation facilities. In summary, 600 sets of fault data can be obtained as shown in Table 1:

[0035] Table 1 Example of multiple groups of fault data for target lanes

[0036]

[0037] In Table 1, each set of fault data corresponding to the 6 serial numbers is e6, e8, e9, e 11 、e 19 、e 22 The corresponding set of fault data, that is, the set of fault data with sequence numbers 1, 2, 3, 4, 5, and 6 are e6, e8, e9, e 11 、e 19 、e 22 The corresponding set of fault data, the set of fault data with serial numbers 7, 8, 9, 10, 11, and 12 are e6, e8, e9, e 11 、e 19 、e 22 A corresponding set of fault data, and so on.

[0038] Step 102 : pre-process the pressure difference data in each set of fault data to obtain multiple sets of target fault data.

[0039] The pressure difference data in each group of target fault data are at the same level.

[0040] As a possible implementation, the pressure difference data in each set of fault data may be standardized so that the pressure difference data in each set of fault data after standardization satisfies a first set value and a second set value. For example, the pressure difference data in each set of fault data may be preprocessed by standardization so that the pressure difference data in each set of fault data after standardization are at the same level, ensuring that the pressure difference data in each set of fault data after standardization has a mean value of 0 and a variance of 1.

[0041] As an example, the fault data in Table 1 above is normalized as shown in Table 2:

[0042] Table 2 Example table of standardized multiple groups of fault data

[0043]

[0044] Therefore, by preprocessing the collected pressure difference data, the error caused by the difference in the values ​​of the pressure difference data at different positions can be reduced, and the accuracy of tunnel fault diagnosis can be improved.

[0045] Step 103 , training a lane fault identification model according to the multiple groups of target fault data to obtain a trained target lane fault identification model.

[0046] Among them, the target lane fault identification model has learned the corresponding relationship between the pressure difference data and the fault lane branches.

[0047] As a possible implementation method, multiple groups of target fault data can be divided into training sets and test sets according to a set ratio; the pressure difference data in each group of target fault data in the training set is used as the input of the lane fault identification model, and the lane identification of the fault lane branch in each group of target fault data in the training set is used as the output of the lane fault identification model. The lane fault identification model is trained to obtain the target lane fault identification model.

[0048] The set ratio may be 7:3 or 8:2, etc., which is not limited in this embodiment. For example, for the above example of obtaining 600 groups of fault data of the target lane through simulated experimental ventilation, 420 groups of the 600 groups of target fault data obtained by preprocessing can be divided into training sets, and 180 groups can be divided into test sets.

[0049] Among them, the training set is used to train the lane fault identification model, the test set does not participate in model training, and the test set is used to evaluate the target lane fault identification model.

[0050] As a possible implementation method, the pressure difference data in each group of target fault data in the test set can be input into the target lane fault identification model to obtain the lane identification of the fault lane branch corresponding to each group of target fault data output by the target lane fault identification model; based on the difference between the lane identification of the fault lane branch corresponding to each group of target fault data output by the target lane fault identification model and the lane identification of the fault lane branch in each group of target fault data in the test set, the performance of the target lane fault identification model is evaluated.

[0051] As an example, for the above example of obtaining 600 sets of fault data, since e6, e8, e9, e 11 、e 19 、e 22 There are 100 groups of fault data for each of the 6 lane branches. Therefore, for any of the 6 lane branches, the number of groups correctly identified by the target lane fault identification model in the 100 groups of fault data can be counted. Then, based on the ratio between the number of groups correctly identified by the statistical target lane fault identification model and the total number of groups (i.e. 100), the accuracy of fault identification of the target lane fault identification model can be determined, thereby evaluating the performance of the target lane fault identification model.

[0052] Step 104, obtaining real-time differential pressure data of ventilation facilities in the target lane, and predicting the location of the faulty lane using the target lane fault identification model based on the real-time differential pressure data.

[0053] In an embodiment of the present invention, a trained target lane fault identification model can be used to perform lane fault diagnosis. Specifically, the acquired real-time pressure difference data of the ventilation facilities in the target lane can be input into the target lane fault identification model to obtain the lane identification of the fault lane branch output by the target lane fault identification model, thereby realizing rapid identification of the fault lane based on the pressure difference data. And because lane fault diagnosis is based on pressure difference data, compared with air volume and air pressure data, the pressure difference data of the ventilation facilities is easier to monitor accurately, thereby effectively ensuring the accuracy of lane fault diagnosis.

[0054] The tunnel fault identification method of the embodiment of the present invention obtains multiple groups of fault data of the target tunnel; wherein each group of fault data consists of the tunnel identification of the selected fault tunnel branch and the pressure difference data of the ventilation facility selected after the fault of the fault tunnel branch; the pressure difference data in each group of fault data is preprocessed to obtain multiple groups of target fault data; wherein the pressure difference data in each group of target fault data is at the same order of magnitude; according to the multiple groups of target fault data, a tunnel fault identification model is trained to obtain a trained target tunnel fault identification model; wherein the target tunnel fault identification model has learned the corresponding relationship between the pressure difference data and the fault tunnel branch; the real-time pressure difference data of the ventilation facility in the target tunnel is obtained, and according to the real-time pressure difference data, the target tunnel fault identification model is used to predict the position of the fault tunnel, thereby, by using the pressure difference data of the ventilation facility to train the tunnel fault identification model, and using the trained target tunnel fault identification model to perform tunnel fault diagnosis, the fault tunnel can be quickly identified based on the pressure difference data. Compared with air volume and air pressure data, the pressure difference data of ventilation facilities is easier to monitor accurately, which can effectively ensure the accuracy of tunnel fault diagnosis. In addition, by preprocessing the collected pressure difference data, the error caused by the difference in the value of pressure difference data at different locations can be reduced, thus improving the accuracy of tunnel fault diagnosis.

[0055] In order to clearly explain how the lane fault identification model is trained in the present invention, the present invention also proposes a lane fault identification method.

[0056] Figure 3 It is a schematic flow chart of a lane fault identification method shown in the third embodiment of the present invention.

[0057] It should be noted that, in the embodiment of the present invention, the tunnel fault identification model is a support vector machine model.

[0058] like Figure 3 As shown, the lane fault identification method includes the following steps:

[0059] Step 301, obtaining multiple groups of fault data of the target lane.

[0060] Step 302 : pre-process the pressure difference data in each set of fault data to obtain multiple sets of target fault data.

[0061] Step 303: divide the multiple groups of target fault data into training sets and test sets according to a set ratio.

[0062] Step 304: determine the main influencing parameters of the tunnel fault identification model and the value range of the main influencing parameters.

[0063] Among them, the main influencing parameters include at least the Gaussian kernel width and the regularization parameter.

[0064] As an example, a support vector machine algorithm can be selected to establish a lane fault identification model, and thus, the lane fault identification model is a support vector machine model. And two main influencing parameters of the support vector machine are selected: Gaussian kernel width a and regularization parameter b. Among them, the value of a in the support vector machine model is {10, 100, 1000, 10000, 100000, 1000000}, and the value of b is {1E-5, 1E-4, 0.001, 0.01, 0.1, 1}.

[0065] Step 305: Determine multiple parameter combinations using a grid search method according to the value ranges of the main influencing parameters.

[0066] Among them, the grid search method is an exhaustive search method, which tries every possible parameter combination among all values ​​of given parameters through exhaustive search.

[0067] As an example, for the example in which a takes the value of {10, 100, 1000, 10000, 100000, 1000000} and b takes the value of {1E-5, 1E-4, 0.001, 0.01, 0.1, 1} in the above support vector machine model, multiple parameter combinations determined by the grid search method are shown in Table 3:

[0068] Table 3 Example of multiple parameter combinations

[0069]

[0070] Step 306 , input the pressure difference data in each group of target fault data in the training set into the lane fault identification model corresponding to each parameter combination, so as to obtain the lane identification of the fault lane branch corresponding to each group of target fault data output by the lane fault identification model corresponding to each parameter combination.

[0071] In an embodiment of the present invention, each parameter combination corresponds to a lane fault identification model, so that for the lane fault identification model corresponding to any parameter combination, the pressure difference data in any group of target fault data in the training set can be input into the model to obtain the lane identification of the fault lane branch corresponding to the group of target fault data output by the model.

[0072] Step 307, for the lane fault identification model corresponding to any parameter combination, the accuracy of the lane fault identification model corresponding to the parameter combination is determined according to the difference between the lane identification of the fault lane branch corresponding to each group of target fault data output by the lane fault identification model corresponding to the parameter combination and the lane identification of the fault lane branch in each group of target fault data in the training set.

[0073] In an embodiment of the present invention, for a lane fault identification model corresponding to any parameter combination, the accuracy of the model can be determined based on the difference between the lane identification of the fault lane branch corresponding to each group of target fault data output by the model and the lane identification of the corresponding fault lane branch in the training set. For example, when the lane identification of the fault lane branch corresponding to any group of target fault data output by the model is different from the lane identification of the fault lane branch in the training set, the output result corresponding to the group of target fault data is recorded as 1, otherwise it is recorded as 0, and then the ratio between the number of groups with an output result of 1 and the total number of groups is determined as the accuracy of the model.

[0074] Step 308 : Determine the lane fault identification model with the highest accuracy among the lane fault identification models corresponding to each parameter combination as the target lane fault identification model.

[0075] As an example, for the example in which a in the above support vector machine model takes the value of {10, 100, 1000, 10000, 100000, 1000000} and b takes the value of {1E-5, 1E-4, 0.001, 0.01, 0.1, 1}, assuming that when the parameter combination a=100, b=0.01, the corresponding lane fault identification model has the highest accuracy, then the support vector machine model with the parameter combination a=100, b=0.01 can be determined as the target lane fault identification model.

[0076] Step 309, obtaining real-time pressure difference data of ventilation facilities in the target lane, and predicting the location of the faulty lane using the target lane fault identification model based on the real-time pressure difference data.

[0077] It should be noted that the execution process of steps 301 to 303 and step 309 can be implemented in any manner in the embodiments of the present invention, and the embodiments of the present invention do not limit this and will not be described in detail.

[0078] The tunnel fault identification method of the embodiment of the present invention determines the main influencing parameters of the tunnel fault identification model and the value range of the main influencing parameters; wherein the main influencing parameters at least include Gaussian kernel width and regularization parameter; according to the value range of the main influencing parameters, a grid search method is used to determine multiple parameter combinations; the pressure difference data in each group of target fault data in the training set is input into the tunnel fault identification model corresponding to each parameter combination to obtain the tunnel identification of the fault tunnel branch corresponding to each group of target fault data output by the tunnel fault identification model corresponding to each parameter combination; for the tunnel fault identification corresponding to any parameter combination, the tunnel fault identification model is used to identify the tunnel fault branch corresponding to each group of target fault data; An identification model is obtained, and the accuracy of the lane fault identification model corresponding to the parameter combination is determined according to the difference between the lane identification of the fault lane branch corresponding to each group of target fault data output by the lane fault identification model corresponding to the parameter combination and the lane identification of the fault lane branch in each group of target fault data in the training set; the lane fault identification model with the highest accuracy among the lane fault identification models corresponding to each parameter combination is determined as the target lane fault identification model, thereby, the target lane fault identification model can be obtained by optimizing the main influencing parameters based on the value range of the main influencing parameters of the selected lane fault identification model.

[0079] In order to clearly illustrate the above embodiment, an example is now given for illustration.

[0080] Figure 4 A schematic diagram of the principle of a lane fault identification method provided in the fourth embodiment of the present invention.

[0081] Based on any embodiment of the present invention, Figure 4 As shown, the lane fault identification method of the embodiment of the present invention can also be implemented according to the following steps:

[0082] Step 401, collecting a pressure difference-resistance change fault location data set.

[0083] Among them, the resistance fault location refers to the fault lane branch.

[0084] Select faulty lane branches and ventilation facilities, and collect faulty lane branches e1, e2, …, e m And the pressure difference data P1, P2, ..., P of the ventilation facilities selected after the failure of each faulty lane branch n As a pressure difference-resistance change fault location data set.

[0085] Among them, a set of pressure difference-resistance change fault location data sets consists of the lane identification of the fault lane branch and the pressure difference data of the ventilation facilities selected after the fault of the fault lane branch.

[0086] The pressure difference data in any set of pressure difference-resistance change fault location data sets is used as an input feature, and the corresponding output feature is the lane identification of the fault lane branch in the corresponding set of pressure difference-resistance change fault location data sets.

[0087] As an example, assume that the simulation experimental ventilation network diagram of the target tunnel is as follows: Figure 2 As shown, Figure 2 In, e1,e2,e3,…,e 32 is the lane branch of the target lane, and the numbers in the circle are the nodes of the lane branch. If e6, e8, e9, e 11 、e 19 、e 22 6 lane branches are selected as fault lane branches, e7 and e 10 、e 16 、e 17 、e 18 、e 24 、e 25 、e 31 Eight ventilation facilities on the tunnel branches are used to monitor the pressure difference data. The locations of the eight ventilation facilities are as follows: Figure 2 As shown in the figure, 600 sets of fault data of the target tunnel are obtained through simulated experimental ventilation, among which e6, e8, e9, e 11 、e 19 、e 22 There are 100 sets of fault data for each of the 6 lane branches. The acquisition process is to obtain e6, e8, e9, e 11 、e 19 、e 22 After any one lane branch is set to fault, collect e7, e 10 、e 16 、e 17 、e 18 、e 24 、e 25 、e 31 The pressure difference data of the eight ventilation facilities on the lane branch are collected, so that a set of fault data is obtained based on the faulty lane branch and the pressure difference data of the eight ventilation facilities. In summary, 600 sets of fault data can be obtained as shown in Table 1 above.

[0088] Step 402: preprocessing of data set.

[0089] In order to reduce the error caused by the size of the pressure difference data at different positions in the input feature and make the pressure difference data in the same input feature at the same level, standardization is used to preprocess the pressure difference data in each input feature. Standardization ensures that the mean value of the pressure difference data in each input feature is 0 and the variance is 1.

[0090] As an example, the fault data in the above Table 1 is normalized as shown in the above Table 2.

[0091] Step 403, divide the data set.

[0092] 70% of the preprocessed data set is divided into a training set and 30% is divided into a test set, wherein the training set is used to train the optimal parameters of the support vector machine fault diagnosis model (i.e., the lane support vector machine fault diagnosis model in the present invention), and the test set does not participate in the training, but is specifically used to evaluate the support vector machine fault diagnosis model under the optimal parameters.

[0093] Step 404: Optimize model parameters on the training set.

[0094] A support vector machine fault diagnosis model is established, and the main influencing parameters and their value ranges are selected, and the parameters are optimized to determine the optimal parameters.

[0095] The support vector machine algorithm was selected to establish the fault diagnosis model, and two main influencing parameters of the support vector machine were selected: Gaussian kernel width a and regularization parameter b. In the support vector machine model, a takes the value of {10, 100, 1000, 10000, 100000, 1000000}, and b takes the value of {1E-5, 1E-4, 0.001, 0.01, 0.1, 1}. Through grid search, the optimal parameters of the support vector machine model are a=100, b=0.01, and the fault diagnosis accuracy of the training set reaches 0.969.

[0096] Step 405: Perform model validation on the test set.

[0097] The support vector machine fault diagnosis model under the optimal parameters is applied to the test set to determine the accuracy of fault identification of the support vector machine fault diagnosis model.

[0098] The support vector machine fault diagnosis model with the optimal parameters of a=100, b=0.01 is applied to the test set. The fault diagnosis results on the test set are shown in Table 4:

[0099] Table 4 Example of fault diagnosis results on the test set

[0100] Fault lane branch Identify the correct number Number of recognition errors Fault identification accuracy <![CDATA[e6]]> 100 0 1 <![CDATA[e8]]> 97 3 0.97 <![CDATA[e9]]> 100 0 1 <![CDATA[e 11 ]]> 100 0 1 <![CDATA[e 19 ]]> 95 5 0.95 <![CDATA[e 22 ]]> 96 4 0.96

[0101] Among them, e6, e9 and e 11 The accuracy even reaches 1, thereby verifying the feasibility of the tunnel fault identification method of the embodiment of the present invention.

[0102] Step 406, model application.

[0103] The real-time pressure difference data of the ventilation facilities in the target tunnel is obtained, and then the location of the faulty tunnel is predicted using the trained model based on the real-time pressure difference data.

[0104] In summary, by simulating ventilation network failures, collecting ventilation facility pressure difference data, and performing standardized preprocessing on the collected pressure difference data, the error caused by the difference in the pressure difference values ​​of ventilation facilities at different locations can be reduced. By using the monitored pressure difference data as input features and the specific fault lane branch as output features, a support vector machine fault diagnosis model is established, which can realize rapid identification of fault lanes and improve the accuracy of fault diagnosis.

[0105] Corresponding to the tunnel fault identification method provided in the above embodiment, the present invention also provides a tunnel fault identification device. Since the tunnel fault identification device provided in the embodiment of the present invention corresponds to the tunnel fault identification method provided in the above embodiment, the implementation method of the tunnel fault identification method is also applicable to the tunnel fault identification device provided in the embodiment of the present invention, and will not be described in detail in the embodiment of the present invention.

[0106] Figure 5 It is a schematic diagram of the structure of a lane fault identification device shown in the fifth embodiment of the present invention.

[0107] like Figure 5 As shown, the lane fault identification device 500 includes: an acquisition module 510 , a preprocessing module 520 , a training module 530 and a processing module 540 .

[0108] Among them, the acquisition module 510 is used to obtain multiple groups of fault data of the target lane; wherein each group of fault data consists of the lane identification of the selected fault lane branch and the pressure difference data of the target lane branch corresponding to the fault of the fault lane branch; the preprocessing module 520 is used to preprocess the pressure difference data in the multiple groups of fault data to obtain multiple groups of target fault data; wherein the pressure difference data in the multiple groups of target fault data are at the same order of magnitude; the training module 530 is used to train the lane fault identification model according to the multiple groups of target fault data to obtain a trained target lane fault identification model; wherein the target lane fault identification model has learned the corresponding relationship between the ventilation facility monitoring pressure difference data and the fault lane branch; the processing module 540 is used to obtain the real-time pressure difference data of the target lane, and according to the real-time pressure difference data, use the target lane fault identification model to predict the position of the fault lane.

[0109] As a possible implementation method of an embodiment of the present invention, the acquisition module 510 is also used to: obtain a simulated experimental ventilation network diagram of the target lane; select at least one faulty lane branch and at least one ventilation facility from the target lane according to the simulated experimental ventilation network diagram; collect the pressure difference data of at least one ventilation facility after any faulty lane branch fails, and use the lane identification of the faulty lane branch and the collected pressure difference data of at least one ventilation facility as a set of fault data.

[0110] As a possible implementation method of an embodiment of the present invention, the training module 530 includes: a division unit, which is used to divide multiple groups of target fault data into training sets and test sets according to a set ratio; a training unit, which is used to use the pressure difference data in each group of target fault data in the training set as the input of the lane fault identification model, and use the lane identification of the fault lane branch in each group of target fault data in the training set as the output of the lane fault identification model, and train the lane fault identification model to obtain the target lane fault identification model.

[0111] As a possible implementation method of an embodiment of the present invention, the tunnel fault identification model is a support vector machine model; the training unit is also used to: determine the main influencing parameters of the tunnel fault identification model and the value range of the main influencing parameters; wherein the main influencing parameters include at least Gaussian kernel width and regularization parameter; according to the value range of the main influencing parameters, a grid search method is used to determine multiple parameter combinations; the pressure difference data in each group of target fault data in the training set is input into the tunnel fault identification model corresponding to each parameter combination to obtain the tunnel identification of the fault tunnel branch corresponding to each group of target fault data output by the tunnel fault identification model corresponding to each parameter combination; for the tunnel fault identification model corresponding to any parameter combination, according to the difference between the tunnel identification of the fault tunnel branch corresponding to each group of target fault data output by the tunnel fault identification model corresponding to the parameter combination and the tunnel identification of the fault tunnel branch in each group of target fault data in the training set, the accuracy of the tunnel fault identification model corresponding to the parameter combination is determined; the tunnel fault identification model with the highest accuracy among the tunnel fault identification models corresponding to each parameter combination is determined as the target tunnel fault identification model.

[0112] As a possible implementation method of an embodiment of the present invention, the above-mentioned device also includes: a testing module, which is used to input the pressure difference data in each group of target fault data in the test set into the target lane fault identification model to obtain the lane identification of the fault lane branch corresponding to each group of target fault data output by the target lane fault identification model; an evaluation module, which is used to evaluate the performance of the target lane fault identification model based on the difference between the lane identification of the fault lane branch corresponding to each group of target fault data output by the target lane fault identification model and the lane identification of the fault lane branch in each group of target fault data in the test set.

[0113] As a possible implementation method of the embodiment of the present invention, the preprocessing module 520 is also used to: perform data standardization on the pressure difference data in each group of fault data so that the average value of the pressure difference data in each group of fault data after standardization is a first set value, and the variance is a second set value.

[0114] The tunnel fault identification device of the embodiment of the present invention obtains multiple groups of fault data of the target tunnel; wherein each group of fault data consists of the tunnel identification of the selected fault tunnel branch and the pressure difference data of the ventilation facility selected after the fault of the fault tunnel branch; pre-processes the pressure difference data in each group of fault data to obtain multiple groups of target fault data; wherein the pressure difference data in each group of target fault data is at the same order of magnitude; trains a tunnel fault identification model based on the multiple groups of target fault data to obtain a trained target tunnel fault identification model; wherein the target tunnel fault identification model has learned the corresponding relationship between the pressure difference data and the fault tunnel branch; obtains the real-time pressure difference data of the ventilation facility in the target tunnel, and predicts the position of the fault tunnel based on the real-time pressure difference data using the target tunnel fault identification model, thereby, by using the pressure difference data of the ventilation facility to train the tunnel fault identification model, and using the trained target tunnel fault identification model to perform tunnel fault diagnosis, it is possible to quickly identify the fault tunnel based on the pressure difference data. Compared with air volume and air pressure data, the pressure difference data of ventilation facilities is easier to monitor accurately, which can effectively ensure the accuracy of tunnel fault diagnosis. In addition, by preprocessing the collected pressure difference data, the error caused by the difference in the value of pressure difference data at different locations can be reduced, thus improving the accuracy of tunnel fault diagnosis.

[0115] In an exemplary embodiment, the present invention also provides an electronic device.

[0116] Among them, electronic equipment includes:

[0117] processor;

[0118] a memory for storing processor-executable instructions;

[0119] The processor is configured to execute instructions to implement the lane fault identification method proposed in any of the aforementioned embodiments.

[0120] As an example, Figure 6 is a schematic diagram of the structure of an electronic device 600 shown in an exemplary embodiment of the present invention. Figure 6 As shown, the electronic device 600 may further include:

[0121] The memory 610 and the processor 620, a bus 630 connecting different components (including the memory 610 and the processor 620), the memory 610 stores a computer program, and when the processor 620 executes the program, the lane fault identification method described in the embodiment of the present invention is implemented.

[0122] Bus 630 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0123] The electronic device 600 typically includes a variety of electronic device readable media. These media can be any available media that can be accessed by the electronic device 600, including volatile and non-volatile media, removable and non-removable media.

[0124] The memory 610 may also include computer system readable media in the form of volatile memory, such as random access memory (RAM) 640 and / or cache memory 650. The server 600 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 660 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 not shown, often referred to as a "hard drive packager"). Although Figure 6 Not shown in the figure, a disk driver package for reading and writing a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk driver package for reading and writing a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each driver package can be connected to the bus 630 via one or more data medium interfaces. The memory 610 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the embodiments of the present invention.

[0125] A program / utility 680 having a set (at least one) of program modules 670 may be stored, for example, in the memory 610, such program modules 670 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 670 generally perform the functions and / or methods of the embodiments described herein.

[0126] The electronic device 600 may also communicate with one or more external devices 690 (e.g., keyboard, pointing device, display 691, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed through an input / output (I / O) interface 692. In addition, the electronic device 600 may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN) and / or public network, such as the Internet) through a network adapter 693. As shown, the network adapter 693 communicates with other modules of the electronic device 600 through a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device driver package, redundant processing unit, external disk driver package array, RAID system, tape driver package, and data backup storage system, etc.

[0127] The processor 620 executes various functional applications and data processing by running the programs stored in the memory 610 .

[0128] It should be noted that the implementation process and technical principles of the electronic device of this embodiment refer to the aforementioned explanation of the lane fault identification method of the embodiment of the present invention, and will not be repeated here.

[0129] In an exemplary embodiment, the present invention further provides a computer-readable storage medium including instructions, such as a memory including instructions, and the above instructions can be executed by a processor of an electronic device to complete the lane fault identification method proposed in any of the above embodiments. Optionally, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0130] In an exemplary embodiment, the present invention further provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the lane fault identification method proposed in any of the above embodiments.

[0131] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art not disclosed by the present invention. The description and examples are to be regarded as exemplary only, and the true scope and spirit of the present invention is indicated by the following claims.

[0132] It should be understood that the present invention is not limited to the exact construction that has been described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for identifying tunnel faults, characterized in that: include: Acquire multiple groups of fault data of the target lane; wherein each group of fault data is composed of a lane identification of a selected fault lane branch and pressure difference data of a ventilation facility selected after the fault of the fault lane branch; Preprocessing the pressure difference data in each group of the fault data to obtain multiple groups of target fault data; wherein the pressure difference data in each group of the target fault data are at the same magnitude; According to the multiple groups of target fault data, a lane fault identification model is trained to obtain a trained target lane fault identification model; wherein the target lane fault identification model has learned the corresponding relationship between the pressure difference data and the fault lane branch; Real-time pressure difference data of ventilation facilities in the target lane are obtained, and based on the real-time pressure difference data, the target lane fault identification model is used to predict the location of the fault lane.

2. The method according to claim 1, characterized in that: The step of obtaining multiple groups of fault data of the target lane includes: Obtaining a simulation experimental ventilation network diagram of the target tunnel; According to the simulation experiment ventilation network diagram, selecting at least one faulty lane branch and at least one ventilation facility from the target lane; The pressure difference data of the at least one ventilation facility after any of the faulty lane branches fails is collected, and the lane identification of the faulty lane branch and the collected pressure difference data of the at least one ventilation facility are taken as a set of fault data.

3. The method according to claim 1, characterized in that The training of the lane fault identification model according to the multiple groups of target fault data to obtain a trained target lane fault identification model includes: Dividing the multiple groups of target fault data into training sets and test sets according to a set ratio; The pressure difference data in each group of the target fault data in the training set is used as the input of the lane fault identification model, and the lane identification of the fault lane branch in each group of the target fault data in the training set is used as the output of the lane fault identification model. The lane fault identification model is trained to obtain the target lane fault identification model.

4. The method according to claim 3, characterized in that The lane fault identification model is a support vector machine model; the pressure difference data in each group of the target fault data in the training set is used as the input of the lane fault identification model, the lane identification of the fault lane branch in each group of the target fault data in the training set is used as the output of the lane fault identification model, and the lane fault identification model is trained to obtain the target lane fault identification model, including: Determine the main influencing parameters of the lane fault identification model and the value range of the main influencing parameters; wherein the main influencing parameters at least include Gaussian kernel width and regularization parameter; According to the value range of the main influencing parameters, a grid search method is used to determine multiple parameter combinations; Inputting the pressure difference data in each group of the target fault data in the training set into the lane fault identification model corresponding to each parameter combination to obtain the lane identification of the fault lane branch corresponding to each group of the target fault data output by the lane fault identification model corresponding to each parameter combination; For the lane fault identification model corresponding to any of the parameter combinations, the accuracy of the lane fault identification model corresponding to the parameter combination is determined according to the difference between the lane identification of the fault lane branch corresponding to each group of the target fault data output by the lane fault identification model corresponding to the parameter combination and the lane identification of the fault lane branch in each group of the target fault data in the training set; The lane fault identification model with the highest accuracy among the lane fault identification models corresponding to the parameter combinations is determined as the target lane fault identification model.

5. The method according to claim 3, characterized in that: The method further comprises: Inputting the pressure difference data in each group of the target fault data in the test set into the target lane fault identification model to obtain the lane identification of the fault lane branch corresponding to each group of the target fault data output by the target lane fault identification model; The performance of the target lane fault identification model is evaluated based on the difference between the lane identification of the faulty lane branch corresponding to each group of the target fault data output by the target lane fault identification model and the lane identification of the faulty lane branch in each group of the target fault data in the test set.

6. The method according to any one of claims 1 to 5, characterized in that The preprocessing of the pressure difference data in each group of the fault data to obtain multiple groups of target fault data includes: The pressure difference data in each group of the fault data are standardized so that the average value of the pressure difference data in each group of the fault data after standardization is a first set value and the variance is a second set value.

7. A tunnel fault identification device, characterized in that: include: An acquisition module is used to acquire multiple groups of fault data of a target lane; wherein each group of fault data is composed of a lane identification of a selected fault lane branch and pressure difference data of a target lane branch corresponding to the fault lane branch after the fault occurs; A preprocessing module, used for preprocessing the pressure difference data in the multiple groups of fault data to obtain multiple groups of target fault data; wherein the pressure difference data in the multiple groups of target fault data are at the same order of magnitude; A training module, used for training a lane fault identification model according to the plurality of sets of target fault data, so as to obtain a trained target lane fault identification model; wherein the target lane fault identification model has learned the corresponding relationship between the ventilation facility monitoring pressure difference data and the fault lane branch; The processing module is used to obtain the real-time pressure difference data of the target lane, and predict the position of the faulty lane using the target lane fault identification model according to the real-time pressure difference data.

8. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method as claimed in any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises a computer program, which implements the method according to any one of claims 1 to 6 when being executed by a processor.

Images