Intelligent Cable Fault Location System for Utility Tunnel

Abstract

The present invention discloses an intelligent cable fault location system for an integrated utility tunnel, which relates to the technical field of cable fault location. The system includes a flaw detection module, an environmental monitoring module, a cable performance information acquisition module, a first neural network model module, a model update module, and a cable fault intelligent prediction module. By setting up the flaw detection module, the environmental monitoring module, the cable performance information acquisition module, the first neural network model module, the model update module, and the cable fault intelligent prediction module, the intelligent cable fault location system for the integrated utility tunnel can analyze the influence of historical environmental change amounts on mechanical damage, thereby analyzing the cable loss rate according to the environmental change amounts. When the predicted cable predicted loss rate is less than the set first loss rate threshold, the cable in the corresponding area is actively detected again, and the cable loss rate prediction model is trained and updated by using the newly added active detection data to ensure the accuracy of the model.

Description

Technical Field

[0001] The present invention relates to the technical field of cable fault location, and particularly to an intelligent cable fault location system for an integrated utility tunnel. Background Art

[0002] An integrated utility tunnel refers to a tunnel space built underground in a city, which integrates various engineering pipelines such as power, communication, heat, and water supply. The above-ground attachments are facilities such as access ports and ventilation ports. In order to ensure the communication and power supply of the city, a large number of cables are laid in the utility tunnel. The large scale and complexity of the cables make it difficult to monitor and locate cable faults.

[0003] The prior art with the publication number CN116540028B discloses an intelligent cable fault location method and system, which relates to the technical field of data processing. It detects the cables in the first target area to obtain the first reflected waveform signal and performs anomaly recognition, intercepts the abnormal waveform signal and identifies the first signal feature, inputs the first signal feature comparison model for comparison to obtain the first feature comparison result; intercepts the first adjacent waveform signal and the second adjacent waveform signal and inputs them into the second signal feature comparison model for comparison, outputs the second feature comparison result, and combines the first feature comparison result to obtain the first fault location result, solving the technical problem in the prior art that the cable fault location method relies too much on the equipment status, resulting in insufficient fault location accuracy and the risk of detection deviation. By segmenting the detection feedback signal, performing targeted independent analysis respectively, and overlapping and proofreading the analysis results for verification, the accuracy of fault detection and location can be effectively improved.

[0004] However, the above prior art needs to actively emit flaw detection signals to obtain feedback signals for analysis, and obtain the detection results to judge and locate faults accordingly. Active emission of flaw detection signals and reception of feedback signals require relevant equipment. Since cables generally have a long length, in order to ensure the monitoring accuracy, usually only a certain section of the cable is selected for flaw detection during signal flaw detection. Therefore, to complete the monitoring of all cables, either a large number of flaw detection devices are needed to detect each section of the cable, or the flaw detection device is used to detect each section of the cable in turn, which requires a large economic cost or time cost; and maintaining the operation of the flaw detection device also requires a large economic cost, making it difficult to achieve real-time monitoring of the entire section of the cable, reducing the timeliness of cable fault monitoring and location. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent cable fault location system for an integrated utility tunnel to solve the above deficiencies in the prior art.

[0006] To achieve the above object, the present invention provides the following technical solutions: An intelligent cable fault location system for an integrated pipe gallery, comprising a flaw detection module, an environmental monitoring module, a cable performance information acquisition module, a first neural network model module, a model update module, and a cable fault intelligent prediction module;

[0007] The flaw detection module is used to actively detect the area of the faulty cable identified by the fault identification model based on detection equipment, and obtain active detection data, where the active detection includes detecting by emitting ultrasonic waves and electromagnetic waves, and is used to improve the monitoring accuracy of cable mechanical damage;

[0008] The cable performance information acquisition module is used to obtain the cable usage time and the current time, and calculate the cable usage duration based on the current time and the cable usage time;

[0009] The environmental monitoring module is used to monitor the environment of the pipe gallery where the cable is located, and obtain environmental data, where the environmental data includes temperature, humidity, and dust content in the air;

[0010] The first neural network model module is used to obtain the cable usage duration at the time of cable fault, and based on the cable usage duration at the time of cable fault and the active detection data, obtain the cable loss rate. Then, based on the environmental data and the cable loss rate, train the first deep neural network model to obtain a cable loss rate prediction model, which is used to output the predicted cable loss rate based on the input environmental data. The cable loss rate can obtain the mechanical damage of the cable through the active detection data, and then calculate the cable loss rate based on the mechanical damage data of the cable and the size data when the corresponding item of the mechanical damage is intact. For example, if a 0.1 - millimeter crack is detected on the surface of the cable insulation layer, and the surface thickness of the cable insulation layer is 0.8 millimeters, the corresponding cable loss rate at this point is

[0011] The model update module is used to control the flaw detection module to actively detect the area where the cable part with a predicted loss rate less than the set first loss rate threshold is located when the predicted cable loss rate is less than the set first loss rate threshold, obtain new active detection data, and control the first neural network model module to train and update the cable loss rate prediction model based on the new active detection data;

[0012] The cable fault intelligent prediction module is used to give an early warning to the cable in the area where the cable part with a predicted loss rate less than the set second loss rate threshold is located when the predicted cable loss rate is less than the set second loss rate threshold.

[0013] Further, the system further includes a region division module, which is used to divide the cables into regions and set a unique identification code associated with each region. The cables can be divided into regions according to length distances such as the length of the cables. For example, the cables within 0 - 500 (excluding) m from the starting position of the pipe gallery are one region, and the cables within 500 - 1000 (excluding) m are one region, etc.

[0014] Further, the environmental monitoring module is also used to process the environmental data to obtain the change amount of the environmental data in each region, specifically including the following steps:

[0015] Respectively obtain the historical environmental data of each region;

[0016] Draw the change curves of the historical environmental data of each region. The abscissa of each change curve is time, and the ordinate can be temperature, humidity, dust content in the air, etc.;

[0017] Traverse the change curves to determine whether the positive or negative of the slope of the change curves changes;

[0018] If so, determine whether the change amount of the ordinate of the change curve after the slope change is greater than the set corresponding change amount threshold;

[0019] If so, mark the point when the slope change is detected as a node; for example, when the change curve is a temperature change curve, if the initial slope of the temperature change curve is negative, and if it is detected that the slope of the temperature change curve at a certain time is positive, then determine whether the change amount of the temperature of the continuous temperature change curve with a positive slope after the slope change is greater than the set corresponding temperature change amount threshold. If so, mark a node at the point where the slope of the temperature change curve becomes positive;

[0020] Divide the change curves according to the nodes to obtain multiple change periods;

[0021] Calculate the maximum value, minimum value, and duration of each change period.

[0022] Further, the first neural network model module trains a first deep neural network model based on environmental data and cable loss rate to obtain a cable loss rate prediction model, including the following steps:

[0023] Obtain the maximum value, minimum value, and duration of each change period of each region, and calculate the corresponding environmental data change amount;

[0024] Based on the environmental data, the change amount of environmental data, and the cable loss rate for each change period in each region, train the first deep neural network model to obtain a cable loss rate prediction model, which is used to obtain the change amount of environmental data for each change period in each region based on the input environmental data, and based on the environmental data and the change amount of environmental data, output the predicted cable loss rate of the cables in each region.

[0025] Further, the system further includes a real-time monitoring module, a database module, and a second neural network model module;

[0026] The real-time monitoring module is used to monitor the electrical parameter data of the cable in real time, where the electrical parameter data includes voltage, current, frequency, etc. The voltage, current, and frequency are not only the voltage, current, and frequency of the signal transmitted in the cable itself, but also include, for example, monitoring high-frequency pulse current through a high-frequency current sensor or a coupling capacitor; monitoring high-frequency electromagnetic waves through a radio frequency antenna or a receiver; detecting changes in the voltage waveform through a high-frequency voltage sensor; monitoring the charge accumulated on the surface of the cable or equipment through a charge-coupled sensor; monitoring the vibration or sound characteristics of the cable through a vibration sensor or an acoustic sensor;

[0027] The database module is used to obtain the historical electrical parameter data of the cable;

[0028] The second neural network model module is used to set a fault label, use the fault label to mark the historical electrical parameter data, and based on the historical electrical parameter data and the fault label, train the second deep neural network model to obtain a fault identification model, which is used to identify and output the electrical parameter data and the fault label representing the cable fault based on the input electrical parameter data.

[0029] Further, the model update module is further used to, when the second neural network prediction model module outputs a fault label, obtain the region where the cable part corresponding to the fault label is located, control the flaw detection module to perform active detection to obtain new active detection data, and control the first neural network model module to train and update the cable loss rate prediction model based on the new active detection data.

[0030] Further, the model update module is further used to compare the cable loss rate calculated from the new active detection data with the corresponding predicted cable loss rate, and determine whether the error exceeds the set loss rate error threshold. If so, it is determined that there is an error in the cable loss rate prediction model; if not, it is determined as a cable fault not caused by mechanical damage.

[0031] Further, the cable fault intelligent prediction module is further used to give an early warning to the region where the cable part corresponding to the fault label is located when the second neural network model outputs a fault label.

[0032] 1. Compared with the prior art, an intelligent cable fault location system for an integrated utility tunnel provided by the present invention can measure mechanical damage to the cable by setting up a flaw detection module, an environmental monitoring module, a cable performance information acquisition module, a first neural network model module, a model update module, and a cable fault intelligent prediction module. Then, based on the historical environmental data of each regional section in the integrated utility tunnel obtained, the historical environmental change amount of each regional section can be obtained, so that the influence of the historical environmental change amount on the mechanical damage can be analyzed, and thus the cable loss rate can be analyzed according to the environmental change amount.

[0033] 2. Compared with the prior art, an intelligent cable fault location system for an integrated utility tunnel provided by the present invention sets up a model update module. When the predicted cable loss rate is less than the set first loss rate threshold, the cable in the corresponding area is actively detected again, and the cable loss rate prediction model is trained and updated using the newly added active detection data to ensure the accuracy of the model.

[0034] 3. Compared with the prior art, an intelligent cable fault location system for an integrated utility tunnel provided by the present invention can, by setting up a real-time monitoring module, a database module, and a second neural network model module, detect the voltage, current, and frequency of the cable in real time, input them into a fault identification model to locate the existing faults, and actively detect the cable in the fault area. The cable loss rate prediction model is trained and updated using the newly added active detection data to further ensure the accuracy of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0036] Figure 1 It is a schematic diagram of the system structure provided by the embodiment of the present invention;

[0037] Figure 2 It is another schematic diagram of the system structure provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the present invention will be further described in detail below in conjunction with the drawings.

[0039] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the present invention.

[0040] In addition, the terms "first" and "second" are only used for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present invention, the meaning of "a plurality" is two or more, unless otherwise specifically defined. In addition, the terms "mounted", "connected", and "coupled" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0041] Example embodiments will be described more fully hereinafter with reference to the accompanying drawings, but the example embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0042] In the case of no conflict, the various embodiments of the present disclosure and the features in the embodiments may be combined with each other.

[0043] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0044] The terms used herein are only for the purpose of describing particular embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a" and "the" are also intended to include the plural forms unless the context clearly indicates otherwise. It will also be understood that when the terms "comprises" and / or "consists of" are used in this specification, it specifies the presence of the stated features, integers, steps, operations, elements, and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0045] Embodiments described herein may be described with reference to plan views and / or cross-sectional views by means of ideal schematic diagrams of the present disclosure. Therefore, the example illustrations may be modified according to manufacturing techniques and / or tolerances. Therefore, the embodiments are not limited to the embodiments shown in the drawings, but include modifications to the configurations formed based on manufacturing processes. Therefore, the regions illustrated in the drawings have schematic properties, and the shapes of the regions shown in the figures illustrate the specific shapes of the regions of the components, but are not intended to be restrictive.

[0046] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless expressly so defined herein.

[0047] Please refer to Figure 1 - Figure 2 , an intelligent cable fault location system for an integrated pipe gallery, including a flaw detection module, an environmental monitoring module, a cable performance information acquisition module, a first neural network model module, a model update module, and a cable fault intelligent prediction module;

[0048] The flaw detection module is used to actively detect the area of the faulty cable identified by the fault identification model based on detection equipment, and obtain active detection data, where the active detection includes detecting by emitting ultrasonic waves and electromagnetic waves, and is used to improve the monitoring accuracy of cable mechanical damage;

[0049] The cable performance information acquisition module is used to obtain the cable usage time and the current time, and calculate the cable usage duration based on the current time and the cable usage time;

[0050] The environmental monitoring module is used to monitor the environment of the pipe gallery where the cable is located, and obtain environmental data, where the environmental data includes temperature, humidity, and dust content in the air; the environmental monitoring module is also used to process the environmental data to obtain the change amount of the environmental data in each area, specifically including the following steps:

[0051] (1) Obtain the historical environmental data of each area respectively;

[0052] (2) Draw the change curves of the historical environmental data of each area, where the abscissa of each change curve is time, and the ordinate can be temperature, humidity, dust content in the air, etc.;

[0053] (3) Traverse the change curves and determine whether the positive or negative of the slope of the change curve changes;

[0054] (4) If so, determine whether the change amount of the ordinate of the change curve after the slope change is greater than the set corresponding change amount threshold;

[0055] (5) If so, mark the point when the slope changes as a node; in one embodiment, the change curve is a temperature change curve. If the initial slope of the temperature change curve is negative, and if it is detected that the slope of the temperature change curve at a certain time is positive, then determine whether the change amount of the temperature of the temperature change curve with a positive and continuous slope after the slope change is greater than the set corresponding temperature change amount threshold. If so, mark a node at the point where the slope of the temperature change curve becomes positive;

[0056] (6) Divide the change curve according to the nodes to obtain multiple change periods;

[0057] (7) Calculate the maximum value, minimum value, and duration of each change period.

[0058] The first neural network model module is used to obtain the cable usage duration during cable faults, and based on the cable usage duration during cable faults and active detection data, obtain the cable loss rate. Then, based on the environmental data and the cable loss rate, train the first deep neural network model to obtain a cable loss rate prediction model, which is used to output the predicted cable loss rate based on the input environmental data. The cable loss rate can obtain the mechanical damage of the cable through active detection data, and then based on the mechanical damage data of the cable and the size data when the corresponding items of the mechanical damage are intact, calculate the loss rate of the cable; in one embodiment, it is detected that there is a 0.1 - millimeter crack on the surface of the cable insulation layer, and the surface thickness of the cable insulation layer is 0.8 millimeters, then the corresponding cable loss rate at this place is including the following steps:

[0059] (1) Obtain the maximum value, minimum value, and duration of each change period in each area, and calculate the corresponding environmental data change amount;

[0060] (2) Based on the environmental data, environmental data change amount, and cable loss rate of each change period in each area, train the first deep neural network model to obtain a cable loss rate prediction model, which is used to obtain the environmental data change amount of each change period in each area based on the input environmental data, and based on the environmental data and the environmental data change amount, output the predicted cable loss rate of the cables in each area. Because the environment changes repeatedly, such as the repeated increase and decrease of temperature and humidity, it will accelerate the aging and loss of the cable's external protective layer, such as the insulation layer, to a certain extent. Therefore, the loss rate of the cable's external protective layer can be predicted by analyzing the environmental data change amount to prevent cable failures caused by damage to the external protective layer, such as insulation layer damage leading to a decrease in insulation performance, etc.

[0061] The model update module is used to control the flaw detection module to actively detect the area where the cable part with a predicted loss rate less than the set first loss rate threshold is located when the predicted loss rate of the cable is less than the set first loss rate threshold, obtain new active detection data, and control the first neural network model module to train and update the cable loss rate prediction model based on the new active detection data;

[0062] The cable fault intelligent prediction module is used to give an early warning for the cable in the area where the cable part with a predicted loss rate less than the set second loss rate threshold is located when the predicted loss rate of the cable is less than the set second loss rate threshold.

[0063] The system further includes a region division module, which is used to divide the cable into regions and set a unique identification code associated with each region. In one embodiment, the cable can be divided into regions according to a length distance such as the length of the cable. For example, the cable from 0 - 500 (excluding) m from the starting point of the pipe gallery is one region, and the cable from 500 - 1000 (excluding) m is one region, etc. Thus, when outputting the region where the corresponding cable part is located, the identification code corresponding to the region can be output.

[0064] The system further includes a real - time monitoring module, a database module, and a second neural network model module;

[0065] The real - time monitoring module is used to monitor the electrical parameter data of the cable in real time. The electrical parameter data includes voltage, current, frequency, etc. The voltage, current, and frequency are not only the voltage, current, and frequency of the signal transmitted in the cable itself, but also include, for example, monitoring high - frequency pulse current through a high - frequency current sensor or a coupling capacitor; monitoring high - frequency electromagnetic waves through a radio - frequency antenna or a receiver; detecting changes in the voltage waveform through a high - frequency voltage sensor; monitoring the charge accumulated on the surface of the cable or equipment through a charge - coupled sensor; monitoring the vibration or sound characteristics of the cable through a vibration sensor or an acoustic sensor;

[0066] The database module is used to obtain the historical electrical parameter data of the cable;

[0067] The second neural network model module is used to set fault labels, mark the historical electrical parameter data with the fault labels, train the second deep neural network model based on the historical electrical parameter data and the fault labels, and obtain a fault identification model for identifying and outputting the electrical parameter data and fault labels indicating cable faults based on the input electrical parameter data.

[0068] The model update module is further configured to, when the second neural network prediction model module outputs a fault label, obtain the area where the cable part corresponding to the fault label is located, control the flaw detection module to perform active detection, obtain new active detection data, and control the first neural network model module to train and update the cable loss rate prediction model based on the new active detection data. Then, compare the cable loss rate calculated from the new active detection data with the corresponding predicted cable loss rate to determine whether the error exceeds the set loss rate error threshold. If so, it is determined that there is an error in the cable loss rate prediction model; if not, it is determined that the cable fault is not caused by mechanical damage.

[0069] The cable fault intelligent prediction module is further configured to issue a warning for the area where the cable part corresponding to the fault label is located when the second neural network model outputs a fault label.

[0070] Only some exemplary embodiments of the present invention have been described by way of illustration. Undoubtedly, for those of ordinary skill in the art, the described embodiments can be modified in various different ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An intelligent positioning system for cable faults in an integrated pipe gallery, characterized in that: It includes a flaw detection module, an environmental monitoring module, a cable performance information acquisition module, a first neural network model module, a model update module, and a cable fault intelligent prediction module; The flaw detection module is used to actively detect the area of ​​the faulty cable identified by the fault identification model based on the detection equipment to obtain active detection data; The cable performance information acquisition module is used to acquire the cable usage time and the current time, and calculate the cable usage time based on the current time and the cable usage time; The environmental monitoring module is used to monitor the environment of the pipe gallery where the cable is located to obtain environmental data, which includes temperature, humidity, and air dust content; The first neural network model module is used to obtain the cable usage time when the cable fails, and obtain the cable loss rate based on the cable usage time and active detection data when the cable fails, and then train the first deep neural network model based on the environmental data and the cable loss rate to obtain a cable loss rate prediction model, which is used to output a predicted cable loss rate based on the input environmental data; The model updating module is used to control the flaw detection module to actively detect the area where the cable portion whose predicted loss rate is less than the set first loss rate threshold is located when the predicted loss rate of the cable is less than the set first loss rate threshold, to obtain newly added active detection data, and control the first neural network model module to train and update the cable loss rate prediction model based on the newly added active detection data; The cable fault intelligent prediction module is used to issue an early warning to the cables in the area where the cable portion whose predicted loss rate is less than the set second loss rate threshold is located when the predicted loss rate of the cable is less than the set second loss rate threshold; The system further comprises a region division module, the region division module being used to divide the cables into regions and to set a unique identification code associated with each region; The environmental monitoring module is also used to process the environmental data to obtain the change amount of the environmental data of each area, which specifically includes the following steps: Obtain historical environmental data for each region separately; Draw the change curves of each historical environmental data of each region, and the horizontal axis of each change curve is time; Traverse the change curve to determine whether the slope of the change curve changes; If yes, then determine whether the change in the ordinate of the change curve after the slope changes is greater than the set corresponding change threshold; If so, mark the point where the slope changes as a node; Segment the change curve according to the nodes to obtain multiple change periods; Calculate the maximum value, minimum value and duration of each change period; The first neural network model module trains a first deep neural network model based on environmental data and cable loss rate to obtain a cable loss rate prediction model, including the following steps: Obtain the maximum value, minimum value, and duration of each change period in each area, and calculate the corresponding environmental data change; Based on the environmental data of each changing period in each region, the change in environmental data and the cable loss rate, a first deep neural network model is trained to obtain a cable loss rate prediction model, which is used to obtain the change in environmental data of each changing period in each region based on the input environmental data, and output the predicted cable loss rate of the cables in each region based on the environmental data and the change in environmental data.

2. According to claim 1, a comprehensive pipe gallery cable fault intelligent positioning system is characterized by: The system also includes a real-time monitoring module, a database module, and a second neural network model module; The real-time monitoring module is used to monitor the electrical parameter data of the cable in real time; The database module is used to obtain historical electrical parameter data of the cable; The second neural network model module is used to set a fault label, use the fault label to mark the historical electrical parameter data, and train a second deep neural network model based on the historical electrical parameter data and the fault label to obtain a fault identification model, which is used to identify and output electrical parameter data and fault labels indicating cable faults based on the input electrical parameter data.

3. According to claim 2, a comprehensive pipe gallery cable fault intelligent positioning system is characterized by: The model updating module is also used to obtain the area where the cable part corresponding to the fault label is located when the second neural network prediction model module outputs the fault label, control the flaw detection module to perform active detection to obtain newly added active detection data, and control the first neural network model module to train and update the cable loss rate prediction model based on the newly added active detection data.

4. According to claim 3, the intelligent positioning system for cable faults in an integrated pipe gallery is characterized in that: The model update module is also used to compare the cable loss rate calculated by the newly added active detection data with the corresponding predicted cable loss rate to determine whether the error exceeds the set loss rate error threshold. If so, it is determined that there is an error in the cable loss rate prediction model; if not, it is determined that the cable fault is caused by non-mechanical damage.

5. According to claim 2, the intelligent positioning system for cable faults in an integrated pipe gallery is characterized in that: The cable fault intelligent prediction module is also used to issue an early warning to the area where the cable part corresponding to the fault label is located when the second neural network model outputs the fault label.

Images