Mining cable fault monitoring system and data processing method thereof

Abstract

The invention relates to a mining cable fault monitoring system and a data processing method thereof. The monitoring system comprises a detection optical fiber which is laid on the cable in parallel and used for detecting cable information of the cable. The detection host is connected with the detection optical fiber, a light emitter and a demodulation unit are arranged in the detection host, the light emitter is used for emitting detection light to the detection optical fiber, the detection light is used for converting the cable information into optical information, and the demodulation unit is used for obtaining the cable information according to the optical information; and the industrial personal computer is used for receiving the cable information acquired by the detection host, an expert fault diagnosis system is arranged in the industrial personal computer, and the expert fault diagnosis system is used for judging whether abnormal information and fault types corresponding to the abnormal information exist in the cable information or not. The mining cable fault monitoring system has the advantages of being high in detection precision, wide in working frequency band, small in transmission loss, long in measurement distance, free of electromagnetic interference and capable of being miniaturized.

Description

Technical Field

[0001] The present invention relates to the technical field of mining cables; in particular, the present invention relates to a mining cable fault monitoring system and a data processing method thereof. Background Art

[0002] With the continuous updating of equipment in my country's coal mining working faces, the number of super-large mines has increased year by year. In order to ensure the high production and efficiency of the entire fully-mechanized mining system, whether the power supply system of the fully-mechanized mining working face is reasonable is a crucial part of the entire fully-mechanized mining system, which is related to the safety and reliability of the entire mine power grid. With the increasing number of high-height and long-distance working faces year by year, the safe power supply of coal mines faces many new challenges. In order to ensure the safe operation of the power system and reduce the various losses caused by power outages, the speed and accuracy of the fault monitoring and positioning of the transmission line in the ultra-long working face are particularly important.

[0003] At present, the monitoring system for long-distance power supply lines in fully mechanized mining working faces has low positioning speed and accuracy, and cannot automatically conduct unmanned real-time monitoring. Summary of the invention

[0004] In view of this, the present invention provides a mining cable fault monitoring system and a data processing method thereof, so as to solve or at least alleviate one or more of the above-mentioned problems and other problems existing in the prior art.

[0005] In order to achieve the above-mentioned object, a first aspect of the present invention provides a mining cable fault monitoring system, wherein the monitoring system comprises: A detection optical fiber, which is laid parallel to the cable and is used to detect cable information of the cable; A detection host, the detection host is connected to the detection optical fiber, and an optical transmitter and a demodulation unit are arranged in the detection host, the optical transmitter is used to transmit detection light to the detection optical fiber, the detection light is used to convert the cable information into optical information, and the demodulation unit is used to obtain the cable information according to the optical information; An industrial computer, the industrial computer is used to receive the cable information obtained by the detection host, the industrial computer is provided with an expert fault diagnosis system, the expert fault diagnosis system is provided with a detection model, the expert fault diagnosis system is used to extract features from the cable information received by the industrial computer, and determine through the detection model whether there is abnormal information in the cable information after feature extraction and the fault type corresponding to the abnormal information.

[0006] In the monitoring system as described above, optionally, the monitoring system also includes a switch, the industrial computer is located on the ground, the detection optical fiber, the detection host and the switch are located in a coal mine, the switch is connected to the detection host, and the switch is used to realize data exchange between the industrial computer and the detection host.

[0007] In the monitoring system as described above, optionally, the detection optical fiber includes an acoustic wave detection optical fiber and a temperature detection optical fiber, the cable is located in a cable detection area, the cable information includes the temperature of the cable surface and the acoustic waves in the cable detection area, and the detection host includes a DAS host and a DTS host, the DAS host is connected to the acoustic wave detection optical fiber, and the DTS host is connected to the temperature detection optical fiber.

[0008] In the monitoring system as described above, optionally, the demodulation unit includes a DAS demodulation unit and a DTS demodulation unit, the DAS demodulation unit includes a first acousto-optic modulator, a first erbium-doped fiber amplifier, a first circulator, a first fiber coupler and a first photoelectric balanced detector, and the DTS demodulation unit includes a second acousto-optic modulator, a second erbium-doped fiber amplifier, a second circulator, a second fiber coupler and a second photoelectric balanced detector.

[0009] In the monitoring system as described above, optionally, an alarm module is also provided in the expert fault diagnosis system, and the alarm module is used to generate a historical alarm record according to the abnormal information, and the historical alarm record is used to determine the fault type corresponding to the abnormal information and the location where the fault occurs.

[0010] In the monitoring system as described above, optionally, a big data prediction model is also provided in the industrial computer, a neural network decay model is provided in the big data prediction model, the abnormal information includes sound wave abnormal information and temperature abnormal information, and the neural network decay model is used to predict the remaining service life of the cable based on the sound wave abnormal information and the historical alarm records.

[0011] In the monitoring system as described above, optionally, the detection model includes a binary classification model and a multi-classification model, the binary classification model is used to determine whether there is abnormal information in the cable information after feature extraction, and the multi-classification model is used to determine the fault type corresponding to the abnormal information.

[0012] In the monitoring system as described above, optionally, the detection optical fiber is fixedly laid on the cable by rolling tape, or the detection optical fiber and the cable are aggregated by a tinned copper mesh braided layer.

[0013] In order to achieve the aforementioned object, the second aspect of the present invention provides a data processing method of the mining cable fault monitoring system as described in the first aspect, wherein the data processing method comprises: Step A: obtaining cable information during the working process of the cable and integrating it into a cable information data set, and training the detection model through the cable information data set; Step B: the detection host transmits the cable information to the industrial computer, the expert fault diagnosis system preprocesses and extracts features of the cable information, the detection model determines whether there is abnormal information in the cable information through the binary classification model, and determines the fault type corresponding to the abnormal information through the multi-classification model; Step C: When there is abnormal information in the cable information, the alarm module receives the abnormal information and matches the abnormal information with the historical alarm record. When the match is successful, the alarm module outputs the abnormal information, the fault type corresponding to the abnormal information and the location where the fault occurs. When the match fails, the alarm module issues an alarm and updates the abnormal information to the historical alarm record. Step D: The big data prediction model extracts the sound wave abnormality information, and the neural network decay model predicts the remaining service life of the cable based on the sound wave abnormality information and the historical alarm records.

[0014] In the data processing method described above, optionally, the preprocessing includes denoising and smoothing filtering, the feature extraction includes time domain feature extraction and frequency domain feature extraction, the time domain features include amplitude, zero crossing rate, skewness and fuzzy entropy, the frequency domain features include frequency domain kurtosis, frequency domain skewness and energy concentration, and the detection model is a SVM model.

[0015] The present invention provides a mining cable fault monitoring system and a data processing method thereof. The mining cable fault monitoring system of the present invention is based on a distributed optical fiber sensor design, has the advantages of high detection accuracy, wide working frequency band, low transmission loss, long measurement distance, freedom from electromagnetic interference and miniaturization, can be used for precise monitoring in complex environments such as narrow space, strong electromagnetic interference, high temperature, high pressure, strong corrosion, strong radiation, etc., and can realize automatic unmanned real-time monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The disclosure of the present invention will become more apparent with reference to the accompanying drawings. It should be understood that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present invention. In the drawings: Figure 1 It is a structural schematic diagram of an embodiment of a mining cable fault monitoring system of the present invention; Figure 2A schematic block diagram of signal flow in a mining cable fault monitoring system of the present invention; and Figure 3 The figure is a flow chart of an embodiment of the data processing method of the present invention. DETAILED DESCRIPTION

[0017] With reference to the accompanying drawings and specific embodiments, the structure, composition, characteristics and advantages of a mining cable fault monitoring system and a data processing method thereof of the present invention will be described below in an exemplary manner. However, all descriptions should not be used to form any limitation on the present invention.

[0018] In addition, for any single technical feature described or implied in the embodiments mentioned in this document, or any single technical feature displayed or implied in the accompanying drawings, the present invention still allows for continued arbitrary combination or deletion between these technical features (or their equivalents) without any technical obstacles, and thus it should be considered that these more embodiments according to the present invention are also within the scope of the description of this document.

[0019] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features.

[0020] Figure 1 The present invention is a schematic diagram of the structure of an embodiment of a mining cable fault monitoring system of the present invention. In this embodiment, the monitoring system may include a detection optical fiber, a detection host and an industrial computer. The detection optical fiber is laid in parallel with the cable and is used to detect the cable information of the cable. The detection host is connected to the detection optical fiber, and an optical transmitter and a demodulation unit are provided in the detection host. The optical transmitter is used to transmit detection light to the detection optical fiber, and the detection light is used to convert the cable information into optical information. The demodulation unit is used to obtain the cable information according to the optical information. The cable information includes the temperature of the cable surface and the sound waves in the cable detection area.

[0021] The industrial computer is used to receive the cable information obtained by the detection host. The industrial computer of this embodiment is provided with an expert fault diagnosis system, and the expert fault diagnosis system is provided with a detection model. The expert fault diagnosis system is used to extract features of the cable information received by the industrial computer, and determine whether there is abnormal information and the fault type corresponding to the abnormal information in the cable information after feature extraction through the detection model.

[0022] In addition, the industrial computer is also equipped with a big data prediction model, which includes a neural network decay model. The abnormal information includes abnormal sound wave information and abnormal temperature information. The neural network decay model is used to predict the remaining service life of the cable based on the abnormal sound wave information and historical alarm records. The detection model includes a binary classification model and a multi-classification model. The binary classification model is used to determine whether there is abnormal information in the cable information after feature extraction, and the multi-classification model is used to determine the fault type corresponding to the abnormal information.

[0023] from Figure 1 It can be seen that the monitoring system also includes a switch. The mining cable detection system of the present invention is applied to cables in coal mines, so the detection optical fiber, detection host and switch are all located in the coal mine, and the industrial computer is located above the ground. In order to realize data exchange between the industrial computer and the detection host, the present invention is provided with a switch. If there are multiple cables to be monitored, it is necessary to arrange the detection host on each corresponding cable line. The switch will adopt a ring network switch, and the output signal of the detection host will eventually be transmitted to the same industrial computer via the underground ring network.

[0024] The detection optical fiber includes an acoustic wave detection optical fiber and a temperature detection optical fiber. The cable is located in a cable detection area. The cable information includes the temperature of the cable surface and the acoustic waves in the cable detection area. The detection host includes a DAS host and a DTS host. The DAS host is connected to the acoustic wave detection optical fiber, and the DTS host is connected to the temperature detection optical fiber.

[0025] The demodulation unit includes a DAS demodulation unit and a DTS demodulation unit. The DAS demodulation unit includes a first acousto-optic modulator, a first erbium-doped fiber amplifier, a first circulator, a first fiber coupler and a first photoelectric balanced detector. The DTS demodulation unit includes a second acousto-optic modulator, a second erbium-doped fiber amplifier, a second circulator, a second fiber coupler and a second photoelectric balanced detector. The optical transmitter includes a DAS optical transmitter and a DTS optical transmitter. The DAS optical transmitter is used to emit a first detection light, and the DTS optical transmitter is used to emit a second detection light.

[0026] Specifically, the DAS host inputs the first detection light into the first acousto-optic modulator through the DAS optical transmitter. After being amplified by the first erbium-doped fiber amplifier, the first detection light is injected into the acoustic wave detection optical fiber through the first circulator. The first detection light continuously generates back Rayleigh scattered light during the transmission of the acoustic wave detection optical fiber. The back Rayleigh scattered light and the first detection light emitted by the DAS optical transmitter are coupled by the first optical fiber coupler and converted into an electrical signal by the first photoelectric balanced detector. At this time, the DAS host will use a high-speed acquisition card to collect the electrical signal to obtain the acoustic wave information of the cable detection area.

[0027] The DTS host inputs the second detection light into the second acousto-optic modulator through the DTS optical transmitter. The modulated second detection light is injected into the temperature detection optical fiber through the third circulator. The second detection light generates back Raman scattered light during the transmission of the temperature detection optical fiber. The back Raman scattered light and the second detection light emitted by the DTS optical transmitter are coupled through the second coupler, and then the Raman scattering is divided into reference light and back scattered light carrying temperature information (Stokes light intensity signal and anti-Stokes light), which are converted into electrical signals through photodiodes. The DTS host calculates the light intensity of the reference light and the back scattered light (the light intensity of Stokes light and anti-Stokes light), and then obtains the temperature monitoring information of the cable through light intensity information demodulation calculation.

[0028] The operation of fiber Bragg grating temperature sensor is based on the characteristic that the fiber Bragg grating pitch changes with temperature. When the temperature changes, the grating pitch in the fiber Bragg grating will change, causing the wavelength of the reflected wave to change accordingly. This change can be detected and quantified by measuring and analyzing the wavelength of the reflected wave. At this time, if the fiber Bragg grating is not affected by external stress and there is no strain caused by external force, that is, Δε=0 in the formula ΔλB=λB(1-Pe)Δε+λB(αf-ξ)ΔT, then the temperature change ΔT can be calculated according to the formula ΔλB=λB(αf-ξ)ΔT. When the initial temperature is known, the current temperature can be calculated to achieve temperature measurement.

[0029] The operation of fiber Bragg grating acoustic wave sensor is mainly based on the change of the reflected wavelength of fiber Bragg grating (FBG) caused by ultrasound. When the sound wave acts on the fiber Bragg grating, it will cause the change of the grating period or the refractive index of the core, resulting in the change of the wavelength of the reflected light. This change is directly related to the characteristics of ultrasound (such as frequency, amplitude, etc.). By measuring and analyzing the change of the wavelength of the reflected light, the detection and measurement of ultrasound can be achieved.

[0030] Figure 2 FIG. 1 is a schematic block diagram of signal flow in a mining cable fault monitoring system of the present invention. Figure 2 It can be seen that the monitoring system can include a signal acquisition layer, a signal processing layer, a transmission layer and a decision layer. The optical signal in the figure refers to the above-mentioned backscattered Rayleigh light, backscattered Raman light and the first detection light and the second detection light coupled with them respectively. The signal acquisition layer is used to collect signals, that is, the above-mentioned optical transmitter emits detection light and then converts the cable information into optical information; the signal processing layer is completed by the detection host, and the detection host processes the collected signals through the demodulation unit and the data acquisition card; the transmission layer is completed by the switch or the ring network switch; the decision layer is completed by the big data prediction model and expert fault diagnosis system set inside the industrial computer.

[0031] Specifically, in the signal acquisition layer, the detection light in the temperature detection fiber and the acoustic wave monitoring fiber interacts with the external measured parameters, causing changes in the optical properties of the light, such as intensity, wavelength, frequency, phase and polarization state, so that the physical quantity to be measured is transmitted in the form of light waves. Then it is analyzed in the signal processing layer, and the analysis results are finally transmitted from the transmission layer to the expert fault diagnosis system and the big data model prediction system of the decision-making layer to determine whether there is a current fault and whether there is a potential fault, and output the type and location of the current fault and predicted fault.

[0032] It should be noted that an alarm module is also provided in the expert fault diagnosis system. The alarm module is used to generate historical alarm records according to abnormal information. The historical alarm records are used to determine the fault type corresponding to the abnormal information and the location where the fault occurred.

[0033] In addition, since the cable fault monitoring system of the present invention is used for cables in coal mines, there is a lot of electromagnetic interference in the coal mine environment. Therefore, the detection optical fiber of the present invention is fixed to the cable by rolling, or the detection optical fiber and the cable are aggregated by a tinned copper mesh braided layer to prevent electromagnetic interference.

[0034] The present invention also proposes a data processing method for the above-mentioned mining cable fault monitoring system. Figure 3 FIG. 1 is a flow chart of an embodiment of the data processing method of the present invention. Figure 3 It can be seen that the data processing method can include four steps: model training, feature extraction, fault judgment and life prediction.

[0035] Next, each step is described in detail.

[0036] Step A is used for model training. In step A, firstly, the cable information during the cable working process can be acquired at the cable working site by means of sensor detection and the like, and integrated into a cable information data set, and the detection model is trained by means of the cable information data set.

[0037] Step B is used to extract features. After the detection host transmits the cable information to the industrial computer, the expert fault diagnosis system preprocesses and extracts features from the cable information. The detection model uses a binary classification model to determine whether there is abnormal information in the cable information, and uses a multi-classification model to determine the fault type corresponding to the abnormal information.

[0038] It should be noted here that preprocessing may include denoising and smoothing filtering, feature extraction may include time domain feature extraction and frequency domain feature extraction, time domain features may include amplitude, zero crossing rate, skewness and fuzzy entropy, frequency domain features include frequency domain kurtosis, frequency domain skewness and energy concentration, and the detection model is a SVM model.

[0039] Step C is used to determine the fault. When there is abnormal information in the cable information, the alarm module receives the abnormal information and matches the abnormal information with the historical alarm record. When the match is successful, the alarm module outputs the abnormal information, the fault type corresponding to the abnormal information and the location where the fault occurred. When the match fails, the alarm module issues an alarm and updates the abnormal information to the historical alarm record.

[0040] Step D is used to predict the service life. The big data prediction model extracts the abnormal sound wave information, and the neural network decay model predicts the remaining service life of the cable based on the abnormal sound wave information and historical alarm records.

[0041] It should be noted here that the big data model prediction system will establish a historical database of acoustic wave data, and use it to establish a neural network decay model to analyze changes in the structure of insulating materials, thereby detecting problems such as cracks, bubbles, moisture, etc. in the insulating materials, reflecting abnormal conditions of the insulating materials of the power supply lines, and taking timely maintenance measures to reduce the risk of accidents.

[0042] The neural network decay model of the present invention is described here. The neural network decay model of the present invention may include an input layer, a hidden layer, and an output layer. In the input layer, the acoustic wave data is cleaned, denoised, normalized, and other processes to eliminate noise and dimensional inconsistency, and then feature extraction is performed on the data to extract key features related to changes in the performance of the insulating material; in the hidden layer, considering the complexity and nonlinearity of the changes in the performance of the insulating material, a recurrent neural network (RNN) is selected to construct a complex time series model to effectively process time series data and capture long-term dependencies in the data; in the output layer, a sigmoid function is used to perform multi-classification tasks to determine the problems of the cable material (such as cracks, bubbles, moisture, etc.). If the prediction result exceeds the preset threshold, an abnormal alarm is triggered so that maintenance or replacement measures can be taken in time.

[0043] The big data model prediction system of the present invention detects early faults and qualitatively or quantitatively evaluates the degree of faults by utilizing the characteristic data of the acoustic wave data measured by the sensor based on the fusion analysis of historical data and current monitoring data. At the same time, it forms a degradation model of the power supply line material with the help of a neural network to judge the online operation status of the line and predict the health status and remaining service life of the line at future moments.

[0044] In summary, this system is designed based on distributed optical fiber sensors, and has the advantages of high detection accuracy, wide working frequency band, low transmission loss, long measurement distance, freedom from electromagnetic interference and miniaturization. It can be used for precise monitoring of complex environments such as confined spaces, strong electromagnetic interference, high temperature, high pressure, strong corrosion, and strong radiation. At the same time, due to the sensitivity of optical fiber itself to a variety of physical quantities (such as temperature, mechanical waves, angular velocity, acceleration, etc.), it can be used to measure environmental parameters such as sound and temperature, and has the advantages of high sensitivity and fast response speed. The distributed optical fiber sensing system can obtain the one-dimensional distribution map of the measured value in the entire optical fiber area in one measurement. By laying it along the power supply line, the changes in temperature and sound waves on the power supply line can be quickly and accurately obtained, and the fault can be quickly and accurately located and predicted.

[0045] This system uses temperature and sound as the medium for power supply line detection, and combines acoustics with temperature detection to verify each other and improve the accuracy of fault monitoring. In the monitoring of power supply lines, temperature and acoustic detection are non-invasive detections, which can be performed when the power equipment is operating normally without stopping for maintenance. In addition, the high precision and high sensitivity of fiber optic acoustic sensors can monitor the tiny sound wave changes of insulating materials, so as to detect potential insulation faults in advance and avoid further damage to the equipment.

[0046] In addition, the system supports remote real-time monitoring, and can transmit monitoring data to the monitoring center through remote communication, so as to realize the whole process monitoring and centralized control of all power lines, and improve the efficiency and accuracy of monitoring. The expert fault diagnosis system equipped with the system can make decisions on faults detected based on temperature and sound to determine the type of fault.

[0047] The system has established a historical monitoring database and is equipped with a data analysis function to achieve statistics and analysis of historical data, summarize and dig out some potential problems and hidden dangers, achieve continuous optimization of power supply line fault monitoring and early warning, and provide reference for the improvement of power supply lines. In addition, the system of the present invention is scalable and can be used for multiple power supply lines.

[0048] The technical scope of the present invention is not limited to the contents of the above description. Those skilled in the art can make various deformations and modifications to the above implementation without departing from the technical idea of ​​the present invention, and these deformations and modifications should all fall within the scope of the present invention.

Claims

1. A mining cable fault monitoring system, characterized in that: The monitoring system comprises: A detection optical fiber, which is laid parallel to the cable and is used to detect cable information of the cable; A detection host, the detection host is connected to the detection optical fiber, and an optical transmitter and a demodulation unit are arranged in the detection host, the optical transmitter is used to transmit detection light to the detection optical fiber, the detection light is used to convert the cable information into optical information, and the demodulation unit is used to obtain the cable information according to the optical information; An industrial computer, the industrial computer is used to receive the cable information obtained by the detection host, the industrial computer is provided with an expert fault diagnosis system, the expert fault diagnosis system is provided with a detection model, the expert fault diagnosis system is used to extract features from the cable information received by the industrial computer, and determine through the detection model whether there is abnormal information in the cable information after feature extraction and the fault type corresponding to the abnormal information.

2. The monitoring system according to claim 1, characterized in that: The monitoring system also includes a switch, the industrial computer is located on the ground, the detection optical fiber, the detection host and the switch are located in the coal mine, the switch is connected to the detection host, and the switch is used to realize data exchange between the industrial computer and the detection host.

3. The monitoring system according to claim 2, characterized in that: The detection optical fiber includes an acoustic wave detection optical fiber and a temperature detection optical fiber. The cable is located in a cable detection area. The cable information includes the temperature of the cable surface and the acoustic waves in the cable detection area. The detection host includes a DAS host and a DTS host. The DAS host is connected to the acoustic wave detection optical fiber, and the DTS host is connected to the temperature detection optical fiber.

4. The monitoring system according to claim 3, characterized in that: The demodulation unit includes a DAS demodulation unit and a DTS demodulation unit. The DAS demodulation unit includes a first acousto-optic modulator, a first erbium-doped fiber amplifier, a first circulator, a first fiber coupler and a first photoelectric balanced detector. The DTS demodulation unit includes a second acousto-optic modulator, a second erbium-doped fiber amplifier, a second circulator, a second fiber coupler and a second photoelectric balanced detector.

5. The monitoring system according to claim 4, characterized in that: The expert fault diagnosis system is also provided with an alarm module, which is used to generate a historical alarm record according to the abnormal information, and the historical alarm record is used to determine the fault type corresponding to the abnormal information and the location where the fault occurs.

6. The monitoring system according to claim 5, characterized in that: The industrial computer is also provided with a big data prediction model, in which a neural network decay model is provided. The abnormal information includes abnormal sound wave information and abnormal temperature information. The neural network decay model is used to predict the remaining service life of the cable based on the abnormal sound wave information and the historical alarm records.

7. The monitoring system according to claim 6, characterized in that The detection model includes a binary classification model and a multi-classification model. The binary classification model is used to determine whether there is abnormal information in the cable information after feature extraction, and the multi-classification model is used to determine the fault type corresponding to the abnormal information.

8. The monitoring system according to claim 1, characterized in that: The detection optical fiber is fixedly laid on the cable by rolling tape, or the detection optical fiber and the cable are aggregated by a tinned copper mesh braided layer.

9. A data processing method for a mining cable fault monitoring system as claimed in claim 7, characterized in that: The data processing method comprises: Step A: obtaining cable information during the working process of the cable and integrating it into a cable information data set, and training the detection model through the cable information data set; Step B: the detection host transmits the cable information to the industrial computer, the expert fault diagnosis system preprocesses and extracts features of the cable information, the detection model determines whether there is abnormal information in the cable information through the binary classification model, and determines the fault type corresponding to the abnormal information through the multi-classification model; Step C: When there is abnormal information in the cable information, the alarm module receives the abnormal information and matches the abnormal information with the historical alarm record. When the match is successful, the alarm module outputs the abnormal information, the fault type corresponding to the abnormal information and the location where the fault occurs. When the match fails, the alarm module issues an alarm and updates the abnormal information to the historical alarm record. Step D: The big data prediction model extracts the sound wave abnormality information, and the neural network decay model predicts the remaining service life of the cable based on the sound wave abnormality information and the historical alarm records.

10. The data processing method according to claim 9, characterized in that: The preprocessing includes denoising and smoothing filtering, the feature extraction includes time domain feature extraction and frequency domain feature extraction, the time domain features include amplitude, zero crossing rate, skewness and fuzzy entropy, the frequency domain features include frequency domain kurtosis, frequency domain skewness and energy concentration, and the detection model is a SVM model.

Images