Cable fault point positioning method and system

Abstract

The invention discloses a cable fault point positioning method and system, and the method comprises the steps: deploying a nanometer material enhanced temperature sensor, a quantum induction type current sensor and an ultrahigh-sensitivity optical fiber acoustic sensor, and collecting cable data; transmitting the data to a near-end data aggregation node through a self-repairing optical communication network; the node preliminarily judges the running state of the cable based on a long-short term memory network introducing an attention mechanism, if a suspected fault occurs, data is transmitted to a cloud intelligent diagnosis platform, and a fault point is positioned by using a deep diagnosis model; fault point positioning information is synchronized to a 3D-GIS system in real time, an intelligent first-aid repair plan is automatically generated and pushed to first-aid repair personnel, after first-aid repair is completed, the long-short term memory network and deep diagnosis model parameters are reversely optimized by using repaired cable operation data, and a simulation fault test is performed after each optimization. The accuracy and efficiency of cable fault positioning and the overall performance of the system can be effectively improved, and stable power transmission is guaranteed.

Description

Technical Field

[0001] The invention belongs to the technical field of cable fault point locating, and relates to a cable fault point locating method and system. Background Art

[0002] As a vital component of power transmission, the operational status of underground cables directly impacts the stability and safety of a city's power supply. Traditional cable fault location technology relies primarily on manual inspections and simple electrical testing.

[0003] Manual inspections, limited by human resources and time constraints, make it difficult to fully monitor cable operation in real time, making it difficult to detect many potential faults in a timely manner. For example, in the complex underground cable networks of large cities, manual inspections can take weeks. During this time, even subtle cable faults can develop and worsen unnoticed.

[0004] Simple electrical tests can usually only detect relatively obvious fault characteristics. For some complex fault types, such as faults caused by long-term electromagnetic interference of cables and slow development of local insulation aging, their diagnostic accuracy is seriously insufficient.

[0005] When faced with multi-branch, long-distance cable lines, traditional cable fault location technology faces huge challenges in determining the location of the fault point. It often takes a lot of time to check each section, which not only leads to long power outages and huge losses to the social economy, but may also cause a series of secondary problems due to delayed emergency repairs, such as damage to production equipment and interruption of public services. Summary of the Invention

[0006] In order to address the deficiencies in the prior art, the present invention provides a cable fault point locating method and system, which effectively improves the accuracy, efficiency and overall system performance of cable fault locating and ensures stable power transmission.

[0007] The present invention adopts the following technical solutions.

[0008] A first aspect of the present invention provides a method for locating a cable fault point, the method comprising the following steps:

[0009] S1: Deploy nanomaterial-enhanced temperature sensors, quantum-sensing current sensors, and ultra-high-sensitivity fiber-optic acoustic sensors to collect cable temperature, current, and acoustic vibration data;

[0010] S2: Transmit the data collected by the sensor to the near-end data aggregation node through the self-repairing optical communication network;

[0011] S3: The near-end data aggregation node preliminarily determines the cable operating status based on the long short-term memory network with an attention mechanism. If a fault is suspected, the data is forwarded to the cloud-based intelligent diagnosis platform.

[0012] S4: The cloud-based intelligent diagnosis platform uses a deep diagnosis model that integrates multi-source information to make a secondary judgment. If a fault is confirmed, the fault point is located based on the dynamic fault location model.

[0013] S5: The fault point location information is synchronized to the 3D geographic information visualization system in real time, and the fault point is mapped to a 3D real-life map, visually displaying the spatial relationship between the fault point and surrounding geographic elements.

[0014] S6: Automatically generates an intelligent emergency repair plan and sends it to the qualified emergency repair personnel closest to the fault point. The emergency repair personnel provide real-time feedback on the repair progress, enabling two-way interaction.

[0015] S7: After the emergency repair is completed, the repaired cable operation data is used to reversely optimize the long short-term memory network and deep diagnostic model parameters, and simulated fault tests are performed after each optimization.

[0016] Preferably, the S1 deploys a nanomaterial-enhanced temperature sensor, a quantum induction current sensor, and an ultra-high-sensitivity fiber optic acoustic sensor, including:

[0017] Nanomaterial-enhanced temperature sensors are spirally wound on the cable surface at 1-meter intervals to sense subtle temperature gradient changes in the cable in real time. The temperature sensor housing is made of heat-insulating and electromagnetically shielding materials.

[0018] Install quantum induction current sensors at the starting point of cable branches and key nodes to measure current fluctuations as low as milliamperes based on the quantum Hall effect;

[0019] Ultra-high-sensitivity fiber optic acoustic sensors are attached to cable joints and bends to capture abnormal acoustic vibrations caused by faults.

[0020] Preferably, the self-repairing optical communication network in S2 automatically switches to a backup optical path when a break occurs in the optical fiber. The backup optical path is a micro-nano optical fiber redundant structure with a switching delay of no more than 5 seconds, ensuring uninterrupted flow of data to the proximal data aggregation node, and at the same time adopts quantum key distribution technology to encrypt data.

[0021] Preferably, the process of introducing the long short-term memory network with attention mechanism in S3 to preliminarily judge the operating status of the cable is as follows:

[0022] i t =σ(W i ·[h t-1 ,x t ]+bi )

[0023] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0024] o t =σ(W o ·[h t-1 ,x t ]+b o )

[0025] g t =σ(W g ·[h t-1 ,x t ]+b g )

[0026] c t =f t c t-1 +i t ·g t

[0027]

[0028] e t =Attention(h t ,x t )

[0029] Among them, x t is the input data at time t;

[0030] e t is the output after attention weighting;

[0031] h t-1 、h t are the hidden states at time t-1 and time t respectively;

[0032] σ is the sigmoid function;

[0033] i t 、f t 、o t 、g t They are the values ​​of the input gate, forget gate, output gate, and candidate memory unit respectively;

[0034] W i 、W f 、W o 、W g are all weight matrices;

[0035] bi 、b f 、b o 、b g are bias vectors;

[0036] c t 、c t-1 They are the memory unit of the current moment and the old memory unit;

[0037] is the memory unit c at the current moment t The result after being processed by the hyperbolic tangent function;

[0038] Attention(,) represents the attention mechanism.

[0039] Preferably, when training the deep diagnostic model in S4, a generative adversarial network (GAN) is used to expand training samples.

[0040] Preferably, the specific formula for locating the fault point by the dynamic fault location model in S4 is:

[0041]

[0042] Among them, L fauit is the distance between the fault point and the reference end;

[0043] V eff is the equivalent wave velocity after considering the environmental impact;

[0044] Where, ΔT waveform is the time delay of the fault characteristic waveform at the monitoring end; β is the environmental correction coefficient; ΔV peok is the peak voltage change.

[0045] Preferably, the equivalent wave velocity after considering the environmental influence is calculated as follows:

[0046] V eff =a0+a1T+a2H+a3ρ+a4W+a5E

[0047] Where T is temperature, H is humidity, ρ is soil resistivity; W is cable material, E is electromagnetic interference, a1,

[0048] a2, a3, a4, a5 are coefficients;

[0049] Preferably, the three-dimensional geographic information visualization system in S5 is linked to the engineering database in real time to timely update the information on the construction and reconstruction of underground pipelines, with the update delay not exceeding 24 hours.

[0050] Preferably, the plan in S6 is prioritized according to the severity of the fault, required professional skills, and tool and equipment list information, and when pushed to the emergency repair personnel, it is accompanied by a detailed virtual real-life image of the fault site, which is generated by a drone equipped with a high-definition camera shooting 50 meters above the fault point.

[0051] A second aspect of the present invention provides a cable fault point locating system, comprising:

[0052] Nanomaterial-enhanced temperature sensors, quantum-sensing current sensors, and ultra-high-sensitivity fiber-optic acoustic sensors for collecting cable temperature, current, and acoustic vibration data;

[0053] A self-healing optical communication network for transmitting data collected by sensors to a proximal data aggregation node;

[0054] The proximal data aggregation node is used to preliminarily determine the cable operating status based on a long short-term memory network with an attention mechanism. If a fault is suspected, the data is forwarded to the cloud-based intelligent diagnosis platform.

[0055] The cloud-based intelligent diagnosis platform uses a deep diagnostic model that integrates multi-source information to make secondary judgments. If a fault is confirmed, the fault point is located based on a dynamic fault location model.

[0056] A 3D geographic information visualization system is used to map the fault point onto a 3D real-life map, visually displaying the spatial relationship between the fault point and surrounding geographic elements;

[0057] Smart terminals are used to automatically generate intelligent emergency repair plans and push them to qualified repair personnel closest to the fault point. Repair personnel provide real-time feedback on repair progress, enabling two-way interaction.

[0058] The optimization module is used to reversely optimize the parameters of the long short-term memory network and the deep diagnostic model using the operation data of the repaired cable after the emergency repair is completed, and to perform simulated fault tests after each optimization.

[0059] A third aspect of the present invention provides a terminal, comprising a processor and a storage medium; the storage medium is used to store instructions; and the processor is used to operate according to the instructions to execute the steps of the method.

[0060] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when executed by a processor.

[0061] Compared with the prior art, the beneficial effects of the present invention include at least:

[0062] The present invention collects cable data from multiple dimensions by deploying nanomaterial-enhanced temperature sensors, quantum-inductive current sensors, and ultra-high-sensitivity fiber-optic acoustic sensors. Nanomaterial-enhanced temperature sensors, quantum-inductive current sensors, and ultra-high-sensitivity fiber-optic acoustic sensors can accurately monitor temperature gradients, low current fluctuations, and acoustic vibrations, respectively.

[0063] The present invention uses an optical communication network with self-repair function, adopts high-quality optical fiber and redundant link design, and can quickly switch to the backup optical path when the optical fiber fails, ensuring uninterrupted data transmission and maintaining the system's continuous monitoring capability of the cable status. It also encrypts data with quantum key distribution technology, which can fundamentally guarantee the security of data transmission, prevent information leakage and malicious tampering, provide reliable data support for fault diagnosis and processing, and enhance the overall stability and credibility of the system.

[0064] The present invention uses an improved long short-term memory network combined with a preliminary analysis module of the attention mechanism and a cloud-based deep diagnosis model that integrates multi-source information to conduct in-depth data mining and analysis, accurately judge faults and precisely locate them, greatly improving the accuracy of fault diagnosis and positioning precision. It can effectively reduce the scope and time of investigation, reduce power outage losses, monitor the operating status of cables in real time, promptly discover potential fault risks, and perform early warning and preventive maintenance.

[0065] The dynamic fault location model of the present invention can more accurately locate the fault point by considering the influence of wave velocity caused by factors such as temperature, humidity, cable material, and electromagnetic interference, introducing an environmental correction coefficient and considering the change of the fault voltage peak.

[0066] This invention integrates fault location information with a three-dimensional geographic information visualization system (3D-GIS) in real time, providing repair personnel with intuitive and accurate geographic information, assisting in developing optimal repair routes and improving repair efficiency. Intelligent repair plans intelligently allocate resources based on fault severity and deliver them to appropriate repair personnel. Repair personnel also receive real-time progress feedback, enabling two-way interaction, optimizing the repair process, shortening outages, improving power supply reliability and customer satisfaction, and ensuring stable power system operation.

[0067] The present invention proposes a fault backtracking and learning mechanism, which uses repaired data to retrain the model and continuously optimizes the parameters of the long short-term memory network and deep diagnostic model. This can improve the system's adaptability and recognition capabilities to complex fault scenarios, enabling it to cope with the ever-changing cable operating environment and fault types, extend the system's effective service life, maintain advanced performance, and reduce subsequent maintenance costs and manpower investment. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0069] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without making creative efforts are all within the scope of protection of the present invention.

[0070] Embodiment 1 of the present invention provides a method for locating a cable fault point, such as Figure 1 As shown, the following steps are included:

[0071] S1: Deploy nanomaterial-enhanced temperature sensors, quantum-sensing current sensors, and ultra-high-sensitivity fiber-optic acoustic sensors to collect cable temperature, current, and acoustic vibration data;

[0072] Furthermore, it is preferred to use an intelligent composite sensor array comprising nanomaterial-enhanced temperature sensors, quantum-inductive current sensors, and ultra-sensitive fiber-optic acoustic sensors to collect cable data in all directions. First, deploy an intelligent composite sensor array comprising nanomaterial-enhanced temperature sensors, quantum-inductive current sensors, and ultra-sensitive fiber-optic acoustic sensors.

[0073] Nanomaterial-enhanced temperature sensors, with a resolution of up to 0.1°C, are spirally wound around the cable surface at 1-meter intervals to detect subtle temperature gradients in real time. During installation, the sensor must be securely attached to the cable to accurately detect these subtle changes. Specifically, the sensor is installed simultaneously with the cable during laying, secured to the cable sheath with a fixture to prevent loosening or displacement.

[0074] The quantum induction current sensor, based on the quantum Hall effect, can accurately measure current fluctuations as low as milliamperes. It is installed at the starting point of cable branches and key nodes to accurately measure current fluctuations as low as milliamperes. During installation, ensure that the sensor wiring is correct and stable to avoid affecting the measurement accuracy due to poor contact.

[0075] Ultra-high-sensitivity fiber optic acoustic sensors respond sensitively to acoustic signals within the frequency range of 100Hz-10kHz. They are attached to cable joints and bends to capture abnormal acoustic vibrations caused by faults. During the attachment process, highly viscous glue with good insulating properties is used to ensure that the sensor is tightly connected to the cable and is not affected by external interference.

[0076] The temperature sensor housing in the intelligent composite sensor array is made of heat-insulating and electromagnetic shielding materials to ensure that the temperature measurement is not interfered with by the external environment.

[0077] S2: Transmit the data collected by the sensor to the near-end data aggregation node through the self-repairing optical communication network;

[0078] It is further preferred that the data collected by the sensor is transmitted through an optical communication network with self-repair and encryption functions. When the network is built, high-quality optical fiber materials are used and redundant links are set up. When a breakpoint occurs in the optical fiber, the network can automatically start the backup optical path switching mechanism within 10 seconds to ensure that the data flows uninterruptedly to the near-end data aggregation node. The principle is to detect the transmission status of the optical fiber in real time through the intelligent monitoring equipment in the network. Once a fault is found, the backup optical path is immediately activated. The transmission rate is stable at more than 500Mbps, and quantum key distribution technology is used to encrypt the data. At the data sending end, the quantum key generation device is used to generate the key and encrypt the collected data; at the receiving end, the corresponding decryption device is used to restore the data to ensure that the data transmission is absolutely safe.

[0079] The backup optical path of the self-repairing optical communication network adopts a micro-nano optical fiber redundant structure, which is usually in a low-power standby state. When activated, it can quickly take on data transmission tasks, and the switching delay does not exceed 5 seconds.

[0080] S3: The near-end data aggregation node preliminarily determines the cable operating status based on the long short-term memory network with an attention mechanism. If a fault is suspected, the data is forwarded to the cloud-based intelligent diagnosis platform.

[0081] Further preferably, a preliminary analysis module based on an improved long short-term memory network (LSTM) is built into the proximal data aggregation node. The network introduces an attention mechanism to focus on key data features, preliminarily analyze the data, and judge the cable operation status. If a fault is suspected, the suspected fault data is forwarded to the cloud intelligent diagnosis platform;

[0082] The algorithm formula of the preliminary analysis module is:

[0083] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0084] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0085] o t =σ(W o ·[h t-1 ,x t ]+b o )

[0086] g t =σ(W g ·[h t-1 ,x t ]+b g )

[0087] c t =f t c t-1 +i t ·g t

[0088]

[0089] e t =Attention(h t ,x t )

[0090] Among them, x t The input data at time t is obtained by preprocessing the data collected and input by the sensor. Specifically, the temperature data is normalized and the current and acoustic data are filtered.

[0091] h t-1 、h t They are the hidden states at the previous moment and the current moment respectively. The initial state is set based on experience and can also be obtained through training with a small number of samples;

[0092] The W series is the weight matrix, and the b series is the bias vector, which is obtained by training and optimizing a large amount of historical data and known fault data during the model training phase;

[0093] σ is the sigmoid function, e t It is the attention-weighted output, used to enhance the identification of fault features;

[0094] i t 、f t 、o t 、g t They are the values ​​of the input gate, forget gate, output gate, and candidate memory unit, which are used to determine which information in the current input data will be updated to the memory unit, control which old information in the memory unit will be retained or discarded, determine which information in the memory unit will have the output at the current moment, and be used to update the memory unit;

[0095] W i 、W f 、W o 、W gBoth are weight matrices, which are obtained by training and optimizing a large amount of historical data and known fault data during the model training phase. They are used to control the degree of influence of different input information on the current calculation results.

[0096] b i 、b f 、b o 、b g Both are bias vectors, obtained through training optimization during the model training phase, providing additional learnable parameters for model calculation, helping the model to better fit the data;

[0097] c t 、c t-1 They are the memory unit of the current moment and the old memory unit;

[0098] is the memory unit c at the current moment t The result after the hyperbolic tangent function processing is used to calculate the hidden state h at the current moment in the long short-term memory network (LSTM) t , in order to participate in the judgment and analysis of the cable operation status;

[0099] Attention(,) represents the attention mechanism.

[0100] In actual operation, data is continuously input into the module, and the module performs calculations according to the above formula to preliminarily determine the operating status of the cable. If a fault is suspected, the data will be transferred to the cloud-based intelligent diagnosis platform.

[0101] S4: The cloud-based intelligent diagnosis platform uses a deep diagnosis model that integrates multi-source information to make a secondary judgment. If a fault is confirmed, the fault point is located based on the dynamic fault location model.

[0102] Furthermore, the cloud-based intelligent diagnosis platform constructs a deep diagnosis model that integrates multi-source information to process suspected fault data and locate the fault point based on a precise algorithm. Specifically, after the deep diagnosis model confirms the fault, a dynamic fault location model constructed based on multi-dimensional parameters such as cable material, laying environment temperature and humidity, and surrounding electromagnetic interference is activated to accurately locate the fault point. The specific formula for accurately locating the fault point is:

[0103]

[0104] Among them, L fauit is the distance between the fault point and the reference end;

[0105] V eff In order to consider the equivalent wave velocity after environmental influence, a large amount of experimental data of cable wave velocity under different environmental conditions such as temperature, humidity and soil medium is analyzed and modeled. Specifically, V eff=a0+a1T+a2H+a3ρ+a4W+a5E

[0106] T is temperature, H is humidity, ρ is soil resistivity, W is cable material, E is electromagnetic interference, and a0, a1, a2, a3, a4, and a5 are coefficients obtained by fitting a large amount of experimental data.

[0107] ΔT waveform The time delay of the fault characteristic waveform at the monitoring end is obtained by calculating the time difference of the collected waveform data. Precise time synchronization equipment is required to ensure the accuracy of the time delay measurement;

[0108] β is the environmental correction factor, which is obtained by looking up the table based on the specific environmental parameters of the cable, such as soil resistivity and humidity;

[0109] ΔV peok is the voltage peak change, which is calculated and analyzed from the collected voltage data.

[0110] By calculating these parameters, the distance L between the fault point and the reference end is finally determined. fauit .

[0111] The above-mentioned dynamic fault location model uses the propagation time and speed of waves in the cable to preliminarily estimate the distance between the fault point and the monitoring end. It then considers the influence of wave speed caused by factors such as temperature, humidity, cable material, and electromagnetic interference, and introduces a correction coefficient to better conform to the actual situation and accurately reflect the location of the fault point. It is known that the voltage peak change reflects the degree of impact of the fault on the electrical characteristics of the cable. When a fault occurs, the voltage will definitely change. The greater the voltage peak change, the more serious the fault. Therefore, the voltage peak change is introduced to improve the accuracy of fault location.

[0112] The preliminary analysis module regularly synchronizes model parameters with the cloud-based intelligent diagnosis platform to ensure consistency between local judgment and global diagnosis. The synchronization cycle does not exceed 1 hour.

[0113] When training the deep diagnostic network of the cloud-based intelligent diagnostic platform, the generative adversarial network (GAN) is used to expand the training samples to improve the model's ability to identify low-probability faults, with the expansion ratio being no less than 30%.

[0114] S5: The fault point location information is synchronized to the 3D geographic information visualization system in real time, and the fault point is mapped to a 3D real-life map, visually displaying the spatial relationship between the fault point and surrounding geographic elements.

[0115] Furthermore, preferably, the fault point location information is synchronized in real time to a three-dimensional geographic information visualization system (3D-GIS) for accurately mapping the fault point to a three-dimensional real-life map, visually displaying the spatial relationship between the fault point and surrounding geographical elements such as buildings and underground pipelines;

[0116] The location coordinates of the fault point and other information are transmitted to the 3D-GIS system through the data interface. The system is linked to the municipal engineering database in real time to promptly update the information on the construction and reconstruction of underground pipelines, with an update delay of no more than 24 hours.

[0117] During the linkage process, a combination of scheduled data query and event-triggered update is adopted to ensure the timeliness and accuracy of information.

[0118] S6: At the same time, an intelligent emergency repair plan is automatically generated and pushed to the emergency repair personnel closest to the fault point and with corresponding qualifications. The emergency repair personnel provide real-time feedback on the emergency repair progress, achieving two-way interaction;

[0119] Furthermore, it is preferred that an intelligent emergency repair plan is automatically generated, and the plan is prioritized according to the severity of the fault, the required professional skills, and the list of tools and equipment. Specifically, for serious faults, experienced emergency repair personnel and complete equipment are given priority, and the plan is pushed to the emergency repair personnel closest to the fault point and with corresponding qualifications through the intelligent terminal. The emergency repair personnel can provide real-time feedback on the emergency repair progress on the terminal, realizing two-way interaction;

[0120] When the intelligent emergency repair plan is pushed to the repair personnel, it is also accompanied by a detailed virtual real-life image of the fault site. The image is generated by a drone equipped with a high-definition camera shooting 50 meters above the fault point, helping the repair personnel to understand the on-site conditions in advance.

[0121] S7: After the emergency repair is completed, the repaired cable operation data is used to reversely optimize the long short-term memory network and deep diagnostic model parameters, and simulated fault tests are performed after each optimization.

[0122] Further optimization is to establish a fault backtracking and learning mechanism. After the emergency repair is completed, the repaired cable operation data is used to reversely optimize the parameters in the diagnosis model and positioning algorithm to continuously improve the system performance. After each optimization, a simulated fault test is performed to ensure system stability.

[0123] Specifically, cable operation data from the period following repairs is collected and used as new training samples, along with previous fault data and normal operation data. The preliminary analysis module of the near-end data aggregation node and the deep diagnostic model of the cloud-based intelligent diagnostic platform are retrained, adjusting parameters such as the weight matrix and bias vector within the model to continuously improve system performance. After each optimization, simulated fault testing is performed. By constructing simulated fault scenarios, inputting test data, and observing the system's diagnostic and location results, system stability is ensured. Specifically, simulated faults of varying types and severity are set to verify the system's ability to accurately identify and locate them. If any issues are found, the cause is further analyzed and adjustments are made.

[0124] Embodiment 2 of the present invention provides a cable fault point locating system, the system comprising:

[0125] Nanomaterial-enhanced temperature sensors, quantum-sensing current sensors, and ultra-high-sensitivity fiber-optic acoustic sensors for collecting cable temperature, current, and acoustic vibration data;

[0126] A self-healing optical communication network for transmitting data collected by sensors to a proximal data aggregation node;

[0127] The proximal data aggregation node is used to preliminarily determine the cable operating status based on a long short-term memory network with an attention mechanism. If a fault is suspected, the data is forwarded to the cloud-based intelligent diagnosis platform.

[0128] The cloud-based intelligent diagnosis platform uses a deep diagnostic model that integrates multi-source information to make secondary judgments. If a fault is confirmed, the fault point is located based on a dynamic fault location model.

[0129] A 3D geographic information visualization system is used to map the fault point onto a 3D real-life map, visually displaying the spatial relationship between the fault point and surrounding geographic elements;

[0130] Smart terminals are used to automatically generate intelligent emergency repair plans and push them to qualified repair personnel closest to the fault point. Repair personnel provide real-time feedback on repair progress, enabling two-way interaction.

[0131] The optimization module is used to reversely optimize the parameters of the long short-term memory network and the deep diagnostic model using the operation data of the repaired cable after the emergency repair is completed, and to perform simulated fault tests after each optimization.

[0132] Embodiment 3 of the present invention provides a terminal, including a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps according to the method.

[0133] Embodiment 4 of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when executed by a processor.

[0134] Compared with the prior art, the beneficial effects of the present invention include at least:

[0135] 1. Cable data is collected from multiple dimensions through an intelligent composite sensor array. Nanomaterial-enhanced temperature sensors, quantum-sensing current sensors, and ultra-high-sensitivity fiber-optic acoustic sensors accurately monitor temperature gradients, low current fluctuations, and acoustic vibrations, respectively, without missing any subtle anomalies. An improved long-short-term memory network combined with a preliminary analysis module of the attention mechanism and a cloud-based deep diagnosis model that integrates multi-source information deeply mines and analyzes data through complex algorithms, accurately determines faults and precisely locates them. This greatly improves the accuracy of fault diagnosis and positioning precision, effectively reduces the scope and time of investigation, reduces power outage losses, monitors the cable operation status in real time, promptly detects potential fault risks, and provides early warning and preventive maintenance.

[0136] 2. Through an optical communication network with self-repair capabilities, high-quality optical fibers and redundant link designs are used. In the event of an optical fiber failure, the backup optical path can be quickly switched to ensure uninterrupted data transmission and maintain the system's ability to continuously monitor the cable status. Quantum key distribution technology encrypts data, fundamentally ensuring data transmission security, preventing information leakage and malicious tampering, providing reliable data support for fault diagnosis and processing, and enhancing the overall stability and credibility of the system.

[0137] 3. The dynamic fault location model of the present invention can more accurately locate the fault point by considering the influence of wave velocity caused by factors such as temperature, humidity, cable material, and electromagnetic interference, introducing an environmental correction coefficient and considering the change of the fault voltage peak.

[0138] 4. By integrating fault location information with a three-dimensional geographic information visualization system (3D-GIS) in real time, repair personnel are provided with intuitive and accurate geographic information, assisting in developing optimal repair routes and improving repair efficiency. Intelligent repair plans intelligently allocate resources based on fault severity and route them to appropriate repair personnel. Repair personnel also receive real-time progress feedback, enabling two-way interaction and optimizing the repair process. This shortens outages, improves power supply reliability and customer satisfaction, and ensures stable power system operation.

[0139] 5. Through the fault backtracking and learning mechanism, the repaired data is used to retrain the model, continuously optimize the diagnosis and positioning algorithm parameters, and improve the system's adaptability and recognition capabilities to complex fault scenarios, so that it can cope with the ever-changing cable operating environment and fault types, extend the effective service life of the system, maintain advanced performance, and reduce subsequent maintenance costs and manpower investment.

[0140] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0141] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0142] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0143] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., utilizing an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions. The electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A cable fault point location method, characterized in that: The method comprises the following steps: S1: Deploy nanomaterial-enhanced temperature sensors, quantum-sensing current sensors, and ultra-high-sensitivity fiber-optic acoustic sensors to collect cable temperature, current, and acoustic vibration data; S2: Transmit the data collected by the sensor to the near-end data aggregation node through the self-repairing optical communication network; S3: The near-end data aggregation node preliminarily determines the cable operating status based on the long short-term memory network with an attention mechanism. If a fault is suspected, the data is forwarded to the cloud-based intelligent diagnosis platform. S4: The cloud-based intelligent diagnosis platform uses a deep diagnosis model that integrates multi-source information to make a secondary judgment. If a fault is confirmed, the fault point is located based on the dynamic fault location model. S5: The fault point location information is synchronized to the 3D geographic information visualization system in real time, mapping the fault point to a 3D real-life map, visually displaying the spatial relationship between the fault point and surrounding geographic elements; S6: Automatically generates an intelligent emergency repair plan and sends it to the qualified emergency repair personnel closest to the fault point. The emergency repair personnel provide real-time feedback on the repair progress, enabling two-way interaction. S7: After the emergency repair is completed, the repaired cable operation data is used to reversely optimize the long short-term memory network and deep diagnostic model parameters, and simulated fault tests are performed after each optimization.

2. A cable fault point locating method according to claim 1, characterized in that: The S1 deploys nanomaterial-enhanced temperature sensors, quantum-induced current sensors, and ultra-high-sensitivity fiber-optic acoustic sensors, including: Nanomaterial-enhanced temperature sensors are spirally wound on the cable surface at 1-meter intervals to sense subtle temperature gradient changes in the cable in real time. The temperature sensor housing is made of heat-insulating and electromagnetic shielding materials. Install quantum induction current sensors at the starting point of cable branches and key nodes to measure current fluctuations as low as milliamperes based on the quantum Hall effect; Ultra-high-sensitivity fiber optic acoustic sensors are attached to cable joints and bends to capture abnormal acoustic vibrations caused by faults.

3. A cable fault point locating method according to claim 1, characterized in that: The self-repairing optical communication network in S2 automatically switches to a backup optical path when a break occurs in the optical fiber. The backup optical path is a micro-nano optical fiber redundant structure with a switching delay of no more than 5 seconds, ensuring uninterrupted data flow to the proximal data aggregation node. At the same time, quantum key distribution technology is used to encrypt data.

4. A cable fault point locating method according to claim 1, characterized in that: The process of initially judging the cable operation status by the long short-term memory network using the attention mechanism in S3 is as follows: i t =σ(W i ·[h t-1 ,x t ]+b i ) f t =σ(W f ·[h t-1 ,x t ]+b f ) i t =σ(W o ·[h t-1 ,x t ]+b o ) g t =σ(W g ·[h t-1 ,x t ]+b g ) c t =f t ·c t-1 +i t ·g t h t =o t fishy ct e t =Attention(h t ,x t ) Among them, x t is the input data at time t; e t is the output after attention weighting; h t-1 、h t They are the hidden states at time t-1 and time t respectively; σ is the sigmoid function; i t 、f t 、o t 、g t are the values ​​of the input gate, forget gate, output gate, and candidate memory unit respectively; W i 、W f 、W o 、W g are all weight matrices; b i 、b f 、b o 、b g are all bias vectors; c t 、c t-1 are the current memory unit and the old memory unit respectively; h ct is the memory unit c at the current moment t The result after processing by the hyperbolic tangent function; Attention(,) represents the attention mechanism.

5. A cable fault point locating method according to claim 1, characterized in that: When training the deep diagnostic model in S4, the adversarial generative network (GAN) is used to expand the training samples.

6. A cable fault point locating method according to claim 1, characterized in that: The specific formula for locating the fault point using the dynamic fault location model described in S4 is: Among them, L fauit is the distance between the fault point and the reference end; V eff is the equivalent wave velocity after considering the environmental impact; ΔT waveform is the time delay of the fault characteristic waveform at the monitoring end; β is the environmental correction coefficient; ΔV peok is the peak voltage change.

7. A cable fault point locating method according to claim 6, characterized in that: The equivalent wave velocity after considering the environmental impact is calculated as follows; <h2 style=";text-align:left;direction:ltr">V<h2 style=";text-align:left;direction:ltr"> eff <h2 style=";text-align:left;direction:ltr"> = a0+a1T+a2H+a3ρ+a4W+a5E Where T is temperature, H is humidity, ρ is soil resistivity, W is the cable material, E is the electromagnetic interference, a 0， a1, a2, a3, a4, a5 are coefficients.

8. A cable fault point locating method according to claim 1, characterized in that: The three-dimensional geographic information visualization system described in S5 is linked to the engineering database in real time to timely update the information on the construction and reconstruction of underground pipelines, and the update delay shall not exceed 24 hours.

9. A cable fault point locating method according to claim 1, characterized in that: The S6 plan is prioritized based on the severity of the fault, required professional skills, and tool and equipment list information. When it is pushed to the emergency repair personnel, it is accompanied by a detailed virtual real-life image of the fault site. The image is generated by a drone equipped with a high-definition camera shooting around the fault point at 50 meters above it.

10. A cable fault point locating system, characterized in that: The system comprises: Nanomaterial-enhanced temperature sensors, quantum-sensing current sensors, and ultra-high-sensitivity fiber-optic acoustic sensors for collecting cable temperature, current, and acoustic vibration data; A self-healing optical communication network for transmitting data collected by sensors to a proximal data aggregation node; The proximal data aggregation node is used to preliminarily determine the cable operating status based on a long short-term memory network with an attention mechanism. If a fault is suspected, the data is forwarded to the cloud-based intelligent diagnosis platform. The cloud-based intelligent diagnosis platform uses a deep diagnostic model that integrates multi-source information to make secondary judgments. If a fault is confirmed, the fault point is located based on a dynamic fault location model. A 3D geographic information visualization system is used to map the fault point onto a 3D real-life map, visually displaying the spatial relationship between the fault point and surrounding geographic elements; Smart terminals are used to automatically generate intelligent emergency repair plans and push them to qualified repair personnel closest to the fault point. Repair personnel provide real-time feedback on repair progress, enabling two-way interaction. The optimization module is used to reversely optimize the parameters of the long short-term memory network and the deep diagnostic model using the operation data of the repaired cable after the emergency repair is completed, and to perform simulated fault tests after each optimization.

11. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1 to 9.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images