High-voltage cable high-resolution continuous temperature measurement system and device based on few-mode optical fiber

Abstract

The present application relates to the field of high-voltage cable monitoring and optical fiber sensing technology, and particularly relates to a high-voltage cable high-resolution continuous temperature measurement system and device based on a few-mode optical fiber. The present application determines the heat generation temperature increment of each cable node based on the difference between the fundamental mode temperature and the high-order mode temperature of each cable node; determines the noise interference characteristics of each cable node according to the data fluctuation degree of the heat generation temperature increment of each cable node and the adjacent cable node; fuses the heat generation temperature increment and the noise interference characteristics of all cable nodes to determine the real temperature increment of each cable node; according to the spatial distance characteristics between each cable node and each cable joint on the high-voltage cable, in combination with the real temperature increment of each cable node and the equivalent load current in the current temperature measurement period, analyzes the contact thermal resistance parameters of each cable joint; and based on the deviation degree of the contact thermal resistance parameters of each cable joint and the contact thermal resistance reference value of each cable joint in the initial state, judges the health state of each cable joint.

Description

Technical Field

[0001] This invention relates to the field of high-voltage cable monitoring and fiber optic sensing technology, specifically to a high-resolution continuous temperature measurement system and device for high-voltage cables based on few-mode optical fiber. Background Technology

[0002] In continuous operation monitoring scenarios involving 10-kilometer-long underground high-voltage cables, cable joints are prone to localized abnormal heating due to aging and loosening of the internal metal connection surfaces. The industry typically employs distributed fiber optic sensing systems to collect temperature data along the entire cable length to determine the health status of the cable joints.

[0003] To ensure the optical signal energy can cover long-distance detection, the temperature measuring equipment must emit detection optical pulses with a relatively long duration. On the one hand, when this wide optical pulse sweeps through the optical fiber across the cable joint, the actual high temperature of the joint is averaged out by the normal low-temperature cables on both sides, resulting in a substantial decrease in spatial resolution. On the other hand, during periods of high summer temperatures or heavy power grid transmission, the entire cable experiences an overall temperature rise. The high-temperature bulges at the joint, smoothed out by the wide optical pulse, are easily masked by the normal overall cable temperature rise, leading to missed detections of equipment degradation. Therefore, due to the above factors, the accuracy of the assessment index obtained by existing technologies in judging the health status of cable joints is relatively low. Summary of the Invention

[0004] To address the problem of low accuracy in obtaining assessment indices when determining the health status of cable joints in existing technologies, the present invention aims to provide a high-resolution continuous temperature measurement system and device for high-voltage cables based on few-mode optical fiber. The specific technical solution adopted is as follows: This invention proposes a high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber, the device comprising: The data acquisition module is used to acquire the basic mode temperature and higher-order mode temperature of each cable node in the current temperature measurement cycle, as well as the equivalent load current of the high-voltage cable in the current temperature measurement cycle. The interference feature analysis module is used to determine the temperature increment of each cable node based on the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node; and to determine the noise interference characteristics of each cable node based on the data fluctuation of the temperature increment of each cable node and the adjacent cable nodes. The thermal resistance characteristic analysis module is used to integrate the heating temperature increment and noise interference characteristics of all cable nodes to determine the true temperature increment of each cable node; based on the spatial distance characteristics between each cable node and each cable joint on the high-voltage cable, combined with the true temperature increment of each cable node and the equivalent load current, the contact thermal resistance parameters of each cable joint are analyzed. The results output module is used to determine the health status of each cable joint based on the degree of deviation between the contact thermal resistance parameter of each cable joint and the contact thermal resistance reference value of each cable joint in the initial state.

[0005] Furthermore, determining the heating temperature increment of each cable node based on the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node includes: The temperature increment of each cable node is obtained by relating the absolute value of the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node to the rated mode sensitivity difference of the few-mode fiber.

[0006] Furthermore, determining the noise interference characteristics of each cable node based on the data fluctuation of the temperature increment of each cable node compared to adjacent cable nodes includes: Take any cable node in the current temperature measurement cycle as the target cable node, and take the target cable node as the center of the preset spatial window. Based on the cumulative difference of the heating temperature increment between every two adjacent cable nodes in the preset spatial window, the temperature fluctuation characteristics of the target cable node are obtained. Based on the distribution range of temperature fluctuation characteristics of all cable nodes, a normalization operation is performed on the temperature fluctuation characteristics of the target cable node to obtain the noise interference characteristics of the target cable node.

[0007] Furthermore, based on the distribution range of temperature fluctuation characteristics of all cable nodes, the temperature fluctuation characteristics of the target cable node are normalized to obtain the noise interference characteristics of the target cable node, including: The ratio of the temperature fluctuation characteristic of the target cable node to the maximum value among the temperature fluctuation characteristics of all cable nodes is used as the noise interference characteristic of the target cable node.

[0008] Furthermore, the process of fusing the temperature increments and noise interference characteristics of all cable nodes to determine the true temperature increment of each cable node includes: Based on the numerical characteristics of the noise interference features of all cable nodes within the current temperature measurement cycle, noisy cable nodes are screened. Based on the quantity characteristics of the noise cable nodes and the dispersion of the temperature increment of the noise cable nodes, an adaptive penalty coefficient is determined within the current temperature measurement cycle. Based on the temperature increment of each cable node during the current temperature measurement cycle, a local temperature rise column vector is constructed. Based on the noise interference characteristics of each cable node in the current temperature measurement cycle, a noise feature diagonal matrix is ​​constructed; the main diagonal of the noise feature diagonal matrix is ​​the noise interference characteristics of each cable node. The adaptive penalty coefficient, local temperature rise column vector, noise feature diagonal matrix, and preset pulse broadening matrix are fused using a regularized least squares algorithm to obtain the reconstructed node temperature rise column vector; the elements in the reconstructed node temperature rise column vector are the actual temperature increments of each cable node.

[0009] Furthermore, the screening of noisy cable nodes based on the numerical characteristics of noise interference features of all cable nodes within the current temperature measurement cycle includes: If the noise interference characteristics of any cable node are greater than or equal to the preset high-frequency noise threshold, then any cable node will be screened as a noisy cable node.

[0010] Further, determining the adaptive penalty coefficient for the current temperature measurement cycle based on the quantitative characteristics of the noise cable nodes and the dispersion of the temperature increments of the noise cable nodes includes: If the total number of noise cable nodes is less than 2, an adaptive penalty coefficient is determined based on a preset ambient noise floor value; if the total number of noise cable nodes is greater than or equal to 2, an adaptive penalty coefficient is obtained based on the dispersion of the heating temperature increment of the noise cable nodes.

[0011] Furthermore, the analysis of the contact thermal resistance parameters of each cable joint based on the spatial distance characteristics between each cable node and each cable joint on the high-voltage cable, combined with the actual temperature increment of each cable node and the equivalent load current, includes: Take any cable connector as the target connector, and determine the target cable node of the target connector based on the spatial distance characteristics between the target connector and each cable node; the sum of the spatial distances between any two adjacent target cable nodes is greater than the total length of the target connector. Based on the average level of the actual temperature increment of the target cable node, the joint temperature rise characteristics are obtained; The ratio of the joint temperature rise characteristic to the square of the equivalent load current is used as the contact thermal resistance parameter of the target joint.

[0012] Furthermore, the determination of the health status of each cable joint based on the deviation between the contact thermal resistance parameter of each cable joint and the reference value of the contact thermal resistance of each cable joint in the initial state includes: The deterioration deviation is obtained by dividing the difference between the contact thermal resistance parameter of the target joint and the corresponding contact thermal resistance reference value by the corresponding contact thermal resistance reference value. If the degradation deviation is less than or equal to the preset deviation warning threshold, the target connector is in a healthy state; otherwise, the target connector is in an unhealthy state.

[0013] The present invention also proposes a high-resolution continuous temperature measurement system for high-voltage cables based on few-mode optical fibers, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a scheme for a high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fibers.

[0014] The present invention has the following beneficial effects: This invention determines the temperature increment of each cable node based on the difference between the fundamental mode temperature and the higher-order mode temperature. This eliminates the ambient background temperature interference caused by common-mode interference, preventing normal overall cable temperature rise from masking weak localized joint heating. It eliminates the risk of equipment degradation underreporting and false alarms across the entire line, ensuring high immunity and accuracy in the extraction of underlying temperature measurement data. Based on the fluctuation of the temperature increment between each cable node and adjacent cable nodes, the noise interference characteristics of each cable node are determined. This accurately identifies and distinguishes between genuine temperature rise signals and false high-frequency noise interference, providing accurate data support for subsequent elimination of high-frequency instrument noise and prevention of erroneous deletion of true extreme heating values.

[0015] By integrating the temperature increments and noise interference characteristics of all cable nodes, the true temperature increment of each cable node is determined, eliminating the local temperature averaging effect caused by high-frequency instrument noise interference and probe light pulses. This represents the pure temperature value extracted from each cable node, unaffected by random optical background noise. Based on the spatial distance characteristics between each cable node and each cable joint on the high-voltage cable, and combining the true temperature increment of each cable node with the equivalent load current, the contact thermal resistance parameters of each cable joint are analyzed. This effectively smooths and offsets temperature fluctuations caused by poor local fiber optic contact, fundamentally eliminating temperature rise fluctuations caused by current changes in the high-voltage cable. The obtained contact thermal resistance parameters can purely reflect the substantial degradation degree of the internal metal conductive connection surface of the cable joint, improving the accuracy of subsequent assessments of the cable joint's health status.

[0016] Based on the deviation between the contact thermal resistance parameter of each cable joint and the reference value of the contact thermal resistance of each cable joint in the initial state, the health status of each cable joint can be judged. This enables the accurate quantification of hidden faults such as internal oxidation aging or mechanical loosening of cable joints, eliminates the differences in the basic thermal resistance of cable joints of different specifications and models, and improves the alarm accuracy and equipment fault diagnosis reliability of long-term continuous operation monitoring of high-voltage cables. Attached Figure Description

[0017] Figure 1 This is a structural diagram of a high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber, provided in one embodiment of the present invention. Figure 2This is a module structure diagram of a thermal resistance characteristic analysis module provided in one embodiment of the present invention. Detailed Implementation

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details the specific scheme of a high-resolution continuous temperature measurement system and device for high-voltage cables based on few-mode optical fibers provided by the present invention.

[0020] Please see Figure 1 The diagram shows a structural diagram of a high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber according to an embodiment of the present invention. The device includes: a data acquisition module 101, an interference feature analysis module 102, a thermal resistance feature analysis module 103, and a result output module 104.

[0021] The data acquisition module 101 is used to acquire the basic mode temperature and higher-order mode temperature of each cable node in the current temperature measurement cycle, as well as the equivalent load current of the high-voltage cable in the current temperature measurement cycle.

[0022] This invention receives the fundamental mode temperature and higher-order mode temperature of each cable node collected by a distributed fiber optic temperature measurement host in each temperature measurement cycle. In this embodiment, the latest temperature measurement cycle acquired under the absolute system time of the distributed fiber optic temperature measurement host is taken as the current temperature measurement cycle.

[0023] The distributed fiber optic temperature measurement host is deployed in the substation equipment room. The system clock of the distributed fiber optic temperature measurement host emits probe light pulses at fixed time intervals. The emitted probe light pulses enter the few-mode fiber and are transmitted and scattered along the entire length of the fiber. In this embodiment, the absolute timestamp of the emitted probe light pulse is used as the start time of the period. The start time of the period is added to a preset duration to obtain the end time of the period. The time interval between the start time and the end time of the period is used as the temperature measurement period. The preset duration represents the physical flight time of a single probe light pulse for a single round trip along the entire length of the few-mode fiber.

[0024] In one specific implementation of this invention, the preset duration is set to 100 microseconds; and the fixed time interval for emitting probe light pulses is set to 200 microseconds.

[0025] Few-mode optical fiber, used as the sensing medium, is pre-embedded inside the high-voltage cable under test or bundled and laid on its surface. One end of the few-mode fiber is connected to the optical port of the distributed optical fiber temperature measurement host via an optical fiber connector, forming a continuous distributed temperature measurement system; the other end extends to the physical end of the high-voltage cable line. Cable nodes are virtual nodes arranged and assigned indexes (1 to N) sequentially at fixed spatial intervals along the physical laying direction of the few-mode fiber, from the first end connecting to the distributed optical fiber temperature measurement host to the last end. Because few-mode fiber is a continuous linear sensing medium, and the distributed optical fiber temperature measurement host can only calculate the relative distance based on the optical signal return time, and because there are inevitably residual and bending errors in the actual laying of few-mode fiber, it is not possible to directly collect data at fixed points based on the physical location of the cable joints. Therefore, this embodiment sets cable nodes to discretize the continuous analog electrical signal into a unified fixed spatial grid sequence, realizing continuous temperature scanning of the entire high-voltage cable without blind spots. Subsequently, specific cable nodes are aligned and matched with real physical cable joints to obtain the local real temperature rise data and contact thermal resistance parameters of each cable joint, thereby accurately assessing the independent health status of each cable joint.

[0026] In one specific implementation of this invention, the fixed spatial interval for dividing cable nodes is set to 0.4 meters.

[0027] During the transmission of the emitted probe light pulse along the entire few-mode fiber, it is modulated by the local temperature of the fiber, generating backscattered light carrying temperature information. This backscattered light returns along the original path of the few-mode fiber to the distributed fiber optic temperature measurement host. Because the light energy of the fundamental mode is concentrated at the very center of the fiber core, appearing as a single solid spot, while the light energy of higher-order modes deviates from the center of the fiber core, appearing as hollow rings or multi-petal shapes. This fundamental difference in spatial energy distribution allows the mode separation device built into the distributed fiber optic temperature measurement host to physically separate the light signals into various modes. This also results in different modes having drastically different responses to changes in external temperature; that is, each mode has different temperature sensitivities. Each mode includes a unique fundamental mode channel and multiple higher-order mode channels.

[0028] In this embodiment, the temperature sensitivity of each mode is obtained from the manufacturer's instructions for the few-mode fiber. Based on the difference between the temperature sensitivity of the higher-order mode of each higher-order mode channel and the temperature sensitivity of the fundamental mode channel, the specific higher-order mode channel for temperature measurement analysis in this embodiment is obtained.

[0029] Specifically, for any higher-order mode channel, the absolute value of the difference between the higher-order mode temperature sensitivity of the higher-order mode channel and the fundamental mode temperature sensitivity of the fundamental mode channel is taken as the two-mode sensitivity difference; the higher-order mode channel corresponding to the maximum value of the two-mode sensitivity difference is taken as the specific higher-order mode channel, and the maximum two-mode sensitivity difference is taken as the rated mode sensitivity difference of the few-mode fiber.

[0030] In one specific implementation of this invention, the rated mode sensitivity difference obtained from the manufacturer's instructions for the few-mode fiber is 0.015.

[0031] It should be noted that this embodiment is based on the optical signals in the fundamental mode channel of the few-mode fiber and a selected specific high-order mode channel for subsequent analysis and calculation.

[0032] In this embodiment, any temperature measurement cycle is taken as the temperature measurement cycle to be analyzed, and the explanation is based on the temperature measurement cycle to be analyzed.

[0033] The distributed fiber optic temperature measurement host is equipped with independent photodetectors for both the fundamental mode channel and specific higher-order mode channels. These detectors perform photoelectric conversion on the separated fundamental mode and higher-order mode optical signals, respectively, and output corresponding analog electrical signals. The high-speed analog-to-digital converter built into the distributed fiber optic temperature measurement host synchronously samples the analog electrical signals of the fundamental mode channel and specific higher-order mode channel at a preset sampling frequency. It records the delay time (i.e., light return time) of each sampling moment relative to the start time of the temperature measurement cycle being analyzed. Based on the principle of optical time-domain reflectometry, the delay time is converted into spatial distance according to the propagation speed of the probe light pulse in the few-mode fiber, obtaining the actual physical distance corresponding to each sampling moment. The actual physical distance represents the axial length of the fiber from the beginning of the few-mode fiber to the physical location where the corresponding backscattered light is generated.

[0034] It should be understood that the method of using the principle of optical time domain reflectometry (OTDR) to convert the transmission speed, return time and spatial distance of probe light in few-mode fiber is a well-known technology in this field, and the underlying conversion formula will not be elaborated here.

[0035] For any cable node, the physical cabling length from the cable node to the beginning of the few-mode fiber is taken as the distance length. The absolute value of the difference between the actual physical distance and the distance length at each sampling time is calculated. The analog electrical signals of the fundamental mode channel and the specific higher-order mode channel at the sampling time corresponding to the actual physical distance with the smallest absolute value of the difference are assigned to the cable node, thereby completing the accurate alignment and mapping between the continuous analog electrical signals and the discrete cable nodes.

[0036] Subsequently, based on the preset temperature calibration curve inside the distributed fiber optic temperature measurement host, the analog electrical signal intensity values ​​of the fundamental mode channel and the specific higher-order mode channel corresponding to each cable node are converted into Celsius temperature values, thereby obtaining the fundamental mode temperature and higher-order mode temperature of each cable node within the temperature measurement period to be analyzed.

[0037] In one specific implementation of this invention, the sampling frequency of the analog-to-digital converter is preset to 250 MHz (megahertz). The method for obtaining the propagation speed of the probe light pulse in the few-mode fiber is to divide the speed of light in vacuum by the effective refractive index of the few-mode fiber core. Here, the speed of light in vacuum is a known physical constant (approximately 3 × 10⁻⁶). 8 The effective refractive index (m / s) is an inherent physical parameter of the few-mode fiber provided by the manufacturer at the time of manufacture. In this embodiment, the effective refractive index of the few-mode fiber core is 1.5, resulting in a transmission speed of 2 × 10⁻⁶ m / s. 8 m / s.

[0038] In one specific implementation of this invention, the preset temperature calibration curve is a fixed mapping curve obtained by the distributed fiber optic temperature measurement host manufacturer before the equipment leaves the factory. This curve is obtained by testing the intensity of simulated electrical signals under multiple set temperature gradients using a standard constant temperature bath and then calibrating the curve through linear fitting using the least squares method. This curve is then directly burned and stored in the memory of the distributed fiber optic temperature measurement host. One simulated electrical signal intensity value corresponds to one Celsius temperature value.

[0039] This invention sends a query request carrying the start time of the temperature measurement cycle to the substation's backend operation database via a network communication interface. It extracts three-phase operating current data within a preset historical time period, calculated backwards from the cycle start time. Based on this historical three-phase operating current data, it calculates the equivalent load current for the temperature measurement cycle to be analyzed. The three-phase operating current data is collected in real-time by a microcomputer-based protection and control device installed in the substation switchgear, using current transformers at a fixed current sampling frequency. After analog-to-digital conversion within the microcomputer-based protection and control device, the data is continuously uploaded to the historical ledger data in the substation's backend operation database. The absolute timestamp of each current transformer sampling is used as the current sampling time.

[0040] For any historical current sampling moment within the preset historical duration of the temperature measurement cycle to be analyzed, the current values ​​of phases A, B, and C in the three-phase operating current data at that historical current sampling moment are read respectively, and the current value of the phase with the largest value is taken as the representative current. The sum of the squares of the representative currents at each historical current sampling moment within the preset historical duration is divided by the total number of historical current sampling moments to obtain the mean square value of the current. Finally, the square root of the mean square value of the current is calculated, and the result is taken as the equivalent load current of the temperature measurement cycle to be analyzed.

[0041] In this embodiment of the invention, if the equivalent load current of the temperature measurement period to be analyzed is less than or equal to a preset proportion (e.g., 5%) of the rated current carrying capacity of the high-voltage cable as specified by the manufacturer, it indicates that the high-voltage cable is in a power outage or extremely low load state during the temperature measurement period to be analyzed, and the heating characteristics are not obvious. At this time, the evaluation process of the temperature measurement period to be analyzed is directly terminated. If the equivalent load current of the temperature measurement period to be analyzed is greater than a preset proportion of the rated current carrying capacity of the high-voltage cable as specified by the manufacturer, the subsequent evaluation process of the temperature measurement period to be analyzed is executed.

[0042] In one specific implementation of this invention, the historical preset duration is set to 2 hours; the current sampling frequency of the current transformer is set to 1 Hz.

[0043] In this embodiment of the invention, to prevent thermodynamic delay caused by the large thermal inertia of the high-voltage cable, which would lead to a significant lag in temperature change on the surface of the high-voltage cable compared to the change in load current during drastic changes in grid load, thus distorting the physical correspondence between temperature and current and triggering false alarms, this embodiment performs a thermodynamic steady-state verification of the high-voltage cable before calculating the equivalent load current for the temperature measurement period to be analyzed: For any historical current sampling moment within a preset historical time period of the temperature measurement period to be analyzed, the maximum value of the three-phase current values ​​at that historical current sampling moment is taken as the representative current at that historical current sampling moment. The variance of all representative currents within the preset historical time period is calculated as the fluctuation variance. The preset fluctuation threshold represents the upper limit of the allowable stable fluctuation of the grid transmission load over a period of time.

[0044] If the fluctuation variance is greater than the preset fluctuation threshold, it means that the high-voltage cable has recently experienced drastic load changes, and the surface temperature of the cable joint has not yet reached thermal equilibrium due to the delayed heating effect. In this case, the evaluation process for the temperature measurement cycle to be analyzed is directly terminated. If the fluctuation variance is less than or equal to the preset fluctuation threshold, it means that the high-voltage cable is in a thermodynamic steady state, and the subsequent evaluation process for the temperature measurement cycle to be analyzed is executed.

[0045] In one specific implementation of this invention, the method for obtaining the preset fluctuation threshold is as follows: directly read the rated current carrying capacity specified on the nameplate of the high-voltage cable, calculate 5% of that rated current carrying capacity, and square the result to obtain the preset fluctuation threshold. In this embodiment, the preset fluctuation threshold obtained is 2500A. 2 .

[0046] The interference feature analysis module 102 is used to determine the heating temperature increment of each cable node based on the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node; and to determine the noise interference characteristics of each cable node based on the data fluctuation of the heating temperature increment of each cable node and the adjacent cable nodes.

[0047] The temperature increment represents the localized temperature rise caused purely by the cable joint's own fault, eliminating interference from the overall ambient background temperature caused by cable tunnels or heavy power grid loads. The fundamental mode temperature represents the apparent temperature measured through the fiber optic fundamental mode channel, while the higher-order mode temperature represents the apparent temperature measured through the fiber optic higher-order mode channel. Because the ambient background temperature in the same space has the same common-mode interference effect on both the fundamental mode channel and specific higher-order mode channels, and because the fundamental mode channel and specific higher-order mode channels have different temperature sensitivities to the temperature increment, the temperature increment obtained based on the difference between the fundamental mode temperature and the higher-order mode temperature can eliminate the ambient background temperature interference caused by common-mode interference. This prevents normal overall cable temperature rise from masking weak localized joint heating, eliminating the risk of equipment degradation underreporting and false alarms across the entire line from the source, ensuring high immunity and high accuracy of the extracted temperature data.

[0048] The noise interference feature represents the normalized spatial weight of whether the physical location of a specific cable node belongs to a true continuous heating region or a high-frequency random noise region. The closer the noise interference feature is to 1, the more likely the specific cable node is in a high-frequency random noise region; the closer the noise interference feature is to 0, the more likely the specific cable node is in a true continuous heating region. The degree of fluctuation in the temperature increment of each cable node and its adjacent cable nodes represents the degree of disorder and random jumps in the temperature increment within the local spatial range centered on the specific cable node.

[0049] According to the laws of physical heat transfer, when heat is actually conducted from a joint to both ends, the temperature change is smooth and gradual along the spatial axis. The fluctuation of the temperature increment between adjacent cable nodes is minimal, exhibiting a high degree of regularity. However, the background noise generated by the photodetector inside the distributed fiber optic temperature measurement host is completely random and jumps around in space, causing drastic and highly disordered fluctuations in the temperature increment between adjacent cable nodes. Therefore, by quantifying the degree of fluctuation of the temperature increment within a local spatial range, it is possible to accurately identify and distinguish between the real temperature rise signal and spurious high-frequency noise interference, providing accurate data support for subsequent elimination of high-frequency instrument noise and prevention of erroneous deletion of true extreme heating values.

[0050] Preferably, in some implementations of the present invention, determining the heating temperature increment of each cable node based on the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node includes: The temperature increment of each cable node is obtained by relating the absolute value of the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node to the rated mode sensitivity difference of the few-mode fiber.

[0051] Based on the principle of linear superposition of temperature fields, the fundamental mode temperature = ambient background temperature + heat generation temperature increment × fundamental mode temperature sensitivity, and the higher-order mode temperature = ambient background temperature + heat generation temperature increment × higher-order mode temperature sensitivity. The absolute value of the difference between the fundamental mode temperature and the higher-order mode temperature represents the actual temperature difference between different mode channels at the same location. It is the temperature response difference caused purely by local abnormal heating after eliminating the ambient background temperature. The rated mode sensitivity difference represents the difference in physical response between the fundamental mode channel and a specific higher-order mode channel of the few-mode fiber to a unit temperature change of 1 degree Celsius, that is, the absolute value of the difference between the fundamental mode temperature sensitivity of the fundamental mode channel and the higher-order mode temperature sensitivity of the specific higher-order mode channel in the data acquisition module 101.

[0052] Therefore, based on the absolute value of the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node, and the relative relationship between the absolute value of the difference and the rated mode sensitivity difference of the few-mode fiber, the temperature increment of each cable node can be obtained.

[0053] In this embodiment, the ratio of the absolute value of the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node to the rated mode sensitivity difference of the few-mode fiber is used as the heating temperature increment of each cable node; the above ratio reflects the relative relationship between the absolute value of the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node and the rated mode sensitivity difference of the few-mode fiber.

[0054] Preferably, in some implementations of the present invention, the noise interference characteristics of each cable node are determined based on the data fluctuation degree of the temperature increment of each cable node and its adjacent cable nodes, including: Take any cable node in the current temperature measurement cycle as the target cable node, and take the target cable node as the center of the preset spatial window. Based on the cumulative difference of the heating temperature increment between every two adjacent cable nodes in the preset spatial window, the temperature fluctuation characteristics of the target cable node are obtained. Based on the distribution range of temperature fluctuation characteristics of all cable nodes, a normalization operation is performed on the temperature fluctuation characteristics of the target cable node to obtain the noise interference characteristics of the target cable node.

[0055] Temperature fluctuation characteristics represent the intensity of local spatial temperature noise interference at a specific cable node. Higher values ​​indicate more drastic high-frequency fluctuations and random instrument noise, while lower values ​​represent smoother, continuous real physical heating. Normalization maps the temperature fluctuation characteristics to a fixed range (0 to 1), yielding noise interference characteristics. This allows for determining the penalty for denoising in subsequent calculations. Values ​​approaching 1 represent pure noise, requiring strong attenuation; values ​​approaching 0 represent real heating, requiring lossless preservation.

[0056] Taking any cable node within the current temperature measurement cycle as the target cable node (this embodiment uses the target cable node as an example), and using the target cable node as the center of a preset spatial window, the temperature increments of all cable nodes within the preset spatial window are extracted. The absolute value of the difference between the temperature increments of any two adjacent cable nodes within the preset spatial window represents the temperature change gradient or local temperature fluctuation amplitude of the high-voltage cable within the corresponding spatial interval. In a continuous, real physical heating area, this absolute value of the difference is small and smooth; however, in areas with high-frequency instrument noise, this absolute value of the difference exhibits drastic random jumps. Therefore, by accumulating the absolute values ​​of the difference between the temperature increments of every two adjacent cable nodes within the preset spatial window, the temperature fluctuation characteristics that quantify the noise interference intensity within the preset spatial window can be obtained.

[0057] In one specific implementation of this invention, the length of the preset space window is set to 11 cable nodes.

[0058] In this embodiment, if the target cable node is close to the beginning or end of the few-mode fiber, resulting in the number of cable nodes contained in the preset space window corresponding to the target cable node being less than 11, the temperature fluctuation characteristics of the target cable node are calculated using only the actual number of cable nodes contained in the preset space window.

[0059] Furthermore, based on the distribution range of temperature fluctuation characteristics of all cable nodes, a normalization operation is performed on the temperature fluctuation characteristics of the target cable node to obtain the noise interference characteristics of the target cable node, including: The ratio of the temperature fluctuation characteristic of the target cable node to the maximum value among the temperature fluctuation characteristics of all cable nodes is used as the noise interference characteristic of the target cable node.

[0060] The maximum value among the temperature fluctuation characteristics of all cable nodes represents the extreme reference point most severely affected by external environmental or instrument background noise across the entire cable line during the current temperature measurement cycle. Calculating the noise interference characteristics of each cable node based on this maximum value enables a unified quantification of the relative noise interference level of each cable node, facilitating the distinction between genuine temperature rise signals and spurious high-frequency noise interference.

[0061] In this embodiment, to prevent the amplification of weak background fluctuations leading to excessive penalties when there is no strong noise interference along the entire cable and the maximum value is extremely small, this embodiment compares the maximum value of the temperature fluctuation characteristics of all cable nodes with a preset environmental noise floor value before performing the normalization operation. The preset environmental noise floor value represents the baseline temperature fluctuation amplitude that the distributed fiber optic temperature measurement host objectively exists in an ideal noise-free environment. Temperature fluctuation characteristics below the preset environmental noise floor value are considered normal physical noise and should not be treated as destructive high-frequency noise for additional penalties.

[0062] If the maximum value is less than the preset environmental noise floor value, it means that there is no abnormal high-frequency interference noise along the entire cable in the current temperature measurement cycle, and all data fluctuations are inherent harmless baseline noise of the hardware. In order to prevent singularities in subsequent matrix operations, the noise interference characteristics of all cable nodes are set to a preset small constant value (e.g., 0.001).

[0063] In one specific implementation of this invention, the method for obtaining the preset environmental noise floor value is as follows: Under calibration conditions where the distributed fiber optic temperature measurement host is installed and the high-voltage cable is in a power-off, cold state (no load heating, thermal equilibrium state), the temperature fluctuation characteristics of each cable node are continuously calculated over multiple (e.g., 100) temperature measurement cycles according to the method of this embodiment. For any given temperature measurement cycle, the maximum value among the temperature fluctuation characteristics of all cable nodes within that cycle is taken as the representative temperature characteristic. The arithmetic mean of all representative temperature characteristics is taken as the preset environmental noise floor value. The preset environmental noise floor value obtained in this embodiment is 0.3℃.

[0064] The thermal resistance characteristic analysis module 103 is used to integrate the heating temperature increment and noise interference characteristics of all cable nodes to determine the true temperature increment of each cable node; based on the spatial distance characteristics between each cable node and each cable joint on the high-voltage cable, combined with the true temperature increment of each cable node and the equivalent load current, the contact thermal resistance parameters of each cable joint are analyzed.

[0065] The temperature increments of all cable nodes represent the temperature rise data purely caused by internal cable heating after eliminating background temperature interference from the overall power grid environment. The noise interference characteristics of all cable nodes represent the normalized weights used to distinguish whether the temperature increments at each cable node are actual heating or high-frequency instrument noise. The true temperature increment obtained by fusing the temperature increments and noise interference characteristics of all cable nodes represents the pure heating temperature value extracted from each cable node after eliminating high-frequency instrument noise interference and the local temperature averaging effect caused by probe light pulses, which is not masked by random optical background noise.

[0066] Due to manufacturing and transportation limitations, high-voltage cables are typically only a few hundred meters long per cable. In long-distance transmission lines, multiple cable segments must be spliced ​​together end-to-end; this splicing point is called a cable joint. Since real-world high-voltage cable joints have a physical length of approximately 1 meter, and the fixed spatial interval of the cable nodes in this embodiment is 0.4 meters, a single cable joint corresponds to a specific cable node section containing 2 to 3 consecutive cable nodes.

[0067] Based on the spatial distance characteristics between each cable node and each cable joint on the high-voltage cable, the specific cable node interval corresponding to each cable joint on the entire high-voltage cable can be determined. Then, based on the actual temperature increment of each cable node within the specific cable node interval, the actual temperature rise data of each cable joint can be determined.

[0068] According to Joule's law, the heating temperature of a cable joint is proportional to the square of the load current and the resistance. Therefore, based on the actual temperature rise data of the cable joint and the equivalent load current, the interference of dynamic fluctuations in the power grid load on temperature measurement can be eliminated, and the contact thermal resistance parameter, which is only related to the physical degradation state, can be decoupled.

[0069] Contact thermal resistance represents the degree of physical degradation of the internal metal conductive connection surface of a specific cable joint, caused by oxidation, aging, or mechanical loosening, resulting in increased thermal resistance. Equivalent load current represents the current reference of the actual steady-state power grid excitation source that causes cable heating, after excluding the thermal dynamic delay error caused by the enormous heat capacity of the high-voltage cable.

[0070] Preferably, such as Figure 2 As shown, in some implementations of the present invention, the thermal resistance feature analysis module 103 includes: a node filtering unit 201, a coefficient analysis unit 202, a basic vector construction unit 203, a feature matrix construction unit 204, and a feature vector analysis unit 205.

[0071] The node screening unit 201 screens noisy cable nodes based on the numerical characteristics of the noise interference characteristics of all cable nodes in the current temperature measurement cycle.

[0072] Noise cable nodes represent cable nodes located in high-frequency random noise regions where there is no smooth, continuous real heating signal.

[0073] The purpose of screening noisy cable nodes is to obtain the background noise interference caused purely by the external environment or instrument hardware, and to completely physically isolate the pure noise from the signal mixed with real heat. In this way, the adaptive penalty coefficient can be dynamically calculated according to the real-time electromagnetic interference environment to ensure that the penalty applied in subsequent noise reduction is just right, and to achieve adaptive filtering with high fault tolerance for different operating conditions.

[0074] Specifically, based on the numerical characteristics of noise interference features of all cable nodes within the current temperature measurement cycle, noisy cable nodes are screened, including: The preset high-frequency noise threshold represents the normalized critical ratio limit for distinguishing between real heating signals and high-frequency random noise interference signals. If the noise interference characteristics of any cable node are greater than or equal to the preset high-frequency noise threshold, it means that the temperature measurement data at that cable node exhibits drastic and irregular spatial jumps, completely lacking the spatial smooth continuity characteristic that must be present when the cable is physically heating up. This means that there is no real cable heating at that cable node, and the cable node is then screened as a noisy cable node.

[0075] In one specific implementation of this invention, the method for obtaining the preset high-frequency noise threshold is as follows: During the calibration phase at the initial stage of high-voltage cable commissioning, noise interference characteristics of all cable nodes are acquired over a period of time (e.g., 24 hours) during a stable operating period with known no overload heating. These noise interference characteristics are arranged in ascending order of value, and the noise interference characteristics located at the 95th percentile are set as the preset high-frequency noise threshold. In this embodiment, the preset high-frequency noise threshold is 0.9.

[0076] The coefficient analysis unit 202 determines the adaptive penalty coefficient for the current temperature measurement cycle based on the quantity characteristics of the noise cable nodes and the dispersion of the temperature increment of the noise cable nodes.

[0077] The adaptive penalty coefficient represents the penalty applied to suppressing and eliminating high-frequency noise interference signals when reconstructing the true temperature rise data of each cable node. The number of noisy cable nodes represents the total number of cable nodes with purely high-frequency random interference throughout the entire high-voltage cable. The dispersion of the temperature increment of noisy cable nodes represents the overall severity of the high-frequency noise interference signals actually encountered by the high-voltage cable.

[0078] Based on the numerical characteristics of the noise cable nodes and the dispersion of the temperature increment, an adaptive penalty coefficient is determined. This allows for the dynamic determination of the denoising penalty intensity that best matches the current field environment under different operating conditions, such as strong electromagnetic interference or extreme quiet conditions. This ensures that the reconstruction algorithm does not experience program deadlock under extreme quiet conditions while accurately eliminating high-frequency noise interference, thus improving the high fault tolerance and anti-interference capability of the reconstruction algorithm in complex field environments.

[0079] Specifically, based on the quantity characteristics of the noise cable nodes and the dispersion of the temperature increment of the noise cable nodes, an adaptive penalty coefficient for the current temperature measurement cycle is determined, including: If the total number of noisy cable nodes in the current temperature measurement cycle is less than 2, it means that the signal-to-noise ratio of the entire high-voltage cable is extremely high or it is in an absolutely silent state of power outage. There are almost no pure noise samples that are severely interfered with by high-frequency noise. In order to prevent the calculation of a false penalty force that has no statistical significance from causing the subsequent reconstruction algorithm to collapse, this embodiment uses the square of the preset environmental noise floor value as the adaptive penalty coefficient.

[0080] If the total number of noisy cable nodes in the current temperature measurement cycle is greater than or equal to 2, it means that there are enough pure noise samples in the current temperature measurement cycle. In order to make the noise reduction penalty dynamically adjusted to keep up with the on-site environment, this embodiment uses the variance of the temperature increment of all noisy cable nodes as the adaptive penalty coefficient.

[0081] The basic vector construction unit 203 constructs a local temperature rise column vector based on the temperature increment of each cable node in the current temperature measurement cycle.

[0082] The local temperature rise column vector represents the column vector consisting of the temperature increments of all cable nodes on the high-voltage cable caused solely by internal heating, excluding the ambient background temperature.

[0083] Specifically, as can be seen from the data acquisition module 101, from the beginning to the end of the few-mode fiber, the cable nodes are assigned serial number indices from 1 to N. The temperature increment of each cable node is arranged from top to bottom according to the ascending order of the serial number index of the corresponding cable node to construct a local temperature rise column vector.

[0084] The feature matrix construction unit 204 constructs a noise feature diagonal matrix based on the noise interference characteristics of each cable node in the current temperature measurement cycle; the main diagonal of the noise feature diagonal matrix is ​​the noise interference characteristics of each cable node.

[0085] The noise feature diagonal matrix represents the sparse penalty operator that maps one-dimensional spatial weights to two-dimensional matrix operations in the subsequent reconstruction algorithm. Mathematically, this ensures that the denoising penalty can be precisely applied to specific cable nodes, achieving independent weighted interception in the spatial domain.

[0086] Specifically, an N×N blank matrix is ​​constructed, where N is the total number of cable nodes, and the rows and columns of the blank matrix are the indexes of each cable node. All elements in this blank matrix except those on the main diagonal are uniformly assigned the value 0. The noise interference characteristics of each cable node in the current temperature measurement cycle are then sequentially filled into the main diagonal positions of this blank matrix, constructing a noise feature diagonal matrix. Specifically, the noise interference characteristics of the i-th cable node are filled into the i-th row and i-th column position of the noise feature diagonal matrix.

[0087] The feature vector analysis unit 205 uses a regularized least squares algorithm to fuse the adaptive penalty coefficient, the local temperature rise column vector, the noise feature diagonal matrix, and the preset pulse broadening matrix to obtain the reconstructed node temperature rise column vector; the elements in the reconstructed node temperature rise column vector are the actual temperature increments of each cable node.

[0088] The reconstructed node temperature rise column vector represents the column vector composed of the high-resolution pure heating temperature of each cable node after eliminating the local temperature averaging effect caused by the probe light pulse and attenuating the high-frequency instrument noise to the background level.

[0089] The preset pulse broadening matrix represents a mathematical model of the local temperature averaging effect caused by the probe light pulse emitted by the distributed fiber optic temperature measurement host in the actual physical space of the cable. It reflects the objective physical limitations: the energy carried by a single light pulse will inevitably cross and cover multiple cable nodes, causing the high temperature originally concentrated at one cable node to be forcibly averaged and distributed to multiple adjacent cable nodes.

[0090] By integrating the adaptive penalty coefficient, local temperature rise column vector, noise feature diagonal matrix, and preset pulse broadening matrix to obtain the reconstructed node temperature rise column vector, it breaks through the physical limit of pulse width set by the optical hardware factory in three dimensions: spatial morphology physical restoration, environmental noise dynamic tracking, and target feature sparse interception. It can reversely remove the signal spatial broadening effect with high precision under strong interference background, realize the restoration of the local true temperature rise characteristics of cable nodes, eliminate the phenomenon of heat generation being masked or alarm being suppressed due to spatial averaging in traditional temperature measurement methods, and improve the perception sensitivity and detection resolution of cable joint deterioration faults.

[0091] Specifically, in this embodiment, the formula for obtaining the column vector of temperature rise of the reconstructed nodes using the Tikhonov regularized least squares algorithm is as follows: in, This represents the column vector of temperature rise at the reconstructed node during the current temperature measurement cycle; This represents the transpose of the preset pulse broadening matrix; This represents the preset pulse broadening matrix; The diagonal matrix representing the noise characteristics of the current temperature measurement cycle; This represents the adaptive penalty coefficient for the current temperature measurement cycle; This represents the column vector of local temperature rise during the current temperature measurement cycle.

[0092] It should be understood that using the Tikhonov regularized least squares algorithm to perform matrix single-step inverse calculation is a well-known technique in the field of mathematics. However, the substantial contribution of this invention lies in cleverly transforming the optical physical constraints (preset pulse broadening matrix) and the on-site electromagnetic environment (adaptive penalty coefficient and noise feature diagonal matrix) into the input variables of this framework, giving this well-known algorithm the ability to break the optical spatial ambiguity limit and restore the true heating of the cable joint. Here, only a brief explanation of its mathematical solution process will be given, and it will not be elaborated further.

[0093] It should be noted that the elements in the reconstructed node temperature rise column vector are arranged in ascending order from top to bottom according to the index of each cable node. This represents the true, high-resolution temperature increment of each cable node after eliminating the interference of ambient background temperature and high-frequency noise, and eliminating the spatial averaging and flattening effect caused by the width of the probe light pulse.

[0094] In one specific implementation of this invention, the method for constructing the preset pulse broadening matrix is ​​as follows: The factory-rated duration of the probe light pulse of the distributed fiber optic temperature measurement host is read, multiplied by the transmission speed of the probe light pulse in the few-mode fiber to obtain the physical spatial coverage length, then divided by the fixed spatial interval for dividing the cable nodes and rounded down to obtain the number of cable nodes K covered by a single probe light pulse. Next, the standard discrete Gaussian window function, known in signal processing, is called to simulate the spatial coverage pattern of a single probe light pulse with strong energy at the center and weak energy at the edges, directly generating an initial vector of length K with values ​​that are maximized at the center and symmetrically decay towards both ends. The value of each element in the initial vector is divided by the sum of all element values ​​to ensure that the sum of all element values ​​in the normalized vector equals 1, and the normalized vector is used as a one-dimensional Gaussian distribution vector.

[0095] Initialize an N×N blank square matrix, where N is the total number of cable nodes. Use a one-dimensional Gaussian distribution vector as the basic unit and slide it downwards from left to right along the main diagonal of the blank square matrix, aligning the center element of the one-dimensional Gaussian distribution vector with the main diagonal element of the blank square matrix. Truncate and discard any numerical parts that exceed the dimensions of the square matrix during the arrangement process. Fill all uncovered blank positions in the square matrix with the value 0, generating a preset pulse broadening matrix after zero-filling and boundary processing.

[0096] Preferably, in some implementations of the present invention, based on the spatial distance characteristics between each cable node and each cable joint on the high-voltage cable, and combined with the actual temperature increment of each cable node and the equivalent load current, the contact thermal resistance parameters of each cable joint are analyzed, including: Take any cable connector as the target connector, and determine the target cable node of the target connector based on the spatial distance characteristics between the target connector and each cable node; the sum of the spatial distances between any two adjacent target cable nodes is greater than the total length of the target connector. Based on the average level of the actual temperature increment of the target cable node, the joint temperature rise characteristics are obtained; The ratio of the joint temperature rise characteristic to the square of the equivalent load current is used as the contact thermal resistance parameter of the target joint.

[0097] This embodiment uses any cable connector as the target connector, and will be described using the target connector as an example. The target cable node of the target connector represents the cable node in the specific cable node interval corresponding to the target connector.

[0098] Specifically, the physical wiring length from the front end of the target connector to the beginning of the few-mode fiber is extracted from the local database of the distributed fiber optic temperature measurement host as the first distance, and the physical wiring length from the end of the target connector to the beginning of the few-mode fiber is extracted as the second distance. The physical wiring length from each cable node to the beginning of the few-mode fiber is obtained from the local database of the distributed fiber optic temperature measurement host as the distance length of each cable node. The absolute value of the difference between the distance length of each cable node and the first distance is calculated as the first judgment distance for each cable node. For all cable nodes, the cable node whose distance length is less than or equal to the first distance and is the minimum of the first judgment distance is designated as the first node. The absolute value of the difference between the distance length of each cable node and the second distance is calculated as the second judgment distance for each cable node. For all cable nodes, the cable node whose distance length is greater than or equal to the second distance and is the minimum of the second judgment distance is designated as the second node.

[0099] The first node, the second node, and the cable node between the first node and the second node are designated as the target cable nodes of the target connector.

[0100] In this embodiment, if there is no cable node with a distance length less than or equal to the first distance, the cable node corresponding to the minimum value of the first judgment distance is directly taken as the first node; similarly, if there is no cable node with a distance length greater than or equal to the second distance, the cable node corresponding to the minimum value of the second judgment distance is directly taken as the second node.

[0101] The temperature rise characteristics of a cable joint represent the overall thermal severity of the target cable joint. Because abnormal heat generated by poor contact within a single cable joint can be conducted along the metal conductor and insulation material within a certain spatial range, the actual temperature increment of a single cable node is easily affected by accidental measurement errors caused by local fiber optic attachment conditions or external microenvironmental factors (such as uneven conduit or soil covering). Therefore, obtaining the joint temperature rise characteristics based on the actual temperature increments of all target cable nodes corresponding to the target joint can effectively smooth out and offset temperature fluctuations caused by poor local fiber optic contact.

[0102] Specifically, in this embodiment, the arithmetic mean of the actual temperature increments of all target cable nodes is used as the joint temperature rise characteristic; this arithmetic mean reflects the average level of the actual temperature increments of all target cable nodes.

[0103] According to Joule's law, the heating power equals the square of the current multiplied by the resistance. In thermal equilibrium, the steady-state temperature rise of a cable joint is proportional to the heating power. Therefore, it can be deduced that the ratio of the joint temperature rise characteristic to the square of the equivalent load current in this embodiment directly and equivalently characterizes the contact resistance of the corresponding cable joint, i.e., the contact thermal resistance parameter. This operation fundamentally eliminates the temperature rise fluctuations caused by current changes, allowing the obtained contact thermal resistance parameter to purely reflect the substantial deterioration degree of the internal metal conductive connection surface of the cable joint.

[0104] The result output module 104 is used to determine the health status of each cable joint based on the degree of deviation between the contact thermal resistance parameter of each cable joint and the contact thermal resistance reference value of each cable joint in the initial state.

[0105] The deviation of the contact thermal resistance parameter of each cable joint from its initial reference value represents the degree of deterioration of the contact thermal resistance of each cable joint in the current temperature measurement cycle compared to its contact thermal resistance at the initial stage of project delivery and commissioning, when the metal contact surface was in the best condition. Therefore, by judging the health status of each cable joint based on the deviation of its contact thermal resistance parameter from its initial reference value, we can accurately quantify hidden faults such as internal oxidation aging or mechanical loosening of cable joints, eliminate differences in the basic thermal resistance of cable joints of different specifications and models, and improve the alarm accuracy and equipment fault diagnosis reliability of long-term continuous operation monitoring of high-voltage cables.

[0106] In one specific implementation of this invention, taking any cable joint as an example, the method for obtaining the contact thermal resistance reference value is as follows: During the initial stable operation phase of the high-voltage cable project during its first power-on commissioning, the calculation results of the contact thermal resistance parameters of the cable joint for B consecutive temperature measurement cycles (e.g., B=10) are obtained. The arithmetic mean of all contact thermal resistance parameters is taken as the contact thermal resistance reference value of the cable joint. For example, for a healthy cable joint with a rated current carrying capacity of 1000A, the conventional order of magnitude of its contact thermal resistance reference value is typically 1.5 × 10⁻⁶. -6 ℃ / A 2 , where ℃ represents degrees Celsius and A represents amperes.

[0107] Preferably, in some implementations of the present invention, the health status of each cable joint is determined based on the degree of deviation between the contact thermal resistance parameter of each cable joint and the reference value of the contact thermal resistance of each cable joint in the initial state, including: The deterioration deviation is obtained by dividing the difference between the contact thermal resistance parameter of the target joint and the corresponding contact thermal resistance reference value by the corresponding contact thermal resistance reference value. If the degradation deviation is less than or equal to the preset deviation warning threshold, the target connector is in a healthy state; otherwise, the target connector is in an unhealthy state.

[0108] The contact thermal resistance baseline value represents the health benchmark of a specific cable joint's contact thermal resistance parameters at the initial stage of project delivery and commissioning, when the metal contact surface is in the best condition. The degradation deviation represents the actual thermal aging level of a specific cable joint, which is the proportion of deterioration relative to the optimal health condition of the specific cable joint when it was first put into operation. It is a dimensionless value that quantifies the degree of oxidation or mechanical loosening of the internal metal conductive connection surface of a specific cable joint due to long-term operation.

[0109] It should be noted that, due to the inherent resistance of the metal contact surface of the cable joint, the reference value of the contact thermal resistance of each cable joint in the initial state is always greater than zero.

[0110] The preset deviation warning threshold represents the maximum safety margin for the cable joint to transition from the normal operating temperature fluctuation range to the irreversible metal structure degradation stage.

[0111] This embodiment uses any cable connector as the target connector for illustration. If the deterioration deviation of the target connector is less than or equal to the preset deviation warning threshold, it means that although the contact thermal resistance parameter of the target connector in the current temperature measurement cycle may have increased slightly compared to the factory setting, the overall conductivity is still maintained within the safe and reliable design margin range, and therefore its condition is determined to be healthy. If the deterioration deviation of the target connector is greater than the preset deviation warning threshold, it means that the internal metal conductive connection surface of the target connector has undergone substantial and severe oxidation or mechanical loosening, which means that the target connector has lost its normal conductivity and is at risk of overheating and burn-through or causing an electrical fire at any time, and therefore its condition is determined to be unhealthy.

[0112] In one specific implementation of this invention, the preset deviation warning threshold is typically set within the range of 0.10 to 0.30 based on engineering experience. The specific value depends on the importance level of the high-voltage cable and the manufacturing process of the cable joint. When the high-voltage cable is a high-level line supplying power to critical hubs or core loads, and the reliability requirements are extremely high, a smaller preset deviation warning threshold (e.g., 0.10) is used to achieve early warning of minor defects, prioritizing increased inspection frequency to eliminate any potential power outages. When the high-voltage cable is a conventional distribution line, or the cable joint manufacturing process allows for a certain degree of thermodynamic deformation, a larger preset deviation warning threshold (e.g., 0.25) is used to improve anti-interference capabilities and avoid excessively frequent invalid alarms caused by fluctuations in normal operation. In this embodiment, the preset deviation warning threshold is set to 0.15.

[0113] In this embodiment, an independent counter is set for each cable connector. Taking the target connector as an example, if the target connector is judged to be healthy in the current temperature measurement cycle, the counter of the target connector is set to 0; if the target connector is judged to be unhealthy in the current temperature measurement cycle, the counter of the target connector is incremented by 1.

[0114] Furthermore, if the cumulative value of the counter for the target connector reaches the warning statistical threshold (e.g., 20), an equipment degradation judgment instruction containing the geographical location of the target connector will be output. This instruction will be directly sent to the monitoring backend or the maintenance personnel terminal to trigger the operation of tightening or replacing the cable connector at the location of the target connector; otherwise, the subsequent temperature measurement cycle analysis of the target connector will continue.

[0115] In summary, this invention determines the temperature increment of each cable node based on the difference between the base mode temperature and the higher-order mode temperature; it determines the noise interference characteristics of each cable node based on the fluctuation of the temperature increment data between each cable node and its adjacent cable nodes; it determines the true temperature increment of each cable node by integrating the temperature increments and noise interference characteristics of all cable nodes; it analyzes the contact thermal resistance parameters of each cable joint based on the spatial distance characteristics between each cable node and each cable joint on the high-voltage cable, combined with the true temperature increment of each cable node and the equivalent load current of the current temperature measurement cycle; and it judges the health status of each cable joint based on the deviation between the contact thermal resistance parameters of each cable joint and the contact thermal resistance reference value of each cable joint in the initial state.

[0116] The present invention also proposes a high-resolution continuous temperature measurement system for high-voltage cables based on few-mode optical fibers, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a scheme for a high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fibers.

Claims

1. A high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber, characterized in that, The device includes: The data acquisition module is used to acquire the basic mode temperature and higher-order mode temperature of each cable node in the current temperature measurement cycle, as well as the equivalent load current of the high-voltage cable in the current temperature measurement cycle. The interference feature analysis module is used to determine the temperature increment of each cable node based on the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node; and to determine the noise interference characteristics of each cable node based on the data fluctuation of the temperature increment of each cable node and the adjacent cable nodes. The thermal resistance characteristic analysis module is used to integrate the heating temperature increment and noise interference characteristics of all cable nodes to determine the true temperature increment of each cable node; based on the spatial distance characteristics between each cable node and each cable joint on the high-voltage cable, combined with the true temperature increment of each cable node and the equivalent load current, the contact thermal resistance parameters of each cable joint are analyzed. The results output module is used to determine the health status of each cable joint based on the degree of deviation between the contact thermal resistance parameter of each cable joint and the contact thermal resistance reference value of each cable joint in the initial state.

2. The high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber according to claim 1, characterized in that, The determination of the heating temperature increment for each cable node based on the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node includes: The temperature increment of each cable node is obtained by relating the absolute value of the difference between the fundamental mode temperature and the higher-order mode temperature of each cable node to the rated mode sensitivity difference of the few-mode fiber.

3. The high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber according to claim 1, characterized in that, The step of determining the noise interference characteristics of each cable node based on the fluctuation of the temperature increment of each cable node relative to adjacent cable nodes includes: Take any cable node in the current temperature measurement cycle as the target cable node, and take the target cable node as the center of the preset spatial window. Based on the cumulative difference of the heating temperature increment between every two adjacent cable nodes in the preset spatial window, the temperature fluctuation characteristics of the target cable node are obtained. Based on the distribution range of temperature fluctuation characteristics of all cable nodes, a normalization operation is performed on the temperature fluctuation characteristics of the target cable node to obtain the noise interference characteristics of the target cable node.

4. The high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber according to claim 3, characterized in that, The distribution range of temperature fluctuation characteristics of all cable nodes is used to normalize the temperature fluctuation characteristics of the target cable node, resulting in the noise interference characteristics of the target cable node, including: The ratio of the temperature fluctuation characteristic of the target cable node to the maximum value among the temperature fluctuation characteristics of all cable nodes is used as the noise interference characteristic of the target cable node.

5. The high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber according to claim 1, characterized in that, The process of integrating the temperature increments and noise interference characteristics of all cable nodes to determine the true temperature increment of each cable node includes: Based on the numerical characteristics of the noise interference features of all cable nodes within the current temperature measurement cycle, noisy cable nodes are screened. Based on the quantity characteristics of the noise cable nodes and the dispersion of the temperature increment of the noise cable nodes, an adaptive penalty coefficient is determined within the current temperature measurement cycle. Based on the temperature increment of each cable node during the current temperature measurement cycle, a local temperature rise column vector is constructed. Based on the noise interference characteristics of each cable node in the current temperature measurement cycle, a noise feature diagonal matrix is ​​constructed; the main diagonal of the noise feature diagonal matrix is ​​the noise interference characteristics of each cable node. The adaptive penalty coefficient, local temperature rise column vector, noise feature diagonal matrix, and preset pulse broadening matrix are fused using a regularized least squares algorithm to obtain the reconstructed node temperature rise column vector; the elements in the reconstructed node temperature rise column vector are the actual temperature increments of each cable node.

6. The high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber according to claim 5, characterized in that, The method of filtering noisy cable nodes based on numerical characteristics of noise interference features of all cable nodes within the current temperature measurement cycle includes: If the noise interference characteristics of any cable node are greater than or equal to the preset high-frequency noise threshold, then any cable node will be screened as a noisy cable node.

7. The high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber according to claim 5, characterized in that, The step of determining the adaptive penalty coefficient for the current temperature measurement cycle based on the quantity characteristics of the noise cable nodes and the dispersion of the temperature increments of the noise cable nodes includes: If the total number of noise cable nodes is less than 2, an adaptive penalty coefficient is determined based on a preset ambient noise floor value; if the total number of noise cable nodes is greater than or equal to 2, an adaptive penalty coefficient is obtained based on the dispersion of the heating temperature increment of the noise cable nodes.

8. The high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber according to claim 1, characterized in that, The analysis of the contact thermal resistance parameters of each cable joint, based on the spatial distance characteristics between each cable node and each cable joint on the high-voltage cable, combined with the actual temperature increment of each cable node and the equivalent load current, includes: Take any cable connector as the target connector, and determine the target cable node of the target connector based on the spatial distance characteristics between the target connector and each cable node; the sum of the spatial distances between any two adjacent target cable nodes is greater than the total length of the target connector. Based on the average level of the actual temperature increment of the target cable node, the joint temperature rise characteristics are obtained; The ratio of the joint temperature rise characteristic to the square of the equivalent load current is used as the contact thermal resistance parameter of the target joint.

9. A high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber according to claim 8, characterized in that, The health status of each cable joint is determined based on the deviation between its contact thermal resistance parameter and the reference value of its contact thermal resistance in the initial state. This includes: The deterioration deviation is obtained by dividing the difference between the contact thermal resistance parameter of the target joint and the corresponding contact thermal resistance reference value by the corresponding contact thermal resistance reference value. If the degradation deviation is less than or equal to the preset deviation warning threshold, the target connector is in a healthy state; otherwise, the target connector is in an unhealthy state.

10. A high-resolution continuous temperature measurement system for high-voltage cables based on few-mode optical fiber, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the scheme of a high-resolution continuous temperature measurement device for high-voltage cables based on few-mode optical fiber as described in any one of claims 1 to 9.

Images