A cable joint fault diagnosis method and system based on multi-source data fusion

Abstract

The application discloses a cable joint fault diagnosis method and system based on multi-source data fusion, and the method comprises the following steps: selecting a joint diagnosis unit, and taking a cable section in the joint diagnosis unit as a high-frequency electromagnetic wave resonant cavity; obtaining an operation response signal and a wideband transient signal of a target cable line; constructing a reference transfer function; taking the wideband transient signal as an endogenous probe signal; taking a triggering time of the endogenous probe signal as an excitation reference time; collecting an upstream response signal and a downstream response signal; constructing a transient transfer function according to the upstream response signal and the downstream response signal; inputting the endogenous probe signal into the transient transfer function and the reference transfer function respectively to obtain a healthy resonance response result and a real-time resonance response result; combining the healthy resonance response result and the real-time resonance response result to determine an inter-resonant cavity boundary change; and determining a fault cause of the cable joint based on the inter-resonant cavity boundary change. The application can effectively improve the accuracy of the cable joint fault diagnosis.

Description

Technical Field

[0001] This application relates to the field of cable joint fault diagnosis, and in particular to a cable joint fault diagnosis method and system based on multi-source data fusion. Background Technology

[0002] As power systems develop towards larger capacity, longer distances, and higher reliability, the application of cable lines in urban power distribution networks, rail transit, and new energy grid connection scenarios is constantly expanding. As a key node in line connection, cable joints bear the dual functions of electrical signal transmission and insulation protection. Their operating status directly determines the power supply reliability of the entire cable line. Once faults such as insulation deterioration or poor contact occur, it is very easy to cause line outages, or even induce serious accidents such as fires and large-scale power outages. Therefore, accurate and early diagnosis of cable joint faults is an important prerequisite for ensuring the safe and stable operation of the power grid.

[0003] Currently, fault diagnosis of cable joints mainly involves collecting operational data such as temperature, current, voltage, partial discharge, or high-frequency signals at the joint and analyzing the collected data to determine if there are any abnormal conditions at the joint. In practical applications, this diagnostic method typically treats each joint as an independent monitoring object, processing the operational data of each joint separately, and comprehensively evaluating abnormal amplitudes, energy changes, or statistical characteristics to identify and locate cable joint faults. However, the above-mentioned traditional cable joint diagnostic methods often rely on independent analysis of individual joints, ignoring the signal propagation relationship formed by multiple joints in the same cable line through the cable body. This leads to a lag in fault diagnosis, inaccurate judgment of the cause of the fault, and thus affects the accuracy of fault diagnosis.

[0004] There is currently no good solution to the above problems. Summary of the Invention

[0005] This application provides a cable joint fault diagnosis method and system based on multi-source data fusion to improve the accuracy of cable joint fault diagnosis.

[0006] To achieve the above objectives, the embodiments of this application adopt the following technical solutions: Firstly, a fault diagnosis method for cable joints based on multi-source data fusion is provided, applied to cable joints, wherein the cable joints include terminal cable joints and intermediate cable joints, and the number of terminal cable joints and intermediate cable joints is at least two. The method includes: In the target cable line, any intermediate cable joint and two terminal cable joints adjacent to the intermediate cable joint are selected as the joint diagnostic unit, and the cable segment in the joint diagnostic unit is used as the high-frequency electromagnetic wave resonant cavity. The terminal cable joint located on the power supply side is the upstream reflection end of the high-frequency electromagnetic wave resonant cavity, and the terminal cable joint located on the load side is the downstream reflection end of the high-frequency electromagnetic wave resonant cavity. Under the normal operating conditions of the target cable line, acquire the operating response signal and broadband transient signal of the target cable line; Construct a baseline transfer function based on the runtime response signal; Broadband transient signals are used as endogenous probe signals for high-frequency electromagnetic resonant cavities. The triggering time of the endogenous probe signal is used as the excitation reference time; Time synchronization configuration is performed on the sensor devices pre-set at the upstream and downstream reflective ends; Based on the excitation reference time, the sensor device with time synchronization configuration collects the upstream response signal and the downstream response signal of the upstream and downstream reflection ends, respectively. Construct an instantaneous transfer function based on the upstream and downstream response signals; The endogenous probe signal is input into the instantaneous transfer function and the reference transfer function respectively to obtain the healthy resonance response result and the real-time resonance response result; The variation of the inter-cavity boundary of the high-frequency electromagnetic wave resonant cavity is determined by combining the results of healthy resonance response and real-time resonance response. The causes of cable joint failures in the target cable line are determined based on the changes in the boundary of the inter-resonant cavity.

[0007] In one possible implementation of the first aspect, the method of constructing the reference transfer function based on the run response signal includes: When the target cable line is operating normally, a reference diagnostic unit is selected, which includes intermediate cable joints and terminal cable joints. Set key measurement points in the benchmark diagnostic unit; Active excitation signals are applied to key measurement points, and response signals from intermediate cable joints and terminal cable joints are collected respectively. Convert the response signals of intermediate cable joints and terminal cable joints into frequency domain signals; The frequency domain signal corresponding to the active excitation signal is used as the input signal, and the frequency domain response signals at the intermediate cable joint and the terminal cable joint are used as the output signals. The frequency domain transfer function of the reference diagnostic unit is constructed by combining the input signal and the output signal, and the frequency domain transfer function is used to characterize the reference transfer function.

[0008] In one possible implementation of the first aspect, the method of setting key measurement points in the benchmark diagnostic unit includes: Based on the structural composition of the target cable line, the physical range of the benchmark diagnostic unit is determined; Based on the structural characteristics of the benchmark diagnostic unit, the sensitivity of the benchmark diagnostic unit to changes in signal transmission characteristics is analyzed, and candidate key measurement points are selected based on the analysis results. The candidate key measurement point located at the intermediate cable joint is taken as the first key measurement point to characterize the signal transmission characteristics at the intermediate node. The candidate key measurement point located at the terminal cable joint is used as the second key measurement point to characterize the signal transmission characteristics at the terminal node and the response characteristics of the target cable line.

[0009] In one possible implementation of the first aspect, the method of constructing the instantaneous transfer function based on the upstream response signal and the downstream response signal includes: The cross-correlation function is obtained by performing cross-correlation analysis on the upstream and downstream response signals. The maximum time delay value is determined by combining the cross-correlation function of the upstream response signal and the downstream response signal, whereby the maximum time delay value represents the propagation time from the upstream response signal to the downstream response signal; The downstream response signal is time-domain aligned using the maximum time delay value to obtain the aligned downstream response signal; Extract the effective signal of transient events by combining the upstream response signal and the aligned downstream response signal; Perform a Fourier transform on the effective signal of the transient event to obtain the frequency domain complex amplitude values ​​of the upstream and downstream response signals; Calculate the self-power spectral density and cross-power spectral density based on the frequency domain complex amplitude values ​​of the upstream and downstream response signals; Regularization parameters are constructed based on the frequency signal intensity of the self-power spectral density to suppress noise amplification at low signal-to-noise ratio frequencies; The instantaneous transfer function is constructed by combining the self-power spectral density, cross-power spectral density, and regularization parameter.

[0010] In one possible implementation of the first aspect, the method of extracting the effective signal of the transient event by combining the upstream response signal and the aligned downstream response signal includes: Calculate the instantaneous energy envelope curves for the upstream response signal and the aligned downstream response signal respectively; Identify the peak energy and noise dynamics of valid signals in transient events; The event trigger threshold is determined based on the energy peak value and noise dynamic value; The start and end times of transient events are identified using event trigger thresholds and instantaneous energy envelope curves. The start time is the earliest moment when the instantaneous energy is greater than or equal to the event trigger threshold, and the end time is the latest moment when the instantaneous energy is less than the event trigger threshold. Based on the start and end times, extract the valid transient event signals from the upstream response signals and the aligned downstream response signals.

[0011] In one possible implementation of the first aspect, the method of obtaining healthy resonance response results and real-time resonance response results by inputting the endogenous probe signal into an instantaneous transfer function and a reference transfer function, respectively, includes: The endogenous probe signal is normalized to obtain a standardized endogenous probe signal; The frequency domain complex spectrum of the endogenous probe signal is obtained by performing a Fourier transform on the standardized endogenous probe signal. The frequency domain complex spectrum of the endogenous probe signal is input into the instantaneous transfer function to obtain the real-time frequency domain response result. The inverse Fourier transform of the real-time frequency domain response result is then performed to obtain the real-time resonant response result. The complex frequency spectrum of the endogenous probe signal is input into the reference transfer function to obtain the reference frequency domain response result. The inverse Fourier transform of the reference frequency domain response result is then performed to obtain the healthy resonance response result.

[0012] In one possible implementation of the first aspect, the method of determining the inter-cavity boundary change of the high-frequency electromagnetic resonant cavity by combining the healthy resonance response results with the real-time resonance response results through differential analysis includes: The degraded residual signal is constructed by performing time-domain and frequency-domain differential processing on the healthy resonance response results and the real-time resonance response results. Energy calculation is performed on the degraded residual signal to obtain a residual energy index, which is used to quantify the strength of the influence of the inter-cavity boundary change on the overall resonant response. When the residual energy index is greater than the preset energy threshold, frequency domain analysis is performed on the healthy resonance response result and the real-time resonance response result to identify the resonant mode peak generated by the high-frequency electromagnetic resonant cavity. Extract the corresponding modal parameters for any resonant mode peak; Calculate the resonant frequency deviation, resonant damping, and energy reflection changes by combining modal parameters; The boundary variation of the inter-cavity resonant cavity is determined by combining the resonant frequency deviation, resonant damping, and energy reflection variation.

[0013] In one possible implementation of the first aspect, the method of determining the inter-cavity boundary variation of the high-frequency electromagnetic resonant cavity by combining the resonant frequency offset, resonant damping, and energy reflection variation includes: When the change in resonant damping is an increase and the change in resonant frequency deviation is constant, it is determined to be resistive degradation. Based on resistive degradation, the change in the boundary of the inter-resonant cavity is determined to be an increase in contact resistance; When the change in resonant damping is a decrease and the change in resonant frequency deviation is constant, it is determined to be capacitive degradation. Based on the capacitive degradation, the boundary change of the inter-resonant cavity is determined to be an increase in the dielectric constant of the insulation.

[0014] Secondly, this application provides a machine-readable storage medium storing instructions that cause a machine to execute the aforementioned cable joint fault diagnosis method based on multi-source data fusion.

[0015] Thirdly, this application provides an electronic device, comprising: The memory is configured to store instructions; and The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the aforementioned cable joint fault diagnosis method based on multi-source data fusion.

[0016] By selecting specific intermediate and terminal cable joints as diagnostic units, the diagnostic scope can be limited to a specific area, improving the targeting and efficiency of the diagnosis. Using the cable segment as a high-frequency electromagnetic resonant cavity makes it easier to capture minute fault characteristics, improving fault detection sensitivity. Acquiring operational response signals and broadband transient signals provides richer dynamic information about the cable line. Constructing a reference transfer function accurately reveals the signal transmission characteristics of the cable line under normal operating conditions, enabling rapid detection of abnormalities deviating from normal conditions, thus achieving early fault warning and diagnosis. Using broadband transient signals as endogenous probe signals within the high-frequency electromagnetic resonant cavity excites specific frequency responses within the cavity, allowing minute structural changes to produce significant signal changes, thereby improving fault detection sensitivity. Using the trigger time of the endogenous probe signal as the excitation reference time eliminates the impact of signal acquisition time deviation on fault analysis, improving diagnostic accuracy. Time synchronization allows for precise comparison of upstream and downstream response signals, leading to more accurate analysis of signal propagation characteristics and aiding in precise fault location. Acquiring upstream and downstream response signals enhances fault feature extraction and improves fault diagnosis accuracy. Constructing an instantaneous transfer function (ITF) can reflect the signal transmission characteristics of a cable line in real time. Comparing it with a reference transfer function allows for rapid fault detection, enabling dynamic monitoring of the cable line and timely capture of fault development. By inputting the endogenous probe signal into both the ITF and the reference transfer function, the resonant response results under healthy and real-time conditions can be directly obtained, facilitating intuitive comparative analysis and clear identification of fault characteristics. Analyzing changes in the resonant cavity boundary can precisely locate the fault, improving the accuracy of fault diagnosis and helping to determine the nature of the fault, thus providing guidance for subsequent maintenance.

[0017] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0018] Figure 1 A flowchart illustrating a cable joint fault diagnosis method based on multi-source data fusion, provided for an embodiment of this application; Figure 2 This is a schematic diagram of a process for constructing a baseline transfer function, provided as an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0020] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0021] Furthermore, if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0022] Figure 1 The illustration schematically shows a flowchart of a cable joint fault diagnosis method based on multi-source data fusion according to an embodiment of this application. Figure 1 As shown in the embodiment of this application, a cable joint fault diagnosis method based on multi-source data fusion is provided. The method is applied to cable joints, which include terminal cable joints and intermediate cable joints. The number of terminal cable joints and intermediate cable joints is at least two. The method may include the following steps.

[0023] S101. In the target cable line, select any intermediate cable joint and two terminal cable joints adjacent to the intermediate cable joint as the joint diagnostic unit, and use the cable segment in the joint diagnostic unit as the high-frequency electromagnetic wave resonant cavity. The terminal cable joint located on the power supply side is the upstream reflection end of the high-frequency electromagnetic wave resonant cavity, and the terminal cable joint located on the load side is the downstream reflection end of the high-frequency electromagnetic wave resonant cavity. S102. Under the normal operating condition of the target cable line, acquire the operating response signal and broadband transient signal of the target cable line; S103. Construct a reference transfer function based on the operation response signal; S104. Use broadband transient signals as endogenous probe signals for high-frequency electromagnetic resonant cavities. S105. Use the triggering time of the endogenous probe signal as the excitation reference time; S106. Configure the sensor devices pre-set at the upstream and downstream reflective ends for time synchronization; S107. Based on the excitation reference time, the sensor device after time synchronization is used to collect the upstream response signal and the downstream response signal of the upstream reflection end and the downstream reflection end, respectively. S108. Construct an instantaneous transfer function based on the upstream and downstream response signals; S109. Input the endogenous probe signal into the instantaneous transfer function and the reference transfer function respectively to obtain the healthy resonance response result and the real-time resonance response result; S110. Combine the results of healthy resonance response with the results of real-time resonance response to determine the changes in the inter-cavity boundary of the high-frequency electromagnetic resonant cavity. S111. Determine the cause of cable joint failure in the target cable line based on the boundary changes of the inter-resonant cavity.

[0024] First, in the target cable line, based on the continuous spatial arrangement of cable joints, an arbitrary intermediate cable joint is selected. Using this intermediate cable joint as the center, two adjacent terminal cable joints are further selected to jointly construct an inter-joint diagnostic unit. The inter-joint diagnostic unit includes a cable segment located between the two terminal cable joints and an intermediate cable joint disposed on this cable segment. In this embodiment, the cable segment in the inter-joint diagnostic unit is equivalent to a high-frequency electromagnetic wave resonant cavity, used to characterize the propagation and reflection characteristics of high-frequency electromagnetic signals within the cable segment. Specifically, the terminal cable joint located on the power supply side, due to its structural and impedance discontinuities, reflects electromagnetic waves under high-frequency conditions and is defined as the upstream reflecting end of the high-frequency electromagnetic wave resonant cavity; the terminal cable joint located on the load side also reflects high-frequency electromagnetic waves and is defined as the downstream reflecting end of the high-frequency electromagnetic wave resonant cavity. A high-frequency electromagnetic wave resonant cavity structure with a clear reflection boundary is formed between the two terminal cable joints, allowing changes in the structural state of the intermediate cable joint and its adjacent cable segments to be mapped to changes in the electromagnetic response characteristics within the resonant cavity, facilitating the identification and fault diagnosis of the deterioration state of the intermediate cable joint and its adjacent cable segments.

[0025] Secondly, when the target cable line is in normal operation, signal acquisition is performed to obtain basic signal data characterizing its healthy operating characteristics. This involves acquiring the operating response signals naturally generated by the target cable line under actual operating conditions without affecting its normal power supply and load operation. These signals reflect the electromagnetic response characteristics of the cable line under normal power supply conditions. This includes voltage and current time-series data acquired during normal operation, as well as transient response data caused by load changes or operating disturbances. The transient response data can be time-domain statistical characteristic data and frequency-domain response data calculated from the operating response signals, used to characterize the electromagnetic response characteristics of the target cable line under actual operating conditions. Simultaneously, under normal operating conditions, the response signals of the target cable line to broadband transient disturbances are acquired. Broadband transient signals refer to electromagnetic signals with short durations in time and containing multiple frequency components in the frequency domain, used to excite the high-frequency electromagnetic propagation and reflection behavior of the cable line and its joints.

[0026] A reference transfer function is constructed based on the operational response signals. Specifically, under normal operating conditions of the target cable line, a reference diagnostic unit containing intermediate cable joints and terminal cable joints is selected, and key measurement points are set within the reference diagnostic unit. Subsequently, a known active excitation signal is applied to the key measurement points, and the response signals at the intermediate cable joints and terminal cable joints are collected respectively. Further, the collected response signals are converted to the frequency domain to obtain the corresponding frequency domain response signals. The frequency domain signal corresponding to the active excitation signal is used as the input signal, and the frequency domain response signals at the intermediate cable joints and terminal cable joints are used as the output signals. Based on the frequency domain mapping relationship between the input and output signals, the frequency domain transfer function of the reference diagnostic unit is constructed. The frequency domain transfer function is used to characterize the transmission and resonance characteristics of the target cable line under normal operating conditions.

[0027] The cable segments and joints in the selected benchmark diagnostic unit of the target cable line are equated to a high-frequency electromagnetic resonant cavity, and the broadband transient signal is used as the endogenous probe signal of the high-frequency electromagnetic resonant cavity. The endogenous probe signal is a signal that propagates inside the resonant cavity. It both excites the cavity to produce a response and is used to characterize the structural state of the cavity and joints. Propagating inside the resonant cavity, it is reflected by upstream and downstream terminal cable joints and modulated by intermediate cable joints, thereby generating collectable response signals at intermediate and terminal cable joints. By collecting the response signals and analyzing their frequency domain characteristics, the structural state and electromagnetic characteristics of the benchmark diagnostic unit can be characterized. This allows the broadband transient signal to act as both an excitation source and a detection tool without the need for external additional sensors, providing a physical basis for subsequent reference transfer function construction and cable degradation diagnosis.

[0028] The cable segments and joints in the selected reference diagnostic unit within the target cable line are equated to a high-frequency electromagnetic resonant cavity, and the broadband transient signal is used as the intrinsic probe signal of the resonant cavity. Simultaneously, the triggering time of the intrinsic probe signal is defined as the excitation reference time, used to time-align the input excitation signal with the response signals acquired at intermediate and terminal cable joints. By acquiring the response signals and analyzing their frequency domain characteristics, the frequency domain transfer function of the reference diagnostic unit can be constructed. This frequency domain transfer function characterizes the transmission and resonance characteristics of the reference diagnostic unit under healthy conditions.

[0029] Sensor devices are pre-installed at the upstream and downstream reflecting ends of the high-frequency electromagnetic resonant cavity to collect electromagnetic response signals at intermediate and terminal cable joints. The upstream reflecting end is located at the terminal cable joint on the power supply side, which is the entrance to the high-frequency resonant cavity, while the downstream reflecting end is located at the terminal cable joint on the load side, which is the exit of the high-frequency resonant cavity. The sensor devices are time-synchronized to ensure that the response signals collected by the upstream and downstream sensors are perfectly aligned in time under the same broadband transient disturbance, guaranteeing a precise time correspondence between the input and output response signals. All sensors can be connected to a unified data acquisition system, which provides a shared clock to ensure strict synchronization of sampling between the upstream and downstream sensors.

[0030] Using the excitation reference time as the time reference point for the signal's activation, and employing a time-synchronized sensor device, signal responses are acquired at the upstream and downstream reflecting ends of the high-frequency resonant cavity, respectively, yielding electromagnetic response data for the upstream and downstream connectors. This ensures that the data acquired at both ends are time-aligned, providing an accurate input-output mapping relationship. The upstream response signal, acquired at the sensor at the upstream reflecting end of the high-frequency electromagnetic resonant cavity, characterizes the reflection and local modulation characteristics of the endogenous probe signal within the cavity; the downstream response signal, acquired at the sensor at the downstream reflecting end, characterizes the propagation and transmission characteristics of the endogenous probe signal within the cavity.

[0031] Based on the upstream and downstream response signals collected from the upstream and downstream reflecting ends of the high-frequency electromagnetic wave resonant cavity, an instantaneous transfer function is constructed. Specifically, firstly, cross-correlation analysis is performed on the upstream and downstream response signals to obtain the cross-correlation function. The maximum time delay value is then determined using the cross-correlation function, representing the propagation time from the upstream to the downstream response signal. Subsequently, the downstream response signal is time-domain aligned using this maximum time delay value to obtain the aligned downstream response signal. The effective transient event signal is then extracted using the upstream response signal. A Fourier transform is performed on the effective transient event signal to obtain the frequency domain complex amplitude values ​​of the upstream and downstream response signals. Further, the autopower spectral density and cross-power spectral density are calculated based on the frequency domain complex amplitude values. A regularization parameter is constructed using the autopower spectral density to suppress noise amplification at low signal-to-noise ratio frequencies. Finally, by combining the autopower spectral density, cross-power spectral density, and regularization parameter, an instantaneous transfer function is constructed to characterize the frequency domain transmission and resonance characteristics of the reference diagnostic unit under transient events.

[0032] A broadband transient signal propagating within a high-frequency electromagnetic resonant cavity is used as an endogenous probe signal. This signal is input into both the instantaneous transfer function and the reference transfer function to obtain real-time and healthy resonant response results. First, the endogenous probe signal is normalized to obtain a standardized endogenous probe signal, and then subjected to a Fourier transform to obtain a complex frequency domain spectrum. Subsequently, the complex frequency domain spectrum is input into the instantaneous transfer function to obtain the real-time frequency domain response result, which is then obtained through an inverse Fourier transform to obtain the real-time resonant response result. Simultaneously, the complex frequency domain spectrum is input into the reference transfer function to obtain the reference frequency domain response result, which is then obtained through an inverse Fourier transform to obtain the healthy resonant response result. The real-time resonant response result reflects the transient signal propagation and resonance characteristics of the reference diagnostic unit under its current operating state, while the healthy resonant response result reflects the transmission and resonance characteristics of the reference diagnostic unit under healthy conditions.

[0033] Based on the healthy resonance response results and the real-time resonance response results, the boundary changes of the inter-cavity resonators in the high-frequency electromagnetic wave resonator are determined by combining time-domain and frequency-domain differential analysis. Specifically, the healthy resonance response results and the real-time resonance response results are differentially processed to construct a degraded residual signal. The degraded residual signal is obtained by performing time-domain and frequency-domain differential analysis on the healthy resonance response results and the real-time resonance response results, and is used to quantify the structural state changes of the benchmark diagnostic unit under transient events. Subsequently, energy calculation is performed on the degraded residual signal to obtain a residual energy index, which is used to quantify the strength of the influence of the boundary changes of the inter-cavity resonators on the overall resonance response. When the residual energy index is greater than a preset energy threshold, frequency-domain analysis is performed on the healthy resonance response results and the real-time resonance response results to identify the resonant mode peaks generated by the high-frequency electromagnetic wave resonator, and the corresponding modal parameters are extracted for any resonant mode peak. Subsequently, the resonant frequency deviation, resonant damping, and energy reflection changes are calculated based on the modal parameters, and the boundary changes of the inter-cavity resonators in the high-frequency electromagnetic wave resonator are determined by comprehensively considering the characteristic parameters.

[0034] The boundary change of the inter-resonant cavity is formed by the boundary of the high-frequency electromagnetic resonant cavity, which is jointly formed by the upstream end, the intermediate cable joint, the downstream end, and the adjacent cable segment. When the intermediate cable joint or the adjacent cable segment deteriorates, it will change the reflection and transmission characteristics of the cavity, thus forming the boundary change signal characteristics. Specifically, residual energy index, resonant frequency deviation, resonant damping, and energy reflection change are extracted from the residual signal. For example, when the joint shows a large residual energy, an increased reflection amplitude, and a small frequency deviation, it is determined that the joint is loose; when there is a significant resonant peak frequency deviation and an increased damping, it is determined that the joint has poor contact or oxidation; when there is a high-frequency mode anomaly and a sudden change in the boundary change signal, it is determined that the joint conductor is broken or disconnected.

[0035] By selecting specific intermediate and terminal cable joints as diagnostic units, the diagnostic scope can be limited to a specific area, improving the targeting and efficiency of the diagnosis. Using cable segments as high-frequency electromagnetic resonant cavities makes it easier to capture minute fault characteristics, improving fault detection sensitivity. Acquiring operational response signals and broadband transient signals provides richer dynamic information about the cable line. Constructing a reference transfer function accurately reveals the signal transmission characteristics of the cable line under normal operating conditions, enabling rapid detection of abnormalities deviating from normal conditions, thus achieving early fault warning and diagnosis. Using broadband transient signals as endogenous probe signals within the high-frequency electromagnetic resonant cavity excites specific frequency responses within the cavity, allowing minute structural changes to produce significant signal changes, thereby improving fault detection sensitivity. Using the trigger time of the endogenous probe signal as the excitation reference time eliminates the impact of signal acquisition time deviation on fault analysis, improving diagnostic accuracy. Time synchronization allows for precise comparison of upstream and downstream response signals, leading to more accurate analysis of signal propagation characteristics and aiding in precise fault location. Acquiring upstream and downstream response signals enhances fault feature extraction and improves fault diagnosis accuracy. Constructing an instantaneous transfer function (ITF) can reflect the signal transmission characteristics of a cable line in real time. Comparing it with a reference transfer function allows for rapid fault detection, enabling dynamic monitoring of the cable line and timely capture of fault development. By inputting the endogenous probe signal into both the ITF and the reference transfer function, the resonant response results under healthy and real-time conditions can be directly obtained, facilitating intuitive comparative analysis and clear identification of fault characteristics. Analyzing changes in the resonant cavity boundary can precisely locate the fault, improving the accuracy of fault diagnosis and helping to determine the nature of the fault, thus providing guidance for subsequent maintenance.

[0036] Figure 2 This application provides a flowchart illustrating the construction of a baseline transfer function. In one embodiment, the baseline transfer function is constructed based on the running response signal, including: S210. When the target cable line is operating normally, select the reference diagnostic unit. The reference diagnostic unit includes intermediate cable joints and terminal cable joints. S220. Set key measurement points in the benchmark diagnostic unit; S230. Apply active excitation signals to key measurement points and collect response signals from intermediate cable joints and terminal cable joints respectively. S240. Convert the response signals of intermediate cable joints and terminal cable joints into frequency domain signals; S250, The frequency domain signal corresponding to the active excitation signal is used as the input signal and the frequency domain response signals at the intermediate cable joint and the terminal cable joint are used as the output signals; S260. Combine the input signal and the output signal to construct the frequency domain transfer function of the reference diagnostic unit. The frequency domain transfer function is used to characterize the reference transfer function.

[0037] When the target cable line is in normal operating condition, a reference diagnostic unit is selected. This reference diagnostic unit includes intermediate cable joints and terminal cable joints. Specifically, based on the cable line layout, representative intermediate joints are selected, and terminal joints near the power supply end and load end are selected respectively. The intermediate joints, terminal joints, and the cable segments in between are combined to form the reference diagnostic unit. For example, assuming the cable line is 1000 meters long, with a cable joint installed every 200 meters, numbered J1 to J6, then J3, located in the middle of the line, can be selected as the representative intermediate joint, J1 as the upstream terminal joint near the power supply end, and J6 as the downstream terminal joint near the load end. The reference diagnostic unit, composed of the intermediate joints, terminal joints, and the cable segments in between, reflects the overall signal transmission characteristics of the target cable line in a healthy state.

[0038] Subsequently, key measurement points are set in the benchmark diagnostic unit. These key measurement points include a first key measurement point located at the intermediate cable joint and a second key measurement point located at the terminal cable joint. These key measurement points are used to acquire the signal transmission characteristics of the benchmark diagnostic unit in a healthy state. The acquired signals are used to establish a reference transfer function or resonant response reference. The setting of key points can be based on the principles of sensitivity and representativeness. That is, the measurement points should be placed at locations where the signal is most sensitive to changes in the structural state. For example, intermediate joints and their transition sections with the cables are most susceptible to degradation, resulting in the most significant signal changes. Furthermore, the measurement points should reflect the overall state of the entire benchmark diagnostic unit, not just local anomalies. For example, placing the measurement points at intermediate and terminal cable joints ensures clearer upstream and downstream boundary characteristics. Voltage / current sensors can be installed at the intermediate cable joint and cable transition sections to synchronously sample the voltage and current signals of the intermediate joint and both transition sections. The voltage and current signals are then analyzed in the time and frequency domains, respectively. Specifically, measurement points with significant amplitude changes, significant frequency peak shifts, or large phase changes can be identified as key measurement points for acquiring the signal transmission characteristics of the reference diagnostic unit in a healthy state. The signals from these key measurement points are used to establish a reference transfer function or resonant response reference.

[0039] Active excitation signals, which can be high-frequency pulse signals or broadband transient signals, are applied to key measurement points in the reference diagnostic unit to trigger its electromagnetic response. During excitation, response signals are simultaneously acquired at key measurement points at intermediate and terminal cable joints via time synchronization. These response signals reflect the input-output characteristics of the reference diagnostic unit in response to the applied excitation signals. The acquired response signals can be processed in both the time and frequency domains and correlated with the applied excitation signals to construct the reference transfer function or resonant response reference of the reference diagnostic unit. The reference transfer function can be used for subsequent excitation response analysis of endogenous probe signals, as well as for cable joint structural condition assessment and degradation diagnosis.

[0040] The response signals acquired at key measurement points of intermediate and terminal cable joints are converted from the time domain to the frequency domain to obtain the amplitude and phase characteristics of the signals at different frequencies. Specifically, a Fast Fourier Transform (FFT) is performed on the acquired time-domain voltage or current signals to convert the response signals of intermediate and terminal cable joints into frequency-domain signals. The resulting frequency-domain complex signals are used as output signals input to the reference transfer function or instantaneous transfer function to describe the input-output characteristics of the reference diagnostic unit.

[0041] The frequency domain representation of the active excitation signal applied to key measurement points, after frequency domain transformation, is used as the input signal. Simultaneously, the frequency domain representations of the response signals acquired at intermediate and terminal cable joints, after frequency domain transformation, are used as the output signals. A clear input-output correspondence is established between the input and output signals to construct the reference transfer function of the reference diagnostic unit. This transfer function characterizes the frequency response characteristics of the intermediate and terminal cable joints. Through the reference transfer function, comparative analysis of real-time acquired signals can be achieved, thereby determining the structural state and degradation of the cable joints, as well as the changes in the inter-cavity boundary of the high-frequency electromagnetic resonator. First, the time-domain response signals acquired at key measurement points are preprocessed, including DC component removal and normalization. Next, a fast Fourier transform is performed on the input and output signals to obtain a frequency-domain complex signal containing amplitude and phase information. For each frequency point, the amplitude of the output signal at that frequency point is divided by the amplitude of the input signal to obtain the amplitude ratio at that frequency point; simultaneously, the phase of the output signal at that frequency point is compared with the phase of the input signal to obtain the phase difference at that frequency point. Subsequently, the amplitude ratios at each frequency point are arranged sequentially according to frequency to form the amplitude response curve; the phase differences at each frequency point are arranged sequentially according to frequency to form the phase response curve. The amplitude response curve and the phase response curve together describe the input-output characteristics of the reference diagnostic unit across the entire frequency range, forming a complete frequency domain transfer function that can comprehensively reflect the gain and phase response characteristics of intermediate cable joints and terminal cable joints.

[0042] By constructing a reference transfer function, the reference transfer function of the cable line under normal operating conditions can be effectively obtained, which helps to promptly identify and locate potential problems in the cable line, and improve the operational reliability and maintenance efficiency of the cable line.

[0043] In one embodiment of this invention, key measurement points are set in the benchmark diagnostic unit, including: S310. Based on the structural composition of the target cable line, determine the physical range of the reference diagnostic unit; S320. Based on the structural characteristics of the reference diagnostic unit, perform a sensitivity analysis on the signal transmission characteristics of the reference diagnostic unit, and select candidate key measurement points based on the analysis results. S330. The candidate key measurement point located at the intermediate cable joint is taken as the first key measurement point to characterize the signal transmission characteristics at the intermediate node. S340. The candidate key measurement point located at the terminal cable joint is used as the second key measurement point to characterize the signal transmission characteristics at the terminal node and the response characteristics of the target cable line.

[0044] First, based on the structural composition of the target cable line, the physical scope of the benchmark diagnostic unit is determined. This involves analyzing the intermediate cable joints and their relative positions to the power supply and load-side terminal joints in the cable line, selecting a typical intermediate cable joint, its adjacent terminal joints, and cable segments to form the benchmark diagnostic unit. The physical scope of the benchmark diagnostic unit includes the central area where the intermediate cable joint is located, the upstream boundary area where the power supply-side terminal joint is located, the downstream boundary area where the load-side terminal joint is located, and the cable segments in between. In this embodiment, the physical scope of the benchmark diagnostic unit refers to a segment of the actual physical structure selected in the target cable line to establish a health status benchmark. The physical scope of the benchmark diagnostic unit may include an intermediate cable joint, the power supply-side terminal cable joint and the load-side terminal cable joint adjacent to the intermediate cable joint, and the cable segments between these joints. For example, the target cable line, along the power transmission direction, sequentially includes a power supply-side terminal cable joint T1, cable segment L1, intermediate cable joint J2, cable segment L2, intermediate cable joint J3, cable segment L3, and load-side terminal cable joint T4. Based on the cable line's structural layout, intermediate cable joint J2, located in the middle of the line and operating normally, is selected as the reference joint. The terminal cable joint T1 on the power supply side of intermediate cable joint J2 and the terminal cable joint T4 on the load side are selected as the physical boundaries of the reference diagnostic unit. Therefore, intermediate cable joint J2, the aforementioned power supply-side terminal cable joint T1, the load-side terminal cable joint T4, and the connecting cable segments L1, L2, and L3 are collectively defined as the physical range of the reference diagnostic unit, forming a high-frequency electromagnetic resonant cavity structure with clearly defined upstream and downstream reflecting ends.

[0045] Based on the structural characteristics of the benchmark diagnostic unit, a sensitivity analysis of signal transmission characteristic changes is performed, and candidate key measurement points are selected accordingly. Specifically, considering the structural composition of the intermediate cable joints, terminal cable joints, and cable segments in the benchmark diagnostic unit, the transmission signal at each structural location of each cable segment is obtained. A Fast Fourier Transform is performed on the preprocessed time-domain signal to extract the phase information corresponding to each frequency point. Subsequently, the phase difference at each structural location is quantified to obtain the response amplitude index of signal transmission characteristic changes, and the sensitivity of each structural location is determined based on the response amplitude index. Locations with large response amplitudes and obvious change characteristics are identified as high-sensitivity areas. Then, within the high-sensitivity areas, multiple response signals are collected at each location, and the consistency of signal shape or amplitude changes at each location is compared. Locations with stable signal changes and good repeatability are retained as candidate point locations; and among the candidate point locations, several evenly distributed measurement points that can cover the entire benchmark diagnostic unit are selected as candidate key measurement points.

[0046] The candidate key measurement point located at the intermediate cable joint is designated as the first key measurement point to characterize the signal transmission characteristics at the intermediate node. The voltage or current response signal acquired at the first key measurement point reflects the transmission characteristics of the intermediate cable joint and its adjacent cable segments under high-frequency or transient signal influences, including changes in amplitude attenuation, phase shift, and energy reflection. As the core measurement location, the first key measurement point provides a benchmark for subsequent instantaneous transfer function construction, resonant response analysis, and structural condition diagnosis.

[0047] The candidate critical measurement point located at the terminal cable joint is designated as the second critical measurement point to characterize the signal transmission characteristics at the terminal node and the overall response characteristics of the target cable line. The voltage or current response signals acquired at the second critical measurement point reflect the signal transmission characteristics of the terminal cable joint and its adjacent cable segments under high-frequency or transient signal influences, including changes in amplitude attenuation, phase shift, and energy reflection. The signals at the second critical measurement point can be compared and analyzed with those at the corresponding locations of the first critical measurement point, thereby achieving a comprehensive assessment of the transmission characteristics, structural condition, and degradation information of the target cable line.

[0048] By setting key measurement points, it is possible to monitor minute structural changes in intermediate joints and obtain the overall response characteristics of terminal nodes and the entire cable line, thereby enabling the monitoring and deterioration diagnosis of the structural status of cable joints and their adjacent cable sections.

[0049] In one embodiment of this example, constructing an instantaneous transfer function based on the upstream response signal and the downstream response signal includes: S410. Based on the cross-correlation analysis of the upstream and downstream response signals, the cross-correlation function is obtained; S420. Determine the maximum time delay value by combining the cross-correlation function of the upstream response signal and the downstream response signal. The maximum time delay value represents the propagation time from the upstream response signal to the downstream response signal. S430. Use the maximum time delay value to perform time-domain alignment on the downstream response signal to obtain the aligned downstream response signal; S440. Combine the upstream response signal and the aligned downstream response signal to extract the effective signal of the transient event; S450. Perform a Fourier transform on the effective signal of the transient event to obtain the frequency domain complex amplitude values ​​of the upstream and downstream response signals; S460. Calculate the self-power spectral density and cross-power spectral density based on the frequency domain complex amplitude values ​​of the upstream and downstream response signals. S470. Construct regularization parameters based on the frequency signal strength of the self-power spectral density to suppress noise amplification at low signal-to-noise ratio frequencies; S480, constructing the instantaneous transfer function by combining the self-power spectral density, cross-power spectral density, and regularization parameter.

[0050] Cross-correlation analysis is performed based on the upstream and downstream response signals to obtain the cross-correlation function. Sensors are installed at the upstream and downstream reflection ends of the high-frequency electromagnetic wave resonant cavity to collect synchronous transient response signals. After preprocessing the collected signals, the cross-correlation between the upstream and downstream response signals is calculated to obtain the cross-correlation function. Taking the upstream signal x(t) as a reference, the cross-correlation function Rxy(τ) between it and the downstream signal y(t) is calculated:

[0051] Where τ is the time delay, τ is obtained by cross-correlation analysis to obtain the cross-correlation function between the upstream response signal and the downstream response signal, and to determine the time delay corresponding to the peak position of the cross-correlation function.

[0052] Next, using MATLAB software, the peak position is found in the cross-correlation function curve. The time offset corresponding to this peak is the maximum time delay value, representing the propagation time of the signal from the upstream reflector to the downstream reflector. Based on the maximum time delay value, the downstream response signal is time-domain aligned to ensure the temporal consistency between the upstream and downstream signals.

[0053] The downstream response signal is time-domain aligned based on the maximum time delay value obtained from cross-correlation analysis. Specifically, the downstream response signal is shifted along the time axis by the entire time length, so that the transient characteristics of the downstream signal correspond to the propagation delay of the upstream response signal, thus obtaining the aligned downstream response signal. The aligned downstream response signal can be directly used for instantaneous transfer function calculation and frequency domain analysis with the upstream response signal to accurately characterize the signal transmission characteristics of cable joints and adjacent cable sections in the target cable line, providing a reliable basis for resonant response analysis and structural condition diagnosis.

[0054] The effective signal of the transient event is extracted by combining the upstream response signal and the aligned downstream response signal. First, the upstream response signal and the aligned downstream response signal are preprocessed separately. Second, short-time energy calculations are performed on the preprocessed upstream and downstream signals to obtain the corresponding energy sequences. These energy sequences are then combined to construct a joint energy sequence, which enhances the transient response characteristics triggered by the endogenous probe signal. The calculation can be performed using the following formula:

[0055]

[0056] Where Ex(i) represents the short-time energy sequence of the upstream response signal; Ey(i) represents the short-time energy sequence of the downstream response signal; k represents the sampling point index variable within the sliding window; L represents the half-window length of the short-time energy calculation window, i.e., L sampling points are taken before and after the current sampling point to participate in the energy calculation, thus forming an energy calculation window of length 2L+1; y(k) represents the signal amplitude of the aligned downstream response signal at the k-th sampling point. Subsequently, a joint energy sequence is constructed, that is, the energy sequence E(i) equals Ex(i) plus Ey(i). Since the transient event triggered by the endogenous probe signal will simultaneously produce energy abrupt changes in the upstream and downstream signals, the joint energy sequence can more clearly highlight the transient response characteristics. The noise characteristics of the joint energy sequence are statistically analyzed in the background time period without transient events, and the mean and standard deviation of the noise energy are calculated. This can be achieved using the mean and standard deviation formulas in existing technologies. An adaptive detection threshold is set based on the mean and standard deviation of noise energy. The adaptive detection threshold T = mean noise energy + n (standard deviation of noise energy), where n is the threshold coefficient, which can be set experimentally. This involves collecting signal samples containing transient events during system debugging, while simultaneously covering different noise environments the system might encounter. For example, data can be collected under low noise, normal noise, and high noise environments to ensure sample representativeness. Several candidate threshold coefficients are selected as test values, either at fixed intervals or based on experience. Each candidate threshold coefficient is applied to the collected sample signals to perform transient event detection, and the results of each detection are recorded, including the number of detected transient events and the number of false positives. Statistical analysis is performed on the detection results for each candidate threshold coefficient to calculate the detection accuracy and false positive rate. Based on the statistical analysis results, a threshold coefficient that ensures both high detection accuracy and low false positive rate is selected as the threshold coefficient, which can be chosen according to enterprise requirements. When the joint energy sequence satisfies E(i) > T, a transient event is determined to exist at the current time. When the joint energy sequence first exceeds the threshold T, the transient event starting point ts is determined. When the combined energy falls below the threshold again and remains below the threshold for a preset duration, the event is defined as ending at point te. The time interval of the transient event is thus defined as [ts, te], corresponding to the transient response triggered by the endogenous probe signal. When multiple energy peaks are detected, event separation can be achieved through peak interval and energy extreme value analysis. All local energy peaks are detected in the combined energy sequence E(i); the time interval between adjacent peaks is calculated; if the time interval between adjacent peaks is greater than the preset minimum event interval Tgap, they are determined to be two independent transient events; the start and end points of each event are then determined. When the interval between two events is small and partially overlaps, the events can be segmented based on the location of local energy valleys, thereby achieving separation of aliased events.The preset minimum event interval Tgap can be obtained by collecting typical transient event samples in the system and statistically analyzing the average duration of the transient events. Based on the determined time interval, corresponding signal segments are extracted from the upstream response signal and the aligned downstream response signal. The effective signal of the upstream transient event is xe(t) = x(t), t∈[ts,te]; the effective signal of the downstream transient event is ye(t) = y(t), t∈[ts,te], thus obtaining the effective signals of the upstream and downstream transient events.

[0057] Next, Fourier transforms are performed on the effective transient event signals extracted from the upstream response signal and the downstream response signal aligned according to the maximum time delay value, converting the time-domain signals into frequency-domain signals, thus obtaining the frequency-domain complex amplitude values ​​of the upstream and downstream response signals. The frequency-domain complex amplitude values ​​include the amplitude and phase information at each frequency point, reflecting the characteristics of the transient event in the frequency domain. Based on the frequency-domain complex amplitude values, the instantaneous transfer function, self-power spectral density, and cross-power spectral density can be further constructed and used to analyze the signal transmission characteristics and resonant response features of cable joints and adjacent cable segments in the target cable line.

[0058] Based on the frequency domain complex amplitudes of the upstream and aligned downstream response signals, the auto-power spectral density and cross-power spectral density of the upstream and downstream response signals are calculated respectively. Based on the frequency domain complex amplitude, the frequency domain complex amplitude of the upstream response signal is multiplied by its conjugate complex number to obtain the auto-power spectral density of the upstream response signal, which characterizes the energy distribution characteristics of the signal at various frequency points. Simultaneously, the frequency domain complex amplitudes of the upstream and downstream response signals are multiplied by their conjugate complex numbers to obtain the cross-power spectral density between the upstream and downstream response signals, which characterizes their energy correlation and phase transfer relationship in the frequency domain. The cross-power spectral density is used to characterize the energy correlation and phase transfer relationship between the upstream and downstream signals in the frequency domain.

[0059] When calculating the instantaneous transfer function, not all frequency points are equally reliable; those frequencies with very weak energy (low self-power spectral density) typically have a high noise content. If these frequency points are directly used for ratio calculations, the noise will be severely amplified. Therefore, by utilizing the signal strength reflected by the self-power spectral density, a protection term is set for each frequency point to prevent low signal-to-noise ratio frequencies from having an excessive impact on the results. Specifically, a regularization parameter sequence that varies with frequency is constructed based on the self-power spectral density. The regularization parameter can be based on a linear mapping method, and its specific expression is shown below:

[0060] Where λ(k) represents the regularization parameter, which is the weight value of the final output. The larger the value, the stronger the suppression; the smaller the value, the weaker the suppression. α represents the adjustment coefficient, which is manually set and used to globally control the strength of the suppression. η represents the average power level of the noise, which can be calculated by collecting noise data during periods without signal. The minimum value is used to prevent calculation errors caused by the denominator becoming zero when P(k) approaches 0. The adjustment coefficient can be determined based on the ratio of the maximum to minimum value of the self-power spectral density and its standard deviation, resulting in a relatively weaker regularization intensity at high signal-to-noise ratio frequencies. The minimum value can be set as a fixed proportion of the minimum self-power spectral density, and the specific fixed proportion can be set according to the company's accuracy requirements. The adjustment coefficient α is used to uniformly adjust the overall intensity of the regularization parameter, and its value is set according to the system's signal-to-noise ratio range. The self-power spectral density sequence p(k) of the current signal spectrum is statistically analyzed, and its maximum value pmax and minimum value pmin are calculated to obtain the power spectrum dynamic range R = pmax / pmin. Next, the adjustment coefficient α is set according to the dynamic range, specifically α = log(R). Under the current settings, when the dynamic range of the signal power spectrum is large, the value of α increases accordingly, thereby increasing the overall regularization parameter λ(k) to enhance the suppression capability of low signal-to-noise ratio (SNR) frequency points. When the dynamic range of the power spectrum is small, the value of α decreases relatively, thereby weakening the regularization strength to avoid excessive suppression of high SNR frequency components. Furthermore, the average noise power η is calculated from the noise data acquired by the system under conditions of no effective signal input. Specifically, when acquiring the time-domain noise sequence n(i), a fast Fourier transform is performed on the noise sequence to obtain the noise spectrum N(k), and the noise power spectral density is calculated. Subsequently, the average noise power level is obtained by averaging across all frequency points. Where K is the number of frequency points, and the obtained η reflects the overall energy level of system noise, which is used to compensate for the noise impact in the calculation of regularization parameters. Since η, p(k) and ε have the same power dimension, η / (p(k)+ε) is a dimensionless quantity, and the adjustment coefficient α is also a dimensionless coefficient, thus ensuring that the regularization parameter λ(k) is a dimensionless weight value.

[0061] Finally, the instantaneous transfer function is constructed by combining the self-power spectral density, cross-power spectral density, and regularization parameter, as shown in the following expression:

[0062] Where H(K) represents the instantaneous transfer function, reflecting the amplitude attenuation and phase shift characteristics of high-frequency electromagnetic waves propagating in the cable at that frequency; Pxy(k) represents the cross power spectral density; Pxx(k) represents the self power spectral density; and λ(k) represents the regularization parameter.

[0063] By introducing a regularization parameter λ(k) into the denominator, this method can effectively avoid abnormal amplification of the transfer function at low signal-to-noise ratio frequencies due to the small value of Pxx(k), significantly improving the noise resistance and stability of the transfer function estimation. It is especially suitable for the need to dynamically track the transmission characteristics of electromagnetic waves in high-frequency resonant cable diagnostic scenarios.

[0064] By constructing regularization parameters based on the frequency signal intensity of the self-power spectral density, noise amplification at low signal-to-noise ratio frequencies is effectively suppressed. Finally, by combining these parameters to construct the instantaneous transfer function, the signal characteristics of transient events can be accurately captured, effectively suppressing noise interference and thus improving the accuracy of the instantaneous transfer function construction.

[0065] In one embodiment of this invention, the effective signal of the transient event is extracted by combining the upstream response signal and the aligned downstream response signal, including: S510. Calculate the instantaneous energy envelope curves for the upstream response signal and the aligned downstream response signal respectively. S520: Identify the peak energy and noise dynamics of valid signals for transient events; S530. Determine the event trigger threshold based on the energy peak value and noise dynamic value; S540. Identify the start and end times of transient events using the event trigger threshold and the instantaneous energy envelope curve, where the start time is the earliest moment when the instantaneous energy is greater than or equal to the event trigger threshold, and the end time is the latest moment when the instantaneous energy is less than the event trigger threshold. S550. Based on the start and end times, extract the valid transient event signals from the upstream response signals and the aligned downstream response signals.

[0066] Instantaneous energy envelope curves are calculated for both the upstream and aligned downstream response signals. This involves extracting energy envelope features reflecting the time-varying signal intensity from the upstream and downstream response signals to characterize the energy transfer and attenuation of the transient signal between the upstream and downstream sides. Specifically, Hilbert transforms are applied to both the upstream and downstream response signals xu(t), using existing Hilbert transform formulas to obtain analytic signals. These analytic signals fully preserve the amplitude information of the original signal while formatting the phase information for easier subsequent processing. Next, the modulus (absolute value) of the analytic signal is taken to obtain the instantaneous amplitude, which can be understood as the contour of the original signal waveform, eliminating details of high-frequency oscillations. Energy is proportional to the square of the instantaneous amplitude; squaring the instantaneous amplitude yields the instantaneous energy sequence. Connecting the instantaneous energy sequences in chronological order creates the final instantaneous energy envelope curve, which primarily characterizes the overall contour and dynamic features of signal energy changes over time.

[0067] Based on the instantaneous energy envelope curve obtained above, the energy peak value and noise dynamic value of the effective signal of the transient event are identified. Specifically, the time interval in the instantaneous energy envelope curve where no transient events occur and energy changes are relatively stable is selected as the noise reference segment. Statistical analysis is performed on the energy values ​​within the noise reference segment to determine the noise dynamic value reflecting the background noise level under the current operating conditions. The time interval where no transient events occur is determined by judging the energy stability of the instantaneous energy envelope curve. Multiple consecutive candidate time windows are divided along the time axis in the instantaneous energy envelope curve, and statistical analysis is performed on the energy values ​​within each candidate time window to obtain the energy mean of the corresponding time window. When the energy change amplitude within a candidate time window is lower than the energy mean and no obvious energy surge is observed, it is determined that no transient event has occurred within the candidate time window, and this time window is determined as the noise reference segment. Subsequently, the energy changes are scanned in the instantaneous energy envelope curve to identify candidate energy peaks formed by energy surges. The candidate energy peaks are compared with the noise dynamic value. When the candidate energy peak is higher than the noise dynamic value and meets the preset duration condition, the corresponding energy peak is determined to be the effective signal energy peak of the transient event.

[0068] After obtaining the energy peak value corresponding to the transient event and the noise dynamic value reflecting the background noise level under the current operating conditions, the event trigger threshold is determined based on the energy peak value and the noise dynamic value. Specifically, using the noise dynamic value as the background energy benchmark, a preset threshold coefficient is introduced to proportionally amplify the noise dynamic value, thereby obtaining the event trigger threshold. When the energy peak value corresponding to the transient event is greater than or equal to the event trigger threshold, the transient event is determined to be triggered; when the energy peak value corresponding to the transient event is less than the event trigger threshold, the energy change is determined to be a noise fluctuation. Through this method, the event trigger threshold can be dynamically adjusted according to changes in the background noise level, improving the reliability and adaptability of transient event trigger determination under different noise environments while ensuring a simple and clear implementation. The preset threshold coefficient can be calculated by statistically analyzing the noise segments in the historical instantaneous energy envelope curve where no transient events occurred and the energy peak values ​​corresponding to historical transient events, calculating the ratio of the average event peak value to the average noise dynamic value, and determining it as a fixed value in combination with a safety margin; or the adaptive threshold coefficient can be dynamically calculated based on the real-time noise dynamic value and the historical energy peak value, so that the event triggering threshold can be dynamically adjusted with the changes in background noise level and transient event amplitude, thereby ensuring the reliability of the judgment while achieving simple operation and feasible implementation.

[0069] Based on the aforementioned instantaneous energy envelope curve and event trigger threshold, the start and end times of transient events are identified. The start time of a transient event is the point in time when the instantaneous energy first exceeds or equals the event trigger threshold, and the end time is the point in time when the instantaneous energy last falls below the event trigger threshold, thus determining the complete duration of the transient event in time. Specifically, selecting the earliest threshold breach as the start time captures the very beginning of the event without losing information; selecting the last drop as the end time captures the complete features of the event's tail. Linking with the threshold allows for adaptive noise control, preventing noise fluctuations from being mistaken for events. This provides a reliable time reference for subsequent transient event feature extraction, upstream and downstream energy comparison analysis, and state determination, while ensuring the accuracy and stability of event identification under different noise environments.

[0070] Based on the start and end times of the transient events identified above, signals within corresponding time periods are extracted from the upstream response signal and the aligned downstream response signal. Signals within these time periods are then identified as valid transient event signals. Specifically, a segment from the start to the end time is extracted from the upstream response signal as the valid upstream transient event signal; similarly, a segment from the start to the end time is extracted from the aligned downstream response signal as the valid downstream transient event signal. This method ensures that the upstream and downstream signals contain only the signal information of the actual transient events, eliminating background noise and interference from non-target disturbances. This provides a reliable data foundation for subsequent energy analysis, feature extraction, and upstream-downstream comparative analysis of transient events.

[0071] By accurately identifying and extracting transient events, efficient extraction of transient events can be achieved, improving the accuracy and reliability of signal processing and providing high-quality data support for subsequent signal analysis and applications.

[0072] In one embodiment of this invention, the endogenous probe signal is input into the instantaneous transfer function and the reference transfer function respectively to obtain the healthy resonance response result and the real-time resonance response result, including: S610. Normalize the endogenous probe signal to obtain a standardized endogenous probe signal. S620. Perform Fourier transform on the standardized endogenous probe signal to obtain the frequency domain complex spectrum of the endogenous probe signal; S630. Input the frequency domain complex spectrum of the endogenous probe signal into the instantaneous transfer function to obtain the real-time frequency domain response result, and perform an inverse Fourier transform on the real-time frequency domain response result to obtain the real-time resonant response result. S640. Input the frequency domain complex spectrum of the endogenous probe signal into the reference transfer function to obtain the reference frequency domain response result, and perform an inverse Fourier transform on the reference frequency domain response result to obtain the healthy resonance response result.

[0073] The acquired endogenous probe signals are normalized to eliminate differences in signal amplitude and unify signal dimensions, thereby obtaining standardized endogenous probe signals. Normalization can be achieved through a minimum-maximum linear transformation, mapping the minimum value of the endogenous probe signal within a preset time period to 0 and the maximum value to 1; or through a mean-variance standardization method, subtracting the mean from the endogenous probe signal and dividing by the standard deviation to obtain a standardized endogenous probe signal with a mean of 0 and a standard deviation of 1.

[0074] In this embodiment, a Fourier transform is performed on the standardized endogenous probe signal to convert the time-domain signal into a frequency-domain signal, thereby obtaining the complex spectrum of the endogenous probe signal. Specifically, the standardized endogenous probe signal is used as input and processed by Discrete Fourier Transform or Fast Fourier Transform to obtain complex spectrum data containing frequency component amplitude and phase information. The complex spectrum can reflect the energy distribution and phase characteristics of the signal at various frequencies, providing basic data for subsequent frequency domain feature extraction, frequency analysis, and system state determination of transient events, while improving the accuracy of analysis and the reliability of determination.

[0075] The complex frequency spectrum of the endogenous probe signal is used as input, and the real-time frequency domain response is calculated using the instantaneous transfer function (ITF), which characterizes the system's amplitude and phase response to each frequency component. Subsequently, an inverse Fourier transform is performed on the real-time frequency domain response to convert the signal back to the time domain, yielding the real-time resonant response. Through this process, the real-time resonant response reflects the system's transient response characteristics under current operating conditions, providing reliable time-domain response data for subsequent transient event analysis, energy feature extraction, and system state determination, while ensuring the integrity of amplitude and phase information.

[0076] The complex frequency spectrum of the endogenous probe signal is used as input, and the reference frequency domain response is calculated using a reference transfer function. This reference transfer function characterizes the amplitude and phase response of the system to each frequency component under healthy conditions. Subsequently, an inverse Fourier transform is performed on the reference frequency domain response to convert the frequency domain signal back to the time domain, yielding the healthy resonance response. Through this processing, the healthy resonance response reflects the transient response characteristics of the system under healthy conditions, providing reliable time-domain reference data for subsequent differential analysis between real-time and healthy resonance response results, transient event feature extraction, and system health status determination, while ensuring the integrity of amplitude and phase information.

[0077] By determining the healthy resonance response results and the real-time resonance response results, we can not only accurately reflect the real-time state of the system, but also compare and analyze it with the healthy state, providing strong technical support for system state monitoring and fault diagnosis, and improving the system's reliability and maintenance efficiency.

[0078] In one embodiment of this invention, the variation of the inter-cavity boundary of the high-frequency electromagnetic resonant cavity is determined by combining the healthy resonance response results with the real-time resonance response results, including: S710. Based on the healthy resonance response results and the real-time resonance response results, perform time-domain and frequency-domain differential processing to construct the degraded residual signal; S720. Perform energy calculation on the degraded residual signal to obtain the residual energy index. The residual energy index is used to quantify the strength of the influence of the change in the resonant cavity boundary on the overall resonant response. S730. When the residual energy index is greater than the preset energy threshold, frequency domain analysis is performed on the healthy resonance response result and the real-time resonance response result to identify the resonance mode peak generated by the high-frequency electromagnetic wave resonant cavity. S740. Extract the corresponding modal parameters for any resonant mode peak; S750, calculate the resonant frequency deviation, resonant damping, and energy reflection changes respectively based on the modal parameters; S760. Combine the changes in resonant frequency deviation, resonant damping, and energy reflection to determine the boundary changes of the inter-resonant cavity of the high-frequency electromagnetic wave resonant cavity.

[0079] By performing time-domain and frequency-domain differential processing on the real-time resonant response results and the healthy resonant response results, a degradation residual signal of the system is constructed. Specifically, the healthy resonant response result is subtracted point-by-point from the real-time resonant response result in the time domain to obtain the time-domain residual signal; simultaneously, the two are Fourier transformed and then differentially analyzed in the frequency domain to obtain the frequency-domain residual signal. The combination of the time-domain residual and the frequency-domain residual forms the degradation residual signal, which reflects the transient response changes and frequency response characteristic shifts of the system relative to its healthy state under the current operating conditions. This provides a reliable basis for subsequent system degradation type determination, health status assessment, and maintenance decisions, and effectively characterizes the degree and type of system degradation.

[0080] Energy calculations are performed on the deteriorated residual signal to obtain a residual energy index. Specifically, the amplitudes of each sampling point of the deteriorated residual signal are squared and summed to obtain the overall energy value of the system's deviation from its healthy state. The residual energy index reflects the overall influence of changes in the inter-cavity boundary on the system's resonant response. A larger residual energy value indicates a more significant impact of system degradation on the resonant response, while a smaller residual energy value indicates a weaker impact. The residual energy index provides a reliable quantitative basis for subsequent system health status assessment, degradation type determination, and maintenance decisions, and effectively reflects the overall changing characteristics of the system's resonant response under deterioration conditions.

[0081] When the residual energy index exceeds a preset energy threshold, further frequency domain analysis can be performed on the healthy resonance response results and the real-time resonance response results to identify the resonant mode peaks generated by the high-frequency electromagnetic resonant cavity. Specifically, the healthy resonance response results and the real-time resonance response results are subjected to Fourier transforms to obtain their frequency domain amplitude spectra, and the corresponding local peaks of the high-frequency electromagnetic resonant cavity are identified in the frequency domain amplitude spectra. The resonant mode peaks include peak frequency and amplitude information. By comparing the resonant mode peaks of the healthy response and the real-time response, the impact of system degradation on the high-frequency resonant cavity can be analyzed, revealing the changes in resonance characteristics. The preset energy threshold is set as the maximum value or high percentile value of the healthy residual energy under the healthy state by collecting multiple sets of healthy resonance response signals, constructing a healthy degradation residual signal, calculating its residual energy, and statistically obtaining the distribution of residual energy under the healthy state.

[0082] For any resonant mode peak generated by a high-frequency electromagnetic resonant cavity, its corresponding modal parameters can be extracted. The resonant mode peak is identified in the frequency domain amplitude spectrum, and its resonant frequency and amplitude are determined. Simultaneously, the damping of the mode is calculated using the peak half-width at half-maximum (FWHM) or fitting methods, and the phase information at the corresponding frequency is obtained. The modal parameters are used to characterize the resonant mode and can be compared with the modal parameters under healthy conditions to analyze the impact of system degradation on specific resonant modes.

[0083] The resonant frequency offset, resonant damping, and energy reflection changes are calculated by combining modal parameters. The peak characteristics of each resonant mode obtained from frequency domain analysis include, but are not limited to, the resonant frequency fr, peak amplitude Ar, damping ζ, and phase. The resonant frequency offset describes the deviation of the current modal frequency relative to the healthy modal frequency, and is calculated as Δfr = fr 实时 -fr 健康 The resonant damping degree describes the rate of modal decay, and it is calculated as Δζ = ζ. 实时 -ζ 健康 Energy reflection variation describes the reflection or retention of modal energy in the resonant cavity. It can be represented by the relative change in peak amplitude or peak energy, and is calculated as ΔEr = Ar. 实时 -Ar 健康A decrease in energy indicates increased resistive loss, while an increase in energy indicates reduced energy attenuation. By comparing the resonant mode frequency, amplitude, and damping under healthy conditions with the corresponding modal parameters under real-time conditions, the changes in resonant frequency deviation, resonant damping, and energy reflection are calculated. Specifically, the resonant frequency deviation is the difference between the real-time modal frequency and the healthy modal frequency, reflecting the modal frequency shift; the resonant damping is the difference between the real-time modal damping and the healthy modal damping, reflecting the rate of modal attenuation; and the energy reflection change is the difference between the real-time modal amplitude and the healthy modal amplitude, reflecting the retention or loss of resonant energy within the cavity. The calculation results can be used to further analyze the impact of system degradation on each resonant mode and to determine the changes in the inter-cavity boundary and degradation type of the high-frequency electromagnetic wave resonant cavity, providing a reliable basis for system health status assessment and maintenance decisions.

[0084] By combining the changes in resonant frequency deviation, resonant damping, and energy reflection of the high-frequency electromagnetic wave resonator, the boundary changes of the inter-cavity resonators can be determined. Specifically, when the resonant damping increases and the resonant frequency deviation remains unchanged, the system is determined to have resistive degradation, and the boundary change of the inter-cavity resonators is further determined to be an increase in contact resistance. When the resonant damping decreases and the resonant frequency deviation remains unchanged, the system is determined to have capacitive degradation, and the boundary change of the inter-cavity resonators is further determined to be an increase in the dielectric constant of the insulation. This method can accurately identify the type of boundary change of the inter-cavity resonators in high-frequency electromagnetic wave resonators based on the variation law of resonant parameters, providing a reliable basis for system health status assessment, degradation type determination, and maintenance decisions, and effectively reflecting the changes in the resonant characteristics of the system under different degradation conditions.

[0085] By combining the results of healthy resonance response with the results of real-time resonance response for differential analysis, the system achieves accurate detection and quantitative evaluation of the boundary changes between high-frequency electromagnetic wave resonant cavities. It can not only accurately identify boundary changes, but also quantify their impact, providing a comprehensive and accurate technical means for health monitoring and fault diagnosis of high-frequency electromagnetic wave resonant cavities, and significantly improving the reliability and maintenance efficiency of the system.

[0086] In one embodiment of this invention, the change in the inter-cavity boundary of the high-frequency electromagnetic resonant cavity is determined by combining the resonant frequency deviation, resonant damping, and energy reflection changes, including: S810. When the change in resonant damping is increasing and the change in resonant frequency deviation is constant, it is determined to be resistive degradation. S820, based on resistive degradation, the change in the boundary of the inter-resonant cavity is determined to be an increase in contact resistance; S830. When the change in resonant damping is a decrease and the change in resonant frequency deviation is constant, it is determined to be capacitive degradation. S840, based on capacitive degradation, determines that the boundary change of the inter-resonant cavity is due to an increase in the dielectric constant of the insulation.

[0087] By analyzing the real-time and healthy resonance response results, the system's resonant damping and resonant frequency deviation are obtained. When the resonant damping increases while the resonant frequency deviation remains constant, the system is considered to have undergone resistive degradation. The resonant damping characterizes the system's vibration attenuation, while the resonant frequency deviation characterizes the shift in the system's resonant frequency. Through joint analysis of these parameter changes, the system degradation type can be identified as resistive degradation, meaning that the system's damping characteristics change while its frequency characteristics remain relatively stable.

[0088] When a system is determined to have resistive degradation, it can be further determined that the boundaries of the resonant cavities between systems have changed, specifically manifested as an increase in contact resistance. Resistive degradation is obtained by analyzing the real-time resonant response and the healthy resonant response, and is characterized by an increase in resonant damping while the resonant frequency deviation remains unchanged. By correlating resistive degradation with the boundary characteristics of the resonant cavities between systems, the increase in contact resistance at system interfaces or connectors can be identified. This provides a reliable basis for subsequent system fault location, interface maintenance, and health status assessment, and effectively reflects the changes in system transmission characteristics under resistive degradation.

[0089] By analyzing the real-time and healthy resonance response results, the system's resonance damping and resonant frequency deviation are obtained. When the resonance damping decreases while the resonant frequency deviation remains unchanged, capacitive degradation of the system is determined. The resonance damping characterizes the system's vibration attenuation characteristics, while the resonant frequency deviation characterizes the shift in the system's resonant frequency. Through joint analysis of the changes in these parameters, the system degradation type can be identified as capacitive degradation, meaning the system's energy storage capacity decreases while its frequency characteristics remain relatively stable.

[0090] When a system is determined to have capacitive degradation, it can be further determined that the boundaries of the inter-system resonant cavities have changed, specifically manifested as an increase in the dielectric constant of the insulation. Capacitive degradation is obtained by analyzing the real-time resonant response and the healthy resonant response, and is characterized by a decrease in resonant damping while the resonant frequency deviation remains unchanged. By correlating capacitive degradation with the boundary characteristics of the inter-system resonant cavities, an increase in the capacitance characteristics of the system's insulating material or dielectric can be identified. This provides a reliable basis for subsequent system health status assessment, dielectric maintenance, and fault location, and effectively reflects the changes in system transmission characteristics under capacitive degradation.

[0091] By accurately judging the changes in the resonant cavity boundaries between high-frequency electromagnetic wave resonant cavities, different types of degradation can be accurately distinguished, providing clear guidance for the maintenance and optimization of high-frequency electromagnetic wave resonant cavities, effectively improving the reliability and stability of the system, and reducing the risk of performance degradation caused by boundary changes.

[0092] This application also provides a machine-readable storage medium storing instructions that cause a machine to execute the above-described cable joint fault diagnosis method based on multi-source data fusion.

[0093] This application also provides an electronic device, including: The memory is configured to store instructions; and The processor is configured to retrieve instructions from memory and, when executing instructions, to implement the aforementioned method for diagnosing cable joint faults based on multi-source data fusion.

[0094] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0095] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0096] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0097] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0098] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0099] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, like read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0100] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0101] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0102] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A cable joint fault diagnosis method based on multi-source data fusion, characterized in that, Applied to cable joints, the cable joints include terminal cable joints and intermediate cable joints, the number of which is at least two, including: In the target cable line, any intermediate cable joint and two terminal cable joints adjacent to the intermediate cable joint are selected as the joint diagnostic unit, and the cable segment in the joint diagnostic unit is used as the high-frequency electromagnetic wave resonant cavity. The terminal cable joint located on the power supply side is the upstream reflection end of the high-frequency electromagnetic wave resonant cavity, and the terminal cable joint located on the load side is the downstream reflection end of the high-frequency electromagnetic wave resonant cavity. Under the normal operating conditions of the target cable line, acquire the operating response signal and broadband transient signal of the target cable line; Construct a baseline transfer function based on the runtime response signal; Broadband transient signals are used as endogenous probe signals for high-frequency electromagnetic resonant cavities. The triggering time of the endogenous probe signal is used as the excitation reference time; Time synchronization configuration is performed on the sensor devices pre-set at the upstream and downstream reflective ends; Based on the excitation reference time, the sensor device with time synchronization configuration collects the upstream response signal and the downstream response signal of the upstream and downstream reflective ends, respectively. Construct an instantaneous transfer function based on the upstream and downstream response signals; The endogenous probe signal is input into the instantaneous transfer function and the reference transfer function respectively to obtain the healthy resonance response result and the real-time resonance response result; The variation of the inter-cavity boundary of the high-frequency electromagnetic wave resonant cavity is determined by combining the results of healthy resonance response and real-time resonance response. The causes of cable joint failures in the target cable line are determined based on the changes in the boundary of the inter-resonant cavity.

2. The method according to claim 1, characterized in that, The construction of the reference transfer function based on the operation response signal includes: When the target cable line is operating normally, a reference diagnostic unit is selected, which includes intermediate cable joints and terminal cable joints. Set key measurement points in the benchmark diagnostic unit; Active excitation signals are applied to key measurement points, and response signals from intermediate cable joints and terminal cable joints are collected respectively. Convert the response signals of intermediate cable joints and terminal cable joints into frequency domain signals; The frequency domain signal corresponding to the active excitation signal is used as the input signal, and the frequency domain response signals at the intermediate cable joint and the terminal cable joint are used as the output signals. The frequency domain transfer function of the reference diagnostic unit is constructed by combining the input signal and the output signal, and the frequency domain transfer function is used to characterize the reference transfer function.

3. The method according to claim 2, characterized in that, Setting key measurement points in the benchmark diagnostic unit includes: Based on the structural composition of the target cable line, the physical range of the benchmark diagnostic unit is determined; Based on the structural characteristics of the benchmark diagnostic unit, the sensitivity of the benchmark diagnostic unit to changes in signal transmission characteristics is analyzed, and candidate key measurement points are selected based on the analysis results. The candidate key measurement point located at the intermediate cable joint is taken as the first key measurement point to characterize the signal transmission characteristics at the intermediate node. The candidate key measurement point located at the terminal cable joint is used as the second key measurement point to characterize the signal transmission characteristics at the terminal node and the response characteristics of the target cable line.

4. The method according to claim 1, characterized in that, The construction of the instantaneous transfer function based on the upstream and downstream response signals includes: The cross-correlation function is obtained by performing cross-correlation analysis on the upstream and downstream response signals. The maximum time delay value is determined by combining the cross-correlation function of the upstream response signal and the downstream response signal, whereby the maximum time delay value represents the propagation time from the upstream response signal to the downstream response signal; The downstream response signal is time-domain aligned using the maximum time delay value to obtain the aligned downstream response signal; Extract the effective signal of transient events by combining the upstream response signal and the aligned downstream response signal; Perform a Fourier transform on the effective signal of the transient event to obtain the frequency domain complex amplitude values ​​of the upstream and downstream response signals; Calculate the self-power spectral density and cross-power spectral density based on the frequency domain complex amplitude values ​​of the upstream and downstream response signals; Regularization parameters are constructed based on the frequency signal intensity of the self-power spectral density to suppress noise amplification at low signal-to-noise ratio frequencies; The instantaneous transfer function is constructed by combining the self-power spectral density, cross-power spectral density, and regularization parameter.

5. The method according to claim 4, characterized in that, The extraction of effective transient event signals by combining upstream response signals and aligned downstream response signals includes: Calculate the instantaneous energy envelope curves for the upstream response signal and the aligned downstream response signal respectively; Identify the peak energy and noise dynamics of valid signals in transient events; The event trigger threshold is determined based on the energy peak value and noise dynamic value; The start and end times of transient events are identified using event trigger thresholds and instantaneous energy envelope curves. The start time is the earliest moment when the instantaneous energy is greater than or equal to the event trigger threshold, and the end time is the latest moment when the instantaneous energy is less than the event trigger threshold. Based on the start and end times, extract the valid transient event signals from the upstream response signals and the aligned downstream response signals.

6. The method according to claim 1, characterized in that, The step of inputting the endogenous probe signal into the instantaneous transfer function and the reference transfer function respectively to obtain the healthy resonance response result and the real-time resonance response result includes: The endogenous probe signal is normalized to obtain a standardized endogenous probe signal; The frequency domain complex spectrum of the endogenous probe signal is obtained by performing a Fourier transform on the standardized endogenous probe signal. The frequency domain complex spectrum of the endogenous probe signal is input into the instantaneous transfer function to obtain the real-time frequency domain response result. The inverse Fourier transform of the real-time frequency domain response result is then performed to obtain the real-time resonant response result. The complex frequency spectrum of the endogenous probe signal is input into the reference transfer function to obtain the reference frequency domain response result. The inverse Fourier transform of the reference frequency domain response result is then performed to obtain the healthy resonance response result.

7. The method according to claim 1, characterized in that, The method of determining the inter-cavity boundary variation of the high-frequency electromagnetic wave resonant cavity by combining the healthy resonance response results with the real-time resonance response results includes: The degraded residual signal is constructed by performing time-domain and frequency-domain differential processing on the healthy resonance response results and the real-time resonance response results. Energy calculation is performed on the degraded residual signal to obtain a residual energy index, which is used to quantify the strength of the influence of the inter-cavity boundary change on the overall resonant response. When the residual energy index is greater than the preset energy threshold, frequency domain analysis is performed on the healthy resonance response result and the real-time resonance response result to identify the resonant mode peak generated by the high-frequency electromagnetic resonant cavity. Extract the corresponding modal parameters for any resonant mode peak; Calculate the resonant frequency deviation, resonant damping, and energy reflection changes by combining modal parameters; The boundary variation of the inter-cavity resonant cavity is determined by combining the resonant frequency deviation, resonant damping, and energy reflection variation.

8. The method according to claim 7, characterized in that, The determination of the inter-cavity boundary variation of the high-frequency electromagnetic wave resonant cavity by combining the resonant frequency deviation, resonant damping, and energy reflection variation includes: When the change in resonant damping is an increase and the change in resonant frequency deviation is constant, it is determined to be resistive degradation. Based on resistive degradation, the change in the boundary of the inter-resonant cavity is determined to be an increase in contact resistance; When the change in resonant damping is a decrease and the change in resonant frequency deviation is constant, it is determined to be capacitive degradation. Based on the capacitive degradation, the boundary change of the inter-resonant cavity is determined to be an increase in the dielectric constant of the insulation.

9. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores instructions for causing the machine to perform a cable joint fault diagnosis method based on multi-source data fusion according to any one of claims 1 to 8.

10. An electronic device, characterized in that, include: The memory is configured to store instructions; as well as The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement a cable joint fault diagnosis method based on multi-source data fusion according to any one of claims 1 to 8.

Images