Early warning parameter determination method and device for cable insulation fault, and electronic equipment

Abstract

The invention discloses a method and device for determining early warning parameters of cable insulation faults and electronic equipment. The method comprises the following steps: acquiring a current signal sequence of a target cable; determining an amplitude distribution characteristic corresponding to the target cable according to the first signal sequence; determining an insulation state characteristic corresponding to the target cable according to the second signal sequence; determining an insulation fluctuation index corresponding to the target cable according to the amplitude distribution characteristics and the insulation state characteristics; and determining a target early warning parameter corresponding to the target cable according to the insulation fluctuation index. According to the method and the device, the technical problem that the insulation fault early warning parameters are not accurately determined when the insulation fault early warning parameters of the cable are determined in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of cable fault detection, and more specifically, to a method, apparatus, and electronic device for determining early warning parameters of cable insulation faults. Background Technology

[0002] In related technologies, cables, as critical power transmission equipment, directly affect the safety and reliability of the entire system due to the stability of their insulation performance. To ensure the reliability of effective assessment and early warning of cable insulation conditions, it is necessary to determine insulation fault early warning parameters for cables. However, in related technologies, there is a technical problem of inaccurate determination of these parameters.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method, apparatus, and electronic device for determining early warning parameters of cable insulation faults, in order to at least solve the technical problem in the related art where the determination of early warning parameters for cable insulation faults is inaccurate.

[0005] According to one aspect of the present invention, a method for determining early warning parameters for cable insulation faults is provided, comprising: acquiring a current signal sequence of a target cable, wherein the current signal sequence includes a first signal sequence and a second signal sequence, wherein the first signal sequence is a signal sequence in which the current signal frequency belongs to a first frequency range, the second signal sequence is a signal sequence in which the current signal frequency belongs to a second frequency range, and the minimum current signal frequency in the first frequency range is greater than the maximum current signal frequency in the second frequency range; determining an amplitude distribution characteristic corresponding to the target cable based on the first signal sequence; determining an insulation state characteristic corresponding to the target cable based on the second signal sequence; determining an insulation fluctuation index corresponding to the target cable based on the amplitude distribution characteristic and the insulation state characteristic; and determining a target early warning parameter corresponding to the target cable based on the insulation fluctuation index.

[0006] Optionally, determining the target warning parameter corresponding to the target cable based on the insulation fluctuation index includes: determining a first determination result of whether the insulation fluctuation index is greater than a fluctuation threshold; if the first determination result is that the insulation fluctuation index is greater than the fluctuation threshold, determining a predicted signal sequence corresponding to the current signal sequence, wherein the time period corresponding to the predicted signal sequence and the time period corresponding to the current signal sequence have a time correlation relationship; determining the insulation attenuation characteristic corresponding to the target cable based on the current signal sequence and the predicted signal sequence; and determining the target warning parameter corresponding to the target cable based on the insulation attenuation characteristic.

[0007] Optionally, determining the insulation attenuation characteristics corresponding to the target cable based on the current signal sequence and the predicted signal sequence includes: splicing the current signal sequence and the predicted signal sequence to obtain a target signal sequence, wherein the target signal sequence includes signal feature values ​​corresponding to multiple time nodes; determining multiple adjacent node pairs based on the multiple time nodes, wherein each of the multiple adjacent node pairs includes two adjacent time nodes; for any target node pair among the multiple adjacent node pairs, determining the sub-attenuation characteristics corresponding to the target node pair based on the signal feature values ​​corresponding to the two adjacent time nodes in the target node pair; determining the sub-attenuation characteristics corresponding to other adjacent node pairs among the multiple adjacent node pairs besides the target node pair by using the method of determining the sub-attenuation characteristics corresponding to the target node pair; and determining the insulation attenuation characteristics corresponding to the target cable based on the sub-attenuation characteristics corresponding to the multiple adjacent node pairs.

[0008] Optionally, determining the predicted signal sequence corresponding to the current signal sequence includes: retrieving a target model, wherein the target model includes a first feature extraction layer and a second feature extraction layer, the first feature extraction layer being used to extract time features of the input data according to a first time sequence, and the second feature extraction layer being used to extract time features of the input data according to a second time sequence, the first time sequence being the reverse of the second time sequence; obtaining a first time feature based on the current signal sequence and the first feature extraction layer of the target model; obtaining a second time feature based on the current signal sequence and the second feature extraction layer of the target model; and determining the predicted signal sequence corresponding to the current signal sequence based on the first time feature and the second time feature.

[0009] Optionally, determining the target warning parameter corresponding to the target cable based on the insulation attenuation characteristics includes: if the insulation attenuation characteristics include an insulation attenuation index, determining a second determination result of whether the insulation attenuation index is greater than an attenuation threshold; if the second determination result is that the insulation attenuation index is greater than the attenuation threshold, determining the current correlation relationship corresponding to the first signal sequence and the harmonic distortion characteristics corresponding to the second signal sequence, wherein the current correlation relationship is used to represent the correlation between the phase angle and the amplitude of the current signal corresponding to the first signal sequence, and the harmonic distortion characteristics represent the distortion characteristics of the signal waveform corresponding to the second signal sequence; and determining the target warning parameter corresponding to the target cable based on the current correlation relationship and the harmonic distortion characteristics.

[0010] Optionally, before acquiring the current signal sequence of the target cable, the method further includes: determining an insulation fault detection device corresponding to the target cable, wherein the insulation fault detection device includes a first detection device and a second detection device, the first detection device being disposed at a current signal acquisition point at the beginning of the target cable, and the second detection device being disposed at a current signal acquisition point at the end of the target cable, the current signal acquisition point including a grounding wire corresponding to the target cable; acquiring current signals at the current signal acquisition point at the beginning of the target cable using the first detection device to obtain a beginning signal sequence; acquiring current signals at the current signal acquisition point at the end of the target cable using the second detection device to obtain an end signal sequence; and determining the current signal sequence corresponding to the target cable based on the beginning signal sequence and the end signal sequence.

[0011] Optionally, after determining the target warning parameter corresponding to the target cable based on the insulation fluctuation index, the method further includes: if the target warning parameter includes a target warning level, determining a control command corresponding to the target warning level; and controlling the execution device to perform the target operation on the target cable based on the control command.

[0012] According to one aspect of the present invention, a device for determining early warning parameters for cable insulation faults is provided, comprising: an acquisition module for acquiring a current signal sequence of a target cable, wherein the current signal sequence includes a first signal sequence and a second signal sequence, wherein the first signal sequence is a signal sequence in which the current signal frequency belongs to a first frequency range, the second signal sequence is a signal sequence in which the current signal frequency belongs to a second frequency range, and the minimum current signal frequency in the first frequency range is greater than the maximum current signal frequency in the second frequency range; a first determination module for determining an amplitude distribution feature corresponding to the target cable based on the first signal sequence; a second determination module for determining an insulation state feature corresponding to the target cable based on the second signal sequence; a third determination module for determining an insulation fluctuation index corresponding to the target cable based on the amplitude distribution feature and the insulation state feature; and a fourth determination module for determining a target early warning parameter corresponding to the target cable based on the insulation fluctuation index.

[0013] According to one aspect of the present invention, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the cable insulation fault early warning parameter determination method described in any of the preceding claims.

[0014] According to one aspect of the present invention, a computer-readable storage medium is provided, comprising: when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enabling the electronic device to perform the cable insulation fault early warning parameter determination method described in any of the preceding claims.

[0015] In this embodiment of the invention, a current signal sequence of the target cable is obtained, wherein the current signal sequence includes a first signal sequence and a second signal sequence, wherein the first signal sequence is a signal sequence in which the current signal frequency belongs to a first frequency range, and the second signal sequence is a signal sequence in which the current signal frequency belongs to a second frequency range, wherein the minimum current signal frequency in the first frequency range is greater than the maximum current signal frequency in the second frequency range; based on the first signal sequence, the amplitude distribution characteristics corresponding to the target cable are determined; based on the second signal sequence, the insulation state characteristics corresponding to the target cable are determined; based on the amplitude distribution characteristics and the insulation state characteristics, the insulation fluctuation index corresponding to the target cable is determined; based on the insulation fluctuation index, the target warning parameter corresponding to the target cable is determined. By separately collecting two types of current signal sequences of the target cable—high frequency (first frequency range, higher than the second frequency range) and low frequency (second frequency range)—and extracting amplitude distribution characteristics from the high frequency signal sequence, transient fault information in the cable can be captured. Insulation status characteristics can be extracted from the low frequency signal sequence, and the long-term aging of the cable can be assessed. By combining amplitude distribution characteristics and insulation status characteristics, the fluctuation of cable insulation can be comprehensively assessed. Based on the obtained insulation fluctuation index, the target early warning parameters of the target cable can be accurately determined, thus solving the technical problem in related technologies where the determination of insulation fault early warning parameters for cables is inaccurate. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a flowchart of a method for determining early warning parameters for cable insulation faults according to an embodiment of the present invention; Figure 2 This is a flowchart of a method for determining early warning parameters for cable insulation faults in an optional embodiment of the present invention; Figure 3 This is a schematic diagram of the BiLSTM gated unit structure in an optional embodiment of the present invention; Figure 4 This is a schematic diagram of the basic structure of the target model in an optional embodiment of the present invention; Figure 5 This is a structural block diagram of a cable insulation fault early warning parameter determination device according to an embodiment of the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] First, some nouns or terms that appear in the description of the embodiments of this application shall be interpreted as follows: LSTM: LSTM stands for Long Short-Term Memory Network. It controls the flow of information by introducing "gates" (such as input gate, forget gate, and output gate), thereby enabling it to effectively learn and remember long-term dependencies.

[0020] BiLSTM: BiLSTM is a bidirectional long short-term memory network, an extension of LSTM. It contains two LSTM layers: one forward propagation (from input to output) and one backward propagation (from output to input).

[0021] Mann-Kendall test: The Mann-Kendall test is a nonparametric statistical test that does not require assumptions about the distribution of the data.

[0022] Vmax-φ spectrum: The Vmax-φ spectrum is a spectrum used to analyze partial discharge (PD) signals. It is a graph that reflects the relationship between the maximum discharge amplitude (Vmax) and the discharge phase angle (φ).

[0023] Example 1 According to an embodiment of the present invention, an embodiment of a method for determining early warning parameters of cable insulation faults is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0024] Figure 1 This is a flowchart of a method for determining early warning parameters for cable insulation faults according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps: S102, acquire the current signal sequence of the target cable, wherein the current signal sequence includes a first signal sequence and a second signal sequence, wherein the first signal sequence is a signal sequence whose current signal frequency belongs to a first frequency range, and the second signal sequence is a signal sequence whose current signal frequency belongs to a second frequency range, and the minimum current signal frequency in the first frequency range is greater than the maximum current signal frequency in the second frequency range.

[0025] In step S102 of this application, the current signal sequence of the target cable is obtained.

[0026] This involves a target cable, which is used for insulation fault early warning detection and is used to transmit electrical energy in a power system.

[0027] This involves a current signal sequence, which is current signal data collected from the target cable, including a first signal sequence and a second signal sequence.

[0028] This involves a first signal sequence, which is a current signal frequency within a first frequency range. The first frequency range typically refers to a high-frequency range and is used to capture transient fault information in cables.

[0029] This involves a second signal sequence, which is a current signal frequency within a second frequency range. The second frequency range typically refers to a low-frequency range and is used to assess the long-term aging of cables.

[0030] This involves a first frequency range, which is the frequency interval used to collect high-frequency fault signals.

[0031] This involves a second frequency range, which is the frequency interval used to collect low-frequency steady-state signals.

[0032] By acquiring a current signal sequence that includes a first signal sequence and a second signal sequence, it is helpful to comprehensively evaluate the early warning parameters of cable insulation faults from current signals in different frequency ranges.

[0033] S104, Based on the first signal sequence, determine the amplitude distribution characteristics corresponding to the target cable.

[0034] In step S104 provided in this application, the amplitude distribution characteristics corresponding to the target cable are determined based on the first signal sequence.

[0035] This involves amplitude distribution characteristics, which are features used to reflect the distribution of current signal intensity of the target cable in the first frequency range (high frequency range), such as the maximum amplitude distribution.

[0036] By determining the amplitude distribution characteristics corresponding to the target cable, we can gain a more comprehensive understanding of the cable's operating status in the high-frequency range, thereby providing data for subsequent insulation fault early warning.

[0037] S106, Based on the second signal sequence, determine the insulation state characteristics corresponding to the target cable.

[0038] In step S106 provided in this application, the insulation state characteristics corresponding to the target cable are determined based on the second signal sequence.

[0039] This includes insulation condition characteristics, which are used to reflect the long-term changes in the cable insulation performance reflected by the current signal in the second frequency range (low frequency range), such as the effective value of the grounding current.

[0040] By determining the amplitude distribution characteristics corresponding to the target cable, we can gain a more comprehensive understanding of the cable's operating status in the low-frequency range, thereby providing data for subsequent insulation fault early warning.

[0041] S108. Based on the amplitude distribution characteristics and insulation condition characteristics, determine the insulation fluctuation index corresponding to the target cable.

[0042] In step S108 provided in this application, the insulation fluctuation index corresponding to the target cable is determined based on the amplitude distribution characteristics and insulation state characteristics.

[0043] This involves an insulation fluctuation index, which reflects the insulation fluctuation state of the target cable. The insulation fluctuation index includes the mean of the maximum amplitude distribution of the fundamental characteristic quantity of the high-frequency current sensor (i.e., the amplitude distribution characteristics of the first signal sequence) and the mean of the effective value of the grounding current of the fundamental characteristic quantity of the low-frequency current sensor (i.e., the insulation state characteristics of the second signal sequence).

[0044] Based on the amplitude distribution characteristics and insulation condition characteristics, the insulation fluctuation index corresponding to the target cable is determined. This avoids relying solely on high-frequency characteristics and ignoring the overall aging trend of the insulation, as well as relying solely on low-frequency characteristics and missing early transient defects. This provides a comprehensive and unified quantitative basis for the subsequent accurate determination of target early warning parameters.

[0045] S110, based on the insulation fluctuation index, determines the target warning parameters corresponding to the target cable.

[0046] In step S110 of this application, the target warning parameters corresponding to the target cable are determined based on the insulation fluctuation index.

[0047] This includes target early warning parameters, which are used to assess the cable insulation condition and trigger corresponding early warning measures. These parameters reflect the health status of the cable insulation and determine whether and what maintenance or repair measures are needed. These target early warning parameters include early warning level and repair method.

[0048] The insulation fluctuation index can comprehensively quantify the dynamic changes of cable insulation under both transient defects and steady-state aging. Therefore, based on the insulation fluctuation index, the target warning parameters corresponding to the target cable can be accurately determined.

[0049] Through the above steps S102-S110, the current signal sequence of the target cable is obtained, wherein the current signal sequence includes a first signal sequence and a second signal sequence, wherein the first signal sequence is a signal sequence whose current signal frequency belongs to a first frequency range, and the second signal sequence is a signal sequence whose current signal frequency belongs to a second frequency range, and the minimum current signal frequency in the first frequency range is greater than the maximum current signal frequency in the second frequency range; based on the first signal sequence, the amplitude distribution characteristics corresponding to the target cable are determined; based on the second signal sequence, the insulation state characteristics corresponding to the target cable are determined; based on the amplitude distribution characteristics and the insulation state characteristics, the insulation fluctuation index corresponding to the target cable is determined; based on the insulation fluctuation index, the target warning parameter corresponding to the target cable is determined. By separately collecting two types of current signal sequences of the target cable—high frequency (first frequency range, higher than the second frequency range) and low frequency (second frequency range)—and extracting amplitude distribution characteristics from the high frequency signal sequence, transient fault information in the cable can be captured. Insulation status characteristics can be extracted from the low frequency signal sequence, and the long-term aging of the cable can be assessed. By combining amplitude distribution characteristics and insulation status characteristics, the fluctuation of cable insulation can be comprehensively assessed. Based on the obtained insulation fluctuation index, the target early warning parameters of the target cable can be accurately determined, thus solving the technical problem in related technologies where the determination of insulation fault early warning parameters for cables is inaccurate.

[0050] As an optional embodiment, determining the target warning parameters corresponding to the target cable based on the insulation fluctuation index includes: determining whether the insulation fluctuation index is greater than a fluctuation threshold (a first determination result); if the first determination result is that the insulation fluctuation index is greater than the fluctuation threshold, determining a predicted signal sequence corresponding to the current signal sequence, wherein the time period corresponding to the predicted signal sequence and the time period corresponding to the current signal sequence have a time correlation relationship; determining the insulation attenuation characteristics corresponding to the target cable based on the current signal sequence and the predicted signal sequence; and determining the target warning parameters corresponding to the target cable based on the insulation attenuation characteristics.

[0051] This embodiment describes the specific steps for determining the target warning parameters corresponding to the target cable based on the insulation fluctuation index.

[0052] This involves a fluctuation threshold, a preset value used to determine whether the insulation fluctuation index is within the normal range. This fluctuation threshold can be a first warning value, defined as: the average maximum amplitude distribution of the high-frequency current sensor being greater than the high-frequency warning threshold, and the average effective value of the grounding current of the low-frequency current sensor being greater than the grounding current threshold.

[0053] This involves a first determining result, which is the result of judging whether the insulation fluctuation index is greater than the fluctuation threshold. This is used to determine whether further analysis is needed, such as determining the predicted signal sequence and insulation attenuation characteristics.

[0054] This involves a predicted signal sequence, which predicts signal characteristics over a future period based on the current signal sequence. This predicted signal sequence can be determined using a target model, such as a BiLSTM model.

[0055] This involves a temporal correlation, which is the relationship between the time period corresponding to the predicted signal sequence and the time period corresponding to the current current signal sequence. For example, this temporal correlation can be used to ensure that the predicted signal sequence and the current signal sequence are continuous and correlated in time. That is, the time period corresponding to the predicted signal sequence is a future time period of the time period corresponding to the current signal sequence.

[0056] When the first determination result is greater than the fluctuation threshold, a predicted signal sequence with current and future time correlation with the current signal sequence is determined. Then, the insulation attenuation characteristics are determined based on the current and predicted signal sequences. This can quantify the degree of insulation performance degradation from both historical state and future trend dimensions, avoiding the inaccuracy of early warning parameters caused by single time dimension analysis.

[0057] As an optional embodiment, determining the insulation attenuation characteristics corresponding to the target cable based on the current signal sequence and the predicted signal sequence includes: splicing the current signal sequence and the predicted signal sequence to obtain a target signal sequence, wherein the target signal sequence includes signal feature values ​​corresponding to multiple time nodes; determining multiple adjacent node pairs based on the multiple time nodes, wherein each of the multiple adjacent node pairs includes two adjacent time nodes; for any target node pair among the multiple adjacent node pairs, determining the sub-attenuation characteristics corresponding to the target node pair based on the signal feature values ​​corresponding to the two adjacent time nodes in the target node pair; determining the sub-attenuation characteristics corresponding to other adjacent node pairs besides the target node pair by using the method of determining the sub-attenuation characteristics corresponding to the target node pair; and determining the insulation attenuation characteristics corresponding to the target cable based on the sub-attenuation characteristics corresponding to the multiple adjacent node pairs.

[0058] In this embodiment, specific steps are described to determine the insulation attenuation characteristics corresponding to the target cable based on the current signal sequence and the predicted signal sequence.

[0059] This involves splicing, which is the operation of integrating a current signal sequence (reflecting the current and historical states) and a prediction signal sequence (reflecting the future states) into a continuous target signal sequence in chronological order.

[0060] Current signal sequence , can be represented as:

[0061] in: The first high-frequency current data (first signal sequence) The mean of the distribution of the maximum amplitude corresponding to each time window; The first signal in the ground current data (second signal sequence) The mean of the valid values ​​corresponding to each time window.

[0062] Predicted signal sequence , can be represented as:

[0063] in: The BiLSTM model predicts the first high-frequency current data (first signal sequence). The mean of the distribution of the maximum amplitude corresponding to each time window; The BiLSTM model predicts the first ground current data (second signal sequence). The mean of the valid values ​​corresponding to each time window.

[0064] The target signal sequence obtained after splicing is then... , means as follows:

[0065] This involves signal feature values, which are used to characterize the cable insulation state at each time point in the target signal sequence.

[0066] This involves multiple time points, which are discrete time points arranged in chronological order within the target signal sequence.

[0067] This involves sub-attenuation features, which are local features that reflect the degree of insulation attenuation during a time period. These features are obtained by calculating the difference and rate of change of the corresponding signal feature values ​​for two adjacent time nodes.

[0068] Based on the sub-attenuation characteristics corresponding to multiple adjacent node pairs, the insulation attenuation characteristics corresponding to the target cable can be determined using the Mann-Kendall test method.

[0069] By calculating the characteristic changes of adjacent time nodes, the system captures the dynamic degradation process of insulation status, realizing a refined and continuous analysis of insulation attenuation trends. This avoids the limitations of single-time-point data, improves the accuracy of insulation attenuation characteristic assessment, and provides a more reliable basis for determining target early warning parameters, thus helping to solve the problem of inaccurate early warning parameter determination.

[0070] As an optional embodiment, determining the predicted signal sequence corresponding to the current signal sequence includes: retrieving a target model, wherein the target model includes a first feature extraction layer and a second feature extraction layer, the first feature extraction layer being used to extract the time features of the input data according to a first time sequence, and the second feature extraction layer being used to extract the time features of the input data according to a second time sequence, the first time sequence being the reverse of the second time sequence; obtaining a first time feature based on the current signal sequence and the first feature extraction layer of the target model; obtaining a second time feature based on the current signal sequence and the second feature extraction layer of the target model; and determining the predicted signal sequence corresponding to the current signal sequence based on the first time feature and the second time feature.

[0071] In this embodiment, the specific steps for determining the predicted signal sequence corresponding to the current signal sequence are described.

[0072] This involves a target model, which is used to predict future signal sequences based on current signal sequences. This target model is capable of simultaneously extracting temporal features from input data in both forward and reverse time directions, and includes a first feature extraction layer and a second feature extraction layer. This target model can be a BiLSTM model.

[0073] This involves a first feature extraction layer, which is a network layer that processes the input current signal sequence according to a first time sequence (such as a forward time sequence from the past to the present) to extract the signal variation pattern in the forward time dimension.

[0074] This involves a second feature extraction layer, which is a network layer that processes the input current signal sequence according to a second time order (such as a reverse time order from now to the past) to extract the network layer that correlates the signal features in the reverse time dimension.

[0075] This involves the first time sequence, which is the forward time flow from historical time nodes to the current time node.

[0076] This involves a second time sequence, which is the reverse of the first time sequence, and is a reverse time flow from the current time point back to the historical time point.

[0077] This involves the first time feature, which is a feature extracted by the first feature extraction layer from the current signal sequence in the first time order, reflecting the positive time change law of the signal (such as the increasing / decreasing trend of the signal from the past to the present, and the stage fluctuation pattern).

[0078] This involves a second time feature, which is a feature extracted by the second feature extraction layer from the current signal sequence according to the second time order, reflecting the reverse time correlation law of the signal (such as the dependence of the current signal on the historical signal, and the fluctuation correlation of reverse tracing).

[0079] As an optional embodiment, the target warning parameters corresponding to the target cable are determined based on the insulation attenuation characteristics, including: if the insulation attenuation characteristics include an insulation attenuation index, determining whether the insulation attenuation index is greater than an attenuation threshold; if the second determination result is that the insulation attenuation index is greater than the attenuation threshold, determining the current correlation relationship corresponding to the first signal sequence and the harmonic distortion characteristics corresponding to the second signal sequence, wherein the current correlation relationship is used to represent the correlation between the phase angle and the amplitude of the current signal corresponding to the first signal sequence, and the harmonic distortion characteristics represent the distortion characteristics of the signal waveform corresponding to the second signal sequence; and the target warning parameters corresponding to the target cable are determined based on the current correlation relationship and the harmonic distortion characteristics.

[0080] This embodiment describes the specific steps for determining the target warning parameters corresponding to the target cable based on insulation attenuation characteristics.

[0081] This involves an attenuation threshold, which is a preset critical value used to determine whether the degree of cable insulation attenuation exceeds the normal range. If the insulation attenuation index exceeds the threshold, it indicates that the insulation performance has deteriorated and further precise analysis is needed, such as further determination of derived characteristic quantities (i.e., current correlation and harmonic distortion characteristics).

[0082] This involves a second determination result, which is the result of judging whether the insulation attenuation index is greater than the attenuation threshold. This result is used to determine whether more detailed features (current correlation, harmonic distortion features) need to be introduced to determine the warning parameters.

[0083] This involves current correlation, which is the correspondence between the phase angle and amplitude of the current signal in the first signal sequence (high-frequency signal). It can be presented in the form of Vmax-φ spectrum, which can reflect the phase distribution law of partial discharge (such as the difference in discharge characteristics between positive and negative half cycles), and thus determine the type of insulation defect (such as air gap discharge, surface discharge).

[0084] This involves harmonic distortion characteristics, which are the degree to which the signal waveform in the second signal sequence (low-frequency signal) deviates from the standard sine wave. It is commonly quantified by indicators such as total harmonic distortion (THD), which can reflect the change in the harmonic content of leakage current caused by insulation aging (e.g., the more severe the aging, the higher the proportion of low-order harmonics).

[0085] This involves the phase angle of the current signal, which is the phase difference between the current signal and a reference signal (such as a voltage signal) at a certain moment during the periodic change of the current signal.

[0086] This involves the amplitude of the current signal, which reflects the strength of the current signal and can be used to help determine the intensity of partial discharge and the magnitude of leakage current.

[0087] By determining whether the insulation attenuation index exceeds the attenuation threshold (the second determining result), cables whose insulation performance degradation may exceed the normal range are screened out. If the threshold is exceeded, the current correlation (the relationship between the phase angle and amplitude of high-frequency signals) and harmonic distortion characteristics (the degree of waveform distortion of low-frequency signals) are further analyzed to more accurately determine the target warning parameters.

[0088] As an optional embodiment, before acquiring the current signal sequence of the target cable, the method further includes: determining an insulation fault detection device corresponding to the target cable, wherein the insulation fault detection device includes a first detection device and a second detection device, the first detection device being set at a current signal acquisition point at the beginning of the target cable, and the second detection device being set at a current signal acquisition point at the end of the target cable, the current signal acquisition point including a grounding wire corresponding to the target cable; acquiring current signals at the current signal acquisition point at the beginning of the target cable using the first detection device to obtain a beginning signal sequence; acquiring current signals at the current signal acquisition point at the end of the target cable using the second detection device to obtain an end signal sequence; and determining the current signal sequence corresponding to the target cable based on the beginning signal sequence and the end signal sequence.

[0089] In this embodiment, specific steps prior to acquiring the current signal sequence of the target cable are described.

[0090] This includes insulation fault detection equipment, which is used to collect the current signal of the target cable.

[0091] This involves a first detection device, which is a current signal acquisition point set at the beginning of the target cable, used to collect the current signal (including high-frequency and low-frequency signals) at the beginning of the cable.

[0092] This involves a second detection device, which is a current signal acquisition point set at the end of the target cable to collect the current signal (including high-frequency and low-frequency signals) at the end of the cable.

[0093] This involves a current signal acquisition point, which is a specific location for deploying detection equipment and collecting cable current signals. This current signal acquisition point can be set on the grounding wire corresponding to the target cable.

[0094] This involves the first-end signal sequence, which is a set of current data continuously acquired in chronological order by the first detection device at the grounding wire acquisition point at the first end of the cable.

[0095] This involves the end signal sequence, which is a set of current data continuously acquired in chronological order by the second detection device at the grounding wire acquisition point at the end of the cable.

[0096] The insulation fault detection equipment corresponding to the target cable is determined (including the first and second detection equipment set at the beginning and end collection points respectively, with the collection points selected on the grounding wire). The signal sequences at the beginning and end are collected by the two types of equipment respectively, and then integrated into the current signal sequence of the target cable. This avoids the signal partiality caused by a single collection point and ensures the integrity and reliability of the collected signals.

[0097] As an optional embodiment, after determining the target warning parameters corresponding to the target cable based on the insulation fluctuation index, the method further includes: if the target warning parameters include a target warning level, determining a control command corresponding to the target warning level; and controlling the execution device to perform the target operation on the target cable according to the control command.

[0098] This embodiment describes the specific steps after determining the target warning parameters corresponding to the target cable based on the insulation fluctuation index.

[0099] This includes target warning levels, which are used to clarify the degree of risk of cable insulation faults. These levels include low risk (e.g., "normal operation"), medium risk (e.g., "enhanced monitoring"), and high risk (e.g., "emergency maintenance"), which can intuitively reflect the urgency of the cable insulation status and are the core basis for generating subsequent control instructions.

[0100] This involves control commands, which are specific instruction signals generated based on the target warning level to guide the actions of the executing equipment.

[0101] This involves execution devices, which are devices that receive and execute control commands, including alarm devices, switching devices, etc.

[0102] After determining the target early warning parameters, corresponding control instructions are determined for the target early warning level, and the execution equipment is controlled to perform the target operation to ensure that the risk of cable insulation failure can be responded to in a timely manner (such as quickly cutting off the fault circuit in case of high risk). This avoids the early warning information remaining only at the analysis level without being effectively dealt with, thereby improving the dynamic control capability of cable insulation failure and ensuring the safe operation of the power system.

[0103] Based on the above embodiments and optional embodiments, an optional implementation method is provided, which is described in detail below.

[0104] In related technologies, cables, as critical power transmission equipment, directly affect the safety and reliability of the entire system due to the stability of their insulation performance. To ensure the reliability of effective assessment and early warning of cable insulation conditions, it is necessary to determine insulation fault early warning parameters for cables. However, in related technologies, there is a technical problem of inaccurate determination of these parameters.

[0105] There is currently no effective solution to the above problems.

[0106] In view of this, an optional embodiment of the present invention provides a method for determining early warning parameters for cable insulation faults, which can effectively solve the above-mentioned technical problems.

[0107] Figure 2 This is a flowchart of a method for determining early warning parameters for cable insulation faults in an optional embodiment of the present invention, such as... Figure 2 As shown, a detailed description follows.

[0108] S1. Obtain the signal measurement values ​​of the sensors in the cable insulation defect fault early warning group; The sensors in the cable insulation defect fault early warning group (i.e., insulation fault detection equipment) include high-frequency current and low-frequency grounding current sensors.

[0109] Insulation fault detection equipment is installed on the grounding wires at both ends of the target cable, which can be the grounding wire at the beginning and the grounding wire at the end of the target cable. Based on the high-frequency signal data (i.e., the first signal sequence) and low-frequency signal data (i.e., the second signal sequence) collected by sensors (i.e., the first detection device and the second detection device) on the grounding wires at both ends of the cable, the average value of the collected high-frequency current signal and low-frequency grounding current signal is calculated.

[0110] The formula for calculating the mean of the maximum amplitude distribution of high-frequency current per unit time is as follows:

[0111] in: It is the mean of the distribution with the largest amplitude. Number of data collection cycles; For the first Maximum discharge amplitude in each acquisition cycle.

[0112] The formula for calculating the average effective value of grounding current per unit time is as follows:

[0113] in: This is the average value of the effective value of the grounding current; For the first Valid values ​​in each collection cycle.

[0114] S2. When the signal measurement value of the sensor in the fault warning group meets the first warning value, execute S3; If the mean value of the maximum amplitude distribution of the high-frequency current per unit time is greater than 1V, and the mean value of the effective value of the grounding current per unit time is greater than 15mA, then proceed with S3.

[0115] S21. Calculate the signal measurement values ​​(i.e., insulation fluctuation index) of the sensors in the fault warning group.

[0116] The first signal sequence is acquired by a high-frequency current sensor, and the second signal sequence is acquired by a low-frequency current sensor.

[0117] Further, based on the first signal sequence, the amplitude distribution characteristics (i.e., the maximum amplitude distribution) corresponding to the target cable are determined. Based on the second signal sequence, the insulation state characteristics (i.e., the effective value of the grounding current) corresponding to the target cable are determined. Based on the amplitude distribution characteristics and the insulation state characteristics, the insulation fluctuation index corresponding to the target cable is determined, wherein the insulation fluctuation index includes the mean of the maximum amplitude distribution as the fundamental characteristic quantity of the high-frequency current sensor, and the mean of the effective value of the grounding current as the fundamental characteristic quantity of the low-frequency current sensor.

[0118] S22. Determining the target early warning parameters corresponding to the target cable based on the insulation fluctuation index also includes: judging whether the signal measurement value of the sensor in the fault early warning group meets the first early warning value (i.e., the fluctuation threshold).

[0119] The first warning value is: the mean value of the maximum amplitude distribution of the high-frequency current sensor is greater than the high-frequency warning threshold, and the mean value of the effective value of the ground current of the low-frequency current sensor is greater than the ground current threshold. When the first warning value is met, S3 is executed.

[0120] Otherwise, based on the insulation fluctuation index, determine the target warning parameters corresponding to the target cable.

[0121] S3. Obtain the signal sequence (i.e., current signal sequence) of the sensor in the fault warning group. Use the sliding window method to process the signal sequences of the high-frequency current sensor and the low-frequency sensor respectively. Use the BiLSTM algorithm to make short-term predictions on the basic characteristic quantities of the high-frequency current sensor and the basic characteristic quantities of the low-frequency sensor to obtain the predicted signal sequence. Combine the current signal sequence and the predicted signal sequence to form a signal time series set. S31. The signal sequences of the high-frequency current sensor and the low-frequency sensor are processed separately using the sliding window method. The amount of data in the window is set to N, and the sliding step size is [missing value]. The fundamental characteristic quantities of the high-frequency current and low-frequency grounding current within a unit time t in each window are calculated and formed into a vector set. ; The collected signal sequences from the high-frequency current sensor and the low-frequency sensor are normalized using the following formula:

[0122] in: The normalized sample amplitude takes values ​​of [0, 1]. It is the minimum value in the sample; It represents the maximum value in the sample.

[0123] Determining the predicted signal sequence corresponding to the current signal sequence includes: The mean of the maximum amplitude distribution of the collected high-frequency current data (i.e., the first signal sequence) and the mean of the effective values ​​of the ground current (i.e., the second signal sequence) are combined to form a data matrix Z as shown below.

[0124]

[0125] in: The first high-frequency current data (first signal sequence) The mean of the distribution of the maximum amplitude corresponding to each time window; The first signal in the ground current data (second signal sequence) The mean of the valid values ​​corresponding to each time window.

[0126] S32. Use the BiLSTM algorithm to make short-term predictions on the basic feature quantities in each window after processing, and obtain the predicted signal sequence, that is, obtain the time series set composed of historical data and predicted values.

[0127] Specifically, this includes: constructing a BiLSTM model and using the BiLSTM algorithm to make short-term predictions of the fundamental characteristic quantities of high-frequency current sensors and low-frequency sensors, forming a signal time series set.

[0128] Figure 3 This is a schematic diagram of the BiLSTM gating unit structure in an optional embodiment of the present invention, as shown below. Figure 3 As shown, This represents the input gate, which controls how much of the current input can be retained in the cell state. If it is 1, the input is allowed; if it is 0, it is not allowed, thus discarding some useless input information. The forget gate controls the degree to which the unit information from the previous moment is forgotten. It determines how much historical information can be saved to the current state. If it is 1, it is saved; if it is 0, the historical data stored in the current node is cleared. This represents the output gate, which determines the output of information. If it is 1, the output value of the current node will be applied to the next layer; if it is 0, the output value of the current node will not be output.

[0129] The gating structure of an LSTM neural network is controlled by a sigmoid activation function, which uses... This indicates that its function is to map variables to the interval [0, 1], thus defining the degree to which information passes through. The function is shown below:

[0130] in: The output value of the sigmoid activation function; This is the input variable for the sigmoid activation function.

[0131] The calculations for the forget gate, input gate, and output gate in the hidden layer are as follows:

[0132]

[0133]

[0134] in: The row vector input of matrix Z at time t; This represents the hidden layer state at the previous moment; , and These represent the calculation results of the forget gate, input gate, and output gate at time t, respectively. , and These represent the weight matrices for the forget gate, input gate, and output gate, respectively. , and These represent the bias terms for the forget gate, input gate, and output gate, respectively.

[0135] The final output of a neuron is determined by the output gate state value. and unit state Let's decide together, like this:

[0136]

[0137]

[0138] in: , These represent the weight matrix and bias term of the input unit state, respectively; , and These represent the candidate state, current state, and hidden layer state of the cell at time t, respectively. tanh represents the hyperbolic tangent activation function, which maps variables to [–1, 1]. Operators This indicates element-wise multiplication.

[0139] The tanh activation function is shown below:

[0140] in: This is the output value of the tanh activation function; This is the input value for the tanh activation function.

[0141] Determining the predicted signal sequence corresponding to the current signal sequence includes: retrieving the target model, wherein the target model includes a first feature extraction layer and a second feature extraction layer, the first feature extraction layer being used to extract the time features of the input data according to a first time sequence, and the second feature extraction layer being used to extract the time features of the input data according to a second time sequence, the first time sequence being the reverse of the second time sequence; obtaining the first time feature based on the current signal sequence and the first feature extraction layer of the target model; obtaining the second time feature based on the current signal sequence and the second feature extraction layer of the target model; and determining the predicted signal sequence corresponding to the current signal sequence based on the first time feature and the second time feature.

[0142] The target model can be a BiLSTM network (i.e., a BiLSTM model), containing two independent LSTM layers (corresponding to the first feature extraction layer and the second feature extraction layer, respectively). The first feature extraction layer performs forward propagation (from input to output), and the second feature extraction layer performs backward propagation (from output to input). Bidirectional computation is added to the LSTM, and the hidden layer output value of the BiLSTM at time t is... From the positive hidden layer state and reverse hidden layer state composition.

[0143] Figure 4 This is a schematic diagram of the basic structure of the target model in an optional embodiment of the present invention, such as... Figure 4 As shown, the basic training steps are as follows: The feature data (current signal sequence) is input in forward order into the feedforward LSTM (i.e., the first feature extraction layer), and simultaneously input in reverse order into the feedback LSTM (i.e., the second feature extraction layer). The output value of each neuron is then calculated in the forward direction. , , , and Five parameters; The error value of each neuron is calculated in reverse, including backpropagation over time and propagation of the error to the next layer of the network. Based on the corresponding error terms, the gradient of each weight is calculated, and the weights are updated using the gradient descent method.

[0144] Obtain the hidden layer state As shown below:

[0145] S33. Using a sliding window, the signal is divided according to time. Each row vector in the data matrix Z is divided into training, validation, and test sample data, with the proportions of the three parts being 75%, 15%, and 10%, respectively. These are used for training, validation, and testing of the BiLSTM model to obtain the corresponding prediction matrix. As shown below:

[0146] in: The BiLSTM model predicts the first high-frequency current data (first signal sequence). The mean of the distribution of the maximum amplitude corresponding to each time window; The BiLSTM model predicts the first ground current data (second signal sequence). The mean of the valid values ​​corresponding to each time window.

[0147] S4. Use the Mann-Kendall test to examine the trend of the signal time series set. When the trend characteristics meet the warning value, calculate the derived characteristic quantities of the sensor signal measurements within the fault warning group.

[0148] Trend analysis, also known as insulation attenuation testing, determines the insulation attenuation characteristics of the target cable, including the rate of change of these characteristics. and standardized statistics In this case, the following steps are used to determine: S41. Apply the Mann-Kendall (MK) test to the time series dataset for trend analysis and calculate the trend characteristics. The current signal sequence and the predicted signal sequence are concatenated to obtain the target signal sequence, that is, the Z and Y data are combined to form the K sequence (i.e., d), and the matrix K is shown below:

[0149] For a sample size of Calculate the median slope of a time series K. The calculation formula is as follows:

[0150] in: Indicates the rate of change; Median means taking the median; , These are the partial discharge characteristic values ​​of time series i and j (i.e., adjacent node pairs, such as the signal characteristic values ​​corresponding to two adjacent time nodes in the target node pair).

[0151] When the slope A value greater than 0 indicates an upward trend, while a value less than 0 indicates a downward trend.

[0152] S42. When the trend characteristic quantity meets the warning value, calculate the derived characteristic quantity of each sensor signal measurement value in the fault warning group. The derived characteristic quantity for high frequency is the correlation coefficient cc of the Vmax-φ spectrum, and the derived characteristic quantity for low frequency sensors is the total harmonic distortion (THD).

[0153] Calculate the MK statistic ,variance As shown below:

[0154] in: It is the core statistic of the MK test, used to preliminarily determine the trend direction of a signal time series.

[0155]

[0156] in: It represents the number of nodes in the time series, indicating the number of repetitions; It is the width of the p-th node, and represents the number of repetitions.

[0157] The sgn function is shown below:

[0158] in: This is the output value of the sgn function; This is the input value for the sgn function.

[0159] Calculate the standardized statistic :

[0160] Give the significance level , A value of 1.96 indicates that the 95% significance test is passed.

[0161] By calculating the row vectors in the K matrix respectively and and the corresponding standardized statistics and Table 1 is a trend test category table. As shown in Table 1, it is used to determine the trend characteristics of high-frequency current and low-frequency grounding current signals. When the trend characteristics exist... >0 and The trend characteristics then meet the warning threshold.

[0162] Table 1

[0163] S43. Calculate the derived characteristic quantities of the signal measurement values ​​of each sensor in the fault early warning group.

[0164] The derived characteristic of a high-frequency current sensor is the correlation coefficient of the Vmax-φ spectrum. (That is, the current correlation), the calculation formula is shown below.

[0165]

[0166] in: and These represent the magnitudes of the maximum amplitude detected by the i-th window during the positive and negative half-cycles, respectively. It refers to the number of windows that are opened in half a circumference; This indicates the difference in shape between the positive and negative half-circles. If the positive and negative half-circles are exactly the same, then... ,on the contrary .

[0167] The derived characteristic quantity of a low-frequency grounding current sensor is the total harmonic distortion (THD) (i.e., harmonic distortion characteristic, which can also be expressed as...). (represented), the calculation formula is shown below.

[0168]

[0169] , , These are the effective values ​​of the 2nd, 3rd, and 4th harmonics of the collected low-frequency grounding current signal, respectively. This is the effective value of the fundamental frequency.

[0170] S5. When the derived characteristic quantities of the sensors in the fault warning group all meet the second warning value, the cable is confirmed to be a cable at risk of breakdown. The cable at risk of breakdown is then disconnected and repaired. Otherwise, suggestions such as continuing normal operation, strengthening monitoring and attention, and arranging power outage repairs as soon as possible are given according to the predicted time nodes.

[0171] The second warning value is the correlation coefficient cc (i.e., current correlation) of the Vmax-φ spectrum, which falls within the range of severe cable insulation conditions, and the harmonic distortion rate of the low-frequency current sensor. (That is, the harmonic distortion characteristics) are greater than the threshold value under severe cable insulation conditions. When setting the threshold range near the breakdown stage, the correlation coefficient threshold range of the Vmax-φ spectrum of the high-frequency current signal is 0.2-0.5, and the harmonic distortion rate of the low-frequency current sensor is... The threshold value is 0.2 when the cable insulation condition is severely compromised.

[0172] The above optional implementation methods can achieve at least the following beneficial effects: (1) Compared with related technologies, the present invention enhances the comprehensiveness of diagnosis and improves the accuracy of fault identification by providing a multi-dimensional feature extraction and hierarchical early warning mechanism. First-level early warning: Based on the real-time measurement values ​​of high-frequency amplitude, phase width and low-frequency grounding current, the breakdown fault that has occurred can be quickly identified and the faulty cable can be cut off immediately to avoid the accident from escalating. Second-level early warning: By using BiLSTM prediction and Mann-Kendall trend test, potential breakdown risks can be analyzed and early warning can be given before the fault occurs to reduce sudden outages.

[0173] (2) Compared with related technologies, the present invention optimizes data segmentation processing by using the sliding window method, divides continuous signals into dynamic windows, avoids delays caused by global analysis, is suitable for capturing transient anomalies, and can detect progressive faults earlier by combining feature prediction within the window. (3) Compared with related technologies, the BiLSTM algorithm adopted in this invention can effectively capture the dynamic change law of high frequency and low frequency current signals, improve the short-term prediction accuracy, and adopt the Mann-Kendall trend test, which does not depend on sample data, does not need to follow a specific distribution, and is not affected by a few outliers. It is suitable for categorical variables, ordinal variables and non-normally distributed data, and can robustly detect the changing trend of signals.

[0174] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0175] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0176] Example 2 According to embodiments of the present invention, an apparatus for implementing the above-described method for determining early warning parameters for cable insulation faults is also provided. Figure 5 This is a structural block diagram of a cable insulation fault early warning parameter determination device according to an embodiment of the present invention, such as... Figure 5 As shown, the device includes: an acquisition module 502, a first determination module 504, a second determination module 506, a third determination module 508, and a fourth determination module 510. The device will be described in detail below.

[0177] The acquisition module 502 is used to acquire the current signal sequence of the target cable, wherein the current signal sequence includes a first signal sequence and a second signal sequence, wherein the first signal sequence is a signal sequence whose current signal frequency belongs to a first frequency range, and the second signal sequence is a signal sequence whose current signal frequency belongs to a second frequency range, and the minimum current signal frequency in the first frequency range is greater than the maximum current signal frequency in the second frequency range; the first determination module 504 is connected to the acquisition module 502, and is used to determine the amplitude distribution characteristics corresponding to the target cable based on the first signal sequence; the second determination module 506 is connected to the first determination module 504, and is used to determine the insulation state characteristics corresponding to the target cable based on the second signal sequence; the third determination module 508 is connected to the second determination module 506, and is used to determine the insulation fluctuation index corresponding to the target cable based on the amplitude distribution characteristics and the insulation state characteristics; the fourth determination module 510 is connected to the third determination module 508, and is used to determine the target warning parameters corresponding to the target cable based on the insulation fluctuation index.

[0178] It should be noted here that the above-mentioned acquisition module 502, first determination module 504, second determination module 506, third determination module 508 and fourth determination module 510 correspond to steps S102 to S110 in the method for determining early warning parameters of cable insulation faults. The multiple modules and the corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in the above embodiment 1.

[0179] Example 3 According to another aspect of the present invention, an electronic device is also provided, comprising: a processor; and a memory for storing processor-executable instructions, wherein the processor is configured to execute instructions to implement the cable insulation fault early warning parameter determination method of any of the above embodiments.

[0180] Example 4 According to another aspect of the present invention, a computer-readable storage medium is also provided, which, when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the cable insulation fault early warning parameter determination method described above.

[0181] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0182] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0183] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0184] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0185] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0186] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0187] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for determining a pre-alarm parameter of a cable insulation fault, characterized in that, The method comprises: obtaining a current signal sequence of a target cable, wherein the current signal sequence comprises a first signal sequence and a second signal sequence, the first signal sequence is a signal sequence in which the current signal frequency belongs to a first frequency range, the second signal sequence is a signal sequence in which the current signal frequency belongs to a second frequency range, and the minimum current signal frequency in the first frequency range is greater than the maximum current signal frequency in the second frequency range; determining an amplitude distribution feature corresponding to the target cable according to the first signal sequence; determining an insulation state feature corresponding to the target cable according to the second signal sequence; determining an insulation fluctuation index corresponding to the target cable according to the amplitude distribution feature and the insulation state feature; determining a target warning parameter corresponding to the target cable according to the insulation fluctuation index.

2. The method of claim 1, wherein, The method of determining a target warning parameter corresponding to the target cable according to the insulation fluctuation index comprises: determining a first determination result of whether the insulation fluctuation index is greater than a fluctuation threshold value; in the case that the first determination result is that the insulation fluctuation index is greater than the fluctuation threshold value, determining a predicted signal sequence corresponding to the current signal sequence, wherein the time period corresponding to the predicted signal sequence has a time correlation relationship with the time period corresponding to the current signal sequence; determining an insulation attenuation feature corresponding to the target cable according to the current signal sequence and the predicted signal sequence; determining a target warning parameter corresponding to the target cable according to the insulation attenuation feature.

3. The method of claim 2, wherein, The method of determining an insulation attenuation feature corresponding to the target cable according to the current signal sequence and the predicted signal sequence comprises: performing splicing processing on the current signal sequence and the predicted signal sequence to obtain a target signal sequence, wherein the target signal sequence comprises signal feature values corresponding to a plurality of time nodes respectively; determining a plurality of adjacent node pairs according to the plurality of time nodes, wherein the plurality of adjacent node pairs respectively comprise two adjacent time nodes; for any target node pair in the plurality of adjacent node pairs, determining a sub-attenuation feature corresponding to the target node pair according to the signal feature values corresponding to the two adjacent time nodes in the target node pair; determining sub-attenuation features corresponding to other adjacent node pairs except the target node pair in the plurality of adjacent node pairs by using the manner of determining the sub-attenuation feature corresponding to the target node pair; determining an insulation attenuation feature corresponding to the target cable according to the sub-attenuation features corresponding to the plurality of adjacent node pairs respectively.

4. The method of claim 2, wherein, The method of determining a predicted signal sequence corresponding to the current signal sequence comprises: calling a target model, wherein the target model comprises a first feature extraction layer and a second feature extraction layer, the first feature extraction layer is used to extract time features of input data in a first time sequence, and the second feature extraction layer is used to extract time features of input data in a second time sequence, and the first time sequence is opposite to the second time sequence; obtaining a first time feature according to the current signal sequence and a first feature extraction layer of the target model; obtaining a second time feature according to the current signal sequence and a second feature extraction layer of the target model; determining a predicted signal sequence corresponding to the current signal sequence according to the first time feature and the second time feature.

5. The method of claim 2, wherein, The determining the target early warning parameter corresponding to the target cable according to the insulation attenuation feature comprises: in the case that the insulation attenuation feature comprises an insulation attenuation index, determining a second determination result that whether the insulation attenuation index is greater than an attenuation threshold value; in the case that the second determination result is that the insulation attenuation index is greater than the attenuation threshold value, determining a current correlation corresponding to the first signal sequence and a harmonic distortion feature corresponding to the second signal sequence, wherein the current correlation is used to represent a correlation between a phase angle of a current signal corresponding to the first signal sequence and an amplitude of the current signal, and the harmonic distortion feature represents a distortion feature of a signal waveform corresponding to the second signal sequence; determining the target early warning parameter corresponding to the target cable according to the current correlation and the harmonic distortion feature.

6. The method of claim 1, wherein, Before the obtaining the current signal sequence of the target cable, the method further comprises: determining an insulation fault detection device corresponding to the target cable, wherein the insulation fault detection device comprises a first detection device and a second detection device, the first detection device is arranged at a current signal collection point of a head end of the target cable, the second detection device is arranged at a current signal collection point of a tail end of the target cable, and the current signal collection point comprises a grounding wire corresponding to the target cable; collecting, by the first detection device, a current signal at the current signal collection point of the head end of the target cable to obtain a head end signal sequence; collecting, by the second detection device, a current signal at the current signal collection point of the tail end of the target cable to obtain a tail end signal sequence; determining the current signal sequence corresponding to the target cable according to the head end signal sequence and the tail end signal sequence.

7. The method according to any one of claims 1 to 6, characterized in that, After the determining the target early warning parameter corresponding to the target cable according to the insulation fluctuation index, the method further comprises: in the case that the target early warning parameter comprises a target early warning level, determining a control instruction corresponding to the target early warning level; controlling, by the control instruction, an execution device to perform a target operation on the target cable.

8. An apparatus for determining a pre-alarm parameter of a cable insulation fault, characterized in that The method comprises: an obtaining module, configured to obtain a current signal sequence of a target cable, wherein the current signal sequence comprises a first signal sequence and a second signal sequence, wherein the first signal sequence is a signal sequence of which a current signal frequency belongs to a first frequency range, the second signal sequence is a signal sequence of which a current signal frequency belongs to a second frequency range, and a minimum current signal frequency in the first frequency range is greater than a maximum current signal frequency in the second frequency range; a first determining module, configured to determine an amplitude distribution feature corresponding to the target cable according to the first signal sequence; a second determining module, configured to determine an insulation state feature corresponding to the target cable according to the second signal sequence; a third determining module, configured to determine an insulation fluctuation index corresponding to the target cable according to the amplitude distribution feature and the insulation state feature; a fourth determining module, configured to determine a target warning parameter corresponding to the target cable according to the insulation fluctuation index.

9. An electronic device, comprising: comprise: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement the method for determining a warning parameter of cable insulation fault according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, When the instructions in the computer readable storage medium are executed by the processor of the electronic device, the electronic device is enabled to perform the method for determining a warning parameter of cable insulation fault according to any one of claims 1 to 7.

Images