A method and system for all-weather monitoring and early warning of double-layer GIS prefabricated cabin equipment

Abstract

This invention discloses a method and system for all-weather monitoring and early warning of double-layer GIS prefabricated cabin equipment, belonging to the field of power equipment condition monitoring technology. The method includes the following steps: collecting electromagnetic field and partial discharge signals inside the cabin, extracting interference characteristic parameters to establish an interference characteristic baseline; matching and comparing the original monitoring signal with the interference characteristic baseline, and obtaining a correction signal through anti-interference processing; cross-correcting the correction signal through a redundant fusion module, utilizing the temporal consistency and spatial complementarity between multiple channels; constructing a temporal characteristic curve based on the corrected signal sequence, dynamically comparing it with historical operating modes, and generating risk warning information based on the deviation results. This invention achieves accurate monitoring of partial discharge and the operating environment by establishing an anti-interference signal acquisition and redundant fusion correction mechanism, thereby solving the technical problem of false alarms or missed alarms in existing systems under strong electromagnetic interference.

Description

Technical Field

[0001] This invention relates to the field of power equipment condition monitoring technology, and more specifically, to a method and system for all-weather monitoring and early warning of double-layer GIS prefabricated cabin equipment. Background Technology

[0002] Gas-insulated switchgear (GIS) is widely used in high-voltage power transmission and transformation projects due to its compact structure and reliable operation. With the increasing demand for intelligent operation and maintenance, real-time monitoring methods based on wireless sensors are gradually being introduced into GIS prefabricated cabins to collect electromagnetic field distribution, partial discharge signals, and operating parameters, thereby achieving all-weather monitoring and early warning of equipment status.

[0003] However, during GIS operation, switching operations, current surges, and partial discharges generate complex electromagnetic fields and transient radio waves. These electromagnetic interference signals propagate and superimpose along multiple paths within the cabin space, easily affecting the wireless sensor communication links and data acquisition modules deployed inside the cabin, leading to sensor output signal distortion, falsification, or even interruption. Existing technologies typically rely on signal amplitude thresholds or single filtering algorithms for anti-interference processing. However, in complex interference environments, these methods struggle to accurately distinguish between genuine fault characteristics and electromagnetic interference, easily resulting in false alarms or missed alarms, and failing to provide reliable support for the safe operation of GIS.

[0004] The above-disclosed technical solutions have at least the following technical problems: the electromagnetic field and partial discharge waves generated during the operation of high-voltage GIS will interfere with wireless sensor communication and data acquisition. The existing system lacks an effective anti-interference acquisition and signal correction mechanism, which is prone to false alarms or missed alarms.

[0005] To address the above problems, this invention proposes a solution. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method and system for all-weather monitoring and early warning of double-layer GIS prefabricated cabin equipment. By establishing an anti-interference signal acquisition and redundancy fusion correction mechanism, it achieves accurate monitoring of partial discharge and operating environment, thereby solving the technical problem that existing systems are prone to false alarms or missed alarms under strong electromagnetic interference.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] On the one hand, a method for all-weather monitoring and early warning of double-layer GIS prefabricated cabin equipment includes the following steps: collecting electromagnetic field and partial discharge signals inside the cabin, extracting interference characteristic parameters to establish an interference characteristic baseline; matching and comparing the original monitoring signal with the interference characteristic baseline, and obtaining a correction signal through anti-interference processing; passing the correction signal through a redundant fusion module, and performing cross-correction using the temporal consistency and spatial complementarity between multiple channels; constructing a temporal characteristic curve based on the corrected signal sequence, dynamically comparing it with historical operating modes, and generating risk warning information based on the deviation results.

[0009] In a preferred embodiment, the acquisition of electromagnetic fields and partial discharge signals within the cabin, and the extraction of interference characteristic parameters, specifically involves: deploying multi-channel sensors according to the structural layout of the cabin to acquire electromagnetic fields and partial discharge signals in different spatial regions; synchronously recording the discharge signals under no-load and various operating conditions, and introducing controlled excitation signals to distinguish between environmental noise and equipment body signals; filtering, de-drifting, and time-series segmentation of the acquired signals to obtain segmented datasets labeled with operating conditions; and calculating the spatial covariance matrix and phase delay distribution under different frequency bands based on the segmented datasets to characterize the correlation and coupling characteristics between various spatial locations.

[0010] In a preferred embodiment, establishing the interference characteristic baseline specifically involves: identifying the electromagnetic transfer function between the cabin structure and the sensor based on the spatial correlation, coupling characteristics, and controlled excitation operating condition data of the signal, and constructing a time-frequency transfer function matrix; using the transfer function matrix and phase delay characteristics, performing sparse representation and basis function decomposition on the complex interference signal to obtain interference basis functions with spatial and spectral interpretability; and performing statistical and compression processing on the coefficient distribution of the interference basis functions under different operating conditions to generate an interference characteristic baseline for signal identification.

[0011] In a preferred embodiment, the matching and comparison of the original monitoring signal with the interference feature baseline specifically involves: performing point-by-point differential calculations on the original monitoring signal acquired in real time and the interference feature vector in the interference feature baseline to obtain the differential residual of the fixed-mode interference; performing cross-correlation operations on the original signal and the baseline signal in the time-frequency domain to extract the joint distribution characteristics of the phase drift and the time-frequency energy offset; and generating a dynamic difference mapping matrix based on the differential residual and the joint distribution characteristics.

[0012] In a preferred embodiment, obtaining the correction signal through anti-interference processing specifically involves: performing differential matching and phase decoupling operations on the real-time acquired monitoring signal based on the dynamic difference mapping matrix to suppress fixed-mode interference and common-mode coupling interference; introducing sparse constraints during the multi-channel signal correction process to suppress redundant interference components and maintain partial discharge characteristics, thereby generating an anti-interference corrected monitoring signal.

[0013] In a preferred embodiment, the step of passing the correction signal through a redundant fusion module and performing cross-correction using the temporal consistency and spatial complementarity among multiple channels specifically involves: constructing a temporal consistency constraint model based on the correction signals of each channel, calculating the temporal deviation of each channel signal within the same window, and establishing a temporal consistency matrix; comparing the variation trend of the same parameter in different compartments in the matrix, and correcting for abnormal deviations by weighted averaging when detected, and outputting a reference signal; constructing a spatial residual compensation vector based on the spatial complementarity of the reference signal and the correction signals of each channel, and performing cross-correction on missing or distorted channel signals; and weighted fusion of the signals that meet the temporal consistency constraints and have undergone spatial complementarity correction to obtain the redundantly fused target monitoring signal.

[0014] In a preferred embodiment, the step of constructing a spatial residual compensation vector based on the spatial complementarity of the reference signal and the correction signals of each channel, and performing cross-correction on the missing or distorted channel signals, specifically involves: establishing spatial correlation relationships between the monitoring signals of multiple channels; performing differential operations on the reference signals and correction signals of adjacent nodes to obtain a spatial residual vector; weighting and fusing the spatial residual vector according to spatial distribution weights and signal correlation weights to generate a spatial residual compensation vector; and for channel signals with missing or distorted signals, superimposing the correction signal of the corresponding channel with the spatial residual compensation vector to obtain the cross-corrected channel signal.

[0015] In a preferred embodiment, the step of constructing a time-series characteristic curve based on the corrected signal sequence, dynamically comparing it with historical operating modes, and generating risk warning information based on the deviation results specifically involves: extracting key feature indicators from the monitoring signal sequence after redundant fusion and arranging them in chronological order to form a time-series characteristic curve; performing statistical analysis on the time-series characteristic curve using a dynamic sliding window, calculating the mean and volatility of the signal within the window, and comparing it with historical operating modes to determine abnormal trends; continuously evaluating the abnormal trends to confirm whether they exhibit a continuous deviation or deterioration trend; and generating and outputting differentiated risk warning information for different compartments when a continuous deviation or deterioration characteristic is confirmed.

[0016] On the other hand, a two-layer GIS prefabricated cabin-type equipment all-weather monitoring and early warning system includes the following modules: an interference baseline construction module: used to collect electromagnetic field and partial discharge signals inside the cabin, extract interference characteristic parameters, and establish an interference characteristic baseline; a signal anti-interference correction module: used to match and compare the original monitoring signal with the interference characteristic baseline, and obtain a correction signal through anti-interference processing; a multi-channel redundant fusion module: used to cross-correct the correction signal through the redundant fusion module, utilizing the temporal consistency and spatial complementarity between multiple channels; and a temporal feature analysis and early warning module: used to construct a temporal feature curve based on the corrected signal sequence, dynamically compare it with historical operating modes, and generate risk early warning information based on the deviation results.

[0017] The technical effects and advantages of the all-weather monitoring and early warning method and system for double-layer GIS prefabricated cabin equipment of this invention are as follows:

[0018] 1. This invention proposes a method for optimizing the deployment of multi-channel sensors based on the topology of the cabin structure, injecting controlled excitation conditions, and identifying electromagnetic transfer functions. This method can establish a dynamic interference characteristic baseline, decompose complex electromagnetic interference, and perform position-dependent correction, thereby significantly improving the anti-interference capability and signal acquisition accuracy of partial discharge monitoring. It also overcomes the shortcomings of traditional static threshold methods, which are prone to false alarms and missed alarms in complex cabin environments.

[0019] 2. By introducing a cross-channel temporal consistency and spatial complementarity fusion mechanism, combined with a dynamic sliding window anomaly trend judgment strategy, this invention can not only achieve adaptive correction of missing or distorted signals, but also provide differentiated early warning outputs for the risk status of upper and lower compartments, thereby ensuring reliable operation monitoring and risk prevention and control of the double-layer GIS prefabricated cabin under all working conditions and all weather conditions. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating an all-weather monitoring and early warning method for a double-layer GIS prefabricated cabin-type equipment according to the present invention.

[0021] Figure 2 This is a schematic diagram of the structure of a two-layer GIS prefabricated cabin-type equipment all-weather monitoring and early warning system according to the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0023] Example 1, Figure 1This invention provides a method for all-weather monitoring and early warning of double-layer GIS prefabricated cabin equipment, comprising the following steps:

[0024] S1, collect electromagnetic field and partial discharge signals inside the cabin, extract interference characteristic parameters and establish interference characteristic baseline;

[0025] In this embodiment, the process of collecting electromagnetic field and partial discharge signals inside the acquisition chamber, extracting interference characteristic parameters, and establishing an interference characteristic baseline specifically involves:

[0026] Based on the structural layout of the double-layer GIS prefabricated cabin, multi-channel sensors are deployed in different spatial areas to achieve zoned acquisition of electromagnetic fields and partial discharge signals inside the cabin.

[0027] The signals collected by the sensors are recorded synchronously under no-load and different operating conditions, and a controllable small-amplitude interference is introduced through controlled excitation conditions to distinguish between environmental noise and equipment signals.

[0028] The multi-condition acquired signals are filtered, de-drifted, and time-series segmented to obtain segmented datasets with condition labels.

[0029] Based on segmented datasets, spatial covariance matrices and phase delay distributions in different frequency bands are calculated to characterize the correlation and coupling features between spatial locations.

[0030] It should be noted that controlled excitation conditions refer to applying specific excitation signals under controllable conditions to observe the response of a system or device in order to obtain analyzable data. The specific explanation is as follows:

[0031] Controlled: Parameters such as the magnitude, frequency, and duration of the excitation can be set and adjusted manually to ensure that experimental or monitoring conditions are repeatable and comparable.

[0032] Excitation: refers to the input signal applied to the system, such as voltage pulses, load disturbances, frequency changes, signal interference, etc., which is used to induce the system's response characteristics.

[0033] Operating condition: refers to the state or condition of a system or equipment under excitation, such as no-load condition, full-load condition, partial-load condition, etc.

[0034] Objective: By using controlled excitation conditions, system characteristic parameters can be extracted, abnormal modes can be identified, monitoring systems can be calibrated, or control strategies can be verified.

[0035] The formation of the interference characteristic baseline specifically refers to:

[0036] Based on correlation and coupling characteristics, and combined with controlled excitation data, the electromagnetic transfer function between the cabin structure and sensors is analyzed. Identification is performed to form a time-frequency transfer function matrix that reflects the actual coupling path, where, Let f be the coupling coefficient from sensor x to sensor y at frequency f. Let be the spectrum of the response signal of sensor y. Let be the spectrum of the excitation signal of sensor x;

[0037] Based on the transfer function matrix and phase delay characteristics, sparse representation and basis function learning of signals are performed on signals, and complex electromagnetic interference is decomposed into several interference basis functions with spatial and spectral interpretability.

[0038] The coefficient distribution of the interference basis function under different operating conditions is statistically analyzed and compressed to form an interference characteristic baseline containing a position-dependent correction vector. This baseline is then bound to each sensor channel for subsequent difference mapping, dynamic compensation, and real-time early warning threshold correction.

[0039] The deployment of multi-channel sensors specifically involves: based on the double-layer cabin structure diagram and the topology of the main electrical components (grounding electrodes, busbars, disconnect switches, etc.), and following the principles of minimum coverage redundancy and maximum spatial resolution, planning a set of multi-channel sensor locations. This includes:

[0040] High-frequency pulse / electromagnetic sensor array (for partial discharge, high sampling rate fs_pd≥200 kHz)

[0041] Broadband electric / magnetic field probe (for electromagnetic field distribution measurement)

[0042] Clock synchronization reference node (PTP / NTP) and location coordinates.

[0043] It should be noted that the innovation of this embodiment lies in the following: to address the technical pain point of insufficient monitoring accuracy in the complex electromagnetic environment of a double-layer GIS prefabricated cabin, an integrated solution is proposed that integrates structural topology sensing, active interference identification and dynamic baseline construction. Its core innovations are reflected in the following aspects: First, it pioneers a multi-channel sensor optimization deployment method based on the cabin structure and electrical component topology. It plans the coordinated layout of high-frequency pulse sensor arrays and broadband electromagnetic field probes according to the principle of "minimum redundancy - maximum spatial resolution," and introduces PTP / NTP clock synchronization nodes, solving the signal attenuation and spatial sampling blind zone problems caused by the double-layer shielding structure, overcoming the limitations of traditional random sensor deployment or single-type configuration. Second, it innovatively adopts controlled excitation condition injection and multi-condition synchronous recording technology. By actively introducing controllable small-amplitude interference and combining it with time-series segmented processing, it achieves accurate differentiation between environmental noise and equipment signals, breaking through the bottleneck of traditional methods relying on passive filtering to eliminate complex interference. Third, it proposes an electromagnetic coupling feature extraction mechanism based on spatial covariance matrix and phase delay distribution. Combined with time-frequency transfer function matrix identification technology, it decomposes complex interference into basis functions with spatial-spectral interpretability, thereby constructing a position-dependent dynamic interference baseline bound to the sensor channel, replacing the traditional static threshold scheme, and achieving real-time compensation and threshold adaptive correction under all operating conditions, significantly improving the anti-interference capability and early warning accuracy of partial discharge monitoring.

[0044] S2, the original monitoring signal is matched and compared with the interference characteristic baseline, and the correction signal is obtained through anti-interference processing;

[0045] The anti-interference processing includes differential matching and phase decoupling;

[0046] In this embodiment, the matching and comparison of the original monitoring signal with the interference characteristic baseline specifically involves:

[0047] The point-by-point differential calculation is performed between the real-time acquired raw monitoring signal and the interference feature vector in the interference feature baseline to obtain the differential residual of the fixed mode interference.

[0048] In the time-frequency domain, cross-correlation is performed on the original signal and the baseline signal to extract the joint distribution of phase drift and time-frequency energy offset.

[0049] A dynamic difference mapping matrix is ​​generated based on the differential residual and joint distribution to characterize the degree of interference deviation under different channels and operating conditions.

[0050] The point-by-point difference calculation is specifically as follows:

[0051]

[0052] The cross-correlation operation is specifically as follows:

[0053] In the time-frequency domain, firstly... and Perform a short-time Fourier transform:

[0054]

[0055]

[0056]

[0057]

[0058]

[0059]

[0060] in, The differential residual sequence is a fixed-mode interference. The raw monitoring signals are collected in real time. This is the baseline interference feature vector. For window functions, The time center of the window function ensures that only a local segment is used for the Fourier transform. For frequency, For time, It is a complex exponential kernel function. The imaginary unit ( ), The cross-correlation function represents the correlation between the original signal and the baseline signal in the time-frequency domain. The complex conjugate of the spectrum of the baseline signal. This is the phase shift amount. This is the time-frequency energy offset. It is a joint distribution.

[0061] The process of obtaining the correction signal through anti-interference processing specifically involves:

[0062] Based on the dynamic difference mapping matrix, differential matching and phase decoupling operations are performed on the real-time acquired monitoring signals. Differential matching is used to cancel fixed-mode interference, and phase decoupling is used to filter out phase synchronization interference components introduced by electromagnetic field common-mode coupling.

[0063] Sparse constraints are introduced during the multi-channel signal correction process to suppress redundant interference components and maintain the effective characteristics of partial discharge, thereby generating an anti-interference correction monitoring signal.

[0064] In this embodiment, the differential matching and phase decoupling operation performed on the real-time acquired monitoring signal based on the dynamic difference mapping matrix specifically includes:

[0065] Using the dynamic difference mapping matrix as a constraint, the real-time signal is differentially compared with the baseline interference signal at the same time point or frequency band.

[0066] By matching the difference results with the interference residual distribution in the mapping matrix, fixed-mode interference components with consistent offsets across multiple sampling points are identified.

[0067] For the identified interference components, amplitude cancellation is performed based on the matching coefficients, thereby reducing the interference of the fixed mode.

[0068] After differential matching is completed, the instantaneous phase sequence in the time-frequency domain is extracted from the signal.

[0069] The phase sequence is compared with the common-mode phase trajectory in the baseline, and the phase synchronization index (e.g., cross-correlation coefficient or Hilbert instantaneous phase difference) is calculated.

[0070] When a high synchronization interval is detected, it is determined that common-mode coupling interference exists;

[0071] For such intervals, a decoupling operation is applied: specifically, an orthogonal basis decomposition is introduced in the frequency domain, the common phase components are projected and removed, and only the independent phase components related to the partial discharge characteristics are retained.

[0072] In this embodiment, the introduction of sparse constraint conditions during multi-channel signal correction specifically refers to:

[0073] After differential matching and phase decoupling are completed, a preliminary correction signal for multiple channels is obtained;

[0074] In the joint modeling of multi-channel signals, a sparse constraint term (such as the L1 norm) is introduced to force the retention of only the feature components that appear sparsely and uniformly across multiple channels.

[0075] By iteratively optimizing the solution, redundant interference components that are widely distributed, have low energy, and lack sparsity consistency are suppressed, highlighting the peak characteristics of the partial discharge signal.

[0076] S3, the correction signal is passed through a redundant fusion module, and cross-correction is performed using the temporal consistency and spatial complementarity among multiple channels, specifically:

[0077] A timing consistency constraint model is constructed based on the correction signals of each channel, the timing deviation of any two channel signals within the same time window is calculated, and a timing consistency matrix is ​​established between the upper and lower layer sensors.

[0078] By comparing the variation trend of the same monitoring parameter in different cabin layers in the time series consistency matrix, when an abnormal deviation (greater than the preset threshold) is detected, the deviation value is corrected by weighted averaging and the corrected reference signal is output.

[0079] Based on the spatial complementarity between the corrected reference signal and the corrected signals of each channel, a spatial residual compensation vector is constructed to cross-correct the missing or distorted channel signals.

[0080] The signals that meet the temporal consistency constraints and have undergone spatial complementarity correction are weighted and fused to obtain the target monitoring signal after redundant fusion.

[0081] In this embodiment, the spatial residual compensation vector is constructed based on the spatial complementarity between the corrected reference signal and the corrected signals of each channel, and the missing or distorted channel signals are cross-corrected. Specifically, this involves:

[0082] A spatial adjacency graph is established among the monitoring signals of multiple channels. The reference signal and the correction signal of adjacent sensor nodes are differentially processed to obtain the spatial residual vector.

[0083] The spatial residual vector is weighted and superimposed according to the sensor geometric distribution weight and the channel signal correlation weight to generate the spatial residual compensation vector.

[0084] For any channel signal that is detected to be missing or distorted, the correction signal of the corresponding channel is superimposed and corrected with the spatial residual compensation vector to obtain the cross-corrected channel signal.

[0085] The spatial residual vector is specifically:

[0086]

[0087] The spatial residual compensation vector is specifically:

[0088]

[0089]

[0090] The cross-corrected channel signal is specifically as follows:

[0091]

[0092] in, For spatial residual vectors, This is the correction signal for adjacent channels. This is the correction signal for the i-th channel. For the set of adjacent channels, These are weighting coefficients. These are the components of the spatial residual vector. This is the spatial residual compensation vector. The spatial distance between passageways. This refers to signal correlation (Pearson correlation coefficient). , The preset adjustment coefficient, This is the channel signal after cross-correction.

[0093] S4 constructs a time-series characteristic curve based on the corrected signal sequence, dynamically compares it with historical operating modes, and generates risk warning information based on the deviation results.

[0094] In this embodiment, the step of constructing a time-series characteristic curve based on the corrected signal sequence, dynamically comparing it with historical operating modes, and generating risk warning information based on the deviation results specifically involves:

[0095] Based on the monitoring signal sequence after redundancy fusion, key feature indicators, including signal mean, variance, peak value and energy, are extracted in time order, and the features of each time window are arranged in sequence to form a time series feature curve;

[0096] A dynamic sliding window is constructed to statistically analyze the signal mean and volatility of the time series characteristic curve within each window;

[0097] The mean within the sliding window is compared with the preset normal fluctuation range, and the volatility is compared with the historical standard deviation threshold. When the mean continuously deviates from the preset range and the volatility exceeds the historical standard deviation threshold, it is judged as an abnormal trend.

[0098] Continuously evaluate the identified abnormal trends to confirm whether they show a continuous deviation or deterioration trend;

[0099] When an abnormal trend matches the characteristics of continuous deviation or deterioration, it triggers the output of differentiated risk warning information for the upper and lower compartments, enabling all-weather reliable monitoring of the two-layer GIS prefabricated compartment.

[0100] The output of the differentiated risk warning information is specifically as follows:

[0101] Based on the differentiated risk levels of the upper and lower compartments, environmental interference warnings, electrical anomaly warnings, and comprehensive risk warnings are output respectively. The warning results are then correlated with external meteorological data to achieve reliable alarms under extreme weather conditions.

[0102] Example 2, Figure 2 This invention discloses an all-weather monitoring and early warning system for a double-layer GIS prefabricated cabin-type equipment, comprising the following modules:

[0103] Interference baseline construction module: used to collect electromagnetic field and partial discharge signals inside the cabin, extract interference characteristic parameters and establish interference characteristic baseline;

[0104] Signal anti-interference correction module: used to match and compare the original monitoring signal with the interference characteristic baseline, and obtain the correction signal through anti-interference processing;

[0105] Multi-channel redundant fusion module: used to cross-correct the correction signal by passing it through the redundant fusion module and utilizing the timing consistency and spatial complementarity between multiple channels.

[0106] The timing feature analysis and early warning module is used to construct timing feature curves based on the corrected signal sequence, dynamically compare them with historical operating modes, and generate risk warning information based on the deviation results.

[0107] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0108] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0109] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0110] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0111] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0112] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for all-weather monitoring and early warning of double-layer GIS prefabricated cabin equipment, characterized in that, Includes the following steps: Collect electromagnetic field and partial discharge signals inside the cabin, extract interference characteristic parameters, and establish an interference characteristic baseline; The original monitoring signal is matched and compared with the interference characteristic baseline, and the correction signal is obtained through anti-interference processing; The correction signal is processed by a redundant fusion module, and cross-correction is performed using the temporal consistency and spatial complementarity among multiple channels. Specifically, a temporal consistency constraint model is constructed based on the correction signals of each channel, the temporal deviation of each channel signal within the same window is calculated, and a temporal consistency matrix is ​​established. The variation trend of the same parameter in different compartments is compared in the matrix. When an abnormal deviation is detected, correction is performed by weighted averaging, and a reference signal is output. Based on the spatial complementarity between the reference signal and the correction signals of each channel, a spatial residual compensation vector is constructed to cross-correct the missing or distorted channel signals; the signals that meet the temporal consistency constraints and have been corrected by spatial complementarity are weighted and fused to obtain the target monitoring signal after redundant fusion. A time-series characteristic curve is constructed based on the corrected signal sequence, and it is dynamically compared with the historical operating mode. Risk warning information is generated based on the deviation results.

2. The all-weather monitoring and early warning method for double-layer GIS prefabricated cabin equipment according to claim 1, characterized in that, The acquisition of electromagnetic field and partial discharge signals within the acquisition chamber, and the extraction of interference characteristic parameters, specifically includes: Multi-channel sensors are deployed in different zones according to the structural layout of the cabin to collect electromagnetic field and partial discharge signals in different spatial areas. The discharge signal is recorded synchronously under no-load and various operating conditions, and a controlled excitation signal is introduced to distinguish between environmental noise and equipment body signals. The acquired signals are filtered, de-drifted, and segmented according to time sequence to obtain segmented datasets with working condition labels. Based on segmented datasets, spatial covariance matrices and phase delay distributions in different frequency bands are calculated to characterize the correlation and coupling features between spatial locations.

3. The all-weather monitoring and early warning method for double-layer GIS prefabricated cabin equipment according to claim 2, characterized in that, The establishment of the interference characteristic baseline specifically involves: Based on the spatial correlation and coupling characteristics of the signal and the controlled excitation condition data, the electromagnetic transfer function between the cabin structure and the sensor is identified, and the time-frequency transfer function matrix is ​​constructed. By utilizing the transfer function matrix and phase delay characteristics, sparse representation and basis function decomposition are performed on the complex interference signal to obtain interference basis functions with spatial and spectral interpretability. The coefficient distribution of the interference basis function under different operating conditions is statistically analyzed and compressed to generate an interference characteristic baseline for signal identification.

4. The all-weather monitoring and early warning method for double-layer GIS prefabricated cabin equipment according to claim 3, characterized in that, The process of matching and comparing the original monitoring signal with the interference characteristic baseline specifically involves: The point-by-point differential calculation is performed between the real-time acquired raw monitoring signal and the interference feature vector in the interference feature baseline to obtain the differential residual of the fixed mode interference. Cross-correlation is performed on the original signal and the baseline signal in the time-frequency domain to extract the joint distribution characteristics of phase drift and time-frequency energy offset. A dynamic difference mapping matrix is ​​generated based on the difference residuals and joint distribution characteristics.

5. The all-weather monitoring and early warning method for double-layer GIS prefabricated cabin equipment according to claim 4, characterized in that, The process of obtaining the correction signal through anti-interference processing specifically involves: Based on the dynamic difference mapping matrix, differential matching and phase decoupling operations are performed on the real-time acquired monitoring signals to suppress fixed-mode interference and common-mode coupling interference; In the process of multi-channel signal correction, sparse constraints are introduced to suppress redundant interference components and maintain partial discharge characteristics, thereby generating an anti-interference correction monitoring signal.

6. The all-weather monitoring and early warning method for double-layer GIS prefabricated cabin equipment according to claim 5, characterized in that, The spatial residual compensation vector is constructed based on the spatial complementarity between the reference signal and the correction signals of each channel, and cross-correction is performed on the missing or distorted channel signals, specifically as follows: Establish spatial correlation between monitoring signals from multiple channels, and perform differential operations on the reference signals and correction signals of adjacent nodes to obtain spatial residual vectors; The spatial residual vector is weighted and fused according to the spatial distribution weight and the signal correlation weight to generate the spatial residual compensation vector. For channel signals with missing or distorted signals, the correction signal of the corresponding channel is superimposed with the spatial residual compensation vector to obtain the cross-corrected channel signal.

7. The all-weather monitoring and early warning method for double-layer GIS prefabricated cabin equipment according to claim 6, characterized in that, The process involves constructing a time-series characteristic curve based on the corrected signal sequence, dynamically comparing it with historical operating patterns, and generating risk warning information based on the deviation results. Specifically: Based on the monitoring signal sequence after redundancy fusion, key feature indicators are extracted and arranged in chronological order to form a time-series feature curve. Statistical analysis of time series characteristic curves is performed using a dynamic sliding window. The mean and volatility of the signal within the window are calculated and compared with historical operating patterns to determine abnormal trends. Continuously assess abnormal trends to confirm whether they exhibit a persistent deviation or deterioration trend; When a persistent deviation or deterioration is confirmed, differentiated risk warning information is generated and output for different cabins.

8. A system using the all-weather monitoring and early warning method for a double-layer GIS prefabricated cabin-type equipment as described in any one of claims 1-7, characterized in that, Includes the following modules: Interference baseline construction module: used to collect electromagnetic field and partial discharge signals inside the cabin, extract interference characteristic parameters and establish interference characteristic baseline; Signal anti-interference correction module: used to match and compare the original monitoring signal with the interference characteristic baseline, and obtain the correction signal through anti-interference processing; Multi-channel redundant fusion module: used to cross-correct the correction signal by passing it through the redundant fusion module and utilizing the timing consistency and spatial complementarity between multiple channels. The timing feature analysis and early warning module is used to construct timing feature curves based on the corrected signal sequence, dynamically compare them with historical operating modes, and generate risk warning information based on the deviation results.

Images