Method and system for vibration testing and state monitoring of in-operation high-voltage circuit breaker

Abstract

The invention discloses a vibration test and state monitoring method and system for an in-operation high-voltage circuit breaker, and the method comprises the steps: carrying out the energy propagation time sequence test of a multi-measurement-point vibration signal, constructing an energy flow direction topological graph, and precisely positioning a vibration source region through energy ratio analysis and path backtracking; acquiring a response time constant by adopting an impact response attenuation test, and converting a vibration source into a fault space positioning coordinate through space grid division and rate matching; a frequency-space mapping relation is established based on a fault position, and vibration characteristics are comprehensively described through peak frequency time sequence tracking and nonlinear component identification such as frequency multiplication, modulation and transient state; performing dispersion analysis on the continuous action impact peak value to form an action consistency deviation, and fusing a nonlinear feature set to quantitatively evaluate the mechanical degradation degree; the grading test density is generated according to the degradation state, the time period test unit is determined, the state grading test benchmark is established to realize mechanical state grade judgment, and the equipment operation reliability and the monitoring efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of power equipment condition monitoring technology, and in particular to a method and system for vibration testing and condition monitoring of in-operation high-voltage circuit breakers. Background Technology

[0002] High-voltage circuit breakers play a crucial role in power systems by breaking and protecting circuits. Their mechanical components experience wear and deterioration during long-term opening and closing operations. Vibration signals can reflect the mechanical operating status of circuit breakers and are an important source of information for equipment health monitoring.

[0003] Existing vibration monitoring methods suffer from the following shortcomings: monitoring points are scattered, lacking a holistic analysis of the vibration propagation process, making it difficult to accurately determine the source of abnormal vibrations; feature extraction mainly relies on single-dimensional amplitude or frequency indicators, failing to fully utilize the time-frequency evolution and system response characteristics inherent in the vibration signal; equipment condition assessment lacks quantitative evaluation standards, relying heavily on empirical judgment; and the fixed monitoring cycle cannot adapt to the monitoring needs of different deterioration stages. These problems limit the effectiveness of vibration monitoring technology in circuit breaker condition assessment and fault prediction. Summary of the Invention

[0004] This invention provides a method and system for vibration testing and condition monitoring of in-operation high-voltage circuit breakers. It aims to achieve precise location of vibration sources by constructing an energy flow topology map, determine fault spatial coordinates by matching the response time constant with the rate of spatial attenuation distribution, comprehensively extract vibration characteristics by establishing a frequency-space mapping relationship and performing nonlinear component detection, quantitatively assess the degree of mechanical degradation by analyzing action consistency deviations, and finally generate graded test densities based on the degradation state. This enables differentiated and intelligent condition monitoring, providing comprehensive and accurate technical support for predictive maintenance and intelligent operation and maintenance decisions for high-voltage circuit breakers.

[0005] The first aspect of this invention proposes a method for vibration testing and condition monitoring of in-operation high-voltage circuit breakers, comprising the following steps: Vibration signal data from multiple measuring points are acquired during the opening and closing operations of the high-voltage circuit breaker in operation. Energy propagation timing tests are performed on the vibration signal data from the multiple measuring points to construct an energy flow topology map. In the energy flow topology map, the vibration source location area is determined by energy ratio analysis. The vibration signal in the vibration source location area is subjected to impact response attenuation test to generate a response time constant. The response time constant is combined with the vibration source location area to convert it into fault space location coordinates. Based on the fault spatial location coordinates, a frequency-space mapping relationship is established, the spectral distribution sequence of the region corresponding to the frequency-space mapping relationship is extracted, peak frequency time-series tracking is performed on the spectral distribution sequence to generate a frequency drift index, and nonlinear component detection is performed based on the frequency drift index and the spectral distribution sequence to form a nonlinear feature set; The mechanical degradation degree of the fault spatial location coordinates is determined to identify the degradation influence area. Continuous action impact peaks are extracted from the degradation influence area to generate a peak sequence. Discreteness analysis is performed on the peak sequence to form action consistency deviation. The action consistency deviation is fused with the nonlinear feature set to generate a graded test density. Based on the graded test density, time period test units are determined, tests are performed on the time period test units to establish a state graded test benchmark, and a mechanical state level determination is generated based on the state graded test benchmark.

[0006] A second aspect of this invention provides a vibration testing and condition monitoring system for in-operation high-voltage circuit breakers, comprising: The energy testing module is used to acquire vibration signal data from multiple measuring points during the opening and closing operations of the high-voltage circuit breaker in operation, and to perform energy propagation timing tests on the vibration signal data from the multiple measuring points to construct an energy flow topology map. The coordinate generation module is used to determine the vibration source location area through energy ratio analysis in the energy flow topology map, perform impact response attenuation test on the vibration signal of the vibration source location area to generate a response time constant, and combine the response time constant with the vibration source location area to convert it into fault space location coordinates. The frequency optimization module is used to establish a frequency-space mapping relationship based on the fault spatial location coordinates, extract the spectral distribution sequence of the region corresponding to the frequency-space mapping relationship, perform peak frequency time-series tracking on the spectral distribution sequence to generate a frequency drift index, and perform nonlinear component detection based on the frequency drift index and the spectral distribution sequence to form a nonlinear feature set. The degradation analysis module is used to determine the degree of mechanical degradation of the fault spatial location coordinates, identify the degradation influence area, extract peak values ​​of continuous action impacts in the degradation influence area to generate a peak sequence, perform dispersion analysis on the peak sequence to form action consistency deviation, and fuse the action consistency deviation with the nonlinear feature set to generate a graded test density. The result determination module is used to determine the time period test unit according to the graded test density, perform tests on the time period test unit to establish a state graded test benchmark, and generate a mechanical state level determination based on the state graded test benchmark.

[0007] The beneficial effects of this invention are reflected in the following points: First, by constructing an energy flow topology map to realize the networked expression of the vibration propagation process, and combining energy ratio analysis and path backtracking methods to identify the vibration source region, the cross-domain matching technology of response time constant and spatial decay rate is adopted to convert the vibration source from a qualitative region to quantitative spatial coordinates. This breaks through the limitation of traditional single-point monitoring in tracing the source, and provides a complete technical path from propagation mechanism analysis to spatial coordinate mapping for the accurate location of faulty components, narrowing the scope of fault investigation and shortening the diagnosis time. Second, by establishing a mapping model between frequency characteristics and spatial location, and combining peak frequency time-series tracking technology to capture the frequency evolution law, the corresponding nonlinear feature components (dominant, modulation, transient) are extracted for different drift modes (monotonic, periodic, abrupt change), forming a multi-dimensional feature system covering the frequency domain, time domain, and nonlinear domain. This overcomes the shortcomings of traditional methods that only focus on a single amplitude or frequency parameter, and realizes a comprehensive characterization of the dynamic evolution process and nonlinear response mechanism of the vibration system, improving the ability to identify early degradation characteristics and abnormal modes. Third, by quantitatively analyzing the discrete characteristics of the peak sequence during continuous operation, a multi-parameter evaluation index including deviation amplitude, deviation frequency, and continuous deviation length was established. By integrating vibration nonlinear characteristics, a degradation grading standard was formed, realizing the configuration of differentiated monitoring strategies from slight to severe degradation. This matched the monitoring frequency and testing depth with the actual condition requirements of the equipment, solving the contradiction between insufficient early-stage degradation monitoring and resource redundancy in the later stages under the traditional fixed-cycle monitoring mode, and improving the timeliness and economy of condition monitoring. Attached Figure Description

[0008] The accompanying drawings illustrate specific examples of the technical solutions described in this invention and, together with the detailed embodiments, form part of the specification, serving to explain the technical solutions, principles, and effects of this invention.

[0009] Unless otherwise specified or otherwise, the same reference numerals in different figures represent the same or similar technical features, and different reference numerals may be used to represent the same or similar technical features.

[0010] Figure 1 This is a flowchart of the method for vibration testing and condition monitoring of high-voltage circuit breakers in operation, as described in this invention.

[0011] Figure 2 This is a structural block diagram of the vibration testing and condition monitoring system for in-operation high-voltage circuit breakers according to the present invention. Detailed Implementation

[0012] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0013] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0014] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0015] The technical solutions of the embodiments of this application are described below.

[0016] like Figure 1 As shown, this embodiment of the invention provides a method for vibration testing and condition monitoring of in-operation high-voltage circuit breakers, including the following steps S110-S150: Step S110: Obtain vibration signal data from multiple measuring points during the opening and closing operation of the high-voltage circuit breaker in operation, and perform energy propagation timing test on the vibration signal data from multiple measuring points to construct an energy flow topology map.

[0017] Specifically, vibration signal data from multiple measurement points were acquired during the opening and closing operations of the high-voltage circuit breaker. Vibration sensor arrays were deployed at key components of the high-voltage circuit breaker body, including the operating mechanism, drive shaft, contact system, and support structure, to collect vibration signal data in real time during the opening and closing operations. Accelerometer-type vibration sensors were used, with a sensitivity set to 100mV / g to 500mV / g and a frequency response range set to 0.5Hz to 10kHz. Measurement points were placed at the energy storage spring seat, trip unit, and buffer locations of the operating mechanism; at the main bearing seat, linkage mechanism, and insulating tie rod locations of the drive shaft; at the moving contact support, stationary contact support, and arc contact locations of the contact system; and at the base mounting surface, insulator flange, and shell connection of the support structure. The spacing between adjacent measurement points was controlled between 200 and 500 mm, and the sampling frequency was set to above 20kHz. The vibration acceleration values, sampling timestamps, and spatial coordinates of each measurement point were recorded. When a high-voltage circuit breaker performs opening or closing operations, the energy storage spring of the operating mechanism rapidly releases energy to drive the transmission shaft. The transmission shaft transmits the motion to the contact system, causing the moving contact to separate from or make contact with the stationary contact, generating a mechanical impact. The impact energy propagates along the circuit breaker structure in the form of vibration waves, and different measuring points sequentially sense the vibration signal. The collected vibration acceleration values, time information, and spatial coordinates of each measuring point are classified and organized according to the measuring point number to form multi-measuring-point vibration signal data.

[0018] A time-series energy propagation test was performed on vibration signal data from multiple measurement points to construct an energy flow topology map. The vibration acceleration time-domain waveforms of each measurement point were extracted from the multi-point vibration signal data, and the acceleration signals were converted into vibration energy signals. Vibration energy was represented by the square of the root mean square value of vibration acceleration, calculated using the formula E=(A_rms)², where E is the vibration energy and A_rms is the root mean square value of acceleration. Time-series analysis was performed on the vibration energy signals of each measurement point to identify the time sequence of energy arrival at each point. The energy arrival time was defined as the moment when the vibration energy at that measurement point first exceeded the energy threshold, which was set to three times the average background noise energy. By comparing the order of energy arrival times at each measurement point, the propagation direction of vibration energy in the circuit breaker structure was determined. Measurement points with earlier energy arrival times were identified as upstream nodes, and those with later energy arrival times were identified as downstream nodes. A set of measurement point nodes was established, with all measurement points used as nodes in the energy flow topology map. Node labels included the measurement point number, spatial coordinates, and energy arrival time. Directed connections are established between nodes based on the energy propagation time sequence, forming directed edges from nodes where energy arrives earlier to nodes where energy arrives later. The weight of the directed edge is represented by the energy propagation intensity between adjacent nodes, which is the ratio of the peak energy of the downstream node to the peak energy of the upstream node. During the opening and closing process of a high-voltage circuit breaker, the operating mechanism generates vibration energy earliest. This energy propagates along the transmission axis to the contact system. The impact energy generated by the opening and closing of the contacts then propagates along the support structure to the circuit breaker base and housing, forming multiple energy propagation paths. An energy flow topology graph is constructed, which describes the propagation network of vibration energy in the circuit breaker structure in the form of nodes and edges. Nodes represent the location of measuring points, directed edges represent the energy flow direction, and the weight of the edge represents the energy propagation intensity.

[0019] Step S120: Determine the vibration source location area through energy ratio analysis in the energy flow topology map, perform impact response attenuation test on the vibration signal in the vibration source location area to generate a response time constant, and combine the response time constant with the vibration source location area to convert it into fault space location coordinates.

[0020] In some embodiments, determining the vibration source location area through energy ratio analysis in the energy flow topology map includes: extracting node energy from the energy flow topology map to form an energy distribution sequence; identifying the energy ratio of adjacent nodes based on the energy distribution sequence; performing energy ratio analysis on the energy ratio of adjacent nodes to identify energy convergence paths; and using the energy convergence paths to backtrack and determine the vibration source location area.

[0021] Energy distribution sequences are formed by extracting node energy from the energy flow topology map. The peak energy data of each node is read from the energy flow topology map, and the node number and corresponding peak energy value are extracted. The nodes are sorted according to their spatial position within the circuit breaker structure. The sorting method can be based on the propagation path from the operating mechanism to the contact system and then to the support structure, or a vertical order from bottom to top, or a radial order from inside to outside. The sorted node peak energy values ​​are then combined to form an energy distribution sequence, where the i-th element corresponds to the peak energy value of the i-th node. For example, during the closing process of a high-voltage circuit breaker, the starting node, the trip unit of the operating mechanism, has a relatively high peak energy value. The energy gradually attenuates as it propagates along the transmission axis to the contact system, rebounds slightly upon reaching the main bearing seat (due to multi-path convergence), attenuates again as it continues to propagate towards the contact support, and finally reaches its lowest value at the edge of the circuit breaker housing. This energy value sequence from the starting point to the end point is the energy distribution sequence, which visually reflects the fluctuations in energy along the spatial propagation path. The energy distribution sequence reflects the amplitude variation of vibration energy during propagation along the circuit breaker structure. In an energy distribution sequence, nodes with higher energy peaks are usually vibration source locations or energy convergence locations, while nodes with lower energy peaks are the end of energy propagation or energy dissipation locations.

[0022] Identify the energy ratio of adjacent nodes based on the energy distribution sequence. Calculate the ratio of adjacent elements in the energy distribution sequence: energy ratio R_i = E_(i+1) / E_i, where E_i is the energy peak value of the i-th node, and E_(i+1) is the energy peak value of the (i+1)-th node. An energy ratio R_i greater than 1 indicates that the energy increases during propagation from node i to node i+1; an energy ratio R_i less than 1 indicates that the energy decreases during propagation; and an energy ratio R_i equal to 1 indicates that the energy remains constant during propagation. Statistically calculate the energy ratio values ​​of all adjacent node pairs to form an energy ratio sequence. Analyze the numerical distribution characteristics of the energy ratio sequence to identify the extreme values ​​and trends of the energy ratios. Node pairs with energy ratios significantly greater than 1 indicate energy amplification or convergence in the propagation path, while node pairs with energy ratios significantly less than 1 indicate rapid energy decay or dissipation in the propagation path.

[0023] Energy ratio analysis is performed on the energy ratios of adjacent nodes to identify energy convergence paths. Node pairs with energy ratios greater than a threshold of 1.2 are selected from the energy ratio sequence. These node pairs are connected according to the propagation direction to form propagation chains of energy growth. The convergence characteristics of these propagation chains are analyzed to identify cases where multiple propagation chains converge to the same node. A convergence node is defined as a node with at least two upstream propagation chains pointing to it; this node is the convergence point of the energy convergence path. The convergence intensity of the convergence node is evaluated, expressed as the ratio of the peak energy of the convergence node to the average peak energy of all upstream nodes. A higher convergence intensity indicates a more significant energy convergence phenomenon. The complete path from the starting node through the energy growth propagation chain to the convergence node is extracted; this path is the energy convergence path. For example, when abnormal vibration occurs in the contact system of a high-voltage circuit breaker, the vibration energy generated by the contact impact propagates simultaneously in two directions: one path propagates upward along the moving contact support to the insulating tie rod (energy ratio R=1.3 indicates energy enhancement), and the other path propagates downward along the linkage mechanism to the main bearing housing of the drive shaft (energy ratio R=1.4 indicates energy enhancement). These two energy-increasing propagation chains converge at the main bearing housing, making it the energy convergence point. Comparison reveals that the energy peak at this convergence point is significantly higher than the value expected from a single path, with a convergence intensity exceeding 1.5. This phenomenon of multiple paths converging at the same point and generating energy superposition is a typical characteristic of an energy convergence path. The energy convergence path records the node sequence of the starting point, intermediate nodes, and convergence point of the energy convergence.

[0024] The vibration source location area is determined by tracing back along the energy convergence path. Starting from the convergence node of the energy convergence path, tracing back in the opposite direction of energy propagation. During the tracing process, the upstream node with the largest energy ratio is selected as the next tracing node, and the tracing continues until the node with the largest energy peak in the energy distribution sequence is reached. The node with the largest energy peak corresponds to the initial location of the vibration energy, which is the core location of the vibration source. During the tracing process, if a branch path is encountered, i.e., a node has multiple upstream nodes, the energy ratios of each upstream path are compared, and the path with the largest energy ratio is selected to continue tracing. All nodes on the tracing path are recorded; these nodes constitute the main propagation channels of the vibration energy. The area consisting of the core node of the vibration source and its directly adjacent nodes is defined as the vibration source location area. The vibration source location area typically contains 3 to 5 adjacent nodes, covering an area of ​​approximately 500 mm around the core location of the vibration source. In high-voltage circuit breakers, if there is a mechanical loosening fault at the trip unit position of the operating mechanism, abnormal vibration is generated when the trip unit operates. This vibration energy propagates to the surrounding area and converges at the drive shaft position. Through tracing analysis, the location of the trip unit can be accurately located as the vibration source location area.

[0025] Impact response attenuation tests were performed on the vibration signals in the vibration source location area to generate response time constants. The time-domain waveforms of the vibration signals at each measuring point within the vibration source location area were extracted, and the vibration response segment generated by the opening and closing impacts was selected for analysis. The vibration response segment was defined as the time range from the peak impact to the vibration amplitude attenuating to 10% of the peak value. The vibration amplitude of the response segment was fitted with an exponential decay function, A(t) = A0·e^(-t / τ), where A(t) is the vibration amplitude at time t, A0 is the initial peak amplitude, τ is the response time constant, and e is the base of the natural logarithm. The fitting parameters were optimized using the least squares method to obtain the response time constant τ corresponding to each measuring point location. The response time constant τ reflects the attenuation rate of vibration energy at that location; a larger τ value indicates slower vibration energy attenuation, lower structural damping at that location, or higher vibration transmission efficiency. In high-voltage circuit breakers, the metal support location of the operating mechanism typically has a larger vibration response time constant due to its high structural rigidity and low damping, while the insulator connection location has a smaller vibration response time constant due to material damping. The response time constant values ​​of all measuring points within the vibration source location area are statistically analyzed to form a response time constant dataset.

[0026] In some embodiments, the step of combining the response time constant with the vibration source location area to convert it into fault spatial location coordinates includes: dividing the vibration source location area into a spatial grid to establish a grid cell array; measuring the attenuation rate of each cell based on the grid cell array to generate a spatial attenuation distribution; performing rate matching between the response time constant and the spatial attenuation distribution to form a matching position index; and using the matching position index to convert and determine the fault spatial location coordinates.

[0027] A spatial grid array of mesh cells is established by dividing the vibration source location area into spatial grids. Based on the spatial extent of the vibration source location area, a three-dimensional Cartesian coordinate system is established, with the origin set at the geometric center of the area. The x-axis is along the axial direction of the circuit breaker body, the y-axis is horizontal, and the z-axis is vertical. The vibration source location area is uniformly gridded within the three-dimensional coordinate system, with each grid cell being a cube and having a side length of 50 mm. The grid density is determined based on the size of the vibration source location area, ensuring that the area contains at least 8×8×8 grid cells. For irregularly shaped vibration source location areas, grid cells may be truncated at the area boundaries; truncated cells are assigned weight coefficients according to the proportion of their volume to the volume of a complete cell. The grid cells are numbered sequentially according to the ascending order of the x, y, and z coordinates. The center coordinates, cell number, and cell boundary coordinates of each grid cell are recorded to establish the mesh cell array. The mesh cell array provides a spatially discretized representation of the vibration source location area, providing a spatial reference for subsequent attenuation rate determination.

[0028] Spatial attenuation distribution is generated by measuring the attenuation rate of each cell in a grid cell array. Reference measuring points are selected within the vibration source location area, with the reference measuring point being the one with the highest energy peak. Using the reference measuring point as the excitation source location, the attenuation law of vibration energy propagating to surrounding grid cells is analyzed. The vibration energy peak value of the corresponding measuring point in each grid cell is extracted. If there is no actual measuring point in a certain grid cell, the energy values ​​of the surrounding measuring points are used for spatial interpolation estimation. The ratio of the energy at this location to the energy at the reference measuring point is calculated. The energy attenuation coefficient is defined as α = E_ref / E_i, where E_ref is the energy peak value of the reference measuring point, and E_i is the energy peak value at the corresponding location of the i-th grid cell. The energy attenuation rate of each grid cell is calculated as β = (α-1) / d, where d is the spatial distance from the center of the grid cell to the reference measuring point, calculated using Euclidean distance. The attenuation rate β reflects the degree of energy attenuation per unit distance during spatial propagation; a larger attenuation rate indicates faster energy attenuation. For grid cells without measuring point coverage, an inverse distance-weighted interpolation method is used to calculate the estimated value based on the attenuation rate of the surrounding cells. A spatial decay distribution is generated, which uses a grid cell array as a framework and records the decay rate value of each cell.

[0029] For example, the step of rate matching between the response time constant and the spatial decay distribution to form a matching location index includes: extracting the gradient of the spatial decay distribution to generate a decay rate gradient field; identifying contour regions in the decay rate gradient field based on the response time constant; performing multi-scale spatial analysis on the contour regions to form a candidate location set; and determining the matching location index using the gradient extrema points in the candidate location set.

[0030] Gradient extraction is performed on the spatial decay distribution to generate a decay rate gradient field. The decay rate values ​​for each grid cell are extracted from the spatial decay distribution, and the decay rate difference between adjacent grid cells is calculated. The gradient component of the decay rate in the x-direction is... Where β_(i,j,k) is the decay rate of the grid cell (i,j,k), and Δx is the spacing of the grid cells in the x-direction. Similarly, the gradient components in the y- and z-directions are calculated. and The gradient vector of the synthesized decay rate is given by a magnitude of . The gradient vector points in the direction of the fastest increase in attenuation rate, and the direction angle is calculated using the arctangent function of each component. The attenuation rate gradient vector is calculated for all mesh elements, forming an attenuation rate gradient field. This field describes the variation of the attenuation rate in three-dimensional space. Regions with larger gradient magnitudes correspond to locations where the attenuation rate changes drastically; these locations are typically interfaces between different materials or structures. In high-voltage circuit breakers, significant attenuation rate gradients are generated at the interfaces between metal and insulating components, and between normal and faulty components.

[0031] Identifying contour regions in a decay rate gradient field based on the response time constant. The response time constant is converted into a time-domain decay rate γ = 1 / τ, and this time-domain decay rate is used as the target feature value to perform contour searches in the three-dimensional spatial structure of the decay rate gradient field. Utilizing the spatial decay rate distribution information provided by the gradient field, the spatial decay rate β value is extracted at each grid cell location corresponding to the gradient field. Cells that meet the matching conditions are identified by comparing the numerical relationship between β and γ. In practical applications, a numerical tolerance is used to determine the contour conditions, with the tolerance set at 5% of the time-domain decay rate. Grid cells that satisfy |β-γ| / γ < 0.05 are included in the contour region. The gradient vector direction of the gradient field is used as a search guide, prioritizing the search for contour cells in regions with gentle gradient changes, because locations with drastic gradient changes are usually material or structural interfaces and are not suitable as contour references for fault location. For example, in a high-voltage circuit breaker buffer aging fault scenario, the vibration response time constant of the aging rubber pad is larger than that of the normal location (assuming τ = 0.02s, corresponding to a time-domain decay rate γ = 50s). -1 By searching along the spatial location of the aging region in the decay rate gradient field, it was found that the spatial decay rate β in this region is also approximately 50 s. -1 The area around the edge (where energy decay slows down due to aging) shows a small gradient value, indicating a gradual change in the decay rate. This confirms it as a uniform aging region rather than an abrupt boundary, thus identifying it as a contour region. This contour region corresponds to the spatial distribution range of buffer aging. The contour region may contain multiple discontinuous spatial sub-regions, each corresponding to a possible fault location. Boundary mesh cells of the contour region are extracted, and their coordinates and decay rate values ​​are recorded. The spatial distribution morphology of the contour region is analyzed; regularly shaped contour regions typically correspond to uniform fault distribution, while irregularly shaped contour regions correspond to localized fault defects.

[0032] Multi-scale spatial analysis of the contour regions was conducted to form a candidate location set. A multi-level scale analysis strategy, from coarse to fine, was adopted to progressively analyze the contour regions at different spatial resolutions. The first scale layer was a global coarse-scale analysis, in which the entire contour region was analyzed for spatial connectivity using a 100 mm × 100 mm coarse grid. The number of large spatially connected regions was identified, and a three-dimensional connected domain labeling algorithm was used to mark adjacent spatially connected units as the same connected domain. This determined the number and approximate location of major fault distribution areas at the large scale, and large regions containing more than 8 grid units were selected as the first-level candidates. The second scale layer was a local meso-scale analysis, in which the large regions selected in the first layer were refined using a 50 mm × 50 mm medium-scale grid. Secondary connected sub-regions were identified within each large region, and the geometric center coordinates and spatial compactness (the ratio of the sub-region volume to the volume of its circumscribed cube) of each sub-region were calculated. The geometric center coordinates were the average of the center coordinates of all grid units within the sub-region. Meso-scale sub-regions with a spatial compactness greater than 0.3 and more than 3 units were selected. The third-scale layer performs fine-grained confidence analysis. For each sub-region selected in the second layer, the original grid resolution (50 mm side length for each grid cell) is restored for precise location. The confidence level of each sub-region is evaluated based on a combination of factors, including the number of grid cells in the sub-region, the degree of matching between the decay rate and the target value, and the spatial compactness of the sub-region. Sub-regions with more cells, higher matching rates, and more compact spaces have higher confidence levels. Sub-regions with confidence levels exceeding a threshold of 50% of the maximum confidence level are selected. The geometric center coordinates of the selected sub-regions are used as candidate fault locations, forming a candidate location set.

[0033] The matching location index is determined using gradient extrema points in the candidate location set. The gradient magnitude of the decay rate gradient field is extracted around each candidate location in the candidate location set. Local maxima of the gradient magnitude are searched within ±2 grid cells around the candidate location using a sliding window comparison method, comparing the gradient magnitude of the central cell with the gradient magnitudes of its 26 neighboring cells (a 3×3×3 neighborhood excluding the center). Gradient maxima correspond to the locations where the decay rate changes most drastically; these locations are typically the boundary between the faulty component and the normal structure, or the fault center. The gradient maxima corresponding to each candidate location are compared, and the extremum point with the largest gradient magnitude is selected as the final matching location. If the gradient magnitudes of multiple extrema points are close, the extremum point closest to the geometric center of the contour region is selected. The cell number of the grid cell containing this gradient extremum point is extracted as the matching location index. In high-voltage circuit breakers, if the rubber pad of the buffer ages, the buffering performance will decrease. The vibration attenuation rate at the aging location will be lower than that at the normal location, forming a gradient peak at the edge of the aging area. The location of the aging buffer can be accurately determined by locating the gradient extreme point, and the positioning accuracy can reach the grid cell size level.

[0034] The spatial location coordinates of the fault are determined using a matching position index transformation. The center coordinates of the matching cells are obtained by querying the grid cell array based on the matching position index. The center coordinates of the matching cells represent the position of the fault source in the three-dimensional space of the vibration source location area. The three-dimensional coordinates are then converted to absolute coordinates in the circuit breaker body coordinate system. The conversion method involves adding the origin coordinates of the grid coordinate system to the relative coordinates of the matching cells. The conversion formulas are: X_abs = X_grid + X_origin, Y_abs = Y_grid + Y_origin, Z_abs = Z_grid + Z_origin, where X_abs, Y_abs, and Z_abs are the absolute coordinates of the fault in the circuit breaker body coordinate system, X_grid, Y_grid, and Z_grid are the relative coordinates of the fault in the grid coordinate system (the local coordinate system of the vibration source location area), and X_origin, Y_origin, and Z_origin are the position coordinates of the origin of the grid coordinate system in the circuit breaker body coordinate system. The location error range of the fault spatial location coordinates is calculated; the error range is equal to half the side length of the grid cell, i.e., 25 mm. The system generates spatial coordinates for fault location, including x, y, and z components and a location error range. In high-voltage circuit breakers, if a main bearing experiences a wear fault, the vibration response time constant at the wear location is relatively large, resulting in a smaller time-domain decay rate. By using rate matching, the fault can be located to the corresponding grid cell position of the main bearing housing, achieving precise spatial fault location.

[0035] Step S130: Establish a frequency-space mapping relationship based on the fault spatial location coordinates, extract the spectral distribution sequence of the region corresponding to the frequency-space mapping relationship, perform peak frequency time-series tracking on the spectral distribution sequence to generate a frequency drift index, and perform nonlinear component detection based on the frequency drift index and the spectral distribution sequence to form a nonlinear feature set.

[0036] In some embodiments, establishing a frequency-space mapping relationship based on the fault spatial location coordinates includes: extracting frequencies from the fault spatial location coordinates to generate dominant frequency components; performing multi-location frequency sampling based on the fault spatial location coordinates to establish a frequency-location sample set; performing correlation analysis between the dominant frequency components and the frequency-location sample set to form a mapping rule base; and structuring the mapping rule base to form a frequency-space mapping relationship.

[0037] Frequency extraction is performed on the fault spatial location coordinates to generate the dominant frequency component. Time-domain data of the vibration signal is extracted from the measurement point corresponding to the fault spatial location coordinates, and spectral analysis is performed on the vibration signal. A Fast Fourier Transform (FFT) is used to convert the time-domain signal to a frequency-domain signal to obtain the spectral distribution of the vibration signal. The frequency component with the largest amplitude is identified in the spectral distribution; this frequency component is the dominant frequency component. The dominant frequency component reflects the frequency characteristics where the vibration energy is most concentrated at the fault location, usually corresponding to the excitation frequency or resonant frequency of the fault. The frequency value, amplitude, and phase information of the dominant frequency component are calculated. The stability of the dominant frequency component is analyzed by calculating the mean and standard deviation of the dominant frequency through multiple measurements. A standard deviation less than 5% of the mean indicates that the dominant frequency is stable. In high-voltage circuit breakers, the dominant frequency generated by contact impact is typically in the range of 200 to 800 Hz, while the dominant frequency of the operating mechanism transmission system is typically in the range of 10 to 100 Hz. Different fault types correspond to different dominant frequency characteristics.

[0038] A frequency-location sample set is established based on multi-location frequency sampling using fault spatial location coordinates. Multiple sampling locations are selected within the surrounding space, centered on the fault spatial location coordinates. The sampling locations are selected using a radial expansion method, with multiple rings of sampling points set outwards from the fault center at intervals of 50 mm, 100 mm, and 150 mm. Each ring contains 8 to 12 sampling points, evenly distributed around the circumference. Vibration signals are extracted from the corresponding measurement points at each sampling location, and spectral analysis is performed to obtain the frequency characteristics of each location. The dominant frequency, secondary dominant frequency, and frequency amplitude information for each sampling location are recorded. The dominant frequency is the frequency component with the largest amplitude, and the secondary dominant frequency is the frequency component with the second largest amplitude. The spatial coordinates of each sampling location are recorded, including the distance and azimuth relative to the fault center. The frequency characteristics of the sampling locations are paired with the spatial coordinates to form frequency-location sample pairs. In high-voltage circuit breakers, the closer the location is to the fault center, the larger the amplitude of the dominant vibration frequency; the amplitude of the dominant frequency gradually decreases with increasing distance, but the frequency value remains essentially unchanged. Integrate the frequency-location sample pairs of all sampling locations to establish a frequency-location sample set.

[0039] A mapping rule base is formed by correlation analysis between the dominant frequency component and the frequency-location sample set. The frequency value of the dominant frequency component is extracted as the reference frequency, and samples close to the reference frequency are searched in the frequency-location sample set. The criterion for frequency proximity is that the relative error between the sample frequency f_sample and the reference frequency f_ref is less than 10%, calculated as |(f_sample-f_ref) / f_ref|<0.1. Samples meeting the criteria are selected, and their spatial distribution characteristics are statistically analyzed. The distance distribution range of the sample locations is calculated to identify the spatial range of the dominant frequency's influence. The azimuth distribution of the samples is analyzed to identify the propagation characteristics of the dominant frequency in different directions. For each different frequency value, the above analysis process is repeated to establish the correspondence between frequency values ​​and spatial distribution ranges. In high-voltage circuit breakers, the influence range of low-frequency vibrations (below 50Hz) is usually large, reaching more than 1 meter, while the influence range of high-frequency vibrations (above 500Hz) is smaller, usually within 0.5 meters. The correspondence between frequency values, spatial ranges, and propagation characteristics is organized to form mapping rules. The mapping rules corresponding to all frequency values ​​are integrated to establish a mapping rule base.

[0040] The mapping rule base is structured to form a frequency-space mapping relationship. The mapping rules in the rule base are categorized and organized according to frequency range into low-frequency, mid-frequency, and high-frequency rules. Low-frequency rules correspond to the frequency range of 0 to 100 Hz, mid-frequency rules to the frequency range of 100 to 1000 Hz, and high-frequency rules to the frequency range above 1000 Hz. Each type of rule is parameterized to establish a functional relationship between the frequency value f and the spatial range r. This functional relationship can be represented by an exponential decay model r = r0·e^(-k·f), where r0, k, and n are fitting parameters. The least squares method is used to fit the data in the mapping rule base to determine the functional parameters of each type of rule. The classification rules, parameterized functions, and fitting parameters are integrated to form a mathematical expression of the frequency-space mapping relationship. The frequency-space mapping relationship describes the quantitative relationship between vibration frequency and spatial location within a fault area and can be used to predict vibration frequency characteristics at any location or to infer the location of the vibration source based on frequency characteristics.

[0041] Extract the spectral distribution sequence of the region corresponding to the frequency-space mapping relationship. Based on the established frequency-space mapping relationship, determine the spatial range covered by the mapping relationship. Within the region corresponding to the mapping relationship, extract vibration signal spectral data at multiple time points in chronological order. The time sampling interval is set to the time it takes for the circuit breaker to complete one full opening and closing operation, typically 2 to 5 seconds. Perform a short-time Fourier transform on the vibration signal at each time point to obtain the spectral distribution at that time. The time window length of the short-time Fourier transform is set to 0.1 to 0.5 seconds, the window function is a Hanning window or a Hamming window, and the frequency resolution is set to 1 to 5 Hz. Extract the amplitude distribution data of the spectrum at each time point, dividing the frequency range into several frequency bands, with a bandwidth set to 10 to 50 Hz. Calculate the average amplitude within each frequency band to form a spectral distribution vector. For example, in the wear fault monitoring of high-voltage circuit breaker operating mechanisms, a frequency-space mapping region centered on the fault location coordinates covers the main bearing and surrounding transmission components of the operating mechanism. During 100 consecutive opening and closing operations, spectral data is extracted once for each operation, forming a spectral distribution vector sequence containing 100 time points. This sequence records the evolution trajectory of the vibration spectral characteristics of the fault area over time during continuous operation. The spectral distribution vectors at each time point are arranged in chronological order to form a spectral distribution sequence.

[0042] Peak frequency time-series tracking is performed on the spectral distribution sequence to generate a frequency drift index. Peak frequency data for each time moment is extracted from the spectral distribution sequence; the peak frequency is defined as the frequency component with the largest amplitude in the spectrum at that moment. The peak frequencies at each moment are arranged in chronological order to form a peak frequency time-series curve. The changing trend of the peak frequency time-series curve is analyzed to identify frequency drift characteristics. Frequency drift refers to the shift of peak frequency over time; the drift direction may be frequency increase or frequency decrease. The difference between peak frequencies at adjacent moments is calculated, with the frequency change rate being Δf / Δt, where Δf is the frequency difference and Δt is the time interval. The mean and standard deviation of the frequency change rate are statistically analyzed; the mean reflects the overall trend of frequency drift, and the standard deviation reflects the severity of frequency fluctuations. The cumulative frequency change is calculated by summing the frequency differences at each moment, reflecting the overall shift in frequency from the initial state to the current moment. In high-voltage circuit breakers, increased clearance due to wear of the operating mechanism gradually lowers the dominant vibration frequency, and decreased stiffness due to spring fatigue also causes frequency drift to lower frequencies. The frequency drift index is generated, which includes the peak frequency time series curve, the average frequency change rate, the cumulative frequency change, and the frequency drift direction, providing a quantitative basis for assessing the degree of fault degradation.

[0043] In some embodiments, the step of performing nonlinear component detection based on the frequency drift index and the spectral distribution sequence to form a nonlinear feature set includes: extracting trends based on the frequency drift index to generate drift trend features, the drift trend features including monotonic drift, periodic drift, and abrupt drift; performing correlation matching between the drift trend features and the spectral distribution sequence to obtain correlation strength; filtering nonlinear features based on the correlation strength, detecting harmonic components for the monotonic drift, detecting modulation components for the periodic drift, and detecting transient components for the abrupt drift; and integrating the harmonic components, the modulation components, and the transient components to form a nonlinear feature set.

[0044] Trend extraction is performed based on frequency drift indices to generate drift trend features, including monotonic drift, periodic drift, and abrupt drift. Peak frequency versus time curves are extracted from the frequency drift indices, and trend analysis is performed on these curves. Linear regression is used to fit the frequency time-series curves to obtain the linear trend of frequency change. If the slope of the fitted line is significantly non-zero (absolute slope greater than 0.1 Hz / s) and the goodness of fit R² is greater than 0.8, it is determined to be monotonic drift. Monotonic drift is divided into two types: monotonically increasing and monotonically decreasing. A positive slope indicates a monotonically increasing frequency, and a negative slope indicates a monotonically decreasing frequency. Periodicity is tested on the frequency time-series curves using autocorrelation analysis or spectral analysis to identify periodic components. If significant periodic components exist (periodic intensity greater than 0.6) and the period duration is within the range of 10 to 100 seconds, it is determined to be periodic drift. Periodic drift reflects the periodic fluctuation characteristics of the frequency, usually caused by periodic excitation or modulation of the system. Abrupt change points are detected in the frequency time-series curve using a sliding window variance test or a CUSUM cumulative sum test to identify the location of abrupt changes. If the difference between the frequency mean and the frequency standard deviation before and after a certain moment exceeds three times the frequency standard deviation, that moment is determined to be an abrupt change point. If the frequency time-series curve has an abrupt change point and the magnitude of the change is greater than 10% of the fundamental frequency, it is determined to be an abrupt drift. In high-voltage circuit breakers, mechanical wear causes a monotonically decreasing frequency, thermal expansion and contraction causes periodic frequency fluctuations, and component breakage or detachment causes abrupt frequency changes. Drift trend characteristics are generated, and drift type, drift parameters, and drift time characteristics are recorded.

[0045] The correlation strength is obtained by matching the drift trend characteristics with the spectral distribution sequence. Spectral distribution sequence data for the time period corresponding to the drift trend characteristics are extracted, and the synchronous change relationship between the spectral characteristics and the drift characteristics is analyzed. For monotonic drift, the correlation coefficient between the amplitude of the spectral peak and the frequency drift is calculated using the Pearson correlation coefficient. If the absolute value of the correlation coefficient is greater than 0.7, the amplitude of the spectral peak is considered strongly correlated with the frequency drift. For periodic drift, the existence of modulation frequency components corresponding to the drift period in the spectrum is analyzed. The frequency drift curve is subjected to Fourier transform to extract the periodic frequency of the drift, and frequency components near this frequency are searched in the spectral distribution sequence. If a modulation frequency component with a significant amplitude (amplitude greater than 20% of the peak amplitude) exists, the spectral modulation component is considered strongly correlated with the periodic drift. For example, during the fault evolution of a high-voltage circuit breaker, spring fatigue causes a monotonic drift in the dominant vibration frequency while the spectral energy distribution continuously changes, showing a synchronous evolution relationship. The periodic fluctuations in the dominant frequency caused by the periodic jamming of the operating mechanism correspond perfectly to the spectral modulation sideband structure, demonstrating a strong correlation between different types of drift characteristics and spectral characteristics. For abrupt drift, the difference in spectral structure before and after the abrupt change is analyzed. The similarity of the spectra before and after the abrupt change is calculated using the cosine similarity of the spectral vectors. If the similarity is less than 0.7, the spectrum is considered to have changed significantly at the abrupt change, indicating a strong correlation between abrupt drift and spectral change. The correlation strength is defined, expressed as a correlation coefficient or similarity value, ranging from 0 to 1, with larger values ​​indicating a stronger correlation.

[0046] Nonlinear features are screened based on correlation strength: harmonic components are detected for monotonically drifting, modulation components for periodically drifting, and transient components for abruptly drifting. Drift features with a correlation strength greater than 0.6 are selected, as these features are significantly correlated with spectral nonlinear components. For monotonically drifting features, harmonic components are detected in the spectral distribution sequence. The fundamental frequency f0 of the spectrum is extracted, and frequency components are searched at harmonic positions such as 2f0, 3f0, and 4f0. If a frequency peak with an amplitude exceeding the noise threshold (peak amplitude greater than 10% of the fundamental frequency amplitude) exists at a harmonic position, it is identified as a harmonic component. The presence of harmonic components indicates the existence of nonlinear stiffness or geometric nonlinearity in the system. For periodically drifting features, modulation components are detected in the spectral distribution sequence. The carrier frequency f_c and modulation frequency f_m in the spectrum are identified, and frequency components are searched at sideband positions such as f_c±f_m and f_c±2f_m. If a significant frequency peak exists at a sideband position, it is identified as a modulation component. The presence of modulation components indicates the existence of periodic excitation or parametric excitation in the system. For abrupt drift characteristics, transient components are detected in the spectral distribution sequence. Short-time spectra at the moment of abrupt change are extracted, and the broadband energy distribution characteristics within the spectrum are analyzed. If the spectral energy is uniformly distributed over a wide frequency range (bandwidth exceeding 100Hz) and its duration is less than 0.1 seconds, it is identified as a transient component. The presence of transient components indicates that the system has experienced transient events such as impacts or collisions. In high-voltage circuit breakers, monotonic frequency drops caused by contact wear are accompanied by enhanced harmonic components; periodic changes in the operating mechanism clearance lead to periodic drift and modulation components; and component loosening causes frequency abrupt changes and transient impact components.

[0047] The harmonic components, modulation components, and transient components are integrated to form a nonlinear feature set. Characteristic parameters for each type of nonlinear component are extracted. The characteristic parameters for the harmonic components include harmonic order, harmonic amplitude, and harmonic frequency. The characteristic parameters for the modulation components include carrier frequency, modulation frequency, modulation depth, and number of sidebands. The characteristic parameters for the transient components include transient peak amplitude, transient duration, and transient frequency band range. All nonlinear components and their characteristic parameters are organized chronologically, recording the occurrence time and evolution of each component. The correlations between nonlinear components are analyzed to identify situations where multiple nonlinear components coexist or are coupled. For example, in the evolution of a composite fault in a high-voltage circuit breaker, only a second harmonic component caused by slight contact wear is detected in the early stage; in the middle stage, the coexistence of modulation and harmonic components due to increased clearance in the operating mechanism occurs simultaneously; and in the later stage, transient impact components caused by component loosening are superimposed against the background of harmonic and modulation. The temporal evolution and coupling relationships of these three types of nonlinear components completely depict the entire process of the fault from its inception to its deterioration. A hierarchical structure for the nonlinear feature set is established. The first layer consists of the nonlinear component types (frequency doubling, modulation, transient), the second layer consists of the characteristic parameters of each component, and the third layer consists of the temporal evolution information of the components.

[0048] Step S140: Determine the mechanical degradation degree of the fault spatial positioning coordinates to identify the degradation influence area; extract the peak value of continuous action impact in the degradation influence area to generate a peak sequence; perform dispersion analysis on the peak sequence to form action consistency deviation; and integrate the action consistency deviation with the nonlinear feature set to generate a graded test density.

[0049] Specifically, the degree of mechanical degradation is determined by measuring the spatial location coordinates of the fault to identify the degradation-affected area. Three-dimensional coordinate data of the fault location is extracted from the spatial location coordinates output in step S120. A dedicated degradation test sensor is deployed at the fault location, or an existing vibration sensor is used to determine the degree of degradation. Degradation determination methods include vibration amplitude growth rate measurement, frequency offset measurement, and damping attenuation rate measurement. The vibration amplitude growth rate is defined as the ratio of the current vibration amplitude to the initial normal state vibration amplitude minus 1; a growth rate exceeding 50% indicates significant degradation. The frequency offset is defined as the difference between the current dominant frequency and the initial normal state dominant frequency; an offset exceeding 10% of the fundamental frequency indicates significant degradation. The damping attenuation rate is defined as the reciprocal of the vibration response time constant; a decrease in the attenuation rate exceeding 30% indicates deterioration of damping performance. The degree of mechanical deterioration is calculated by combining three indicators, using a weighted summation method: L = w1·R_amp + w2·R_freq + w3·R_damp, where L is the degree of mechanical deterioration, R_amp is the vibration amplitude growth rate, R_freq is the frequency deviation rate, R_damp is the damping attenuation rate change rate, and w1, w2, and w3 are weighting coefficients. In high-voltage circuit breakers, wear of the operating mechanism leads to increased vibration amplitude and decreased frequency, spring fatigue leads to decreased damping attenuation rate, and contact erosion leads to increased impact vibration amplitude. Deterioration levels are classified according to the numerical value: 0 to 0.3 is slight deterioration, 0.3 to 0.6 is moderate deterioration, and above 0.6 is severe deterioration. The radius of the deterioration influence area is determined by extending outward from the fault location: 200 mm for slight deterioration, 400 mm for moderate deterioration, and 600 mm for severe deterioration. Mark the spatial boundaries of the degradation impact area to form degradation impact area data.

[0050] Peak sequence is generated by continuously extracting impact peaks from the deterioration-affected area. Representative measuring points are selected within the deterioration-affected area, with the locations of highest deterioration and the boundaries of the affected area chosen as representative points. The high-voltage circuit breaker is controlled to perform continuous opening and closing actions, with the number of consecutive actions set to 10 to 50 and the action interval set to 5 to 10 seconds. The time-domain waveforms of vibration signals from representative measuring points are collected during each opening and closing action to identify the impact peak for each action. The impact peak is defined as the maximum amplitude of the vibration signal at the instant of the opening or closing action. The opening impact peak corresponds to the vibration peak at the instant the moving contact separates from the stationary contact, and the closing impact peak corresponds to the vibration peak at the instant the moving contact contacts the stationary contact. A peak detection algorithm is used to identify local maxima in the vibration signal, and maxima exceeding a threshold are selected as impact peaks. The threshold is set to 5 times the root mean square value of the background vibration amplitude. The impact peak values ​​and corresponding action numbers for each action are extracted, and the impact peak data are arranged in chronological order to form a peak sequence. The peak value sequence reflects the variation pattern of impact vibration amplitude during continuous operation, and the fluctuation characteristics of the peak values ​​in the sequence reflect the repeatability and consistency of mechanical action. In high-voltage circuit breakers, the impact peak value during continuous operation should remain stable under normal conditions, with the peak value fluctuation range being less than 15% of the average value. If mechanical deterioration exists, the peak value will show an increasing trend or the fluctuation range will expand.

[0051] In some embodiments, the step of performing dispersion analysis on the peak sequence to form an action consistency deviation includes: extracting peak statistical features from the peak sequence; evaluating the deviation between the peak statistical features and a preset benchmark peak to generate a deviation distribution; performing dispersion quantification processing based on the deviation distribution to determine a dispersion index; and using the dispersion index to convert and form an action consistency deviation.

[0052] Extracting peak statistical features from the peak sequence. This involves extracting statistical parameters describing the peak distribution characteristics from the peak sequence data. The mean, variance, and standard deviation of the peak sequence are obtained. The mean reflects the central location of the impact intensity, and the standard deviation reflects the dispersion of the peak data. Skewness and kurtosis are extracted from the peak sequence. Skewness reflects the symmetry of the peak distribution; positive skewness indicates a right-skewed distribution with high peak anomalies, while negative skewness indicates a left-skewed distribution with low peak anomalies. Kurtosis reflects the sharpness of the peak distribution; positive kurtosis indicates a sharper distribution than a normal distribution with more extreme values, while negative kurtosis indicates a flatter distribution than a normal distribution. In high-voltage circuit breakers, the peak distribution under normal conditions is close to a normal distribution with skewness and kurtosis close to 0. Mechanical degradation causes the peak distribution to exhibit skewness or thick tails. The maximum and minimum peak values ​​are extracted, and the peak range is the difference between the maximum and minimum values, reflecting the fluctuation range of the peak. The peak mean, standard deviation, skewness, kurtosis, and range are then processed to form the peak statistical feature data.

[0053] The deviation distribution is generated by evaluating the deviation between the peak statistical characteristics and a preset benchmark peak value. Preset benchmark peak value data is obtained; the benchmark peak value represents the impact peak statistical characteristics under normal operating conditions at the time of the circuit breaker's manufacture or during initial commissioning. The benchmark peak value includes the benchmark mean, benchmark standard deviation, and benchmark range. The deviation of the current peak statistical characteristics from the benchmark peak value is calculated. The mean deviation, standard deviation deviation, and range deviation are all calculated using a relative deviation method; a larger deviation value indicates a more severe deviation. For each peak data point in the peak sequence, its deviation from the benchmark mean is calculated as d_i = |p_i - μ_ref| / μ_ref, where d_i is the deviation of the i-th peak, p_i is the i-th peak data point, and μ_ref is the benchmark mean. The deviation values ​​of each peak data point are statistically analyzed, and then sorted and grouped according to the magnitude of the deviation. The deviation values ​​are divided into several intervals, with an interval width set to 0.1 or 0.2. The number of peak data points within each deviation interval is counted, forming a deviation distribution histogram. In high-voltage circuit breakers, under slight degradation, most peak deviations are less than 0.2, while under severe degradation, the peak deviation distribution range expands and abnormal peaks with deviations exceeding 0.5 appear. A deviation distribution is generated, which records the frequency and relative frequency of each deviation interval in the peak sequence.

[0054] For example, the step of determining the dispersion index by performing discrete quantification processing based on the deviation distribution includes: performing continuity analysis on the deviation distribution to establish a continuous feature library; identifying key discrete features from the continuous feature library, the key discrete features including deviation magnitude, deviation frequency, and continuous deviation length; standardizing the deviation magnitude to obtain a discrete benchmark value; and comprehensively evaluating the deviation frequency and the continuous deviation length using the discrete benchmark value to determine the dispersion index.

[0055] A continuity feature library is established by performing continuity analysis on the deviation distribution. The deviation distribution is arranged in chronological or action sequence order to form a deviation time-series curve. The continuity characteristics of the deviation time-series curve are analyzed to identify the persistent and discontinuous patterns of deviation. The run test method is used to identify continuous rising segments, continuous falling segments, and stationary segments in the deviation time-series curve. A continuous rising segment is defined as a sequence with at least three consecutive data points showing increasing deviation; a continuous falling segment is defined as a sequence with at least three consecutive data points showing decreasing deviation; and a stationary segment is defined as a sequence with deviation fluctuations less than 10% of the mean. The length, start position, and end position of each continuous segment are calculated. For continuous deviation segments, i.e., segments where the deviation consistently exceeds the threshold, the length and maximum deviation value of the continuous deviation segment are extracted. In high-voltage circuit breakers, occasional faults cause abnormal deviations in a single peak value, but the preceding and following peak values ​​are normal; persistent degradation causes multiple consecutive peak deviations to exceed the limit. The periodic characteristics of the deviation time-series curve are analyzed, and autocorrelation analysis is used to identify whether there are periodic fluctuations in the deviation. If significant periodic components are present, the period length and period amplitude are extracted. Organize the features of continuous segments, continuous deviation segments, and periodicity to establish a continuous feature database.

[0056] Key discrete features are identified from the continuous feature library, including deviation magnitude, deviation frequency, and consecutive deviation length. Deviation magnitude is extracted as the maximum deviation value in the peak sequence. The maximum deviation corresponds to the abnormal peak furthest from the normal state, reflecting the maximum range of dispersion. The number of peaks with deviation magnitudes exceeding each threshold level is calculated, with threshold levels set at 0.2, 0.4, and 0.6. The distribution of deviation magnitudes across different intervals is statistically analyzed to identify the main concentrated intervals of deviation magnitudes. Deviation frequency is extracted as the number of times peak data with deviations exceeding a threshold occurs. The threshold is set to twice the baseline deviation or 1.5 times the mean of the deviation distribution. The proportion of deviation frequency to the total number of actions is calculated; a higher proportion indicates more frequent deviations. The temporal distribution of deviations is analyzed to identify whether deviations are random or concentrated. Consecutive deviation length is extracted as the number of peaks contained in the longest segment where the deviation continuously exceeds the threshold. Consecutive deviation length reflects the persistence and stability of degradation; a longer length indicates more severe degradation and a longer duration. In high-voltage circuit breakers, increased gap size leads to occasional large deviations, but the frequency of these deviations is not high. Increased wear leads to an increase in the frequency of deviations and the occurrence of continuous deviations. The numerical values ​​and characteristics of deviation amplitude, frequency, and length of continuous deviations are compiled to form key discrete feature data.

[0057] The deviation magnitude is standardized to obtain discrete baseline values. Standardization maps the deviation magnitude to a uniform standardized range. Maximum value standardization is used, with the formula A_norm = (A - A_min) / (A_max - A_min), where A_norm is the standardized deviation magnitude, A is the original deviation magnitude, A_min is the minimum possible deviation magnitude (usually 0), and A_max is the maximum possible deviation magnitude or the maximum value in the sample. The standardized deviation magnitude ranges from 0 to 1; values ​​closer to 1 indicate more severe deviation. A benchmark-referenced standardization method is then used, comparing the deviation magnitude to a baseline deviation magnitude. The baseline deviation magnitude is the maximum allowable deviation under normal conditions, typically set at 20% of the baseline mean. A standardized value greater than 1 indicates deviation exceeding the normal range. Statistical characteristics of the standardized deviation magnitude are extracted, including the mean, maximum value, and proportion exceeding 1. A weighted average is then applied to the standardized deviation magnitude, using either an arithmetic mean or a geometric mean, to obtain discrete baseline values. The discrete baseline value comprehensively reflects the overall degree and severity of deviation of the peak sequence, with a numerical range of 0 to 1. In high-voltage circuit breakers, the discrete baseline value for slight degradation is below 0.3, the discrete baseline value for moderate degradation is between 0.3 and 0.7, and the discrete baseline value for severe degradation exceeds 0.7.

[0058] The dispersion index is determined by comprehensively evaluating the deviation frequency and continuous deviation length using discrete benchmark values. A comprehensive dispersion evaluation model is established, with the discrete benchmark value, deviation frequency, and continuous deviation length as inputs, and the dispersion index as output. The dispersion index D = w_B·f_B(B) + w_F·f_F(F) + w_L·f_L(L) is calculated using a weighted summation method, where D is the dispersion index, B is the discrete benchmark value, F is the deviation frequency, L is the continuous deviation length, w_B, w_F, and w_L are weighting coefficients, typically taken as w_B = 0.4, w_F = 0.3, and w_L = 0.3, and f_B, f_F, and f_L are normalization functions that map each parameter to the range of 0 to 1. When normalizing the deviation frequency, it is divided by the maximum possible deviation frequency, i.e., the total number of actions; when normalizing the continuous deviation length, it is divided by the maximum possible continuous deviation length. The fuzzy comprehensive evaluation method is used to assess dispersion. Membership functions for each parameter are established, and the comprehensive evaluation result is calculated based on the membership vector and weight matrix. In high-voltage circuit breakers, a dispersion index of 0 to 0.3 is considered good (high operational consistency), 0.3 to 0.6 is acceptable (acceptable operational consistency), 0.6 to 0.8 is a warning level (decreased operational consistency requiring attention), and above 0.8 is a fault level (severely deteriorated operational consistency requiring action). The dispersion index values ​​are calculated, and the dispersion level is determined based on the value range, generating a dispersion assessment report.

[0059] Action consistency deviation is derived using a dispersion index conversion method. A conversion relationship between the dispersion index and action consistency deviation is established, employing either a linear or nonlinear mapping function. The linear conversion formula is C = k·D + b, where C is the action consistency deviation, D is the dispersion index, and k and b are conversion coefficients. The nonlinear conversion formula uses an exponential or power function to map the numerical range of the dispersion index to the engineering evaluation scale of the action consistency deviation. The evaluation scale range for action consistency deviation is 0 to 100, where 0 represents completely consistent action with no deviation, and 100 represents completely inconsistent action with extremely large deviation. The parameters of the conversion function are determined based on engineering experience or historical data, such that a dispersion index of 0.1 corresponds to a deviation of 10, a dispersion index of 0.5 corresponds to a deviation of 50, and a dispersion index of 1 corresponds to a deviation of 100. In high-voltage circuit breakers, an action consistency deviation less than 20 is considered a good condition; a deviation of 20 to 50 indicates a condition requiring increased monitoring; a deviation of 50 to 80 indicates an abnormal condition requiring fault investigation; and a deviation exceeding 80 indicates a serious condition requiring shutdown and maintenance. The action consistency deviation value is calculated, generating the action consistency deviation evaluation result.

[0060] A graded test density is generated by integrating motion consistency deviation and nonlinear feature set. The numerical values ​​of the motion consistency deviation index and the nonlinear feature set data output from step S130 are extracted, and their correlation is analyzed. Motion consistency deviation reflects the stability of mechanical motion, while the nonlinear feature set reflects the degree of nonlinearity of the vibration system. Combining the two allows for a comprehensive assessment of the mechanical system's degradation state. A deviation-feature correlation matrix is ​​established, with matrix rows corresponding to the motion consistency deviation level (low, medium, high) and matrix columns corresponding to the nonlinear feature type (frequency harmonics, modulation, transient). The degradation state level is determined based on the combination of deviation level and feature type. Low deviation and no obvious nonlinear feature correspond to a slight degradation state; medium deviation or the presence of a single nonlinear feature corresponds to a moderate degradation state; and high deviation and the presence of multiple nonlinear features correspond to a severe degradation state. The test density requirement is determined based on the degradation state level: a test density of once a month for slight degradation, once a week for moderate degradation, and once a day for severe degradation. In high-voltage circuit breakers, newly commissioned equipment has a low testing density. Equipment that has been in operation for more than 5 years and shows moderate deterioration requires increased testing frequency, while equipment with severe deterioration requires immediate intensive monitoring. A graded testing density is generated, with data including the deterioration level, recommended testing frequency, and key testing areas, providing a basis for subsequent testing plan development.

[0061] Step S150: Determine the time period test unit based on the graded test density, perform tests on the time period test unit to establish a state graded test benchmark, and generate a mechanical state level determination based on the state graded test benchmark.

[0062] In some embodiments, determining the time period test unit based on the graded test density includes: establishing a grade-density correspondence based on the graded test density; dividing the time period using the grade-density correspondence to form a time period distribution; configuring test execution parameters based on the time period distribution to generate a parameter configuration table; and combining the time period distribution with the parameter configuration table to determine the time period test unit.

[0063] A grade-density correspondence was established based on graded test density. Deterioration state levels and corresponding test density information were extracted from the graded test density data. Deterioration state levels included three categories: slight deterioration, moderate deterioration, and severe deterioration. Test density was expressed as the frequency of tests per unit time. The test density for slight deterioration was once a month (monthly test frequency 1), for moderate deterioration it was once a week (weekly test frequency 1), and for severe deterioration it was once a day (daily test frequency 1). A correspondence table between deterioration levels and test densities was established, recording the name, level code, and corresponding test density value for each deterioration level. The quantitative relationship between test density and deterioration degree was analyzed. Test density increased with increasing deterioration degree; the more severe the deterioration, the higher the required monitoring frequency. In high-voltage circuit breakers, newly commissioned equipment had low deterioration degrees and sparse test densities, while older equipment showed increased deterioration degrees and correspondingly higher test densities. Severe deterioration required intensive monitoring to ensure equipment safety.

[0064] The monitoring cycle is divided into several test periods based on the level-density correspondence. The length of each test period is determined by the test density: 30 days for monthly testing, 7 days for weekly testing, and 1 day for daily testing. For long-term monitoring tasks, the period distribution is planned according to annual or semi-annual cycles. Within a monitoring cycle, the period division is dynamically adjusted based on changes in equipment degradation. If the equipment degradation is stable, the length of each period remains consistent. If the equipment degradation shows a worsening trend, the length of subsequent periods gradually shortens and the testing frequency increases. In high-voltage circuit breaker monitoring, the initial period length is 30 days; after 3 years of operation, as degradation increases, the period length shortens to 7 days; and after abnormal degradation signs are detected, the period length is further shortened to 1 day for emergency monitoring. The start and end times and period numbers of each period are marked to form the period distribution data. The period distribution is displayed as a time axis, showing the period division results within the monitoring cycle and intuitively reflecting the temporal distribution pattern of the testing frequency.

[0065] A parameter configuration table is generated based on the time period distribution to configure test execution parameters. For each time period in the distribution, corresponding test execution parameters are configured. These parameters include test action type, number of test actions, test data acquisition frequency, test sensor layout, and test data processing method. For time periods corresponding to slight degradation, the test action type is set to standard opening and closing test, with one test action, a data acquisition frequency of 20kHz, a basic sensor layout covering the main measurement points, and conventional feature extraction as the data processing method. For time periods corresponding to moderate degradation, the test action type is set to continuous opening and closing test, with 3 to 5 test actions, a data acquisition frequency of 50kHz, additional measurement points in the degradation-affected area added to the sensor layout, and trend analysis and dispersion analysis added to the data processing method. For time periods corresponding to severe degradation, the test action type is set to comprehensive performance test, with more than 10 test actions, a data acquisition frequency of 100kHz, full coverage monitoring achieved by the sensor layout, and deep feature extraction and multi-dimensional analysis as the data processing method. In high-voltage circuit breakers, rapid testing methods are used during periods of slight degradation to minimize the impact on operation, while detailed testing methods are used during periods of severe degradation to comprehensively assess the equipment status. The test execution parameters for each time period are compiled into a parameter configuration table, which records the parameter settings for each time period using the time period number as an index.

[0066] The time period distribution and parameter configuration table are combined to determine the time period test units. The start and end times and time period numbers of each time period in the time period distribution are extracted, and the corresponding test execution parameters are obtained from the parameter configuration table. The time information of each time period is associated and combined with the test execution parameters to form a complete time period test unit. The time period test unit includes the time period number, start and end times, degradation level, test action type, number of test actions, data acquisition frequency, sensor layout scheme, and data processing method. For each time period test unit, a specific test execution time point is planned. The test execution time point is selected in the middle of the time period. If the time period is 30 days long, the test execution time point is set on the 15th day; if the time period is 7 days long, the test execution time point is set on the 4th day; if the time period is 1 day long, the test execution time point is set during the low load period of that day. In high-voltage circuit breaker monitoring, the test execution time point should avoid peak load periods of the power grid, and nighttime or holidays should be prioritized for testing. The determined time period test units are arranged in chronological order to form a complete test plan.

[0067] Establish a state-level test benchmark for the time-period test unit. Perform test actions at designated test times according to the planning requirements of the time-period test unit. Control the high-voltage circuit breaker to perform opening and closing operations according to the specified action type and number of actions, and collect vibration signals, current signals, and mechanical travel signals during the test process. Extract features from the collected test data, including the peak amplitude, dominant frequency, and response time constant of the vibration signal; the peak current and current duration of the current signal; and the travel length and action time of the mechanical travel signal. Analyze the numerical distribution and trends of each characteristic parameter to identify the characteristic parameter ranges corresponding to each degradation level. For slight degradation, the peak vibration amplitude is within 1 to 1.5 times the normal value, the dominant frequency deviation is less than 10%, and the response time constant change is less than 20%. For moderate degradation, the peak vibration amplitude is within 1.5 to 2 times the normal value, the dominant frequency deviation is within 10% to 30%, and the response time constant change is within 20% to 50%. For severely deteriorated conditions, the peak vibration amplitude exceeds twice the normal value, the dominant frequency shift exceeds 30%, and the response time constant changes by more than 50%. In high-voltage circuit breakers, when the operating mechanism experiences slight wear, the vibration amplitude increases slightly but the dominant frequency remains relatively stable. With moderate wear, both vibration amplitude and frequency show significant changes. With severe wear, the vibration characteristics deteriorate significantly, and abnormal noise appears. A condition-level testing benchmark is established, which describes the criteria for each deterioration level in the form of characteristic parameter thresholds, including vibration amplitude thresholds, frequency shift thresholds, and time constant thresholds.

[0068] The mechanical condition level is determined based on the condition grading test benchmark. The characteristic parameter values ​​obtained from the current test are extracted, and each characteristic parameter is compared with the threshold values ​​in the condition grading test benchmark. For the vibration peak amplitude, its corresponding degradation level range is determined. If the vibration peak amplitude is within 1 to 1.5 times the normal value, the vibration amplitude level is determined to be slightly degraded. If the vibration peak amplitude is within 1.5 to 2 times the normal value, the vibration amplitude level is determined to be moderately degraded. If the vibration peak amplitude exceeds 2 times the normal value, the vibration amplitude level is determined to be severely degraded. The same determination method is used to determine the degradation level of the dominant frequency offset and the change in response time constant. Combining the degradation level determination results of each characteristic parameter, the overall mechanical condition level is determined using the most severe level principle. If any characteristic parameter is determined to be severely degraded, the overall mechanical condition is determined to be severely degraded. If all characteristic parameters are determined to be moderately degraded or less, but at least one is moderately degraded, the overall mechanical condition is determined to be moderately degraded. If all characteristic parameters are determined to be slightly deteriorated, the overall mechanical condition is determined to be slightly deteriorated. In high-voltage circuit breakers, if the vibration amplitude and frequency deviation are slightly deteriorated but the response time constant change reaches moderate deterioration, the overall condition is determined to be moderately deteriorated, requiring enhanced monitoring. Finally, a mechanical condition level assessment report is generated, which includes the values ​​of each characteristic parameter, the corresponding deterioration level, the overall mechanical condition level, and recommended maintenance measures.

[0069] To implement the vibration testing and condition monitoring method for in-operation high-voltage circuit breakers corresponding to the above method embodiments, in order to achieve the corresponding functions and technical effects. See also Figure 2 , Figure 2 This diagram illustrates a structural block diagram of a vibration testing and condition monitoring system 200 for in-operation high-voltage circuit breakers according to an embodiment of this application. For ease of explanation, only the parts relevant to this embodiment are shown. The vibration testing and condition monitoring system 200 for in-operation high-voltage circuit breakers provided in this application includes: Energy testing module 201 is used to acquire vibration signal data from multiple measuring points during the opening and closing operation of the high-voltage circuit breaker in operation, and to perform energy propagation timing test on the vibration signal data from the multiple measuring points to construct an energy flow topology map; The coordinate generation module 202 is used to determine the vibration source location area through energy ratio analysis in the energy flow topology map, perform impact response attenuation test on the vibration signal of the vibration source location area to generate a response time constant, and combine the response time constant with the vibration source location area to convert it into fault space location coordinates. Frequency optimization module 203 is used to establish a frequency-space mapping relationship based on the fault spatial location coordinates, extract the spectral distribution sequence of the region corresponding to the frequency-space mapping relationship, perform peak frequency time-series tracking on the spectral distribution sequence to generate a frequency drift index, and perform nonlinear component detection to form a nonlinear feature set based on the frequency drift index and the spectral distribution sequence. The degradation analysis module 204 is used to determine the mechanical degradation degree of the fault spatial positioning coordinates to identify the degradation influence area, extract the peak value of continuous action impact in the degradation influence area to generate a peak sequence, perform dispersion analysis on the peak sequence to form action consistency deviation, and fuse the action consistency deviation with the nonlinear feature set to generate a graded test density. The result determination module 205 is used to determine the time period test unit according to the graded test density, perform tests for the time period test unit to establish a state graded test benchmark, and generate a mechanical state level determination based on the state graded test benchmark.

[0070] The aforementioned vibration testing and condition monitoring system 200 for in-operation high-voltage circuit breakers can implement the vibration testing and condition monitoring method for in-operation high-voltage circuit breakers described in the above-described method embodiments. The options in the above method embodiments are also applicable to this embodiment and will not be detailed here. The remaining content of this application's embodiments can be referred to the content of the above method embodiments, and will not be repeated in this embodiment.

[0071] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A method for vibration test and condition monitoring of an in-service high voltage circuit breaker, characterized in that, The method comprises the following steps: acquiring vibration signal data of multiple measuring points during the opening and closing operation of a high-voltage circuit breaker, performing energy propagation timing test on the vibration signal data of the multiple measuring points to construct an energy flow topology diagram; determining a vibration source positioning area in the energy flow topology diagram through energy ratio analysis, performing impact response attenuation test on the vibration signal of the vibration source positioning area to generate a response time constant, and converting the response time constant and the vibration source positioning area into a fault space positioning coordinate; establishing a frequency-space mapping relationship based on the fault space positioning coordinate, extracting a frequency spectrum distribution sequence of a corresponding area of the frequency-space mapping relationship, performing peak frequency timing tracking on the frequency spectrum distribution sequence to generate a frequency drift index, performing nonlinear component detection based on the frequency drift index and the frequency spectrum distribution sequence to form a nonlinear feature set; measuring the mechanical degradation degree of the fault space positioning coordinate to determine a degradation influence area, performing continuous action impact peak extraction on the degradation influence area to generate a peak sequence, performing dispersion analysis on the peak sequence to form an action consistency deviation, and fusing the action consistency deviation and the nonlinear feature set to generate a hierarchical test density; determining a time period test unit according to the hierarchical test density, performing test to establish a state hierarchical test benchmark, and generating a mechanical state grade determination according to the state hierarchical test benchmark.

2. The method of claim 1, wherein, The method comprises the following steps: extracting node energy from the energy flow topology diagram to form an energy distribution sequence; identifying adjacent node energy ratios based on the energy distribution sequence; performing energy ratio analysis on the adjacent node energy ratios to identify an energy convergence path; determining a vibration source positioning area by backtracking using the energy convergence path.

3. The method of claim 1, wherein, The method comprises the following steps: dividing the vibration source positioning area into a space grid to establish a grid cell array; measuring the attenuation rate of each cell based on the grid cell array to generate a spatial attenuation distribution; performing rate matching on the response time constant and the spatial attenuation distribution to form a matching position index; determining a fault space positioning coordinate by conversion using the matching position index.

4. The method of claim 1, wherein, The method comprises the following steps: extracting a dominant frequency component from the fault space positioning coordinate to generate a dominant frequency component; performing multi-position frequency sampling based on the fault space positioning coordinate to establish a frequency-position sample set; performing correlation analysis on the dominant frequency component and the frequency-position sample set to form a mapping rule library; structuring the mapping rule library to form a frequency-space mapping relationship.

5. The method of claim 1, wherein, The method comprises the following steps: extracting a drift trend feature based on the frequency drift index to generate a drift trend feature, the drift trend feature including monotonic drift, periodic drift, and sudden drift; performing correlation matching on the drift trend feature and the frequency spectrum distribution sequence to obtain a correlation strength; Screening nonlinear features according to the correlation strength, detecting frequency doubling components for the monotonic drift, detecting modulation components for the periodic drift, and detecting transient components for the abrupt drift; Integrating the frequency doubling components, the modulation components, and the transient components to form a nonlinear feature set.

6. The method of claim 1, wherein, The dispersion analysis of the peak sequence forms the motion consistency deviation, including: Extracting peak statistical features from the peak sequence; Performing deviation degree evaluation on the peak statistical features and the preset reference peak to generate a deviation degree distribution; Performing dispersion quantification processing based on the deviation degree distribution to determine a dispersion index; Converting the dispersion index to form the motion consistency deviation.

7. The method of claim 1, wherein, The time period test unit is determined according to the hierarchical test density, including: Establishing a grade-density correspondence relationship based on the hierarchical test density; Dividing time periods to form a time period distribution using the grade-density correspondence relationship; Configuring test execution parameters based on the time period distribution to generate a parameter configuration table; Combining the time period distribution and the parameter configuration table to determine the time period test unit.

8. The method of claim 3, wherein, The rate matching of the response time constant and the spatial attenuation distribution forms a matching position index, including: Extracting gradients of the spatial attenuation distribution to generate an attenuation rate gradient field; Identifying contour line areas in the attenuation rate gradient field based on the response time constant; Performing multi-scale spatial analysis on the contour line areas to form a candidate positioning set; Determining the matching position index using gradient extreme points in the candidate positioning set.

9. The method of claim 6, wherein, The dispersion quantification processing based on the deviation degree distribution determines a dispersion index, including: Performing continuity analysis on the deviation degree distribution to establish a continuity feature library; Identifying key dispersion features from the continuity feature library, the key dispersion features including deviation amplitude, deviation frequency, and continuous deviation length; Performing standardization processing on the deviation amplitude to obtain a dispersion reference value; Performing comprehensive evaluation on the deviation frequency and the continuous deviation length through the dispersion reference value to determine the dispersion index.

10. A vibration testing and condition monitoring system for an operating high voltage circuit breaker, characterized in that, Including: An energy test module configured to acquire vibration signal data of multiple measurement points during the opening and closing actions of a high-voltage circuit breaker in operation, and to perform energy propagation time sequence testing on the vibration signal data of the multiple measurement points to construct an energy flow direction topology graph; A coordinate generation module configured to determine a vibration source positioning area in the energy flow direction topology graph by energy ratio analysis, to perform impact response attenuation testing on the vibration signals of the vibration source positioning area to generate a response time constant, and to convert the response time constant and the vibration source positioning area into a fault space positioning coordinate; A frequency optimization module configured to establish a frequency-space mapping relationship based on the fault space positioning coordinate, to extract a frequency spectrum distribution sequence of a corresponding area of the frequency-space mapping relationship, to perform peak frequency time sequence tracking on the frequency spectrum distribution sequence to generate a frequency drift index, and to perform nonlinear component detection based on the frequency drift index and the frequency spectrum distribution sequence to form a nonlinear feature set; The deterioration analysis module is used for determining the mechanical deterioration degree of the fault space positioning coordinates, determining a deterioration influence area, performing continuous action impact peak value extraction on the deterioration influence area to generate a peak value sequence, performing dispersion analysis on the peak value sequence to form an action consistency deviation, and fusing the action consistency deviation and the nonlinear feature set to generate a hierarchical test density. The result determination module is used for determining a time period test unit according to the hierarchical test density, performing test to establish a state hierarchical test benchmark, and generating a mechanical state grade determination according to the state hierarchical test benchmark.

Images