A method for identifying incorrect operation of high-voltage switchgear

Abstract

The present invention discloses a method for identifying misoperation of a high-voltage switchgear, which relates to the technical field of misoperation identification of a high-voltage switchgear. The present invention addresses sensor failure caused by dust coverage and misjudgment caused by electromagnetic interference. By adopting a physical conversion model from vibration spectrum energy to displacement, when dust obscures the displacement sensor, the cabinet vibration energy is used to reconstruct a highly robust displacement trajectory, breaking through the physical limitations of traditional optical / magnetic induction solutions. For electromagnetic interference scenarios, a load-adaptive ripple envelope extraction algorithm is designed, combining anti-interference feature quantization with frequency domain mask training to maintain stability of current features under strong pulses. Furthermore, the problem of time domain signal drift is resolved by aligning the vibration spectrum with the motor current phase. Ridge regression weight optimization is combined with regularization adaptation to improve the accuracy of virtual displacement mapping. The environmental perception module dynamically switches decision weights to achieve seamless protection in both dust and electromagnetic scenarios.

Description

Technical Field

[0001] The present invention relates to the technical field of high-voltage switch cabinet misoperation identification, in particular to a high-voltage switch cabinet misoperation identification method. Background Art

[0002] In high-dust operation scenarios such as mines and cement plants, high-voltage switchgear is exposed to coal ash and metal dust environments for a long time. These micron-sized particles continuously penetrate into the interior of the cabinet, forming conductive scale. The surfaces of key components such as displacement sensors and photoelectric sensors that existing anti-misoperation technologies rely on are easily covered by dust. For example, when dust accumulates to a thickness of 0.1mm on the sensing window of the trolley position detection device, the scattering rate of the infrared beam increases dramatically, making it impossible for the system to identify the actual displacement of the trolley. Even more troublesome is that after the dust in the humid mine is mixed with water vapor, it will form a solid surface at the interlocking mechanism shaft, causing mechanical jamming.

[0003] Non-contact monitoring solutions that have emerged in recent years, such as millimeter-wave radar and laser ranging, can reduce physical contact, but the diffuse dust in mines still causes severe signal attenuation. For example, the smart switch cabinet in publication number CN216043169U uses dual sound and light alarms. In actual applications, the voice reminders will be drowned out in strong noise environments, and the flashing indicator lights are not easily recognizable in dim tunnels.

[0004] When sensor readings drift, current solutions either directly trigger false alarms or discard data, resulting in protection failure. Although some adaptive filtering algorithms can be dynamically calibrated, they do not model the nonlinear relationship between dust concentration and signal attenuation. The key is that dust is regarded as a pure interference factor, while ignoring the fact that its physical deposition process itself carries equipment status information. Therefore, a method for identifying high-voltage switchgear misoperation is urgently needed to solve this problem. Summary of the Invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] The present invention provides a method for identifying misoperation of a high-voltage switchgear to solve the problems of sensor failure caused by dust in mines, frequent false alarms caused by electromagnetic interference, and the inability of existing solutions to take into account robustness in harsh environments.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] An embodiment of the present invention provides a method for identifying misoperation of a high-voltage switchgear, which includes:

[0009] Step S1, collecting multi-source sensor data, including cabinet door displacement signal, motor drive current waveform, cabinet vibration spectrum and ambient dust concentration data;

[0010] Step S2, dynamically compensating the multi-source sensor data to generate anti-interference operation time series data;

[0011] Step S3, extracting an operation intention feature vector, including a current ripple envelope feature analyzed from the motor drive current waveform, and a predetermined frequency band energy feature analyzed from the cabinet vibration spectrum;

[0012] Step S4: input the operation intention feature vector into the pre-trained operation behavior analysis model and output the misoperation risk level;

[0013] Step S5: performing hierarchical interlocking actions according to the risk level of misoperation.

[0014] As a preferred solution of the method for identifying misoperation of a high-voltage switchgear according to the present invention, the dynamic compensation process in step S2 includes:

[0015] Construct a nonlinear mapping relationship between dust concentration and displacement signal attenuation amplitude;

[0016] When a sudden change in the attenuation amplitude of the displacement signal is detected, the vibration reconstruction unit is activated.

[0017] As a preferred solution of the method for identifying misoperation of a high-voltage switchgear according to the present invention, in the vibration reconstruction unit of step S2, physical conversion of vibration spectrum energy to displacement is performed, including:

[0018] Energy aggregation of continuous frequency bands that have been relevantly screened:

[0019] in, Characteristic band aggregate energy, unit dB, Sampling time, unit s, The lower limit frequency of the characteristic frequency band, in Hz, Characteristic band upper limit frequency, unit Hz, Passband gain compensation coefficient, The frequency of the independent variable of integration, in Hz, Vibration spectrum energy density, unit: dB;

[0020] The aggregate energy is converted into a series linearly related to the mechanical displacement using an exponential weighting model:

[0021] in, Virtual displacement sequence, unit mm, Sampling time, unit s, Frequency component index, dimensionless, The total number of frequency components involved in the mapping, dimensionless, No. Component energy displacement conversion weight, unit , No. Component instantaneous energy density, in dB, No. Component center frequency, in Hz, Background noise power spectrum, in dB, Nonlinear energy index, dimensionless, empirical range 0.8-1.2;

[0022] The weight vector is solved by ridge regression in closed form, and the regularization factor is adaptively expressed as:

[0023] ,

[0024] ,

[0025] in, Weight vector ,unit , The sample energy matrix is ​​given by Composition, unit dB, Indicates the sampling sequence number corresponding to the time, represents the sample index, The transpose of Ridge regression regularization factor, dimensionless, The identity matrix, dimensionless, Actual displacement observation vector, unit mm, Regularization coefficient, empirical range 0.01-0.05, dimensionless, for The maximum singular value of , in dB.

[0026] As a preferred solution of the method for identifying misoperation of a high-voltage switchgear according to the present invention, the execution logic of the vibration reconstruction unit includes:

[0027] Extract characteristic frequency bands that are strongly correlated with mechanical displacement in the cabinet vibration spectrum;

[0028] Based on the phase information of the motor drive current waveform, the energy of the characteristic frequency band is aligned in the time domain;

[0029] Generate a virtual displacement sequence that is linearly related to the actual displacement of the cabinet door;

[0030] The characteristic frequency band that is strongly correlated with mechanical displacement is determined as follows:

[0031] Collect vibration spectrum samples of the switchgear trolley within the standard displacement range;

[0032] Calculate the Pearson correlation coefficient between the energy of each frequency component and the displacement distance;

[0033] Continuous frequency bands whose correlation coefficient exceeds a first threshold and whose signal-to-noise ratio exceeds a second threshold are screened.

[0034] As a preferred solution of the method for identifying misoperation of a high-voltage switchgear according to the present invention, the extraction of the current ripple envelope characteristics in step S3 includes:

[0035] Band-pass filtering is performed on the motor drive current waveform to separate the ripple component;

[0036] Calculate the rising edge slope of the ripple component and the number of zero crossings per unit time.

[0037] As a preferred solution of the method for identifying misoperation of a high-voltage switchgear according to the present invention, in step S3, the anti-interference characteristic calculation of the motor drive current ripple envelope is performed, including:

[0038] Construct a bandpass impulse response whose center frequency adjusts with load:

[0039] ,

[0040] in, For the Point filter impulse response, is the discrete time index, dimensionless, is the ripple center frequency, in Hz, is the half bandwidth, in Hz, is the sampling frequency, in Hz, is the Kaiser window coefficient;

[0041] At the filter output Apply Hilbert transform to obtain the analytical signal:

[0042] ,

[0043] in, is the ripple envelope amplitude, unit is A, is the current waveform after bandpass, unit is A, is the Hilbert transform operator, is an imaginary unit, dimensionless, is the sampling time, unit is s;

[0044] In the observation window The two anti-interference characteristics are calculated internally, and the calculation formula is:

[0045] ,

[0046] ,

[0047] in, is the normalized ratio of the envelope energy, dimensionless, is the envelope rising edge variation coefficient, dimensionless, is the observation window length, unit is s, For the Sampling time, unit s, is the sample index, dimensionless, For the Point envelope amplitude, unit A, is the median absolute deviation operator, is the median operator; the final anti-interference eigenvector is recorded as .

[0048] As a preferred solution of the method for identifying misoperation of a high-voltage switchgear according to the present invention, the training of the operation behavior analysis model in step S4 includes:

[0049] Injecting standard electromagnetic pulse interference samples into historical operational data;

[0050] Frequency domain masking technology is used to enhance the robustness of vibration spectrum features;

[0051] Optimizing convolutional neural network decision boundaries through adversarial training.

[0052] As a preferred solution of the method for identifying misoperation of a high-voltage switchgear according to the present invention, the method further includes:

[0053] Environmental anti-interference mode switch, when the environmental electromagnetic field strength exceeds the standard:

[0054] Suppress the current ripple envelope feature in the operation intention feature vector;

[0055] Increase the decision weight of the cabinet vibration spectrum characteristics to the preset value;

[0056] The triggering conditions for switching the environmental anti-interference mode include:

[0057] Real-time monitoring of the spectrum distribution of environmental electromagnetic field intensity;

[0058] When the energy in the 15kHz-100kHz frequency band continuously exceeds the standard, it is determined to be an electromagnetic interference scenario;

[0059] If the duration of the exceeding standard exceeds the set value, the frequency domain feature priority decision-making mechanism will be activated.

[0060] As a preferred solution of the method for identifying misoperation of a high-voltage switchgear according to the present invention, the hierarchical interlocking action in step S5 includes:

[0061] When the risk level of misoperation is level one, the electromagnetic locking device of the cabinet door is triggered;

[0062] When the misoperation risk level is level 2, the circuit breaker is opened and the grounding switch is closed simultaneously;

[0063] Generates a security event log containing the hash value of the feature vector.

[0064] As a preferred solution of the method for identifying misoperation of a high-voltage switchgear according to the present invention, the method for determining the characteristic frequency band includes:

[0065] Analyze the vibration spectrum distribution during the movement of the switch cabinet trolley;

[0066] Select the frequency band where the correlation coefficient between energy change and displacement distance is greater than 0.8;

[0067] Exclude frequency bands dominated by ambient background noise.

[0068] The beneficial effects of the present invention are as follows: the present invention addresses the problems of sensor failure caused by dust coverage and misjudgment caused by electromagnetic interference. Through the physical conversion model of vibration spectrum energy to displacement, when dust obscures the displacement sensor, the cabinet vibration energy is used to reconstruct a highly robust displacement trajectory, breaking through the physical limitations of traditional optical / magnetic induction solutions; for electromagnetic interference scenarios, a load-adaptive ripple envelope extraction algorithm is designed, combined with anti-interference feature quantization and frequency domain mask training, so that the current characteristics remain stable under strong pulses; in addition, the time domain signal drift problem is solved by aligning the vibration spectrum with the motor current phase; ridge regression weight optimization is combined with regularization adaptation to improve the accuracy of virtual displacement mapping; the environmental perception module dynamically switches decision weights to achieve seamless protection in dust-electromagnetic dual scenarios; and through millisecond-level risk classification interlocking, electromagnetic locking / circuit breaker opening, passive alarms are converted to active interception, significantly reducing the risk of serious accidents.

[0069] The present invention reuses existing industrial sensors and edge computing units without the need to modify the switch cabinet structure. It is particularly suitable for deployment in harsh environments such as mines and substations, providing intelligent protection for the safe operation of high-voltage equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0071] Figure 1 This is a flow chart of the method for identifying misoperation of a high-voltage switchgear in Example 1. DETAILED DESCRIPTION

[0072] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0073] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0074] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0075] Example 1, with reference to Figure 1 This embodiment provides a method for identifying misoperation of a high-voltage switchgear, comprising the following steps:

[0076] Step S1, collecting multi-source sensor data, including cabinet door displacement signal, motor drive current waveform, cabinet vibration spectrum and ambient dust concentration data;

[0077] Step S2, dynamically compensating and processing multi-source sensor data to generate anti-interference operation time series data;

[0078] The motion compensation process in step S2 includes:

[0079] Construct a nonlinear mapping relationship between dust concentration and displacement signal attenuation amplitude;

[0080] When a sudden change in the attenuation amplitude of the displacement signal is detected, the vibration reconstruction unit is activated;

[0081] In the vibration reconstruction unit of step S2, the physical conversion of vibration spectrum energy to displacement is performed, including:

[0082] Energy aggregation of continuous frequency bands that have been relevantly screened:

[0083] ,

[0084] in, Characteristic band aggregate energy, unit dB, Sampling time, unit s, The lower limit frequency of the characteristic frequency band, in Hz, Characteristic band upper limit frequency, unit Hz, Passband gain compensation coefficient, The frequency of the independent variable of integration, in Hz, Vibration spectrum energy density, in dB. Integration compresses the high-dimensional spectrum into a single energy trajectory, dissipating noise within the integration window. Gain compensation corrects for differences in the sensor chain. The resulting trajectory changes monotonically with door travel and remains smooth even in pulse interference scenarios. The impact of dust obstruction on the energy curve is negligible, laying a stable benchmark for subsequent displacement mapping.

[0085] The aggregate energy is converted into a series linearly related to the mechanical displacement using an exponential weighting model:

[0086] ,

[0087] in, Virtual displacement sequence, unit mm, Sampling time, unit s, Frequency component index, dimensionless, The total number of frequency components involved in the mapping, dimensionless, No. Component energy displacement conversion weight, unit , No. Component instantaneous energy density, in dB, No. Component center frequency, in Hz, Background noise power spectrum, in dB, The nonlinear energy index is dimensionless and has an empirical range of 0.8-1.2. The exponentially weighted model maintains slope consistency in both small travel and acceleration phases, addressing the issue of energy-displacement relationship bending under different operating conditions. The noise spectrum is subtracted component by component, and the virtual displacement trajectory is smooth for small signal fluctuations. The lifting and falling phases are well symmetrical, allowing for direct input into the interlocking threshold determination without the need for additional filtering.

[0088] The weight vector is solved by ridge regression in closed form, and the regularization factor is adaptively expressed as:

[0089] ,

[0090] ,

[0091] in, Weight vector ,unit , The sample energy matrix is ​​given by Composition, unit dB, Indicates the sampling sequence number corresponding to the time, represents the sample index, The transpose of Ridge regression regularization factor, dimensionless, The identity matrix, dimensionless, Actual displacement observation vector, unit mm, Regularization coefficient, empirical range 0.01-0.05, dimensionless, for The maximum singular value of , in dB; the closed-form solution avoids oscillations in the iterative process and maintains stable convergence when the sample is sparse or the frequency components are highly correlated; the regularization factor is adaptively adjusted with the matrix spectral norm, which not only limits excessive weights but also prevents underfitting; after calibration, the correlation between the virtual displacement curve and the laser measured curve is greatly improved, and the fluctuations in repeated experiments are controlled within an acceptable range, providing reliable input for the multi-source fusion module;

[0092] The execution logic of the vibration reconstruction unit includes:

[0093] Extract characteristic frequency bands that are strongly correlated with mechanical displacement in the cabinet vibration spectrum;

[0094] Based on the phase information of the motor drive current waveform, the energy of the characteristic frequency band is aligned in the time domain;

[0095] Generate a virtual displacement sequence that is linearly related to the actual displacement of the cabinet door;

[0096] The characteristic frequency band that is strongly correlated with mechanical displacement is determined as follows:

[0097] Collect vibration spectrum samples of the switchgear trolley within the standard displacement range;

[0098] Calculate the Pearson correlation coefficient between the energy of each frequency component and the displacement distance;

[0099] Screening continuous frequency bands whose correlation coefficient exceeds a first threshold and whose signal-to-noise ratio exceeds a second threshold;

[0100] Methods for determining characteristic frequency bands include:

[0101] Analyze the vibration spectrum distribution during the movement of the switch cabinet trolley;

[0102] Select the frequency band where the correlation coefficient between energy change and displacement distance is greater than 0.8;

[0103] Eliminate frequency bands dominated by ambient background noise;

[0104] Step S3, extracting an operation intention feature vector, including a current ripple envelope feature analyzed from the motor drive current waveform, and a predetermined frequency band energy feature analyzed from the cabinet vibration spectrum;

[0105] The extraction of the current ripple envelope characteristics in step S3 includes:

[0106] Band-pass filtering is performed on the motor drive current waveform to separate the ripple component;

[0107] Calculate the rising edge slope of the ripple component and the number of zero crossings per unit time;

[0108] In step S3, the anti-interference characteristic calculation of the motor drive current ripple envelope is performed, including:

[0109] Construct a bandpass impulse response whose center frequency adjusts with load:

[0110] ,

[0111] in, For the Point filter impulse response, is the discrete time index, dimensionless, is the ripple center frequency, in Hz, is the half bandwidth, in Hz, is the sampling frequency, in Hz, is the Kaiser window coefficient; the adaptive center frequency moves synchronously with load current drift to avoid filtering out actual ripple components; the Kaiser window sidelobes decay quickly, suppressing adjacent harmonic leakage, and the half-bandwidth dynamically converges based on the statistical results of the ripple energy distribution, maintaining a stable passband gain during the rising period of electromagnetic interference; the filtered waveform retains only the current commutation ripple, with the fundamental and spike pulses attenuated by more than 30dB, eliminating the need for additional notch compensation in subsequent envelope solutions;

[0112] At the filter output Apply Hilbert transform to obtain the analytical signal:

[0113] ,

[0114] in, is the ripple envelope amplitude, unit is A, is the current waveform after bandpass, unit is A, is the Hilbert transform operator, is an imaginary unit, dimensionless, is the sampling time, in seconds. The analytical signal method provides the instantaneous amplitude and phase at once, avoiding dual-channel errors caused by splitting the upper and lower envelopes. Symmetrical filtering maintains the Hilbert phase shift at 90°, and boundary effects are suppressed by zero-phase extension. The envelope trajectory is highly coupled to the mechanical impact cycle, and sporadic pulses are converted into isolated cusps, which facilitate statistical filtering and improve the credibility of subsequent robust indicators.

[0115] In the observation window The two anti-interference characteristics are calculated internally, and the calculation formula is:

[0116] ,

[0117] ,

[0118] in, is the normalized ratio of the envelope energy, dimensionless, is the envelope rising edge variation coefficient, dimensionless, is the observation window length, unit is s, For the Sampling time, unit s, is the sample index, dimensionless, For the Point envelope amplitude, unit A, is the median absolute deviation operator, is the median operator; the final anti-interference eigenvector is recorded as ; here Comparing the mean square amplitude with the arithmetic mean amplitude can quickly identify the ripple distortion caused by abnormal armature commutation. MAD is used to characterize rising edge jitter, which is naturally robust to pulse interference and occasional glitches. Both indices are dimensionless, making them easy to normalize across devices.

[0119] Step S4: input the operation intention feature vector into the pre-trained operation behavior analysis model and output the misoperation risk level;

[0120] The training of the operation behavior analysis model in step S4 includes:

[0121] Injecting standard electromagnetic pulse interference samples into historical operational data;

[0122] Frequency domain masking technology is used to enhance the robustness of vibration spectrum features;

[0123] Optimizing convolutional neural network decision boundaries through adversarial training;

[0124] Step S5, performing hierarchical interlocking actions according to the risk level of misoperation;

[0125] The hierarchical interlocking action in step S5 includes:

[0126] When the risk level of misoperation is level one, the electromagnetic locking device of the cabinet door is triggered;

[0127] When the misoperation risk level is level 2, the circuit breaker is opened and the grounding switch is closed simultaneously;

[0128] Generate a security event log containing a hash value of the feature vector;

[0129] The method further includes:

[0130] Environmental anti-interference mode switch, when the environmental electromagnetic field strength exceeds the standard:

[0131] Suppress the current ripple envelope feature in the operation intention feature vector;

[0132] Increase the decision weight of the cabinet vibration spectrum characteristics to the preset value;

[0133] The trigger conditions for switching to the environmental anti-interference mode include:

[0134] Real-time monitoring of the spectrum distribution of environmental electromagnetic field intensity;

[0135] When the energy in the 15kHz-100kHz frequency band continuously exceeds the standard, it is determined to be an electromagnetic interference scenario;

[0136] If the duration of the exceeding standard exceeds the set value, the frequency domain feature priority decision-making mechanism will be activated.

[0137] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for identifying misoperation of a high-voltage switchgear, characterized in that: include, Step S1, collecting multi-source sensor data, including cabinet door displacement signal, motor drive current waveform, cabinet vibration spectrum and ambient dust concentration data; Step S2, dynamically compensating the multi-source sensor data to generate anti-interference operation time series data; Step S3, extracting an operation intention feature vector, including a current ripple envelope feature analyzed from the motor drive current waveform, and a predetermined frequency band energy feature analyzed from the cabinet vibration spectrum; Step S4: input the operation intention feature vector into the pre-trained operation behavior analysis model and output the misoperation risk level; Step S5, performing hierarchical interlocking actions according to the risk level of misoperation; The motion compensation process in step S2 includes: Construct a nonlinear mapping relationship between dust concentration and displacement signal attenuation amplitude; When a sudden change in the attenuation amplitude of the displacement signal is detected, the vibration reconstruction unit is activated; In the vibration reconstruction unit of step S2, the physical conversion of vibration spectrum energy to displacement is performed, including: Energy aggregation of continuous frequency bands that have been relevantly screened: Among them, E b (t) represents the aggregate energy of the characteristic frequency band, unit is dB, t represents the sampling time, unit is s, f l Indicates the lower limit frequency of the characteristic frequency band, unit Hz, f u Indicates the upper limit frequency of the characteristic band, unit Hz, G(f) indicates the passband gain compensation coefficient, f indicates the integral independent variable frequency, unit Hz, E v (f,t) represents the vibration spectrum energy density, in dB; The aggregate energy is converted into a series linearly related to the mechanical displacement using an exponential weighting model: Among them, x virt (t) represents the virtual displacement sequence, unit is mm, t represents the sampling time, unit is s, i represents the frequency component index, dimensionless, N represents the total number of frequency components involved in the mapping, dimensionless, w i Represents the energy displacement conversion weight of the i-th component, unit: mm·dB -α , E v (f i ,t) represents the instantaneous energy density of the i-th component, in dB, f i Indicates the center frequency of the i-th component, in Hz, S n (f i ) represents the background noise power spectrum, in dB, α represents the nonlinear energy index, dimensionless, with an empirical range of 0.8-1.2; The weight vector is solved by ridge regression in closed form, and the regularization factor is adaptively expressed.

2. A method for identifying misoperation of a high-voltage switchgear according to claim 1, characterized in that: The weight vector is solved by using ridge regression closed form and the steps of giving an adaptive expression of the regularization factor include: λ=κσ max (E), Where w represents the weight vector Unit: mm·dB -α , E represents the sample energy matrix, [E v (f i ,t j )-S n (f i )] α Composition, unit dB, t j Indicates the sampling sequence number corresponding to the time, j indicates the sample index, represents the transpose of E, λ represents the ridge regression regularization factor, dimensionless, I represents the N×N identity matrix, dimensionless, x represents the actual displacement observation vector, unit mm, κ represents the regularization scale coefficient, empirical range 0.01-0.05, dimensionless, σ max (E) is the maximum singular value of E, in dB.

3. A method for identifying misoperation of a high-voltage switchgear according to claim 1, characterized in that: The execution logic of the vibration reconstruction unit includes: Extract characteristic frequency bands that are strongly correlated with mechanical displacement in the cabinet vibration spectrum; Based on the phase information of the motor drive current waveform, the energy of the characteristic frequency band is aligned in the time domain; Generate a virtual displacement sequence that is linearly related to the actual displacement of the cabinet door; The characteristic frequency band that is strongly correlated with mechanical displacement is determined as follows: Collect vibration spectrum samples of the switchgear trolley within the standard displacement range; Calculate the Pearson correlation coefficient between the energy of each frequency component and the displacement distance; Continuous frequency bands whose correlation coefficient exceeds a first threshold and whose signal-to-noise ratio exceeds a second threshold are screened.

4. A method for identifying misoperation of a high-voltage switchgear according to claim 1, characterized in that: The extraction of the current ripple envelope characteristics in step S3 includes: Band-pass filtering is performed on the motor drive current waveform to separate the ripple component; Calculate the rising edge slope of the ripple component and the number of zero crossings per unit time.

5. A method for identifying misoperation of a high-voltage switchgear according to claim 4, characterized in that: In step S3, the anti-interference characteristic calculation of the motor drive current ripple envelope is performed, including: Construct a bandpass impulse response whose center frequency adjusts with load: Among them, h n is the impulse response of the filter at the nth point, n is the discrete time index, dimensionless, f r is the ripple center frequency, in Hz, Δf is the half bandwidth, in Hz, f s is the sampling frequency in Hz, w(n) is the Kaiser window coefficient; At the filter output i b Perform Hilbert transform on (t) to obtain the analytical signal: Where A(t) is the ripple envelope amplitude, unit is A, i b (t) is the current waveform after bandpass, unit is A, is the Hilbert transform operator, j is the imaginary unit, dimensionless, t is the sampling time, unit is s; Calculate two anti-interference features within the observation window [0, T], and the calculation formula is: Where η1 is the envelope energy normalization ratio, dimensionless, η2 is the envelope rising edge variation coefficient, dimensionless, T is the observation window length, unit s, t k is the kth sampling moment, unit s, k is the sample index, dimensionless, A(t k ) is the envelope amplitude of the kth point, unit A, MAD(·) is the median absolute deviation operator, median(·) is the median operator; the final anti-interference eigenvector is recorded as 6. A method for identifying misoperation of a high-voltage switchgear according to claim 1, characterized in that: The training of the operation behavior analysis model in step S4 includes: Injecting standard electromagnetic pulse interference samples into historical operational data; Frequency domain masking technology is used to enhance the robustness of vibration spectrum features; Optimizing convolutional neural network decision boundaries through adversarial training.

7. A method for identifying misoperation of a high-voltage switchgear according to claim 1, characterized in that: Also includes: Environmental anti-interference mode switch, when the environmental electromagnetic field strength exceeds the standard: Suppress the current ripple envelope feature in the operation intention feature vector; Increase the decision weight of the cabinet vibration spectrum characteristics to the preset value; The triggering conditions for switching the environmental anti-interference mode include: Real-time monitoring of the spectrum distribution of environmental electromagnetic field intensity; When the energy in the 15kHz-100kHz frequency band continuously exceeds the standard, it is determined to be an electromagnetic interference scenario; If the duration of the exceeding standard exceeds the set value, the frequency domain feature priority decision-making mechanism will be activated.

8. A method for identifying misoperation of a high-voltage switchgear according to claim 1, characterized in that: The hierarchical interlocking action in step S5 includes: When the risk level of misoperation is level one, the electromagnetic locking device of the cabinet door is triggered; When the misoperation risk level is level 2, the circuit breaker is opened and the grounding switch is closed simultaneously; Generates a security event log containing the hash value of the feature vector.

9. A method for identifying misoperation of a high-voltage switchgear according to claim 3, characterized in that: The method for determining the characteristic frequency band includes: Analyze the vibration spectrum distribution during the movement of the switch cabinet trolley; Select the frequency band where the correlation coefficient between energy change and displacement distance is greater than 0.8; Exclude frequency bands dominated by ambient background noise.

Images