A non-contact online detection system and method for asynchronous circuit breaker tripping.

Abstract

This invention relates to the field of high-voltage circuit breaker condition monitoring technology, and more particularly to a non-contact online detection system and method for asynchronous operation of circuit breaker contacts. The technical solution includes a multi-channel electro-optic sensing unit, a synchronous signal conditioning module, a high-speed synchronous acquisition module, a timing feature extraction module, an asynchronous operation evaluation module, and a host computer module. The multi-channel electro-optic sensing unit contains at least two independent sensing probes, each correspondingly positioned on the outside of each contact of the multi-contact circuit breaker, for non-contact sensing of transient electric field signals generated during the opening and closing of each contact, and converting them into analog electrical signals for output. This invention achieves online detection without power outages through non-contact sensing, accurately locating and directly reflecting the actual operation of the contact through precise timing abrupt changes, ensuring sampling reliability with dual-trigger redundancy, and supporting preventative proactive maintenance by incorporating historical trend analysis.

Description

Technical Field

[0001] This invention relates to the field of high-voltage circuit breaker condition detection technology, and in particular to a non-contact online detection system and method for detecting asynchronous operation of circuit breaker contacts. Background Technology

[0002] Multi-break circuit breakers are the core switching equipment in UHV converter stations. The synchronicity of the opening and closing actions of each break directly affects the breaking performance and equipment lifespan. When the synchronicity of the break actions exceeds the standard, the first closing / first opening break will be subjected to excessively high transient voltage and arc erosion, which will aggravate insulation deterioration and contact wear. In severe cases, it can lead to major accidents such as breaking failure and equipment breakdown.

[0003] Currently, the detection of asynchronous circuit breaker operation mainly relies on power outage maintenance, which involves measuring the time difference of operation by adding displacement sensors or limit switches to the transmission mechanism. However, this method has drawbacks such as requiring power outage work, long detection cycles, and inability to reflect the operating characteristics under actual operating conditions. Existing online monitoring solutions mostly focus on indirect parameters such as coil current and energy storage motor status, which cannot directly reflect the actual operating sequence of the circuit breaker. On the other hand, electrical quantity monitoring solutions based on voltage transformers have problems such as narrow bandwidth, inability to capture transient changes, and the need to connect to the primary high-voltage circuit.

[0004] Existing non-contact overvoltage monitoring technologies based on BGO crystals mainly focus on overvoltage amplitude acquisition, waveform noise reduction, and type identification. They only treat transient voltage as the monitoring object and fail to explore the correlation between the temporal characteristics of transient signals and the mechanical operating characteristics of circuit breakers, thus failing to achieve online diagnosis of asynchrony in circuit breaker operation. Therefore, this application proposes a non-contact online detection system and method for asynchrony in circuit breaker operation. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a non-contact online detection system and method for the asynchronicity of circuit breaker operation. The system captures transient electric field changes during the operation of each circuit breaker using a BGO non-contact sensor array, and combines high-precision synchronous sampling and time-series feature extraction algorithms to achieve online quantitative detection and status assessment of the asynchronicity of circuit breaker operation.

[0006] In a first aspect, this application provides a non-contact online detection system for the asynchronous operation of a multi-break circuit breaker, including a multi-channel electro-optic sensing unit, a synchronous signal conditioning module, a high-speed synchronous acquisition module, a timing feature extraction module, an asynchronous operation evaluation module, and a host computer module.

[0007] The multi-channel electro-optic sensing unit includes at least two independent sensing probes, which are arranged one-to-one on the outside of each break of the multi-break circuit breaker. They are used to non-contactly sense the transient electric field signals generated during the opening and closing of each break and convert them into analog electrical signals for output.

[0008] The synchronization signal conditioning module is connected to the multi-channel electro-optic sensing unit and is used to condition the analog electrical signals of each channel and output a standardized transient signal.

[0009] The high-speed synchronous acquisition module is connected to the synchronous signal conditioning module and is used to receive the circuit breaker operation trigger signal, start multi-channel synchronous sampling, and output discrete sampling data of each channel.

[0010] The time-series feature extraction module is connected to the high-speed synchronous acquisition module and is used to calculate the sampled data of each channel based on the signal mutation detection algorithm to locate the start mutation time and peak time of the transient signal at each break point.

[0011] The asynchronous action evaluation module is connected to the timing feature extraction module and is used to calculate the difference between the timing feature parameters of each break point action. Combined with the preset asynchronous action threshold, it outputs the asynchronous action level of the break point opening and closing and the mechanical state evaluation results.

[0012] The host computer module is communicatively connected to the action asynchrony evaluation module and is used to display the action timing waveforms, asynchrony time differences, and status evaluation results of each break point.

[0013] Optionally, the multi-channel electro-optic sensing unit senses changes in the spatial electric field based on the electro-optic effect. Each sensing probe has a built-in light source, electro-optic crystal, and photoelectric conversion element, and outputs an electrical signal corresponding to the electric field strength.

[0014] Optionally, the high-speed synchronous acquisition module supports two modes: external triggering and internal threshold triggering. After sampling is triggered, it automatically records the complete waveform data of the first time period before the trigger and the second time period after the trigger.

[0015] Optionally, the signal mutation detection algorithm is a sliding window energy mutation algorithm: the energy of the signal within the window is calculated using a sliding window of fixed length. When the energy sum increment of multiple consecutive windows exceeds a preset threshold, the starting point of the first window exceeding the threshold is determined as the moment of the transient signal mutation. The time corresponding to the extreme value of the signal is searched within a preset time window after the moment of the mutation and determined as the peak time.

[0016] Optionally, the asynchronous action assessment module has built-in thresholds for asynchronous opening and asynchronous closing. When the maximum time difference between the disconnection points is less than the warning threshold, it is judged as normal; when it is between the warning threshold and the alarm threshold, it is judged as a state of attention; and when it is greater than the alarm threshold, it is judged as an abnormal state.

[0017] Optionally, the operation asynchrony assessment module also includes a built-in historical trend analysis unit, which stores the time difference data of different periods of circuit breaker operation, fits the changing trend of different period deviations, and outputs preventive maintenance suggestions.

[0018] Secondly, this application provides a non-contact online detection method for the asynchronicity of tripping action of multi-break circuit breakers, applied to the non-contact online detection system for the asynchronicity of tripping action of multi-break circuit breakers described in the first aspect, comprising the following steps:

[0019] Step 1: Install sensor probes on the outside of each break of the multi-break circuit breaker to complete the physical fixing and signal wiring of the multi-channel sensor unit;

[0020] Step 2: Receive the circuit breaker opening and closing operation trigger signal, start multi-channel synchronous high-speed sampling, and collect the transient electric field signal during the operation of each break point;

[0021] Step 3: Preprocess the sampled data from each channel to obtain standardized transient waveforms;

[0022] Step 4: Using a signal mutation detection algorithm, calculate the standardized waveforms of each channel to locate the start and peak times of the transient signal at each break point.

[0023] Step 5: Calculate the maximum difference between the initial abrupt change time and the maximum difference between the peak time for all break points, and use this as the time difference between opening and closing asynchrony;

[0024] Step 6: Compare the time difference between different periods with the preset threshold to determine the different levels of the fracture action, and output the mechanical condition assessment results and early warning information.

[0025] Step 7: Upload the transient waveform, time difference between different periods, and state assessment results to the host computer for display and storage.

[0026] Optionally, in step two, the trigger signal adopts a dual-trigger redundancy mechanism: it prioritizes receiving external trigger signals; when the external trigger signal fails or is missing, it automatically enables internal amplitude triggering; and when the signal amplitude exceeds a preset amplitude threshold, it initiates high-speed sampling.

[0027] Optionally, in step four, the location of the initial mutation time also includes secondary verification: taking a data segment before and after the initially located mutation time, calculating the signal change rate, and confirming the mutation time as valid when the signal change rate meets the preset conditions; otherwise, determining it as interference and searching again.

[0028] Optionally, step six also includes a degradation assessment based on historical data: retrieve time difference data of different periods of the same type of operation within a historical time period, fit the trend of deviation change, and output an early maintenance warning when it is predicted that the difference between different periods will exceed the alarm threshold within a future predetermined time period.

[0029] Compared with the prior art, this application includes at least one of the following beneficial technical effects:

[0030] Detection can be achieved simply by placing a BGO sensor probe on the outside of the fracture. There is no need to connect to a primary high-voltage circuit or install a contact-type travel sensor. The original structure of the equipment is not changed, and the detection can be completed during normal operation of the equipment.

[0031] Based on the sliding window energy mutation algorithm to capture the initial mutation point of the transient electric field, combined with nanosecond-level synchronous sampling, the actual action time of each break is accurately restored, which can more realistically reflect the mechanical state of the break than indirect parameters such as coil current.

[0032] It supports external operation command triggering and internal amplitude change triggering, avoiding missed detection caused by single trigger failure, and ensuring that transient timing waveforms can be completely recorded for each circuit breaker operation.

[0033] It not only enables single-time different-period level determination, but also uses historical data to fit the deterioration trend, predicts the mechanical performance degradation in advance, and provides data basis for preventive maintenance.

[0034] In summary, this invention achieves online detection without power outages through non-contact sensing, accurately locates and directly reflects the actual action of the break point through time-series abrupt changes, ensures sampling reliability with dual-trigger redundancy, and supports preventive proactive maintenance by combining historical trend analysis. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the overall system structure of the present invention;

[0036] Figure 2 This is a flowchart of the detection method of the present invention;

[0037] Figure 3 A schematic diagram showing the arrangement of sensor probes for a four-break circuit breaker;

[0038] Reference numerals in the attached diagram: 1. BGO crystal probe head one; 2. BGO crystal probe head two; 3. BGO crystal probe head three; 4. BGO crystal probe head four; 5. Circuit breaker housing; 6. Break point; 7. Junction box; 8. High-voltage terminal block. Detailed Implementation

[0039] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0040] Example 1

[0041] This embodiment provides a non-contact online detection system for the asynchrony of the tripping of multi-break circuit breakers.

[0042] See Figure 1 The system includes a multi-channel BGO electro-optic sensing unit, a synchronous signal conditioning module, a high-speed synchronous acquisition module, a timing feature extraction module, an action asynchrony evaluation module, and a host computer module.

[0043] The multi-channel BGO electro-optic sensing unit comprises at least two independent BGO sensing probes, with the number of probes corresponding one-to-one with the number of breaks in the multi-break circuit breaker. Each probe is positioned on the outside of each break in the multi-break circuit breaker, for example, on the outer wall of the break housing, directly opposite the break gap. No direct contact is required between the probe and the break, nor is it connected to the primary high-voltage circuit. It senses the transient electric field signals generated during the opening or closing operations of each break in a non-contact manner. Each probe converts the sensed transient electric field signals into analog voltage signals and outputs them to the back-end processing module via signal transmission cables.

[0044] In the multi-channel BGO electro-optic sensing unit, each BGO sensing probe adopts a reflective integrated structure. Specifically, each sensing probe integrates a 1550nm DFB laser source, a polarization-maintaining fiber, an optical circulator, a BGO crystal, and a reflector. The DFB laser source generates linearly polarized light, which is transmitted through the polarization-maintaining fiber to the optical circulator, and then guided by the circulator to the BGO crystal. The BGO crystal is positioned at the front end of the probe and close to the electric field region of the fracture. When the linearly polarized light passes through the BGO crystal, it is modulated by the electric field of the fracture space, generating a birefringence phase difference based on the Pockels effect, thus changing the polarization state. The modulated beam is reflected by the reflector located at the rear end of the BGO crystal, returns along the original optical path, passes through the BGO crystal again, and is guided by the optical circulator to the photodetector. The photodetector converts the received optical signal into a voltage signal output, the amplitude of which is proportional to the electric field strength at the fracture. Because each probe has an independent light source and photoelectric conversion link, the channels are electrically isolated, effectively avoiding crosstalk between channels.

[0045] The synchronous signal conditioning module is connected to the multi-channel BGO electro-optic sensing unit to receive the analog voltage signals output from each channel and condition them. Specifically, the conditioning operations include differential amplification and bandpass filtering. Differential amplification is used to suppress common-mode electromagnetic interference in the field and increase the signal amplitude to the range of the acquisition module; bandpass filtering is used to filter out ultra-low frequency temperature drift components and ultra-high frequency noise components, retaining the signal components within the effective frequency band of transient changes, and outputting a standardized transient signal to the acquisition module. The channels of the synchronous signal conditioning module maintain strict synchronization to ensure that there is no relative time delay distortion before the signal from each channel enters the acquisition module.

[0046] The high-speed synchronous acquisition module is connected to the synchronous signal conditioning module to receive circuit breaker operation trigger signals and initiate multi-channel synchronous sampling upon receiving the trigger signal. The high-speed synchronous acquisition module performs synchronous analog-to-digital conversion on the standardized transient signals of each channel, ensuring that the sampling synchronization error between channels is no greater than 100ns. After sampling, the high-speed synchronous acquisition module outputs the discrete sampling data corresponding to each channel to the timing feature extraction module. In this embodiment, the high-speed synchronous acquisition module supports two triggering modes: external triggering and internal threshold triggering. In external triggering mode, the high-speed synchronous acquisition module receives the opening and closing command signals of the circuit breaker operating mechanism (e.g., the auxiliary contact signals of the circuit breaker opening and closing control circuit), and immediately initiates multi-channel synchronous sampling upon receiving an opening or closing command. In internal threshold triggering mode, the high-speed synchronous acquisition module monitors the signal amplitude of each channel in real time, and automatically initiates synchronous sampling when the signal amplitude of any channel suddenly exceeds a preset amplitude threshold. Both triggering modes can be enabled simultaneously, providing redundancy. After sampling is triggered, the high-speed synchronous acquisition module automatically records the complete waveform data for 10ms before and 90ms after the trigger, for a total time window of 100ms. This time window is long enough to completely cover the entire transient process of the circuit breaker operation.

[0047] The timing feature extraction module is connected to the high-speed synchronous acquisition module to receive discrete sampled data from each channel. Based on a sliding window energy mutation algorithm, it calculates the starting mutation time and peak time of the transient signal corresponding to each break point. Specifically, for each channel, the timing feature extraction module calculates the signal energy within the window segment by segment using a sliding window, monitors energy mutation points to locate the transient start time, and then searches for the peak time within a preset time window after the start time.

[0048] The asynchrony assessment module is connected to the timing feature extraction module. It receives the start and peak times of each break point, calculates the maximum difference between the start and peak times of all breaks, and calculates the maximum difference between the peak times of all breaks. The asynchrony assessment module stores preset asynchrony thresholds. It compares the calculated maximum difference with the preset thresholds, determines the asynchrony level of the break point opening and closing based on the comparison result, and outputs the corresponding mechanical condition assessment result.

[0049] The host computer module communicates with the action asynchrony evaluation module to receive the action timing waveforms, time differences and status evaluation results of each break point output by the action asynchrony evaluation module, and performs visualization and data storage for operation and maintenance personnel to view and trace.

[0050] As a preferred embodiment of this invention, the sliding window energy mutation algorithm used in the time-series feature extraction module is implemented as follows: For the sampling sequence of any channel, a window of fixed length (e.g., 20 sampling points) slides along the time axis with a step size of one sampling point. The energy sum (i.e., the sum of the squares of the amplitudes of each sampling point within the window) of the signal within each window is calculated one by one. Then, the energy sum increment between adjacent windows is calculated. When the energy sum increment of three consecutive windows exceeds a preset energy mutation threshold, the time corresponding to the starting sampling point of the first window in these three windows is determined as the starting mutation time of the transient signal of that channel. After determining the starting mutation time, the maximum value of the channel signal is searched within a 500μs time window after the starting mutation time, and the time corresponding to the maximum value is determined as the peak time of the transient signal of that channel.

[0051] It should be noted that the asynchrony assessment module has built-in thresholds for both opening and closing. During closing operations, the warning threshold is set to 2ms and the alarm threshold to 5ms; during opening operations, the warning threshold is set to 3ms and the alarm threshold to 6ms. For the current operation type (closing or opening), the asynchrony assessment module compares the maximum difference between the initial abrupt change times of the disconnection points with the corresponding warning and alarm thresholds: when the maximum difference is less than the warning threshold, the asynchrony level is determined to be normal; when the maximum difference is greater than or equal to the warning threshold and less than or equal to the alarm threshold, the asynchrony level is determined to be a warning state, and a warning message is output; when the maximum difference is greater than the alarm threshold, the asynchrony level is determined to be an abnormal state, and an alarm is triggered. For reference, the maximum difference at the peak time can be used as an auxiliary criterion; when the initial abrupt change time determination result is in a critical state, the peak time difference is used as an auxiliary reference for comprehensive judgment.

[0052] In addition, the asynchronous operation assessment module also includes a built-in historical trend analysis unit. This unit stores the time difference data between different periods of circuit breaker operations (including the maximum difference at the start time and / or the maximum difference at the peak time of each operation), and performs trend fitting on the stored asynchronous deviation data based on the time series to obtain the trend of the asynchronous deviation with the number of operations or time. Based on this trend, the historical trend analysis unit can predict the development direction of the asynchronous deviation in the future and output preventive maintenance suggestions in advance when it predicts that the asynchronous deviation will exceed the alarm threshold.

[0053] For method steps not described in detail in the above system embodiments, please refer to the method embodiments below. The specific workflow and control logic of each module in the system embodiments can also be implemented by referring to the corresponding steps in the method embodiments.

[0054] Example 2

[0055] This embodiment provides a non-contact online detection method for the asynchronicity of multi-break circuit breaker contacts. This method is based on the non-contact online detection system for the asynchronicity of multi-break circuit breaker contacts described in the above system embodiment. See also... Figure 2 The method includes the following steps.

[0056] Step 1: Sensor Probe Installation. Install BGO sensor probes on the outer side of each break of the multi-break circuit breaker. Specifically, fix one BGO sensor probe to the outer housing of each break, with the number of sensor probes matching the number of breaks. During installation, adjust the probe orientation so that the sensitive axis of the BGO crystal inside the probe is parallel to the electric field direction of the break gap to obtain maximum electric field coupling sensitivity. After installation, complete the signal cable connections between each sensor probe and the signal conditioning module, as well as the communication and control wiring between the modules. After the system is powered on, perform a self-test to confirm that the signals of each channel are normal and the communication links are unobstructed.

[0057] Step 2: Synchronous Trigger Sampling. When the circuit breaker performs opening and closing operations, the system receives the circuit breaker opening and closing operation trigger signal and, in response to this signal, initiates multi-channel synchronous high-speed sampling to acquire the transient electric field signals generated at each break point during the operation. Specifically, after receiving the trigger signal, the high-speed synchronous acquisition module synchronously performs analog-to-digital conversion on each channel at a sampling rate of not less than 10 MS / s, with a synchronization error between channels not exceeding 100 ns, acquiring discrete sampling data for each channel.

[0058] In this embodiment, a dual-trigger redundancy mechanism is adopted for the trigger signal. Specifically, the high-speed synchronous acquisition module prioritizes receiving the opening and closing command signal of the circuit breaker operating circuit as the external trigger signal, which can be obtained from the auxiliary contacts of the circuit breaker opening and closing control circuit. When the external trigger signal is normal, the system uses the external trigger signal as the sampling start reference; when the external trigger signal fails (e.g., the auxiliary contact signal is missing or abnormal), the system automatically activates the internal amplitude trigger mechanism, and the high-speed synchronous acquisition module monitors the signal amplitude of each channel in real time. When the signal amplitude of any channel exceeds a preset amplitude threshold (e.g., exceeding 1.2 times the steady-state amplitude of the power frequency), it is determined that an operation event has occurred, and high-speed sampling is automatically started. The dual-trigger redundancy mechanism ensures that every circuit breaker operation can be reliably recorded, avoiding missed detections caused by the failure of a single trigger method.

[0059] After each sampling trigger, the system automatically records the complete waveform data for the first time period before the trigger (e.g., 10ms before the trigger) and the second time period after the trigger (e.g., 90ms after the trigger), forming a complete operation transient record containing the waveforms before and after the trigger.

[0060] Step 3: Waveform Preprocessing. The sampled data from each channel is preprocessed to remove baseline drift and high-frequency noise, resulting in a standardized transient waveform. Specifically, preprocessing includes baseline correction and smoothing filtering. Baseline correction removes DC component offsets caused by temperature drift or circuit bias, restoring the signal to zero baseline. Smoothing filtering removes high-frequency spikes and random noise while retaining the effective high-frequency components of transient changes. The filtered signal is the standardized transient waveform, used for subsequent time-series feature extraction.

[0061] Step 4: Timing Feature Point Localization. A sliding window energy mutation algorithm is used to calculate the initial mutation time and peak time of the transient signal at each break point for each channel's standardized waveform. Specifically, for each channel, a fixed-length sliding window slides along the time axis, calculating the energy sum within each window and monitoring the energy sum increment between adjacent windows. When the energy sum increment of multiple consecutive windows (e.g., three consecutive windows) exceeds a preset energy mutation threshold, the time corresponding to the starting point of the first window exceeding the threshold is determined as the initial mutation time of the transient signal for that channel. After determining the initial mutation time, within a preset time window (e.g., 500 μs) following the initial mutation time, the time corresponding to the maximum value of the channel signal is searched, and this time is determined as the peak time of the transient signal for that channel.

[0062] In this embodiment, the location of the initial mutation moment also includes a secondary verification step. Specifically, after initially locating the mutation moment using the sliding window energy mutation algorithm, a data segment (e.g., 1ms data before and after) is taken before and after the initially located mutation moment, and the slope change rate of the signal within the data segment is calculated. The calculated slope change rate is compared with the slope during steady-state operation. If the slope change rate exceeds a preset multiple (e.g., more than 5 times the steady-state slope), the initially located mutation moment is confirmed to be valid and is used as the final initial mutation moment; if the slope change rate does not exceed the preset multiple, the initially located mutation moment is determined to be caused by interference signals, is discarded, and the mutation search is performed again.

[0063] Step 5: Calculation of Time Difference Between Opening and Closing Phases. Calculate the maximum difference in the initial abrupt change times of all break points, and the maximum difference in the peak times of all break points. Use these two differences as the time differences between opening and closing phases. Specifically, the maximum difference in initial times is the difference between the maximum and minimum values ​​among the initial abrupt change times of all break points; the maximum difference in peak times is the difference between the maximum and minimum values ​​among the peak times of all break points. The maximum difference in initial times serves as the primary criterion for evaluating the asynchronicity of break point actions, while the maximum difference in peak times serves as a secondary reference.

[0064] Step Six: Status Assessment and Early Warning. The calculated time difference between different periods (i.e., the maximum difference at the start time) is compared with a preset threshold. Based on the comparison result, the different period level of the circuit breaker action is determined, and the mechanical status assessment result and early warning information are output. Specifically, the corresponding threshold group is selected for comparison according to the current operation type (closing operation or opening operation). For closing operations, the early warning threshold is 2ms, and the alarm threshold is 5ms; for opening operations, the early warning threshold is 3ms, and the alarm threshold is 6ms. When the maximum difference at the start time is less than the early warning threshold, the different period level is determined to be normal; when the maximum difference at the start time is greater than or equal to the early warning threshold and less than or equal to the alarm threshold, the different period level is determined to be a state of alert, and an early warning information is output; when the maximum difference at the start time is greater than the alarm threshold, the different period level is determined to be an abnormal state, and an alarm is triggered.

[0065] In this embodiment, the condition assessment step also includes a degradation assessment based on historical data. Specifically, the system retrieves historical data on the time differences between different periods for the same type of operation (both closing or both opening) of the same circuit breaker within the past 6 months. Using operation time or number of operations as independent variables and the time difference between different periods as the dependent variable, a linear regression method is used to fit the growth trend of the deviation between different periods over time. Based on the fitted trend line, the system predicts the trend of the difference between different periods over the next 3 months. When the predicted value will exceed the alarm threshold within 3 months, a preventative maintenance warning is issued in advance, prompting maintenance personnel to arrange a maintenance plan.

[0066] Step 7: Result Display and Storage. Upload the transient waveforms of each fracture point, the marked initial abrupt change time and peak time, the calculated time differences between different periods, the determined period levels, the mechanical condition assessment results, and the early warning information to the host computer module. The host computer module synchronously displays each waveform and marks the characteristic moments, showing the values ​​and status levels of different periods. Simultaneously, all data is automatically stored in the historical database for subsequent traceability queries and trend analysis.

[0067] Those skilled in the art will understand that the physical installation and fixing of the sensor probe in step one above is usually completed once before the system is put into operation, while steps two through seven are automatically executed in a loop each time the circuit breaker is operated. For circuit breakers with the system already installed, each operation can start directly from step two.

[0068] Example 3

[0069] This embodiment takes the four-break AC filter bank circuit breaker of an ultra-high voltage converter station as the application object, and describes the implementation of the present invention in detail with specific parameters.

[0070] System Overall Architecture: The system consists of four independent BGO sensor probes, a four-channel synchronous signal conditioning module, a four-channel high-speed synchronous acquisition card, an embedded main control unit, and a host computer. The four BGO sensor probes are respectively installed on the outer side of the casing of the four breaks of the circuit breaker. (See attached diagram.) Figure 3 The circuit breaker housing is directly opposite the break gap and connected to the synchronous signal conditioning module via a shielded signal line. After conditioning, it is synchronously sampled by a four-channel high-speed synchronous acquisition card. The acquired data is sent to the embedded main control unit to complete timing calculation and status evaluation. Finally, the evaluation results are uploaded to the host computer for display. The four independent BGO sensor probes include BGO sensor probe 1, BGO sensor probe 2, BGO sensor probe 3, and BGO sensor probe 4. BGO sensor probe 1, BGO sensor probe 2, BGO sensor probe 3, and BGO sensor probe 4 are respectively connected to the circuit breaker housing 5. The four circuit breaker housings 5 ​​are arranged sequentially from top to bottom. One of the circuit breaker housings 5 ​​is connected to the break 6. BGO sensor probe 1 is connected to the junction box 7 after the signal line is connected. The uppermost circuit breaker housing 5 is connected to the high-voltage terminal block 8. The lowest circuit breaker housing 5 is connected to the post insulator, base, and wiring wire in sequence.

[0071] BGO sensor probe design: Each probe employs an integrated reflective package, internally integrating a 1550nm DFB laser source, a three-port optical circulator, a polarization-maintaining fiber, a 3mm×3mm×9mm BGO crystal, and a reflective diaphragm. During operation, linearly polarized light emitted from the DFB laser source enters the optical circulator via the polarization-maintaining fiber and is guided to the BGO crystal. As the linearly polarized light passes through the BGO crystal, it undergoes birefringence and phase difference due to the electric field modulation at the fracture point; the polarization state changes with the electric field strength. The beam is reflected back along its original path after reaching the reflective diaphragm at the end of the BGO crystal, passes through the BGO crystal again, and is guided by the optical circulator to the built-in photodetector. The photodetector converts the optical signal carrying the electric field information into a voltage signal output, with the output amplitude linearly corresponding to the electric field strength at the fracture point.

[0072] Synchronous signal conditioning module: Each channel's conditioning circuit sequentially includes a differential amplification stage and a bandpass filter stage. The differential amplification stage uses a high-performance differential amplifier to suppress common-mode interference in strong electromagnetic environments, with a fixed gain of 10. The bandpass filter stage uses an active bandpass filter with a passband of 100Hz~20MHz. The lower cutoff frequency of 100Hz is used to filter out ultra-low frequency components such as temperature drift, and the upper cutoff frequency of 20MHz is used to filter out high-frequency noise while completely preserving high-frequency components of transient changes, ensuring that timing characteristics are not distorted or delayed during the filtering process.

[0073] High-speed synchronous acquisition module: Employs a four-channel synchronous ADC chip, with a single-channel sampling rate set at 10MS / s. The synchronization error between channels is less than 50ns, meeting the nanosecond-level timing positioning accuracy requirements. The acquisition module is configured with dual trigger logic: In external trigger mode, it connects to the auxiliary contact signal of the circuit breaker's opening and closing control circuit; sampling is immediately initiated when the auxiliary contact actuates. In internal trigger mode, the acquisition module monitors the signal amplitude of each channel in real time; sampling is automatically triggered when the sampled value of any channel exceeds 1.2 times the steady-state peak value of the power frequency. After each trigger, waveform data for 100ms (10ms before trigger and 90ms after trigger) is recorded, fully covering the entire transient process of circuit breaker operation.

[0074] Temporal feature extraction algorithm: The sliding window energy mutation algorithm is run within the embedded main control unit. The specific execution steps are as follows:

[0075] (1) For any channel's sampling sequence x(n), set the window length to 20 sampling points (corresponding to a 2μs time width) and the step size to 1 sampling point (corresponding to 0.1μs). Start sliding the window from the beginning of the sequence, according to the formula E(k)=Σx 2 (i) (i ranges from k to k+19) Calculate the energy sum for each window.

[0076] (2) Calculate the energy difference between adjacent windows ΔE(k) = E(k) - E(k-1).

[0077] (3) Set an energy mutation threshold. When ΔE(k) of three consecutive windows exceeds the threshold, mark the starting sampling point of the first window in these three windows as the candidate mutation time.

[0078] (4) Perform secondary slope verification on candidate abrupt change points: Take 1ms data segments before and after the candidate point and calculate the rate of change of the slope of the signal within the data segment. If the rate of change of the slope exceeds 5 times the slope during steady-state operation, then the candidate point is confirmed as the transient initiation abrupt change point of the fracture. Otherwise, it is judged as an interference signal, removed, and the search is restarted.

[0079] (5) In Within the subsequent 500μs time window, search for the sampling point corresponding to the maximum value of the channel signal, and record the time of this sampling point as the peak time. .

[0080] The logic for evaluating actions at different times calculates the maximum difference between the start and peak times of the four fracture points:

[0081]

[0082]

[0083] Maximum difference at the start time The maximum difference at peak times serves as the primary criterion for different periods. For supplementary reference, the rating rules are as follows:

[0084] Closing operation: This is normal; To monitor the situation, an early warning is issued; This is an anomaly, triggering an alarm.

[0085] Opening operation: This is normal; Pay attention to the status; This is abnormal.

[0086] Simultaneously retrieve similar operations (both closing or both opening) from the historical database over the past 6 months. The data were linearly fitted using the least squares method to obtain... The trend line changes with the number of operations or time, and the changes in the difference value in different periods over the next 3 months are predicted accordingly. If the predicted value will exceed the alarm threshold within 3 months, preventive maintenance suggestions will be issued in advance.

[0087] Complete testing procedure: On-site deployment stage: Fix one set of BGO sensor probes on the outer shell of each break of the four-break circuit breaker, adjust the probe orientation so that the crystal sensitive axis is parallel to the electric field direction of the break; complete the signal wiring between each probe and the signal conditioning module, connect the power supply and communication cables of each module, and power on the system to confirm that all channels are working normally.

[0088] Trigger sampling phase: When the circuit breaker performs opening and closing operations, the system prioritizes receiving the trigger signal from the external auxiliary contact; if the external signal is abnormal or missing, it will be triggered by internal amplitude change (triggered when the amplitude of any channel signal exceeds 1.2 times the steady-state peak value of the power frequency), and start four-channel 10MS / s synchronous sampling to record the complete transient waveform for 100ms from 10ms before the trigger to 90ms after the trigger.

[0089] Waveform preprocessing stage: Baseline correction is performed on the sampled data to remove DC drift; smoothing filtering is performed to filter out high-frequency spike interference and output a standardized transient waveform.

[0090] Timing point location stage: The sliding window energy mutation algorithm is executed on the standardized waveforms of each channel, and the start mutation time and peak time of each break point are obtained after secondary verification.

[0091] Different Calculation Stages: Calculate the maximum difference in the starting time between fracture surfaces. Maximum difference from peak time , as a quantification value for different phases of the action.

[0092] Status assessment phase: Select the corresponding closing or opening threshold group according to the operation type, determine the different levels (normal / caution / abnormal) by comparing with the threshold, and analyze the deterioration trend in combination with historical data to output the assessment results and maintenance suggestions.

[0093] Results display and storage stage: The host computer synchronously displays four transient waveforms, marks the characteristic moments of each channel (initial change moment and peak moment), displays the values ​​and status levels of different periods, and all data is automatically stored in the historical database for subsequent traceability and trend analysis.

[0094] The above specific embodiments are merely several optional embodiments of the present invention. Based on the technical solutions of the present invention and the relevant teachings of the above embodiments, those skilled in the art can make various alternative improvements and combinations to the above specific embodiments.

Claims

1. A non-contact online detection system for asynchronous operation of multi-break circuit breakers, characterized in that, It includes a multi-channel electro-optic sensing unit, a synchronous signal conditioning module, a high-speed synchronous acquisition module, a timing feature extraction module, an action asynchrony evaluation module, and a host computer module; The multi-channel electro-optic sensing unit includes at least two independent sensing probes, which are arranged one-to-one on the outside of each break of the multi-break circuit breaker. They are used to non-contactly sense the transient electric field signals generated during the opening and closing of each break and convert them into analog electrical signals for output. The synchronization signal conditioning module is connected to the multi-channel electro-optic sensing unit and is used to condition the analog electrical signals of each channel and output a standardized transient signal. The high-speed synchronous acquisition module is connected to the synchronous signal conditioning module and is used to receive the circuit breaker operation trigger signal, start multi-channel synchronous sampling, and output discrete sampling data of each channel. The time-series feature extraction module is connected to the high-speed synchronous acquisition module and is used to calculate the sampled data of each channel based on the signal mutation detection algorithm to locate the start mutation time and peak time of the transient signal at each break point. The asynchronous action evaluation module is connected to the timing feature extraction module and is used to calculate the difference between the timing feature parameters of each break point action. Combined with the preset asynchronous action threshold, it outputs the asynchronous action level of the break point opening and closing and the mechanical state evaluation results. The host computer module is communicatively connected to the action asynchrony evaluation module and is used to display the action timing waveforms, asynchrony time differences, and status evaluation results of each break point.

2. The non-contact online detection system for asynchronous operation of multi-break circuit breakers according to claim 1, characterized in that, The multi-channel electro-optic sensing unit senses changes in the spatial electric field based on the electro-optic effect. Each sensing probe has a built-in light source, electro-optic crystal, and photoelectric conversion element, and outputs an electrical signal corresponding to the electric field strength.

3. The non-contact online detection system for asynchronous operation of multi-break circuit breakers according to claim 1, characterized in that, The high-speed synchronous acquisition module supports two modes: external trigger and internal threshold trigger. After sampling is triggered, it automatically records the complete waveform data of the first time period before the trigger and the second time period after the trigger.

4. The non-contact online detection system for asynchronous operation of multi-break circuit breakers according to claim 1, characterized in that, The signal mutation detection algorithm is a sliding window energy mutation algorithm: the energy of the signal within the window is calculated using a sliding window of fixed length. When the energy sum increment of multiple consecutive windows exceeds a preset threshold, the starting point of the first window exceeding the threshold is determined as the moment of the transient signal mutation. The time corresponding to the extreme value of the signal is searched within a preset time window after the moment of the mutation and determined as the peak time.

5. The non-contact online detection system for asynchronous operation of multi-break circuit breakers according to claim 1, characterized in that, The asynchronous action assessment module has built-in thresholds for asynchronous opening and asynchronous closing. When the maximum time difference between the disconnection points is less than the warning threshold, it is judged as normal; when it is between the warning threshold and the alarm threshold, it is judged as a state of attention; and when it is greater than the alarm threshold, it is judged as an abnormal state.

6. The non-contact online detection system for asynchronous operation of multi-break circuit breakers according to claim 1, characterized in that, The asynchronous operation assessment module also has a built-in historical trend analysis unit, which stores the time difference data of different periods of circuit breaker operation, fits the changing trend of different period deviations, and outputs preventive maintenance suggestions.

7. A non-contact online detection method for asynchronous operation of multi-break circuit breakers, applied to the non-contact online detection system for asynchronous operation of multi-break circuit breakers as described in any one of claims 1-6, characterized in that: step: Step 1: Install sensor probes on the outside of each break of the multi-break circuit breaker to complete the physical fixing and signal wiring of the multi-channel sensor unit; Step 2: Receive the circuit breaker opening and closing operation trigger signal, start multi-channel synchronous high-speed sampling, and collect the transient electric field signal during the operation of each break point; Step 3: Preprocess the sampled data from each channel to obtain standardized transient waveforms; Step 4: Using a signal mutation detection algorithm, calculate the standardized waveforms of each channel to locate the start and peak times of the transient signal at each break point. Step 5: Calculate the maximum difference between the initial abrupt change time and the maximum difference between the peak time for all break points, and use this as the time difference between the opening and closing abrupt changes; Step 6: Compare the time difference between different periods with the preset threshold to determine the different levels of the fracture action, and output the mechanical condition assessment results and early warning information. Step 7: Upload the transient waveform, time difference between different periods, and state assessment results to the host computer for display and storage.

8. The non-contact online detection method for asynchronous operation of multi-break circuit breakers according to claim 7, characterized in that, In step two, the trigger signal adopts a dual-trigger redundancy mechanism: it prioritizes receiving external trigger signals; when the external trigger signal fails or is missing, it automatically enables internal amplitude triggering; and when the signal amplitude exceeds the preset amplitude threshold, it starts high-speed sampling.

9. The non-contact online detection method for asynchronous operation of multi-break circuit breakers according to claim 7, characterized in that, In step four, the location of the initial mutation time also includes secondary verification: taking a data segment before and after the initially located mutation time, calculating the signal change rate, and confirming the mutation time as valid when the signal change rate meets the preset conditions; otherwise, it is judged as interference and the search is restarted.

10. A non-contact online detection method for asynchronicity of tripping action of a multi-break circuit breaker according to claim 7, characterized in that, Step six also includes a degradation assessment based on historical data: retrieve time difference data of different periods of the same type of operation within a historical time period, fit the trend of deviation change, and output an early maintenance warning when it is predicted that the difference between different periods within a future predetermined time period will exceed the alarm threshold.

Images