Intelligent circuit breaker multi-parameter state monitoring and fault diagnosis system

Abstract

The application is specifically a multi-parameter state monitoring and fault diagnosis system for intelligent circuit breaker, and relates to the technical field of power equipment state monitoring and fault diagnosis, comprising: a dielectric property micro-change online monitoring module; a surface / body charge and electroluminescence optical monitoring module; a mechanical-electrical performance correlation analysis module; a diagnosis evaluation module.In the application, the dielectric property micro-change module adopts low-frequency superimposed excitation and partial discharge fingerprint spectrum fusion technology, which can sensitively capture the dielectric micro-change of gradual degradation such as dampness and uniform aging; the surface / body charge and electroluminescence module adopts non-contact electrostatic potential array and PMT / APD hybrid optical imaging, which can accurately identify surface charge accumulation and early growth of electrical treeing; the mechanical-electrical correlation module relies on FBG strain monitoring and multi-physical field coupling modeling, which can quantify the impact of thermal mechanical stress accumulation and impact vibration.

Description

Technical Field

[0001] This invention relates to the field of power equipment condition monitoring and fault diagnosis technology, and in particular to a multi-parameter condition monitoring and fault diagnosis system for intelligent circuit breakers. Background Technology

[0002] As a core switching device in a power system, the circuit breaker bears the critical responsibility of controlling the opening and closing of circuits and ensuring the safe operation of the system. Its operating status directly affects the stability and reliability of the power system. With the development of power systems towards higher voltage, larger capacity, and intelligence, higher requirements are placed on the operational reliability and scientific maintenance of circuit breakers.

[0003] Traditional circuit breaker condition monitoring technologies mostly adopt a single-parameter monitoring mode, focusing only on information from a single dimension such as partial discharge, mechanical vibration, or electrical parameters, which has obvious limitations.

[0004] It is impossible to fully capture the characteristic signals of multiple failure processes in circuit breaker solid insulation, such as moisture absorption, aging, electrical tree growth, surface charge accumulation, and mechanical stress accumulation, leading to missed detection of early hidden dangers.

[0005] Single-parameter monitoring is susceptible to environmental interference, has a high false alarm rate, and lacks quantitative analysis of the relationship between mechanical condition and electrical performance, making it difficult to accurately locate and identify fault types.

[0006] In addition, most existing technologies can only make qualitative judgments about faults, but cannot quantitatively assess the severity of faults or predict the remaining life of circuit breakers, resulting in a lack of targeted operation and maintenance work and difficulty in achieving preventive maintenance.

[0007] Therefore, developing a system that can monitor the status of circuit breakers in a multi-dimensional, high-precision, and full-lifecycle manner, and achieve accurate fault diagnosis and remaining life prediction, has become an urgent technical problem to be solved in the field of power equipment operation and maintenance. Summary of the Invention

[0008] The purpose of this invention is to provide a multi-parameter status monitoring and fault diagnosis system for intelligent circuit breakers in order to solve the above-mentioned problems.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] The intelligent circuit breaker multi-parameter condition monitoring and fault diagnosis system includes:

[0011] The online monitoring module for micro-changes in dielectric properties is configured to capture micro-changes in dielectric parameters of solid insulating materials as they gradually deteriorate through synchronous detection of low-frequency superimposed excitation and analysis of partial discharge fingerprint spectrum.

[0012] The surface / volume charge and electroluminescence optical monitoring module is configured to use non-contact electrostatic potential monitoring and high-sensitivity electroluminescence / light emission imaging technology to capture physical signals of the unique failure path of solid insulation;

[0013] The mechanical-electrical performance correlation analysis module is configured to establish quantitative correlation rules between mechanical state and electrical performance based on FBG strain / vibration monitoring and electro-thermal-mechanical multiphysics coupling modeling.

[0014] The diagnostic assessment module is configured to analyze multi-dimensional parameters to obtain dielectric anomaly index, charge anomaly index, light emission anomaly index, and mechanical anomaly index, and then fuse them to obtain a comprehensive anomaly assessment coefficient and predict the remaining life of the circuit breaker.

[0015] Preferably, the online monitoring module for minute changes in dielectric properties specifically includes:

[0016] High-impedance coupler: installed between the busbar and the capacitive voltage divider;

[0017] Low-frequency signal generator: communicates with the main controller via an RS485 interface;

[0018] Synchronous demodulation detection circuit: synchronously acquires the amplitude of test voltage and response current;

[0019] The data acquisition card communicates with the industrial computer via a PCIe interface;

[0020] The collected test voltage With response current Perform analysis and calculate the equivalent capacitance. Dielectric loss factor ;

[0021] After the equipment is put into operation, it will continuously monitor for a preset period of time, remove outliers, and establish an initial baseline value. , ;

[0022] Baseline calibration will be performed quarterly thereafter.

[0023] UHF sensor: Installed in the monitoring window reserved in the solid-sealed electrode housing;

[0024] HFCT sensor: mounted on the grounding lead of the circuit breaker;

[0025] Signal conditioning module: built-in preamplifier and bandpass filter;

[0026] The absolute value of the difference between the calibrated dielectric loss factor and the initial baseline dielectric loss factor obtained from monitoring. >k× This increases the sampling rate and extends the acquisition time of PD monitoring; k is the dielectric performance abnormality triggering threshold determined based on the normal fluctuation range of dielectric properties of solid insulating materials, early degradation test data, and engineering practice experience.

[0027] Simultaneously, PRPD graph analysis is initiated to extract feature parameters, which are then matched with a preset defect fingerprint database. When the matching degree is greater than a preset threshold, a defect type warning is output.

[0028] Preferably, the surface / volume charge and electroluminescence optical monitoring module specifically includes:

[0029] Non-contact electrostatic potentiometer array: one sensor is arranged every 10cm along the axial direction of the solid insulating component, and four sensors are arranged circumferentially to form a 360° full coverage monitoring.

[0030] When the circuit breaker is de-energized or under maintenance, the sensor moves along a preset path to collect the electrostatic potential value of each point on the surface of the insulating component. A two-dimensional surface potential distribution map is drawn using MATLAB to identify the potential peak point.

[0031] After the circuit breaker is opened or closed, continuous monitoring is started to record the potential decay curves of preset key points;

[0032] An exponential fitting algorithm is used to extract the dispersion time constant. ,when >When the preset boundary between normal and deteriorated states is reached, the surface condition is determined to be abnormal, and the ambient humidity at the time of the abnormality is recorded.

[0033] Preferably, the method further includes an electroluminescence / light emission imaging monitoring submodule:

[0034] The optical sensor array adopts a hybrid array design of PMT and APD; PMT is placed in the key area and APD is placed in the remaining area;

[0035] Each sensor is equipped with dual filters, including an ultraviolet filter and a blue light filter;

[0036] A partial light shield is installed between the sensor and the insulating components to form a closed monitoring space; the inner wall of the light shield is coated with a light-absorbing coating; the sensor has a built-in miniature cooling fan.

[0037] Data synchronization is triggered by the zero-crossing point of the power frequency voltage, which synchronizes and correlates the optical signal, the power frequency voltage phase signal, and the PD signal.

[0038] Optical signal acquisition and filtering: continuously acquire optical intensity signals, set thresholds, and remove noise signals; perform pulse counting and amplitude statistics on valid optical signals;

[0039] Early detection of electrical tree branches: If a continuous light signal appears in the preset phase interval and no change in PD signal exceeding the threshold is detected, it is determined as an early warning of electrical tree branch growth.

[0040] Preferably, the mechanical-electrical performance correlation analysis module specifically includes:

[0041] FBG strain / temperature sensor: During the solidified electrode casting process, the FBG sensor is embedded in the key stress point, and the sensor is bonded to the epoxy resin using a silane coupling agent; the optical fiber leading out of the sensor is armored for protection.

[0042] MEMS accelerometers are attached to the solidified electrode housing with epoxy adhesive, and the installation position corresponds one-to-one with the FBG sensor to ensure the consistency of vibration signals.

[0043] Strain and temperature data are stored at preset time intervals. Real-time stress is calculated through stress-strain relationship. The daily maximum, minimum and fluctuation range of stress are statistically analyzed to determine whether there is an abnormality in the accumulation of thermomechanical stress.

[0044] Transient impact vibration monitoring: The trigger condition is the circuit breaker opening and closing auxiliary contact signal, and the acceleration-time curve is recorded; the peak impact acceleration, impact duration and vibration energy are extracted to determine whether the impact vibration is abnormal.

[0045] Preferably, the method further includes a correlation analysis and diagnostic model submodule:

[0046] Electric field equations: Poisson's equation is used;

[0047] Boundary conditions: The conductor surface potential is the system rated voltage, and the metal support is grounded;

[0048] Temperature field equation: The heat conduction equation is used;

[0049] Boundary conditions: Ambient temperature is the measured value, surface heat dissipation coefficient is 10W / ( •℃);

[0050] Stress field equations: The equations of equilibrium of elasticity are adopted;

[0051] Boundary conditions: Fixed constraints on the metal support, free constraints on the epoxy resin surface;

[0052] Coupling terms: The temperature field is converted into thermal strain through the coefficient of thermal expansion, and the stress field affects the electric field distribution through the stress dependence of the material's dielectric constant.

[0053] Preferably, the method further includes:

[0054] Will Divide by The relative rate of change of dielectric parameters is obtained;

[0055] The partial discharge fingerprint spectrum monitoring submodule extracts feature parameters from the acquired PRPD spectrum; it then performs SVM algorithm matching with a preset defect fingerprint database to output the PD matching degree.

[0056] After normalizing the relative change rate of dielectric parameters and the PD matching degree, weighting factors for the relative change rate of dielectric parameters and the PD matching degree are preset, and the dielectric anomaly index is calculated by weighted summation.

[0057] Preferably, the method further includes:

[0058] During static scanning monitoring, a surface potential distribution map of the insulating component is plotted, and the maximum potential value is extracted as... ;

[0059] During dynamic monitoring, the peak potential within 1 second after the circuit breaker is opened or closed is taken as... ;

[0060] Preset dissipation time constants respectively and initial potential Within the normal range, and dissipation time constants that are not within the normal range. and initial potential These are recorded as dissipation time anomalies and potential peak value anomalies, respectively.

[0061] Calculate the absolute value of the difference between the maximum and minimum dissipation time outliers to obtain the dissipation range;

[0062] The absolute value of the difference between the maximum and minimum peak anomalies is calculated to obtain the potential range value;

[0063] After normalizing the dissipation range and the potential range, weighting factors for the dissipation range and the potential range are preset respectively, and the weighted sum is calculated to obtain the charge anomaly index.

[0064] Preferably, the method further includes:

[0065] After acquiring the optical signal, ambient light interference is eliminated, and the optical pulse signal in the preset phase interval is extracted. The average light intensity of three consecutive cycles is calculated as the average light intensity; and the average light intensity in each time period is acquired sequentially.

[0066] Preset the critical light intensity for the early growth of electric trees and the limiting light intensity for the development stage of electric trees;

[0067] If the average light intensity obtained is greater than the critical light intensity for the early growth of electrical trees, then the average light intensity is subtracted from the critical light intensity for the early growth of electrical trees to obtain the light intensity difference.

[0068] Obtain the light intensity difference corresponding to each average light intensity, and subtract the minimum light intensity difference from the maximum light intensity difference to obtain the light intensity range value;

[0069] The light intensity range value is then divided by the limiting light intensity during the electric tree development stage to obtain the light emission anomaly index.

[0070] Preferably, the method further includes:

[0071] Based on the obtained acceleration-time curve, and the maximum value in the curve, the peak impact acceleration is taken as the peak acceleration.

[0072] If multiple peak values ​​occur during a single opening or closing operation, the average of the three largest peak values ​​shall be taken as the peak impact acceleration.

[0073] Thermomechanical stress calculated based on stress-strain relationship is calculated by statistically analyzing the daily maximum stress value and taking the average value of 7 consecutive days as the thermomechanical stress for calculation.

[0074] The peak impact acceleration and thermomechanical stress were obtained for each monitoring time period.

[0075] The maximum allowable value of thermomechanical stress and the allowable range of peak impact acceleration are preset respectively;

[0076] The thermomechanical stress that exceeds the maximum allowable value of thermomechanical stress is recorded as abnormal thermomechanical stress, and the absolute value of the difference between the maximum and minimum abnormal thermomechanical stress is calculated to obtain the abnormal stress difference.

[0077] The difference between each peak impact acceleration and the allowable range of peak impact acceleration is calculated to obtain the acceleration anomaly difference. The maximum acceleration anomaly difference is extracted, and the maximum acceleration anomaly difference and the anomaly stress difference are normalized. The weighting factors of the maximum acceleration anomaly difference and the anomaly stress difference are preset respectively, and the weighted summation is calculated to obtain the mechanical anomaly index.

[0078] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0079] 1. This invention employs a dielectric property micro-change module using low-frequency superposition excitation and partial discharge fingerprint spectrum fusion technology to sensitively capture gradual degradation of dielectric micro-changes such as moisture absorption and uniform aging; the surface / volume charge and electroluminescence module accurately identifies surface charge accumulation and early electrical tree growth through a non-contact electrostatic potential array and PMT / APD hybrid optical imaging; the mechanical-electrical correlation module quantifies the accumulation of thermomechanical stress and the impact of impact vibration based on FBG strain monitoring and multi-physics field coupling modeling.

[0080] 2. This invention constructs a multi-parameter quantitative fusion diagnostic system, transforming four dimensions of parameters—dielectric, charge, light emission, and mechanical—into standardized anomaly indices. Weighted summation yields a comprehensive anomaly assessment coefficient, enabling precise quantification of fault severity. The pre-defined relationship between the coefficients and remaining lifetime allows for direct output of circuit breaker lifetime prediction results. Simultaneously, relying on algorithms such as DFT, SVM, and exponential fitting, along with multi-physics coupling models, it effectively suppresses environmental interference and reduces false alarm and false negative rates. Attached Figure Description

[0081] Further details, features, and advantages of this application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0082] Figure 1 This is a system structure diagram of the present invention. Detailed Implementation

[0083] Several embodiments of this application will now be described in more detail with reference to the accompanying drawings to enable those skilled in the art to implement this application. This application may be embodied in many different forms and for various purposes and should not be limited to the embodiments set forth herein. These embodiments are provided to make this application thorough and complete, and to fully convey the scope of this application to those skilled in the art. The embodiments described do not limit this application.

[0084] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It will be further understood that terms such as those defined in commonly used dictionaries shall be interpreted as having a meaning consistent with their meaning in the relevant field and / or the context of this specification, and shall not be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0085] Example 1

[0086] Its specific implementation method is combined with the appendix Figure 1 Please provide a detailed explanation.

[0087] Appendix Figure 1 The diagram below shows the structure of the intelligent circuit breaker multi-parameter condition monitoring and fault diagnosis system provided in this embodiment of the invention. It illustrates the connection relationship between the online monitoring module for micro-changes in dielectric properties and the diagnostic evaluation module, and marks the main functional interaction flow of each module.

[0088] In this embodiment, it includes:

[0089] The online monitoring module for micro-changes in dielectric properties is configured to capture micro-changes in dielectric parameters of solid insulating materials during gradual deterioration, such as moisture absorption and uniform aging, through a fusion technology of low-frequency superimposed excitation synchronous detection and partial discharge fingerprint spectrum analysis, thereby enabling early warning and defect type identification.

[0090] Specifically, it includes:

[0091] Low-frequency superposition excitation and synchronous detection submodule:

[0092] The dielectric properties (capacitance, dielectric loss) of solid insulating materials are frequency-dependent. Low-frequency signals (VLF / low frequency) can penetrate the interior of insulating materials and more sensitively reflect slow polarization processes such as dipole turning hysteresis and interface charge migration.

[0093] When damp, water molecules, being polar molecules, increase the number of dipoles, leading to an increase in equivalent capacitance. Aging causes polymer chain breakage, reducing molecular polarization resistance and significantly increasing the dielectric loss factor (especially at low frequencies). This module achieves uninterrupted online monitoring by superimposing a low-frequency test mode onto the power frequency signal. Its core challenge is solving the technical problem of isolating and separating the test signal from the power frequency signal.

[0094] High-resistance coupler: It adopts a ceramic dielectric high-resistance design, with a power frequency impedance ≥100MΩ, a low frequency (0.1-10Hz) impedance ≤10kΩ, a voltage withstand rating ≥1.2 times the system rated voltage, and is compatible with circuit breakers of voltage levels from 3.6kV to 40.5kV. It is installed between the busbar and the capacitive voltage divider and is fixed by a flange to ensure mechanical stability.

[0095] Low-frequency signal generator: Output voltage amplitude 0-10kV (continuously adjustable), frequency resolution 0.01Hz, frequency stability ±0.001% / h, waveform distortion ≤0.5%, with sine wave and square wave output modes (sine wave by default), communicates with the main controller via RS485 interface, and supports remote parameter configuration;

[0096] Synchronous demodulation detection circuit: Adopting a dual-channel lock-in amplifier (LIA) core architecture, with an input impedance ≥1MΩ, phase resolution 0.01°, signal-to-noise ratio ≥80dB, and bandwidth 10mHz-100Hz, it can simultaneously acquire the amplitude (accuracy ±0.1%) and phase difference (accuracy ±0.1°) of the test voltage and response current. It has a built-in 50Hz notch filter (attenuation ≥60dB) to effectively suppress power frequency interference.

[0097] Data acquisition card: 16-bit AD resolution, sampling rate ≥1kHz, sampling buffer ≥1MB, supports continuous sampling and triggered sampling modes, trigger threshold can be customized, and communicates with industrial computers via PCIe interface;

[0098] Based on the Discrete Fourier Transform (DFT) algorithm, the acquired test voltage is analyzed. (amplitude) Phase ) and response current (amplitude) Phase ) Perform analysis and calculate the equivalent capacitance , Dielectric loss factor , Output a set of calculation results every 10 seconds;

[0099] After the equipment is put into operation, it is continuously monitored for a preset period of 72 hours. Outliers (such as data from lightning strikes or switching overvoltage periods) are removed, and an initial baseline value is established using the arithmetic mean method. , ;

[0100] Baseline calibration will be performed quarterly thereafter to correct for the effects of ambient temperature (0-60℃). The calibration formula is as follows: ;in, For capacitor temperature coefficient, The temperature coefficient of dielectric loss is a material intrinsic parameter, obtained through factory testing. The equivalent capacitance obtained during the monitoring process (without correction for temperature effects); This is the calibration equivalent capacitance after ambient temperature correction (to eliminate the effect of temperature on capacitance and be used for comparison with the baseline value). The dielectric loss factor is the measured value obtained during the monitoring process; This is the calibration dielectric loss factor after ambient temperature correction (to eliminate the effect of temperature on dielectric loss and to be used for comparison with the baseline value). The actual ambient temperature during monitoring (in °C) is used. The calibration formula uses 25 °C as the reference temperature to correct for parameter deviations at different temperatures.

[0101] Partial Discharge (PD) Fingerprint Monitoring Submodule:

[0102] Air gaps, cracks, or surface contamination inside solid insulation can trigger partial discharge, which generates UHF electromagnetic waves (300MHz-3GHz) and HFCT current pulses (1MHz-1GHz).

[0103] Different types of defects correspond to unique PD fingerprint patterns:

[0104] The PRPD spectrum of internal air gap discharge is distributed in two clusters, with phases concentrated in 0°~90° and 180°~270°; the PRPD spectrum of surface discharge is distributed in a broad manner, with a wide phase coverage and dispersed amplitude.

[0105] By using a linkage mechanism that triggers enhanced PD monitoring through abnormal dielectric parameters, the problem of high false alarm rate in single PD monitoring can be solved.

[0106] The hardware includes:

[0107] UHF sensor: adopts microstrip antenna structure, operates in the frequency band of 300MHz-3GHz, gain ≥5dB, VSWR ≤1.5, is installed in the monitoring window reserved in the solid-sealed pole shell, and is ≥5cm away from the surface of the insulating part, and is connected to the signal conditioning module through a coaxial cable;

[0108] HFCT sensor: adopts Rogowski coil design, inner diameter ≥50mm, bandwidth 1MHz-1GHz, transformation ratio 1:1000, sensitivity ≥1mV / A, and is installed on the grounding lead of the circuit breaker. During installation, ensure that the iron core is closed to avoid magnetic flux leakage.

[0109] Signal conditioning module: Built-in preamplifier (adjustable gain 20-60dB), bandpass filter (UHF channel 300MHz-3GHz, HFCT channel 1MHz-1GHz), noise suppression ratio ≥60dB, output signal amplitude 0-5V (adaptive to data acquisition card input range).

[0110] When the dielectric monitoring unit detects: the absolute value of the difference between the calibrated dielectric loss factor and the initial baseline dielectric loss factor. >k× The sampling rate of PD monitoring is automatically increased from 1MHz to 10MHz, and the acquisition time is extended from 10s / time to 60s / time; k is the dielectric performance abnormality trigger threshold determined based on the normal fluctuation range of dielectric properties of solid insulating materials, early degradation test data, and engineering practice experience.

[0111] Simultaneously, PRPD spectrum analysis is initiated to extract 12 feature parameters (such as discharge phase distribution range, maximum discharge quantity, average discharge quantity, spectrum entropy value, symmetry coefficient, etc.) and match them with the preset defect fingerprint database (internal discharge, surface discharge, corona discharge). When the matching degree is greater than the preset threshold (≥85%), a defect type warning is output.

[0112] The surface / volume charge and electroluminescence optical monitoring module is configured to use non-contact electrostatic potential monitoring and high-sensitivity electroluminescence / light emission imaging technology to accurately capture physical signals of solid insulation failure paths such as surface charge accumulation and early electrical tree growth, filling the gap in traditional monitoring.

[0113] Specifically, it includes:

[0114] Non-contact surface electrostatic potential monitoring submodule:

[0115] Solid insulating materials have high surface resistivity, and during operation, surface charge is prone to accumulate due to electric field induction, charge injection, and other reasons.

[0116] Charge accumulation leads to local electric field distortion (distortion coefficient can reach 1.5-2.0). When the electric field strength exceeds the surface flashover critical value (usually 20-30 kV / cm), surface discharge will be triggered; and the charge dissipation rate is directly related to the surface state.

[0117] Surface contamination, aging, or microcracks can affect the dissipation time constant. Enlarge (under normal circumstances) The duration is 5-30 seconds, but can exceed 100 seconds in abnormal situations.

[0118] This module uses non-contact monitoring, avoiding the damage to the insulating surface caused by traditional contact measurement.

[0119] Non-contact electrostatic potentiometer array: Utilizing the principle of vibration capacitance, the measurement range is ±10kV, the resolution is 1V, the response time is <1ms, the measurement distance is 5-50cm (adjustable), each sensor size is ≤30mm×30mm×20mm, and the protection level is IP67 (suitable for humid and dusty environments inside cabinets); Array arrangement: One sensor is arranged every 10cm along the axial direction of solid insulating components (such as sleeves and insulating rods), and four sensors are arranged circumferentially (at 90° intervals) to form 360° full coverage monitoring;

[0120] When the circuit breaker is de-energized or under maintenance, the sensor moves along a preset path (spiral scanning, step size 0.5cm) to collect the electrostatic potential values ​​at various points on the surface of the insulating component. A two-dimensional surface potential distribution map is then plotted using MATLAB (color mapping: red ≥ 5kV, yellow 1-5kV, green -1-1kV, blue ≤ -1kV). Peak potential points are identified (if the peak value is ≥ 5kV and the area is ≥ 10...). (This was determined to be an abnormal charge accumulation).

[0121] After the circuit breaker is opened or closed (trigger signal: opening and closing auxiliary contact signal), continuous monitoring is started, with a sampling frequency of 10Hz for 300s, and the potential decay curves of preset key points (potential peak point, both ends of the insulation component) are recorded.

[0122] Using exponential fitting algorithm ,in This is the initial potential; For time; extract the discrete time constant ,when >When the preset boundary between normal and deteriorated states is reached, the surface condition is determined to be abnormal, and the ambient humidity at the time of the abnormality is recorded.

[0123] The sensor housing is equipped with a grounded metal shield to suppress power frequency electric field interference; data transmission uses fiber optic communication to avoid electromagnetic interference; if a potential change occurs during monitoring (ΔV / Δt>1kV / s), abnormal data (which may be caused by external electromagnetic pulse interference) will be automatically removed.

[0124] It also includes an electroluminescence / photoluminescence (EL / PL) imaging monitoring submodule:

[0125] The high-sensitivity optical sensor array adopts a hybrid array design of PMT and APD; PMT is arranged in key areas (such as the root of the conductor lead and the middle section of the insulating tie rod), and APD is arranged in other areas. Each sensor has a field of view of ≥60° and a monitoring distance of 10-30cm.

[0126] Optical filtering system: Each sensor is equipped with dual filters (switchable), including an ultraviolet filter (center wavelength 254nm, bandwidth ±10nm, transmittance ≥80%) and a blue light filter (center wavelength 450nm, bandwidth ±20nm, transmittance ≥85%), to filter out ambient light interference (such as industrial fluorescent lamps and sunlight) (ambient light suppression ratio ≥1000:1).

[0127] A partial light shield (made of black polytetrafluoroethylene with a light-blocking rate of ≥99.9%) is installed between the sensor and the insulating components to form a closed monitoring space; the inner wall of the light shield is coated with a light-absorbing coating (absorption rate ≥99%) to avoid light reflection interference; the sensor has a built-in miniature cooling fan (speed 3000r / min) to ensure that the long-term operating temperature is ≤60℃.

[0128] Synchronization triggering conditions: Data synchronization is triggered when the power frequency voltage crosses zero. A GPS synchronization clock (time accuracy ±1μs) is used to synchronize and correlate the optical signal, the power frequency voltage phase signal (acquired through a voltage transformer, phase resolution 0.1°), and the PD signal to generate a three-dimensional phase-light intensity-time map. 1000 light intensity data points are collected per cycle.

[0129] Optical signal acquisition and filtering: Continuously acquire light intensity signals and set thresholds (PMT thresholds). W / APD threshold W / Noise signals are removed; pulse counting (number of light pulses per unit time) and amplitude statistics (maximum light intensity, average light intensity) are performed on the effective light signals.

[0130] Early detection of electrical treeing: If a continuous light signal appears within a preset phase range (e.g., ±30° near the peak of the power frequency voltage), with light pulses detected for three consecutive cycles and the number of pulses ≥ 5 times / cycle, and no PD signal change exceeding the threshold (discharge < 10 pC), it is determined as an early warning of electrical treeing growth. The growth stage of the electrical treeing is determined by combining the light intensity amplitude and phase distribution (initial stage: light intensity < 10 pC). W / cm², growth rate <1μm / h; development stage: light intensity - W / (Growth rate 1-10 μm / h).

[0131] Discharge type-assisted identification: The light signal of surface discharge is mainly concentrated on the surface of the insulating component, with wavelength mainly blue light (around 450nm), and the light pulse is highly synchronized with the PD signal; the light signal of internal discharge is concentrated inside the insulation, with wavelength mainly ultraviolet light (around 254nm), and the light pulse duration is longer (10-100ns); the wavelength distribution, spatial location and temporal characteristics of the light signal are used to assist in the identification of PD type.

[0132] The mechanical-electrical performance correlation analysis module is configured to establish quantitative correlation rules between mechanical state and electrical performance based on FBG strain / vibration monitoring and electro-thermal-mechanical multiphysics coupling modeling.

[0133] Specifically, it includes:

[0134] Micro-stress / vibration sensing submodule:

[0135] The solid-insulated circuit breaker's sealed pole is cast from epoxy resin, conductor, and metal support components. The coefficients of thermal expansion of different materials vary considerably (epoxy resin has a coefficient of thermal expansion of 6-8 × 10⁻⁶). / ℃, copper conductor 17× / ℃), long-term operation with load changes and environmental temperature cycles (-40℃-60℃) will cause internal thermomechanical stress (normal working stress ≤5MPa, abnormal stress can reach more than 10MPa), stress accumulation will cause epoxy resin creep and micro cracks (when the crack width is ≥1μm, the electric field distortion is significant).

[0136] In addition, the mechanical shock (impact acceleration ≤50g) generated by the opening and closing operation of the circuit breaker can cause the interface between the insulation structure and the metal parts to debond, forming an air gap, which in turn can trigger partial discharge.

[0137] This module enables direct monitoring of mechanical stress and vibration through embedded / surface sensors.

[0138] FBG strain / temperature sensor: Employs Bragg grating technology, with a center wavelength of 1550nm, a strain measurement range of ±2000με (corresponding to a stress range of 0-10MPa), a strain resolution of 1με, a temperature measurement range of -40℃-120℃, a temperature resolution of 0.1℃, and sensor dimensions ≤10mm×2mm×0.5mm (fiber diameter 125μm). Installation method: During the casting process of the solidified electrode post, the FBG sensor is embedded in key stress points (the root of the conductor lead, the interface between the metal support and the epoxy resin, and the middle of the electrode post). A silane coupling agent is used to bond the sensor to the epoxy resin to ensure a strain transfer efficiency ≥95%. The sensor's fiber optic cable is armored (stainless steel armor tube, 3mm diameter) to prevent mechanical damage.

[0139] MEMS accelerometer: Utilizes piezoelectric principle, measurement range ±50g, bandwidth 1-1000Hz, acceleration resolution 0.01g, output signal is analog voltage (0-5V), dimensions ≤5mm×5mm×3mm, protection rating IP67. Installation method: Mounted to the solidified electrode housing (corresponding to the critical internal insulation area) using epoxy adhesive. The installation position corresponds one-to-one with the FBG sensor to ensure consistent vibration signal; the shear strength of the cured adhesive is ≥10MPa, suitable for long-term vibration environments.

[0140] Long-term thermomechanical stress monitoring: Sampling frequency 1Hz, storing strain ε and temperature T data at preset time intervals (every 10 minutes), and analyzing the stress-strain relationship. ( The elastic modulus of epoxy resin at 25°C is... Corrected for temperature variations: Calculate real-time stress; statistically analyze the daily maximum, minimum and fluctuation range of stress. When the maximum stress value is ≥8MPa or the fluctuation range is ≥3MPa for 7 consecutive days, it is judged as an abnormal accumulation of thermomechanical stress.

[0141] Transient impact vibration monitoring: The trigger condition is the circuit breaker opening and closing auxiliary contact signal (triggered 10ms in advance), the sampling rate is automatically increased to 10kHz, the acquisition time is 1s (the opening and closing operation lasts for about 500ms), and the acceleration-time curve is recorded; the peak impact acceleration, impact duration, and vibration energy are extracted. , The acceleration function varies with time; Three characteristic parameters (impact duration); when the peak impact acceleration > 30g or the vibration energy > 100... At that time, it was determined to be an abnormal impact vibration.

[0142] It also includes a sub-module for correlation analysis and diagnostic models:

[0143] Model Foundation: A 3D geometric model of the solid-sealed electrode was established based on the ANSYS Workbench platform (including epoxy resin, conductor, metal support, and sensor mounting location). Tetrahedral elements were used for mesh generation, with a mesh size <0.1mm in critical areas (FBG sensor mounting location and interface area), and a total number of elements ≥ Ensure calculation accuracy;

[0144] Electric field equations: Poisson's equation is used. , The dielectric constant of epoxy resin is 2.5-3.0. Potential; Free charge density;

[0145] Boundary conditions: The conductor surface potential is the system rated voltage, and the metal support is grounded;

[0146] Temperature field equation: using the heat conduction equation , The density of epoxy resin; Specific heat capacity; Thermal conductivity; The Joule heat density is calculated from the conductor current.

[0147] Boundary conditions: Ambient temperature is the measured value, surface heat dissipation coefficient is 10W / ( •℃);

[0148] Stress field equations: using the equilibrium equations of elasticity. , These are volume forces (gravity is negligible); the constitutive relation adopts Hooke's law. , It is the elasticity matrix;

[0149] Boundary conditions: Fixed constraints on the metal support, free constraints on the epoxy resin surface;

[0150] Coupling terms: The temperature field is transformed into thermal strain through the coefficient of thermal expansion, and the stress field is transformed into stress dependence of the material's dielectric constant. , The intrinsic dielectric constant of a solid insulating material under no stress (or reference condition); The mechanical stress borne by the solid insulating material; 0.01 is the stress influence coefficient of the dielectric constant, which affects the electric field distribution;

[0151] Through factory testing (applying rated voltage, simulating temperature cycling, and mechanical shock), the measured strain and temperature data of the FBG sensor are collected and compared with the model calculation results. The elastic modulus, dielectric constant, and other parameters of the epoxy resin are then corrected to ensure that the model calculation error is less than the preset threshold.

[0152] The diagnostic assessment module is configured to analyze multi-dimensional parameters to obtain dielectric anomaly index, charge anomaly index, light emission anomaly index, and mechanical anomaly index, and then fuse them to obtain a comprehensive anomaly assessment coefficient and predict the remaining life of the circuit breaker.

[0153] This process requires normalizing the dielectric anomaly index, charge anomaly index, optical emission anomaly index, and mechanical anomaly index, and pre-setting weighting factors for each index. The comprehensive anomaly evaluation coefficient is obtained by multiplying the dielectric anomaly index, charge anomaly index, optical emission anomaly index, and mechanical anomaly index with their corresponding weighting factors and then summing the products.

[0154] Multiple sets of comprehensive anomaly assessment coefficients are preset with their value ranges. Each value range corresponds to the remaining life of a circuit breaker. The comprehensive anomaly assessment coefficient is matched with the value ranges of the multiple sets of comprehensive anomaly assessment coefficients to obtain the remaining life of the circuit breaker corresponding to the comprehensive anomaly assessment coefficient.

[0155] Will Divide by The relative rate of change of dielectric parameters is obtained;

[0156] The partial discharge fingerprint spectrum monitoring submodule extracts 12 feature parameters from the acquired PRPD spectrum; it then performs SVM algorithm matching with the preset defect fingerprint database (internal discharge, surface discharge, corona discharge) to output the PD matching degree.

[0157] After normalizing the relative change rate of dielectric parameters and the PD matching degree, weighting factors for the relative change rate of dielectric parameters and the PD matching degree are preset respectively, and the dielectric anomaly index is calculated by weighted summation.

[0158] By fusing and quantizing the relative change rate of dielectric parameters with the SVM algorithm matching results of partial discharge (PD) fingerprint spectrum, the accuracy and anti-interference ability of dielectric anomaly identification are improved.

[0159] The relative change rate of dielectric parameters can be directly related to the gradual deterioration process of solid insulating materials, such as moisture absorption and aging, and reflect the quantitative trend of the degree of deterioration. The PD matching degree, by extracting 12 characteristic parameters and comparing them with the preset defect fingerprint database, accurately identifies specific defect types such as internal discharge and surface discharge. The combination of the two effectively avoids the false alarm problem of single dielectric parameter fluctuation or single PD monitoring, and provides a more reliable basis for early deterioration warning.

[0160] A key link was established between dielectric monitoring and comprehensive system diagnosis, laying a high-quality foundation for the subsequent calculation of comprehensive anomaly assessment coefficients.

[0161] It eliminates the dimensional differences of different parameters through normalization, and the preset design of the weighting factor can be adapted to the actual needs of circuit breakers of different voltage levels from 3.6kV to 40.5kV, taking into account both versatility and specificity.

[0162] Meanwhile, based on the parameter background corrected by the multiphysics coupling model, the scientific nature and stability of the dielectric anomaly index are ensured. It can accurately map the true level of insulation degradation, provide maintenance personnel with a clear reference for defect types and anomaly degrees, help to accurately formulate maintenance strategies, reduce the risk of insulation failure, and extend the service life of circuit breakers.

[0163] During static scanning monitoring, a surface potential distribution map of the insulating component is plotted, and the maximum potential value is extracted as... ;

[0164] During dynamic monitoring, the peak potential within 1 second after the circuit breaker is opened or closed is taken as... ;

[0165] Preset dissipation time constants respectively and initial potential Within the normal range, and dissipation time constants that are not within the normal range. and initial potential These are recorded as dissipation time anomalies and potential peak value anomalies, respectively.

[0166] Calculate the absolute value of the difference between the maximum and minimum dissipation time outliers to obtain the dissipation range;

[0167] The absolute value of the difference between the maximum and minimum peak anomalies is calculated to obtain the potential range value;

[0168] After normalizing the dissipation range and the potential range, weighting factors for the dissipation range and the potential range are preset respectively, and the weighted sum is calculated to obtain the charge anomaly index.

[0169] By distinguishing between static scanning and dynamic monitoring scenarios, a comprehensive and accurate quantitative assessment system for charge anomalies was constructed, which improved the ability to identify charge-related defects on solid insulating surfaces.

[0170] In static scenarios, the array of sensors can scan 360° to create a two-dimensional surface potential distribution map, which can accurately pinpoint areas of high potential accumulation. In dynamic scenarios, the potential decay curve after switching on and off can be captured and the dissipation time constant τ can be extracted, which can effectively reflect hidden problems such as surface contamination, aging, or micro-cracks.

[0171] Meanwhile, by calculating the dissipation range and potential range, the charge anomaly is transformed from a qualitative judgment into a quantitative indicator, avoiding the random interference of a single data point and making the assessment of the degree of charge-related anomalies more scientific.

[0172] It provides high-quality charge dimension data support for comprehensive system anomaly assessment and further improves the closed-loop logic of multi-parameter fusion diagnosis. Its normalization process eliminates the differences in data dimensions under different monitoring scenarios and voltage levels, and the preset design of the weighting factor can flexibly adapt to the diverse application requirements of 3.6kV-40.5kV circuit breakers, taking into account both versatility and specificity.

[0173] In addition, the monitoring process simultaneously records ambient humidity and automatically removes electromagnetic pulse interference data, ensuring the stability and reliability of the charge anomaly index. This provides a clear reference for maintenance personnel to accurately locate the root cause of surface discharge hazards, helping to take protective measures in advance and reduce the risk of insulation failure.

[0174] After acquiring the optical signal, ambient light interference is eliminated, and optical pulse signals within a preset phase range (±30° near the peak of the power frequency voltage) are extracted. The average light intensity of three consecutive cycles is calculated as the average light intensity; and the average light intensity in each time period is acquired sequentially.

[0175] Preset the critical light intensity for the early growth of electric trees and the limiting light intensity for the development stage of electric trees;

[0176] If the average light intensity obtained is greater than the critical light intensity for the early growth of electrical trees, then the average light intensity is subtracted from the critical light intensity for the early growth of electrical trees to obtain the light intensity difference.

[0177] Obtain the light intensity difference corresponding to each average light intensity, and subtract the minimum light intensity difference from the maximum light intensity difference to obtain the light intensity range value;

[0178] The light intensity range value is then divided by the limiting light intensity during the electric tree development stage to obtain the light emission anomaly index.

[0179] A precise and proprietary early monitoring and quantification system for electrical treeing has been constructed, filling the gap in traditional monitoring of latent failure paths in solid insulation. By locking onto a key phase interval of ±30° near the peak power frequency voltage, it extracts effective light pulse signals from three consecutive cycles. Combined with light intensity thresholds for the early growth and development stages of electrical trees, the latent growth state of electrical trees is transformed into quantifiable light intensity differences and ranges, achieving an upgrade from simply judging presence to identifying stages.

[0180] Meanwhile, relying on anti-interference designs such as dual filters and partial light shields to eliminate the influence of ambient light, early warning can be completed based solely on optical signal characteristics, avoiding the problem of easy omissions when relying on PD signals, making the early detection of electrical treeing, a core cause of insulation failure, more reliable.

[0181] Based on the obtained acceleration-time curve, the maximum value in the curve is taken as the peak impact acceleration (the absolute value is taken, and the impact direction does not affect stress damage).

[0182] If multiple peak values ​​occur during a single opening or closing operation, the average of the three largest peak values ​​shall be taken as the peak impact acceleration.

[0183] Thermomechanical stress calculated based on stress-strain relationship is calculated by statistically analyzing the daily maximum stress value and taking the average value of 7 consecutive days as the thermomechanical stress for calculation.

[0184] The peak impact acceleration and thermomechanical stress were obtained for each monitoring time period.

[0185] The maximum allowable value of thermomechanical stress and the allowable range of peak impact acceleration are preset respectively;

[0186] The thermomechanical stress that exceeds the maximum allowable value of thermomechanical stress is recorded as abnormal thermomechanical stress, and the absolute value of the difference between the maximum and minimum abnormal thermomechanical stress is calculated to obtain the abnormal stress difference.

[0187] The difference between each peak impact acceleration and the allowable range of peak impact acceleration is calculated to obtain the acceleration anomaly difference. The maximum acceleration anomaly difference is extracted, and the maximum acceleration anomaly difference and the anomaly stress difference are normalized. The weighting factors of the maximum acceleration anomaly difference and the anomaly stress difference are preset respectively, and the weighted summation is calculated to obtain the mechanical anomaly index.

[0188] A precise and comprehensive quantitative assessment system for mechanical anomalies has been established, effectively covering the core causes of mechanical failures in circuit breakers and significantly improving the reliability and accuracy of mechanical condition monitoring.

[0189] For transient impact vibration, it avoids the random interference of a single peak by extracting the peak value of the acceleration-time curve (averaging the top 3 peak values ​​for multiple peak values). For long-term thermomechanical stress, it takes the average of the daily maximum stress over 7 consecutive days, which can truly reflect the stress accumulation effect. At the same time, by calculating the abnormal stress difference and the maximum acceleration abnormal difference, it transforms mechanical anomalies such as excessive thermomechanical stress and impact vibration overload into quantifiable indicators, comprehensively capturing the hidden risks caused by mechanical problems such as interface debonding of insulation structures and microcracks.

[0190] It provides high-quality mechanical dimension support for multi-parameter fusion diagnosis of the system, and further improves the closed-loop logic of comprehensive anomaly assessment. Its normalization process eliminates the dimensional differences of mechanical parameters under different monitoring scenarios and voltage levels, and the preset design of the weighting factor can flexibly adapt to the diverse application requirements of 3.6kV-40.5kV circuit breakers, taking into account both versatility and specificity.

[0191] The quantified mechanical anomaly index can accurately map the impact of mechanical state on insulation performance, complementing the anomaly indices in dielectric, charge, and light emission dimensions, making fault diagnosis more comprehensive.

[0192] At the same time, it provides maintenance personnel with a clear reference for the degree of mechanical abnormality, helping them to take targeted maintenance measures in advance (such as adjusting the opening and closing mechanism and reinforcing the structure), reduce the risk of insulation failure caused by mechanical stress, and extend the service life of the circuit breaker.

[0193] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0194] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0195] It should be noted that, in this document, the use of relational terms such as "first" and "second" is merely for distinguishing one entity or operation from another, and does not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0196] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0197] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0198] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0199] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0200] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0201] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0202] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A multi-parameter condition monitoring and fault diagnosis system for intelligent circuit breakers, characterized in that, include: The online monitoring module for micro-changes in dielectric properties is configured to capture micro-changes in dielectric parameters of solid insulating materials as they gradually deteriorate through synchronous detection of low-frequency superimposed excitation and analysis of partial discharge fingerprint spectrum. The surface / volume charge and electroluminescence optical monitoring module is configured to use non-contact electrostatic potential monitoring and high-sensitivity electroluminescence / light emission imaging technology to capture physical signals of the unique failure path of solid insulation; The mechanical-electrical performance correlation analysis module is configured to establish quantitative correlation rules between mechanical state and electrical performance based on FBG strain / vibration monitoring and electro-thermal-mechanical multiphysics coupling modeling. The diagnostic assessment module is configured to analyze multi-dimensional parameters to obtain dielectric anomaly indices, charge anomaly indices, optical emission anomaly indices, and mechanical anomaly indices. After fusion processing, a comprehensive anomaly assessment coefficient is obtained, and the remaining life of the circuit breaker is predicted. Specifically, this includes: The diagnostic assessment module specifically includes: The absolute value of the difference between the dielectric loss factor and the initial baseline dielectric loss factor. Divide by the initial baseline dielectric loss factor The relative rate of change of dielectric parameters is obtained; The partial discharge fingerprint spectrum monitoring submodule extracts feature parameters from the acquired PRPD spectrum; it then performs SVM algorithm matching with a preset defect fingerprint database to output the PD matching degree. After normalizing the relative change rate of dielectric parameters and the PD matching degree, weighting factors for the relative change rate of dielectric parameters and the PD matching degree are preset respectively, and the dielectric anomaly index is calculated by weighted summation. During static scanning monitoring, the maximum potential value of the surface potential distribution map of the insulating component is extracted as... During dynamic monitoring, the peak potential within 1 second after the circuit breaker is opened or closed is taken as the value. ; The dissipation time constant that exceeds the preset normal range and initial potential These are recorded as dissipation time anomalies and potential peak value anomalies, respectively. The absolute value of the difference between the maximum and minimum dissipation time anomalies is calculated to obtain the dissipation range, and the absolute value of the difference between the maximum and minimum peak anomalies is calculated to obtain the potential range. After normalizing the dissipation range and the potential range, respectively, the charge anomaly index is obtained by weighting and summing them with preset weighting factors. Also includes: After acquiring the optical signal, ambient light interference is eliminated, and the optical pulse signal in the preset phase interval is extracted. The average light intensity of three consecutive cycles is calculated as the average light intensity; and the average light intensity in each time period is acquired sequentially. Preset the critical light intensity for the early growth of electric trees and the limiting light intensity for the development stage of electric trees; If the average light intensity obtained is greater than the critical light intensity for the early growth of electrical trees, then the average light intensity is subtracted from the critical light intensity for the early growth of electrical trees to obtain the light intensity difference. Obtain the light intensity difference corresponding to each average light intensity, and subtract the minimum light intensity difference from the maximum light intensity difference to obtain the light intensity range value; The light intensity range value is then divided by the limiting light intensity during the electric tree development stage to obtain the light emission anomaly index. The maximum value in the acceleration-time curve is taken as the peak impact acceleration. If multiple peak values ​​occur in a single opening and closing operation, the average of the three largest peak values ​​is taken. The daily maximum thermomechanical stress calculated based on the stress-strain relationship was statistically analyzed, and the average value of 7 consecutive days was taken as the thermomechanical stress used for calculation. Acquire the peak impact acceleration and thermomechanical stress during each monitoring period; The thermomechanical stress exceeding the preset maximum allowable value is recorded as abnormal thermomechanical stress, and the absolute value of the difference between the maximum and minimum abnormal thermomechanical stress is used to obtain the abnormal stress difference. The difference between each peak acceleration and the preset allowable range is calculated to obtain the acceleration anomaly difference value, and the maximum acceleration anomaly difference value is extracted. After normalizing the maximum acceleration anomaly difference and the anomaly stress difference, the mechanical anomaly index is obtained by weighting and summing them together with preset weighting factors.

2. The intelligent circuit breaker multi-parameter condition monitoring and fault diagnosis system according to claim 1, characterized in that, The online monitoring module for minute changes in dielectric properties specifically includes: High-impedance coupler: installed between the busbar and the capacitive voltage divider; Low-frequency signal generator: communicates with the main controller via an RS485 interface; Synchronous demodulation detection circuit: synchronously acquires the amplitude of test voltage and response current; The data acquisition card communicates with the industrial computer via a PCIe interface; The collected test voltage With response current Perform analysis and calculate the equivalent capacitance. Dielectric loss factor ; After the equipment is put into operation, it will continuously monitor for a preset period of time, remove outliers, and establish an initial baseline value. , ; Baseline calibration will be performed quarterly thereafter. UHF sensor: Installed in the monitoring window reserved in the solid-sealed electrode housing; HFCT sensor: mounted on the grounding lead of the circuit breaker; Signal conditioning module: built-in preamplifier and bandpass filter; The absolute value of the difference between the calibrated dielectric loss factor and the initial baseline dielectric loss factor obtained from monitoring. >k× This increases the sampling rate and extends the acquisition time of PD monitoring; k is the dielectric performance abnormality triggering threshold determined based on the normal fluctuation range of dielectric properties of solid insulating materials, early degradation test data, and engineering practice experience. Simultaneously, PRPD graph analysis is initiated to extract feature parameters, which are then matched with a preset defect fingerprint database. When the matching degree is greater than a preset threshold, a defect type warning is output.

3. The intelligent circuit breaker multi-parameter condition monitoring and fault diagnosis system according to claim 1, characterized in that, The surface / volume charge and electroluminescence optical monitoring module specifically includes: Non-contact electrostatic potentiometer array: one sensor is arranged every 10cm along the axial direction of the solid insulating component, and four sensors are arranged circumferentially to form a 360° full coverage monitoring. When the circuit breaker is de-energized or under maintenance, the sensor moves along a preset path to collect the electrostatic potential value of each point on the surface of the insulating component. A two-dimensional surface potential distribution map is drawn using MATLAB to identify the potential peak point. After the circuit breaker is opened or closed, continuous monitoring is started to record the potential decay curves of preset key points; An exponential fitting algorithm is used to extract the dispersion time constant. ,when >When the preset boundary between normal and deteriorated states is reached, the surface condition is determined to be abnormal, and the ambient humidity at the time of the abnormality is recorded.

4. The intelligent circuit breaker multi-parameter condition monitoring and fault diagnosis system according to claim 3, characterized in that, It also includes an electroluminescence / light emission imaging monitoring submodule: The optical sensor array adopts a hybrid array design of PMT and APD; PMT is placed in the key area and APD is placed in the remaining area; Each sensor is equipped with dual filters, including an ultraviolet filter and a blue light filter; A partial light shield is installed between the sensor and the insulating components to form a closed monitoring space; the inner wall of the light shield is coated with a light-absorbing coating. The sensor has a built-in miniature cooling fan; Data synchronization is triggered by the zero-crossing point of the power frequency voltage, which synchronizes and correlates the optical signal, the power frequency voltage phase signal, and the PD signal. Optical signal acquisition and filtering: continuously acquire optical intensity signals, set thresholds, and remove noise signals; Pulse counting and amplitude statistics are performed on the effective optical signal; Early detection of electrical tree branches: If a continuous light signal appears in the preset phase interval and no change in PD signal exceeding the threshold is detected, it is determined as an early warning of electrical tree branch growth.

5. The intelligent circuit breaker multi-parameter condition monitoring and fault diagnosis system according to claim 1, characterized in that, The mechanical-electrical performance correlation analysis module specifically includes: FBG strain / temperature sensor: During the solidified electrode casting process, the FBG sensor is embedded in the key stress point, and the sensor is bonded to the epoxy resin using a silane coupling agent; the optical fiber leading out of the sensor is armored for protection. MEMS accelerometers are attached to the solidified electrode housing with epoxy adhesive, and the installation position corresponds one-to-one with the FBG sensor to ensure the consistency of vibration signals. Strain and temperature data are stored at preset time intervals. Real-time stress is calculated through stress-strain relationship. The daily maximum, minimum and fluctuation range of stress are statistically analyzed to determine whether there is an abnormality in the accumulation of thermomechanical stress. Transient impact vibration monitoring: The trigger condition is the circuit breaker opening and closing auxiliary contact signal, and the acceleration-time curve is recorded; the peak impact acceleration, impact duration and vibration energy are extracted to determine whether the impact vibration is abnormal.

6. The intelligent circuit breaker multi-parameter condition monitoring and fault diagnosis system according to claim 5, characterized in that, It also includes a sub-module for correlation analysis and diagnostic models: Electric field equations: Poisson's equation is used; Boundary conditions: The conductor surface potential is the system rated voltage, and the metal support is grounded; Temperature field equation: The heat conduction equation is used; Stress field equations: The equations of equilibrium of elasticity are adopted; Boundary conditions: Fixed constraints on the metal support, free constraints on the epoxy resin surface; Coupling terms: The temperature field is converted into thermal strain through the coefficient of thermal expansion, and the stress field affects the electric field distribution through the stress dependence of the material's dielectric constant.

Images