Circuit breaker service life prediction system and method based on multi-source heterogeneous data fusion and dynamic weight correction

Abstract

The invention discloses a circuit breaker service life prediction system and method based on multi-source heterogeneous data fusion and dynamic weight correction, and the method comprises the steps: collecting the current, pressure, displacement, main loop current, voltage, contact temperature, environment temperature and partial discharge of a circuit breaker, and carrying out the multi-source heterogeneous data fusion algorithm, thereby achieving the prediction of the service life of the circuit breaker. Carrying out fusion analysis on the collected data; a dynamic weight correction algorithm is executed according to an analysis result, and different types of fault weights are corrected and adjusted, so that the life prediction model is more accurate; and executing a fault identification algorithm after each action of the circuit breaker, and calculating the service life of the circuit breaker according to the weight and the health index corresponding to each fault. According to the invention, the prediction accuracy of the service life of the short-circuiter is greatly improved, and the use safety of power equipment is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power equipment status monitoring, and in particular relates to a circuit breaker life prediction system and method based on multi-source heterogeneous data fusion and dynamic weight correction. Background Art

[0002] Circuit breakers are one of the most important components in power systems. They not only control the load current when energized, but also provide protection by quickly disconnecting faulty lines when a power system failure occurs. Because circuit breaker failures can often lead to serious power grid incidents, circuit breaker fault diagnosis is crucial.

[0003] Patent CN118731676B in the prior art discloses a system and method for online monitoring and fault diagnosis of molded case circuit breakers. This method does not have circuit breaker life prediction functions, but only has data monitoring and fault diagnosis functions. Patent CN116595863A in the prior art discloses a circuit breaker life prediction method and system using multi-parameter fusion. Although it has a life prediction function, the life prediction model is relatively simple and only predicts the life by calculating the contact wear value, resulting in a relatively weak prediction result accuracy.

[0004] In addition, patent CN119556119A in the prior art proposes an online monitoring device and method for the operating status of circuit breakers based on distribution automation. This device collects circuit breaker data using displacement sensors, acceleration sensors, Hall effect current sensors, and pressure sensors to achieve online monitoring and status assessment of the circuit breaker. This method is primarily based on contact wear, combined with the current, displacement, pressure, and acceleration data of the opening and closing coils, to analyze operating mechanism jamming and vibration, and predict lifespan based on the weights of each characteristic parameter. Although this method involves circuit breaker lifespan prediction, it has the following drawbacks: (1) The partial discharge factor is not considered, and the circuit breaker insulation failure cannot be judged; (2) The impact of the environment on the circuit breaker was not considered, and the accelerated aging of the circuit breaker in harsh environments was ignored; (3) The collected parameters were not subjected to secondary data fusion analysis, and the parameters were directly weighted and calculated, resulting in low accuracy of the calculation results; (4) Parameter weights are preset according to parameter importance. The fault weights may change after the circuit breaker ages. It is impossible to accurately predict the life span by still using the preset weights.

[0005] Therefore, it is necessary to provide a new circuit breaker life prediction system and method based on multi-source heterogeneous data fusion and dynamic weight correction to solve the above technical problems. Summary of the Invention

[0006] The main purpose of the present invention is to provide a circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction, which takes into account the influence of multiple factors on the circuit breaker life and greatly improves the accuracy of circuit breaker life prediction.

[0007] The present invention achieves the above-mentioned object through the following technical solution: a circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction, comprising the following steps: S1. Data collection: The collected data includes the circuit breaker's opening coil current, closing coil current, energy storage current, displacement mechanical waveform data, pressure mechanical waveform data, main circuit current, main circuit voltage, contact temperature, ambient temperature and humidity, and partial discharge data; S2. Data analysis: Data analysis includes opening analysis, closing analysis, energy storage analysis, contact analysis, environment analysis, and insulation analysis; and correspondingly obtains first analysis data, second analysis data, third analysis data, fourth analysis data, fifth analysis data, and sixth analysis data; S3. Weight Correction: Deeply mine the first, second, and fifth analysis data, and dynamically correct the weights of various faults. The corrected weights include a mechanical aging weight W1, a contact wear weight W2, an insulation failure weight W3, and an environmental corrosion health index HI4. W1, W2, W3, and HI4 all have their own initial values. S4, Fault Analysis: Re-analyze and classify the features of the first, second, third, fourth, and sixth analysis data. When a set warning condition is triggered, output a corresponding fault type warning. The fault type warning includes mechanical aging warning A, contact wear warning B, insulation failure warning C, and environmental corrosion warning D. S5. Life assessment: Use the mechanical aging health index HI1, contact wear health index HI2, insulation aging health index HI3, and environmental corrosion health index HI4 to describe various fault conditions respectively. According to the fault type warning obtained in step S4, adjust the health index of the fault type accordingly; normalize and fuse the adjusted health indexes of mechanical aging fault, contact wear fault, and insulation aging fault with the corrected weights obtained in step S3 to obtain a comprehensive health index CHI; then use the adjusted environmental corrosion health index HI4 to calculate the aging coefficient α, and finally calculate the remaining life RUL of the circuit breaker; where, The calculation formula of comprehensive health index CHI is as follows: ; The calculation formula of aging coefficient α is as follows: ; in,α 基准 is the basic aging rate factor, γ is the environmental sensitivity coefficient; The remaining service life (RUL) is calculated as follows: ; Among them, L 初始 is the design life of the circuit breaker.

[0008] Furthermore, the opening analysis includes analyzing the opening coil current, the displacement mechanical waveform data, the pressure mechanical waveform data and the main circuit current to obtain the first analysis data; the first analysis data includes the opening time, the opening speed, the opening distance, the opening overshoot, the peak value of each stage of the opening current, the opening coil current energy spectrum and the main circuit current value.

[0009] Furthermore, the closing analysis includes analyzing the closing coil current, the displacement mechanical waveform data, and the pressure mechanical waveform data to obtain the second analysis data; the second analysis data includes closing time, closing speed, closing distance, closing overshoot, closing bounce, closing times, peak values ​​of closing current at each stage, and closing coil current energy spectrum.

[0010] Furthermore, the energy storage analysis includes performing analysis using the energy storage current to obtain the third analysis data; the third analysis data includes current peak values ​​at each stage and an energy spectrum of the energy storage current.

[0011] Furthermore, the contact analysis includes using the displacement mechanical waveform data, the main circuit current, the main circuit voltage and the contact temperature to perform analysis to obtain the fourth analysis data; the fourth analysis data includes current and temperature rise characteristics and current harmonic analysis data during opening and closing.

[0012] Furthermore, the environmental analysis includes performing analysis using the environmental temperature and humidity to obtain the fifth analysis data; the fifth analysis data includes the environmental temperature, environmental humidity and environmental severity index.

[0013] Furthermore, the insulation analysis includes performing analysis using the main circuit voltage and the partial discharge data to obtain the sixth analysis data; the sixth analysis data includes discharge amount, discharge phase, and discharge frequency.

[0014] Furthermore, step S3 includes: If the closing action reaches the set number of times within the set time, it is judged as frequent closing and the mechanical aging weight W1 is increased; When the tripping action is caused by short circuit or overload, the contact wear weight W2 is increased; When the partial discharge exceeds the threshold value for a set number of times within the set time, the insulation failure weight W3 is increased; When the environmental severity index exceeds the threshold for a set period of time, the environmental corrosion health index HI4 is adjusted.

[0015] Furthermore, the dynamic weight correction algorithm in step S3 includes the following steps: S31. Determine whether the driving algorithm is cycle-driven, event-driven, or a partial discharge over-limit event. If it is cycle-driven, execute steps S32 to S34; if it is event-driven, execute steps S35 to S313; if it is a partial discharge over-limit event, execute steps S314 to S316; S32, reading historical data of the environmental degradation index; S33, determining whether the duration of the environmental severity index exceeding the threshold reaches a set duration, if so, proceeding to the next step, if not, ending the weight correction; S34: Determine that the environmental corrosion risk has increased and adjust the environmental corrosion health index HI4; S35. Determine whether the event is an opening event or a closing event. If it is a closing event, execute steps S36 to S39; if it is an opening event, execute steps S310 to S313. S36, reading the second analysis data obtained from the closing analysis; S37, extracting historical data of closing overshoot, closing bounce, and closing times; S38, determine whether the closing action is too frequent; S39. If it is determined that the closing action is too frequent, the weight of the mechanical failure of the opening and closing coil and the opening and closing mechanism is increased, that is, the mechanical aging weight W1 is increased; otherwise, the weight correction is terminated; S310, reading first analysis data obtained from the opening analysis; S311, extracting historical data of tripping overshoot and tripping current capacity; S312: Determine whether the tripping action is caused by a short circuit or an overload. If so, increase the weight of the contact electrical wear, i.e., increase the contact wear weight W2, and end the weight correction. If not, proceed to the next step. S313, accumulating the tripping current capacity, and increasing the weight of the contact mechanical wear according to the accumulated value of the tripping current capacity, that is, increasing the contact wear weight W2, and ending the weight correction; S314, reading sixth analysis data obtained by insulation analysis; S315, extracting historical data of discharge amount and discharge frequency; S316. Determine whether the number of partial discharge exceeding limits in the most recently set event period exceeds the set ratio in the previous set time period. If so, the partial discharge exceeding limit frequency is significantly increased, and the insulation failure weight W3 is increased; otherwise, end the weight correction.

[0016] Furthermore, when one of the following situations occurs, the mechanical aging warning A is issued: A1. During the opening and closing analysis, if the peak value of the opening and closing current is found to have changed by more than a preset ratio (e.g., 10%), further analysis is performed on the energy changes in the characteristic frequency bands of the opening and closing current energy spectrum to analyze the energy spectrum of the set frequency bands. A2. During the energy storage analysis, if the energy storage time deviation exceeds a preset ratio (e.g., 15%) and the peak value of the spring energy storage phase increases, further analyze the energy changes in the characteristic frequency bands of the energy storage current energy spectrum to analyze the energy spectrum of the set frequency bands. A3. During the analysis of the displacement and pressure waveform characteristics during the opening and closing analysis, it was found that the opening and closing time deviation exceeded the preset ratio (for example, 10%), and the change in the opening and closing speed was greater than the preset ratio (for example, 10%). In the contact analysis, the current harmonic data at the time of closing is analyzed. If the contact temperature rise after closing is abnormal or the current and temperature rise data do not match, a contact wear warning B is issued; In insulation analysis, if the number of consecutive partial discharge data exceeds the threshold within the set time, an insulation failure warning C will be issued; In the environmental analysis, if it is found that the environmental severity index exceeds the threshold for a long time, an environmental corrosion warning D is issued.

[0017] Furthermore, in step S5, the adjustment calculation formula of the environmental corrosion health index HI4 is: ; in, is the initial environmental corrosion health index; EAI is the environmental severity index; EAI_H is the environmental severity index threshold; T is the environmental severity index calculation period; K is the environmental attenuation rate constant.

[0018] Furthermore, step S5 includes the following steps: S51, obtaining initial values ​​of the mechanical aging health index HI1, the contact wear health index HI2, the insulation aging health index HI3, and the environmental corrosion health index HI4, and adjusting the health index of the fault type according to the fault type warning issued in step S4; S52. Normalize and fuse the mechanical aging health index HI1, the contact wear health index HI2, and the insulation aging health index HI3 to obtain a comprehensive health index CHI. The calculation formula is as follows: ; S53. Calculate the aging coefficient α according to the environmental corrosion conditions. The calculation formula is as follows: ; in, α 基准 is the basic aging rate factor, γ is the environmental sensitivity coefficient; S54. Calculate the remaining useful life RUL using the following formula: ; Among them, L 初始 The design life of the circuit breaker is.

[0019] Another object of the present invention is to provide a circuit breaker life prediction system based on multi-source heterogeneous data fusion and dynamic weight correction, which includes: Data acquisition module, which collects the circuit breaker's opening coil current, closing coil current, energy storage current, displacement mechanical waveform data, pressure mechanical waveform data, main circuit current, main circuit voltage, contact temperature, ambient temperature and humidity, and partial discharge data; an analysis module, which performs opening analysis, closing analysis, energy storage analysis, contact analysis, environmental analysis, and insulation analysis based on the data collected by the data collection module; and correspondingly obtains first analysis data, second analysis data, third analysis data, fourth analysis data, fifth analysis data, and sixth analysis data; an analysis result storage module, storing the first analysis data, the second analysis data, the third analysis data, the fourth analysis data, the fifth analysis data, and the sixth analysis data; A weight correction module performs in-depth mining on the first analysis data, the second analysis data, and the fifth analysis data, and performs dynamic weight correction on various faults to obtain a mechanical aging weight W1, a contact wear weight W2, an insulation failure weight W3, and an environmental corrosion health index HI4; A fault type judgment module performs feature reanalysis and classification on the data in the analysis result storage module, compares the features with the set warning trigger conditions, and outputs corresponding fault type warnings when the set warning conditions are triggered, including mechanical aging warning A, contact wear warning B, insulation failure warning C, and environmental corrosion warning D; The life calculation module adjusts the health index of the fault type according to the fault warning type output by the fault type judgment module; normalizes and fuses the adjusted health indexes of mechanical aging fault, contact wear fault and insulation aging fault with the correction weights output by the weight correction module to obtain a comprehensive health index CHI; then uses the adjusted environmental corrosion health index HI4 to calculate the aging coefficient α, and finally calculates the remaining life RUL of the circuit breaker; wherein, The calculation formula of comprehensive health index CHI is as follows: ; The calculation formula of aging coefficient α is as follows: ; in, α 基准 is the basic aging rate factor, γ is the environmental sensitivity coefficient; The remaining service life (RUL) is calculated as follows: ; Among them, L 初始 is the design life of the circuit breaker.

[0020] Furthermore, the data acquisition module is electrically connected to a plurality of Hall sensors, a displacement sensor, a pressure sensor, and a voltage transformer; wherein the data collected by the plurality of Hall sensors include the opening coil current, the closing coil current, the energy storage current, and the main circuit current; the data collected by the displacement sensor is the displacement mechanical waveform data; the data collected by the pressure sensor is the pressure mechanical waveform data; and the data collected by the voltage transformer is the main circuit voltage; The analysis module, the analysis result storage module, the weight correction module, the fault type judgment module and the life calculation module together constitute a data processing module; the data processing module is electrically connected to a contact temperature acquisition module, a partial discharge detection module and an ambient temperature and humidity detection module.

[0021] Compared with the prior art, the circuit breaker life prediction system and method based on multi-source heterogeneous data fusion and dynamic weight correction of the present invention has the following beneficial effects: By collecting various data such as the circuit breaker's opening and closing coil current, pressure, displacement, main circuit current, voltage, contact temperature, ambient temperature, and partial discharge data, the collected data is fused and analyzed through a multi-source heterogeneous data fusion algorithm to obtain further analysis results; based on the analysis results, a dynamic weight correction algorithm is executed to correct and adjust the weights of different types of faults, making the life prediction model more accurate; after each circuit breaker action, a fault identification algorithm is executed to calculate the circuit breaker life based on the weight and the health index corresponding to each fault, greatly improving the accuracy of circuit breaker service life prediction and the safety of power equipment use; it is suitable for the full life cycle health management of smart substation circuit breakers, status assessment of high-voltage transmission system circuit breakers, and fault warning of new energy stations (wind power / photovoltaic) circuit breakers. Specifically: (1) Fusion of multi-source heterogeneous data allows for fault analysis and life prediction of circuit breakers through multiple data sources, improving the accuracy of analysis and prediction. (2) Compared with the single fault analysis function, the life prediction function quantifies the health status of the circuit breaker, allowing users to more intuitively understand the health status of the circuit breaker; (3) The introduction of partial discharge faults improves the circuit breaker fault types; (4) Environmental corrosion faults are introduced to correct the aging rate of the circuit breaker according to the harsh environmental conditions, thereby improving the accuracy of life prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flow chart of an embodiment of the present invention; Figure 2 This is a data flow diagram for multi-source heterogeneous data fusion in an embodiment of the present invention; Figure 3 This is a flow chart of a dynamic weight correction algorithm according to an embodiment of the present invention; Figure 4 This is a block diagram of a fault type warning and life prediction algorithm in an embodiment of the present invention; Figure 5 2 is a system block diagram in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] Example 1: Please refer to Figures 1-4 This embodiment is a circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction, which includes the following steps: S1. Data collection: The collected data include the circuit breaker's opening coil current, closing coil current, energy storage current, displacement mechanical waveform data, pressure mechanical waveform data, main circuit current, main circuit voltage, contact temperature, ambient temperature and humidity, and partial discharge data.

[0024] S2. Data analysis: Data analysis includes opening analysis, closing analysis, energy storage analysis, contact analysis, environmental analysis, and insulation analysis; among which: The trip analysis includes performing a trip analysis using the trip coil current, the displacement mechanical waveform data, the pressure mechanical waveform data, and the main circuit current to obtain first analysis data. The first analysis data includes trip time, initial trip speed, trip distance, trip overshoot, peak values ​​at each stage of the trip current, trip coil current energy spectrum, and main circuit current value.

[0025] When the system collects the primary trip coil current, it assumes the circuit breaker has performed a primary trip and begins trip analysis. The purpose of trip analysis is to generate secondary analysis results from the collected raw data. The system extracts the trip coil current, displacement mechanical waveform, pressure mechanical waveform, and main circuit current waveform from the collected primary trip coil current waveform to begin trip analysis. Based on the trip coil current waveform and displacement mechanical waveform, the system calculates the trip time, initial trip speed, trip distance, trip overshoot, peak values ​​of the trip coil current at each stage, the trip coil current energy spectrum, and the main circuit current value during the trip phase.

[0026] The various data in the first analysis data are explained as follows: The tripping time refers to the time interval from the energization of the tripping coil to the complete extinguishing of the arc when the contacts of all poles separate (the main circuit current returns to 0).

[0027] The initial separation speed refers to the average speed at the moment of contact separation, usually the average speed within 0.01 seconds after separation.

[0028] The opening distance refers to the minimum distance between the moving and static contacts when the circuit breaker is in the opening state.

[0029] Opening overshoot refers to the travel of the contacts exceeding the rated opening distance after the opening operation.

[0030] Peak values ​​of the tripping coil current at each stage - The tripping coil current usually has two peaks, and the peak values ​​of these two peaks are recorded.

[0031] The trip coil current energy spectrum is obtained by Fourier transforming the 20-2000Hz frequency domain component value of the current waveform.

[0032] The closing analysis includes performing a closing analysis using the closing coil current, the displacement mechanical waveform data, and the pressure mechanical waveform data to obtain second analysis data. The second analysis data includes closing time, closing speed, closing distance, closing overshoot, closing bounce, closing times, peak values ​​of closing current at each stage, and closing coil current energy spectrum.

[0033] The energy storage analysis includes performing energy storage analysis using the energy storage current to obtain third analysis data. The third analysis data includes current peak values ​​at each stage and energy spectrum of the energy storage current.

[0034] The contact analysis includes performing contact analysis using the displacement mechanical waveform data, the main circuit current, the main circuit voltage, and the contact temperature to obtain fourth analysis data. The fourth analysis data includes current and temperature rise characteristics and current harmonic analysis data during opening and closing.

[0035] Contact analysis is performed when opening and closing the circuit breaker. During opening, the harmonic components of the main circuit at the moment of opening are analyzed and calculated (the higher harmonic values ​​of the current are obtained using Fourier transform); during closing, the current value and contact temperature rise over a period of time are recorded.

[0036] The environmental analysis includes performing environmental analysis using the environmental temperature and humidity to obtain fifth analysis data. The fifth analysis data includes environmental temperature, environmental humidity, and an environmental severity index.

[0037] In environmental analysis, ambient temperature and humidity are recorded chronologically. The environmental severity index (ESI) is calculated and recorded every set time (for example, 30 minutes). The ESI is calculated based on the average temperature and humidity over the past 30 minutes. The higher the temperature and humidity, the more severe the environment, and the higher the ESI.

[0038] The insulation analysis includes performing insulation analysis using the main circuit voltage and the partial discharge data to obtain sixth analysis data. The sixth analysis data includes discharge amount, discharge phase, and discharge frequency.

[0039] Whenever the partial discharge amount is detected to be greater than the set threshold (for example, 5pC), insulation analysis is started, and the discharge amount and discharge phase (the phase value of the main circuit voltage during discharge) are recorded. Discharge frequency monitoring is started and the number of discharges within a period of time is recorded (the discharge frequency is the number of discharges within a period of time).

[0040] S3. Weight Correction: Deeply mine the first, second, and fifth analysis data to correct the weights of the impact of various faults on the life of the circuit breaker. The weights include mechanical aging weight W1, contact wear weight W2, insulation failure weight W3, and environmental corrosion health index HI4. The weight correction algorithm is divided into two triggering modes: cycle-driven and event-driven.

[0041] Periodic drive refers to the periodic execution of the weight correction algorithm at a set time. After entering the periodic drive branch, the historical data of the environmental severity index stored in the database is read. If the environmental severity index exceeds the threshold for a long time, it means that the circuit breaker has been working in a harsh environment for a long time. The risk of corrosion of circuit breaker components increases, and the environmental corrosion health index is adjusted.

[0042] Event-driven means that the weight correction algorithm will be triggered when a corresponding event occurs in the circuit breaker. The events are divided into opening events, closing events and partial discharge exceeding limit events.

[0043] After a tripping event occurs, the tripping feature data in the database is read. If it is identified that the tripping is caused by an overload or short circuit in the main circuit, the weight of the electrical wear of the contacts is greatly increased (the current is very large during a short circuit or overload, and if the tripping is performed, the contacts will arc severely, causing electrical erosion of the contacts). Otherwise, the tripping current capacity is increased, and the weight of the mechanical wear of the contacts is increased according to the accumulated value of the tripping current.

[0044] After a closing event occurs, the closing feature data in the database is read to determine whether the closing has occurred frequently recently. If the closing has occurred frequently, the mechanical fault weight of the opening and closing coil and the opening and closing mechanism is increased.

[0045] After a partial discharge exceeding limit event occurs, the partial discharge data in the database is read. If the partial discharge exceeding limit frequency increases significantly, the insulation failure weight is increased.

[0046] The weight correction algorithm specifically includes the following steps: S31. Determine whether the driving algorithm is periodic driving, opening / closing event driving, or partial discharge over-limit event driving. If it is periodic driving, execute steps S32 to S34. If it is opening / closing event driving, execute steps S35 to S313. If it is partial discharge over-limit event driving, execute steps S314 to S316. S32, reading historical data of the environmental degradation index; S33: Determine whether the environmental severity index exceeds the threshold for a set period of time (for example, the duration of the environmental severity index exceeding the threshold exceeds one day). If so, proceed to the next step; if not, end the weight adjustment. S34: If the environmental corrosion risk is determined to be increased, the environmental corrosion health index HI4 is adjusted; S35. Determine whether the event is an opening event or a closing event. If it is a closing event, execute steps S36 to S39; if it is an opening event, execute steps S310 to S313. S36, reading the second analysis data obtained from the closing analysis; S37, extracting historical data of closing overshoot, closing bounce, and closing times; S38. Determine whether the closing action is too frequent. The frequent determination rule is: if the number of operations in a day reaches one thousandth of the rated number, it is considered frequent. For example, if the rated number is 10,000 times, if it exceeds 10 times a day, it is considered frequent. S39. If it is determined that the closing action is too frequent, the weight of the mechanical failure of the opening and closing coil and the opening and closing mechanism is increased, that is, the mechanical aging weight W1 is increased; otherwise, the weight correction is terminated; S310, reading first analysis data obtained from the opening analysis; S311, extracting historical data of tripping overshoot and tripping current capacity (i.e., main circuit current value); S312: Determine whether the tripping action is caused by a short circuit or overload. If so, significantly increase the weight of contact electrical wear, i.e., increase the contact wear weight W2, and end the weight correction. If not, proceed to the next step. S313, accumulating the tripping current capacity, and increasing the weight of the contact mechanical wear according to the accumulated value of the tripping current capacity, that is, increasing the contact wear weight W2, and ending the weight correction; S314, reading sixth analysis data obtained by insulation analysis; S315, extracting historical data of discharge amount and discharge frequency; S316. Determine whether the number of partial discharge exceeding limits in the most recent set event segment exceeds the set ratio in the previous set time segment (for example, the number of partial discharge exceeding limits in the most recent event segment is 30% more than that in the previous time segment). If so, the partial discharge exceeding limit frequency is significantly increased, and the insulation failure weight W3 is increased; otherwise, end the weight correction.

[0047] The mechanical aging weight W1, contact wear weight W2, and insulation failure weight W3 all have their own initial weights, which are derived from long-term failure data statistics. For example, if a mechanical structure has 200 components and the failure rate of each individual component is 0.0005%, the overall failure rate of the mechanical structure is 0.0005% * 200 = 0.1%. If a contact structure has 18 components and the failure rate of each individual component is 0.0025%, the failure rate of the contact components is 0.045%. The insulation failure failure rate is 0.015%. Based on the failure rate of each component, the initial weights for each type of fault can be preset: mechanical aging weight W1 = (0.1) / (0.1 + 0.045 + 0.015) = 0.625, contact wear weight W2 = 0.28125, and insulation failure weight W3 = 0.09375.

[0048] When the environmental severity index exceeds the threshold for a long time, it will affect the environmental corrosion health index HI4, and the environmental corrosion health index HI4 needs to be corrected and adjusted. The environmental corrosion health index HI4 is used to calculate the subsequent aging coefficient.

[0049] Example of correction for severe environment: If the severe environment index exceeds the threshold for a long time (for example, it exceeds the threshold for more than 1 day) during a periodic event, the environmental corrosion health index is adjusted based on the duration of the severe environment (the time when the severe environment index is greater than the threshold) and the severe index value. The corrected environmental corrosion health index is recorded as HI4, and the correction formula is as follows: ; in, is the initial environmental corrosion health index, which takes a value of 1; EAI is the environmental severity index; EAI_H is the environmental severity index threshold, which is 0.7 in this embodiment (i.e., when EAI is greater than 0.7, it is considered severe); T is the environmental severity index calculation period, which is 0.0208 days (30 minutes) in this embodiment; K is the environmental attenuation rate constant, which is 0.002 in this embodiment. The environmental attenuation rate constant is defined as the health index HI4 returning to 0 after 500 days in a severe environment with (EAI-EAI_H)=1. The data is obtained through aging experiments.

[0050] An example of weight correction for a short-circuit trip: If the main circuit current I exceeds a set multiple (e.g., 5) of the rated current Ir during tripping, the trip is considered to be caused by a short circuit. Based on the allowable number of short-circuit trips N specified in the circuit breaker's datasheet (e.g., a 1000A rated circuit breaker is allowed to trip 30 times at 25kA), the weight is adjusted based on the cumulative number of short-circuit trips n, to a maximum of 2 times the initial weight and no more than 1. Due to the increase in contact wear weight W2, the weights for mechanical aging and insulation failure need to be reduced proportionally, so that all weights sum to 1.

[0051] ; S4. Fault Analysis: The first, second, third, fourth, and sixth analysis data are re-analyzed and classified. When a pre-set warning condition is triggered, a corresponding fault type warning is output. Fault type warnings include mechanical aging warning A, contact wear warning B, insulation failure warning C, and environmental corrosion warning D.

[0052] When one of the following situations occurs, the mechanical aging warning A is issued: A1. During the opening and closing analysis, if the peak value of the opening and closing current changes by more than 10%, further analysis is performed on the energy changes in the characteristic frequency bands of the opening and closing current energy spectrum to reveal the energy spectrum of the set frequency band (e.g., 50-200 Hz). A2. During the energy storage analysis, if the energy storage time deviation exceeds 15% and the peak value of the spring energy storage stage increases, further analysis is performed on the energy changes in the characteristic frequency band of the energy storage current energy spectrum to reveal the energy spectrum of the set frequency band. A3. In the opening and closing analysis, when analyzing the displacement and pressure waveform characteristics, it was found that the opening and closing time deviation exceeded 10%, and the speed change just after opening and closing was greater than 10%.

[0053] Example of mechanical aging warning: If the peak value of the opening and closing current changes by more than 10%, the energy spectrum of the opening and closing coil current is extracted, and the energy spectrum of this waveform is compared with the reference energy spectrum. If the amplitude of the low-frequency band (less than 200Hz) of the energy spectrum decreases by 15% as a whole, it may be spring fatigue. If the energy spectrum in the high-frequency band (1-2kHz) has a peak (such as a 50% increase in the 1.5k frequency component), it may be contact jamming. If the main lobe of the energy spectrum (that is, the center frequency of the part with an amplitude of more than 50%) deviates by more than 10%, it may be aging and wear of the connecting rod mechanism. There are many subdivisions of mechanical aging, which are collectively referred to as mechanical aging in this embodiment.

[0054] In the contact analysis, the current harmonic data at the time of closing is analyzed. If it is found that the contact temperature rise is abnormal after closing and the current and temperature rise data do not match, a contact wear warning B is issued.

[0055] Example of contact wear warning analysis: The system has a built-in relationship between main circuit current and temperature rise based on the circuit breaker type. For example, the temperature rise is 5°C for a main circuit current of 100A, 10°C for 200A, 20°C for 300A, and 50°C for 500A. If the current main circuit current is 150A, the contact temperature rise should be greater than 5°C and less than 10°C, but the measured contact temperature rise is 15°C. The system then analyzes the main circuit current harmonics at closing. The calculated harmonic values ​​are compared with the current harmonic values ​​at the initial stage of circuit breaker operation, focusing on harmonics above 1kHz. If the harmonics increase abnormally, it is determined to be contact wear.

[0056] In insulation analysis, if it is found that the partial discharge data exceeds the threshold value for a set number of consecutive times within a short period of time (for example, it exceeds the threshold value three times in a row), an insulation failure warning C is issued.

[0057] In the environmental analysis, if it is found that the environmental severity index exceeds the threshold for a long time, an environmental corrosion warning D is issued.

[0058] This embodiment classifies circuit breaker fault types into four types, namely mechanical aging, contact wear, insulation failure, and environmental corrosion. Environmental corrosion will increase the probability of the other three faults occurring. Figure 4 The judgment basis of four types of faults are given respectively.

[0059] Mechanical aging refers to the aging of all mechanical components, including the opening and closing operating mechanism, energy storage mechanism, and opening and closing action mechanism. The aging of the opening and closing operating mechanism can be judged by the opening and closing current. The current can be divided into four stages with a total of two peaks. Compared with the standard waveform, if the peak change is greater than 10%, the current energy spectrum is further analyzed. Mechanical vibration or mechanical jam information can be extracted from the current energy spectrum. For example, the energy spectrum of 50~200Hz can be used to determine mechanical vibration.

[0060] The aging of the energy storage mechanism can be judged by analyzing the energy storage current. If the energy storage time deviation is greater than 15% and the peak value of the spring energy storage stage increases, it indicates that the energy storage mechanism has a jamming problem and the mechanism is aging.

[0061] The opening and closing action mechanism can be judged by combining the opening and closing coil current, opening and closing displacement and pressure waveform. By calculating the opening and closing time and the opening and closing speed, if the opening and closing time deviation is greater than 10% and the opening and closing speed deviation is greater than 10%, it can be judged that the opening and closing mechanism is aging.

[0062] Contact wear can be determined by analyzing the contact temperature rise and main circuit current. The system establishes a database of contact temperature rise and main circuit current characteristics based on historical data. For example, when the current is 400A, the contact temperature rise is 10°C, and 500A corresponds to a temperature rise of 15°C. If the circuit current is 400A and the contact temperature rise reaches 15°C, the current harmonic data at the time of closing is further analyzed. If its high-frequency harmonic component increases, it can be confirmed as contact wear.

[0063] Insulation failure can be detected by analyzing partial discharge data. If the partial discharge exceeds the threshold three times in a row within a short period of time, it is considered to be an insulation failure.

[0064] Environmental corrosion is not a physical fault of the circuit breaker, but it will accelerate mechanical aging, contact wear and insulation failure, and users need to pay attention to it.

[0065] S5. Life assessment: Based on the weight obtained in step S3 and the fault type obtained in step S4, the remaining life of the circuit breaker is assessed. Specifically, the following steps are included: S51. Use the health index HI to describe various fault conditions, including the mechanical aging health index HI1, the contact wear health index HI2, the insulation aging health index HI3, and the environmental corrosion health index HI4. The initial value of each of the above health indices is 1. According to the fault type warning issued in step S4, the health index of the corresponding fault type is adjusted. For example, if a mechanical aging warning and a contact wear warning are issued in step S4, the mechanical aging health index HI1 and the contact wear health index HI2 are correspondingly reduced. The index adjustment range is determined based on the specific cause of the fault and experience.

[0066] S52. Weighted fusion of each health index is performed and normalized to obtain a comprehensive health index, denoted as CHI. The calculation formula is as follows: ; S53. Calculate the aging coefficient α according to the environmental corrosion conditions. The calculation formula is as follows: ; in, α 基准It represents the basic aging rate factor, with a default value of 1.0 (which can be obtained by fitting the historical data of the circuit breaker). HI4 is the environmental corrosion health index. γ is the environmental sensitivity coefficient, ranging from 0.3 to 1.0 (which can be determined according to the corrosion resistance level of the circuit breaker. If it is a coastal environment, the coefficient value can be increased due to the influence of salt spray).

[0067] S54. Calculate the remaining useful life RUL using the following formula: ; Among them, L 初始 The design life of the circuit breaker is (for example 20 years or 10,000 operations).

[0068] Please refer to Figure 5 This embodiment also provides a circuit breaker life prediction system, which includes a data acquisition module and a data processing module; the data processing module includes an analysis module, an analysis result storage module, a weight correction module, a fault type judgment module, and a life calculation module. The data acquisition module includes a multi-channel analog-to-digital conversion module and an FPGA data acquisition module electrically connected to the multi-channel analog-to-digital conversion module. The data acquisition module is electrically connected to a number of Hall sensors, displacement sensors, pressure sensors, and voltage transformers, wherein the data collected by the Hall sensors include the opening coil current, the closing coil current, the energy storage current, and the main circuit current; the data collected by the displacement sensor is the displacement mechanical waveform data; the data collected by the pressure sensor is the pressure mechanical waveform data; and the data collected by the voltage transformer is the main circuit voltage.

[0069] This embodiment uses multiple sensors to comprehensively collect various data from the circuit breaker. Hall current sensors collect the opening and closing coil current and energy storage current. By installing laser displacement sensors and pressure sensors on the circuit breaker, displacement and pressure data during the opening and closing stages are collected. Hall current sensors are used to replace traditional AC current transformers to collect the main circuit current. Due to the need to perform high-frequency harmonic analysis on the current, Hall current sensors with higher bandwidth are required to collect the main circuit current. The main circuit voltage is collected through a voltage transformer.

[0070] All of the above sensors output analog signals, which are converted into digital signals through the multi-channel analog-to-digital conversion module. Since many waveforms require high-frequency harmonic analysis, the sampling frequency needs to reach above 200kHz.

[0071] Since there are more than a dozen analog-to-digital conversion channels and the sampling rate is as high as over 200kHz, traditional processors cannot simultaneously take into account data reading and data processing. Therefore, this embodiment uses an FPGA data acquisition module, which is responsible for reading data from the analog-to-digital conversion module and sending the data to the data processing module through a high-speed bus.

[0072] The data processing module is electrically connected to a contact temperature acquisition module, a partial discharge detection module, and an ambient temperature and humidity detection module.

[0073] The contact temperature acquisition module is used to collect the contact temperature and uses a wireless temperature sensor for measurement. The system receives the contact temperature data measured by the wireless temperature sensor through the wireless receiving module.

[0074] The partial discharge detection module and the ambient temperature and humidity detection module are both connected to the data processing module via a digital bus, and after measuring corresponding data, the data are sent to the data processing module via the bus.

[0075] The data processing module is responsible for processing all data, executing the above-mentioned circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction, accurately identifying various types of circuit breaker faults and predicting life.

[0076] The data acquisition module collects data and then uploads it to the analysis module. The analysis module performs fusion analysis on a plurality of heterogeneous data, that is, executes the data analysis step described in step S2.

[0077] The analysis result storage module is used to store analysis data obtained by the analysis module, including the first analysis data, the second analysis data, the third analysis data, the fourth analysis data, the fifth analysis data and the sixth analysis data.

[0078] The weight correction module is configured to implement the weight correction step described in step S3 and output the corrected weights of various faults, including the mechanical aging weight W1, the contact wear weight W2 and the insulation failure weight W3.

[0079] The fault type determination module is configured to implement the fault analysis step described in step S4, and output a corresponding fault type warning when a preset warning condition is triggered.

[0080] The life calculation module is configured to implement the life assessment step described in step S5, and adjust the mechanical aging health index HI1, contact wear health index HI2, insulation aging health index HI3 and environmental corrosion health index HI4 according to the corresponding fault type warning output by the fault type judgment module; combine the various fault weights output by the weight correction module, normalize the mechanical aging health index HI1, contact wear health index HI2, and insulation aging health index HI3, and then calculate the environmental corrosion aging coefficient α; then calculate the remaining life of the circuit breaker according to the life calculation formula, and output the life result.

[0081] For those skilled in the art, several variations and improvements can be made without departing from the inventive concept of the present invention, and all of these fall within the scope of protection of the present invention.

Claims

1. A circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction, characterized by: The following steps are involved: S1. Data collection: The collected data includes the circuit breaker's opening coil current, closing coil current, energy storage current, displacement mechanical waveform data, pressure mechanical waveform data, main circuit current, main circuit voltage, contact temperature, ambient temperature and humidity, and partial discharge data; S2. Data analysis: Data analysis includes opening analysis, closing analysis, energy storage analysis, contact analysis, environment analysis, and insulation analysis; and correspondingly obtains first analysis data, second analysis data, third analysis data, fourth analysis data, fifth analysis data, and sixth analysis data; S3. Weight Correction: Deeply mine the first, second, and fifth analysis data, and dynamically correct the weights of various faults. The corrected weights include mechanical aging weight W1, contact wear weight W2, insulation failure weight W3, and environmental corrosion health index HI4. W1, W2, W3, and HI4 all have their own initial values; S4, Fault Analysis: Re-analyze and classify the features of the first, second, third, fourth, and sixth analysis data. When a set warning condition is triggered, output a corresponding fault type warning. The fault type warning includes mechanical aging warning A, contact wear warning B, insulation failure warning C, and environmental corrosion warning D. S5. Life assessment: Use the mechanical aging health index HI1, contact wear health index HI2, insulation aging health index HI3, and environmental corrosion health index HI4 to describe various fault conditions respectively. According to the fault type warning obtained in step S4, adjust the health index of the fault type accordingly; normalize and fuse the adjusted health indexes of mechanical aging fault, contact wear fault, and insulation aging fault with the corrected weights obtained in step S3 to obtain a comprehensive health index CHI; then use the adjusted environmental corrosion health index HI4 to calculate the aging coefficient α, and finally calculate the remaining life RUL of the circuit breaker; where, The calculation formula of comprehensive health index CHI is as follows: ； The calculation formula of aging coefficient α is as follows: ； in, α 基准 is the basic aging rate factor, γ is the environmental sensitivity coefficient; The remaining service life (RUL) is calculated as follows: ； Among them, L 初始 is the design life of the circuit breaker.

2. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1 is characterized by: The tripping analysis includes analyzing the tripping coil current, the displacement mechanical waveform data, the pressure mechanical waveform data and the main circuit current to obtain the first analysis data; the first analysis data includes the tripping time, the initial tripping speed, the tripping distance, the tripping overshoot, the peak values ​​of the tripping current at each stage, the tripping coil current energy spectrum and the main circuit current value.

3. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1 is characterized by: The closing analysis includes analyzing the closing coil current, the displacement mechanical waveform data, and the pressure mechanical waveform data to obtain the second analysis data; the second analysis data includes closing time, closing speed, closing distance, closing overshoot, closing bounce, closing times, peak values ​​of closing current at each stage, and closing coil current energy spectrum.

4. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1 is characterized by: The energy storage analysis includes performing analysis using the energy storage current to obtain the third analysis data; the third analysis data includes the current peak value of each stage and the energy spectrum of the energy storage current.

5. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1 is characterized by: The contact analysis includes using the displacement mechanical waveform data, the main circuit current, the main circuit voltage and the contact temperature to analyze and obtain the fourth analysis data; the fourth analysis data includes current and temperature rise characteristics and current harmonic analysis data during opening and closing.

6. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1 is characterized by: The environmental analysis includes performing analysis using the environmental temperature and humidity to obtain the fifth analysis data; the fifth analysis data includes the environmental temperature, environmental humidity, and an environmental severity index.

7. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1 is characterized by: The insulation analysis includes performing analysis using the main circuit voltage and the partial discharge data to obtain the sixth analysis data; the sixth analysis data includes discharge amount, discharge phase, and discharge frequency.

8. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1 is characterized by: Step S3 includes: If the closing action reaches the set number of times within the set time, it is judged as frequent closing and the mechanical aging weight W1 is increased; When the tripping action is caused by short circuit or overload, the contact wear weight W2 is increased; When the partial discharge exceeds the threshold value for a set number of times within the set time, the insulation failure weight W3 is increased; When the environmental severity index exceeds the threshold for a set period of time, the environmental corrosion health index HI4 is adjusted.

9. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1 or 8, characterized in that: The dynamic weight correction algorithm in step S3 includes the following steps: S31. Determine whether the driving algorithm is cycle-driven, event-driven, or a partial discharge over-limit event. If it is cycle-driven, execute steps S32 to S34; if it is event-driven, execute steps S35 to S313; if it is a partial discharge over-limit event, execute steps S314 to S316; S32, reading historical data of the environmental degradation index; S33, determining whether the duration of the environmental severity index exceeding the threshold reaches a set duration, if so, proceeding to the next step, if not, ending the weight correction; S34: Determine that the environmental corrosion risk has increased and adjust the environmental corrosion health index HI4; S35. Determine whether the event is an opening event or a closing event. If it is a closing event, execute steps S36 to S39; if it is an opening event, execute steps S310 to S313. S36, reading the second analysis data obtained from the closing analysis; S37, extracting historical data of closing overshoot, closing bounce, and closing times; S38, determine whether the closing action is too frequent; S39. If it is determined that the closing action is too frequent, the weight of the mechanical failure of the opening and closing coil and the opening and closing mechanism is increased, that is, the mechanical aging weight W1 is increased; otherwise, the weight correction is terminated; S310, reading first analysis data obtained from the opening analysis; S311, extracting historical data of tripping overshoot and tripping current capacity; S312: Determine whether the tripping action is caused by a short circuit or an overload. If so, increase the weight of the contact electrical wear, i.e., increase the contact wear weight W2, and end the weight correction. If not, proceed to the next step. S313, accumulating the tripping current capacity, and increasing the weight of the contact mechanical wear according to the accumulated value of the tripping current capacity, that is, increasing the contact wear weight W2, and ending the weight correction; S314, reading sixth analysis data obtained by insulation analysis; S315, extracting historical data of discharge amount and discharge frequency; S316. Determine whether the number of partial discharge exceeding limits in the most recently set event period exceeds the set ratio in the previous set time period. If so, the partial discharge exceeding limit frequency is significantly increased, and the insulation failure weight W3 is increased; otherwise, end the weight correction.

10. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1, characterized in that: When one of the following situations occurs, the mechanical aging warning A is issued: A1. During the opening and closing analysis, if the peak value of the opening and closing current is found to have changed by more than a preset ratio, further analysis will be conducted on the energy changes in the characteristic frequency bands of the opening and closing current energy spectrum to analyze the energy spectrum of the set frequency bands. A2. During the energy storage analysis, if the energy storage time deviation exceeds the preset ratio and the peak value of the spring energy storage stage increases, further analyze the energy changes in the characteristic frequency band of the energy spectrum of the energy storage current to analyze the energy spectrum of the set frequency band; A3. During the analysis of the displacement and pressure waveform characteristics during the opening and closing phases, it was found that the opening and closing time deviation exceeded the preset ratio, and the change in the speed immediately after opening and closing was greater than the preset ratio. In the contact analysis, the current harmonic data at the time of closing is analyzed. If the contact temperature rise after closing is abnormal or the current and temperature rise data do not match, a contact wear warning B is issued; In insulation analysis, if the number of consecutive partial discharge data exceeds the threshold within the set time, an insulation failure warning C will be issued; In the environmental analysis, if it is found that the environmental severity index exceeds the threshold for a long time, an environmental corrosion warning D is issued.

11. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1 is characterized in that: The adjustment calculation formula of the environmental corrosion health index HI4 is: ； in, is the initial environmental corrosion health index; EAI is the environmental severity index; EAI_H is the environmental severity index threshold; T is the environmental severity index calculation period; K is the environmental attenuation rate constant.

12. The circuit breaker life prediction method based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 1 or 11, characterized in that: Step S5 includes the following steps: S51, obtaining initial values ​​of the mechanical aging health index HI1, the contact wear health index HI2, the insulation aging health index HI3, and the environmental corrosion health index HI4, and adjusting the health index of the fault type according to the fault type warning issued in step S4; S52. Normalize and fuse the mechanical aging health index HI1, the contact wear health index HI2, and the insulation aging health index HI3 to obtain a comprehensive health index CHI. The calculation formula is as follows: ； S53. Calculate the aging coefficient α based on the environmental corrosion health index HI4. The calculation formula is as follows: ； in, α 基准 is the basic aging rate factor, γ is the environmental sensitivity coefficient; S54. Calculate the remaining useful life RUL using the following formula: ； Among them, L 初始 The design life of the circuit breaker is.

13. A circuit breaker life prediction system based on multi-source heterogeneous data fusion and dynamic weight correction, characterized by: It includes: Data acquisition module, which collects the circuit breaker's opening coil current, closing coil current, energy storage current, displacement mechanical waveform data, pressure mechanical waveform data, main circuit current, main circuit voltage, contact temperature, ambient temperature and humidity, and partial discharge data; an analysis module, which performs opening analysis, closing analysis, energy storage analysis, contact analysis, environmental analysis, and insulation analysis based on the data collected by the data collection module; and correspondingly obtains first analysis data, second analysis data, third analysis data, fourth analysis data, fifth analysis data, and sixth analysis data; an analysis result storage module, storing the first analysis data, the second analysis data, the third analysis data, the fourth analysis data, the fifth analysis data, and the sixth analysis data; A weight correction module performs in-depth mining on the first analysis data, the second analysis data, and the fifth analysis data, and performs dynamic weight correction on various faults to obtain a mechanical aging weight W1, a contact wear weight W2, an insulation failure weight W3, and an environmental corrosion health index HI4; A fault type judgment module performs feature reanalysis and classification on the data in the analysis result storage module, compares the features with the set warning trigger conditions, and outputs corresponding fault type warnings when the set warning conditions are triggered, including mechanical aging warning A, contact wear warning B, insulation failure warning C, and environmental corrosion warning D; The life calculation module adjusts the health index of the fault type according to the fault warning type output by the fault type judgment module; normalizes and fuses the adjusted health indexes of mechanical aging fault, contact wear fault and insulation aging fault with the correction weights output by the weight correction module to obtain a comprehensive health index CHI; then uses the adjusted environmental corrosion health index HI4 to calculate the aging coefficient α, and finally calculates the remaining life RUL of the circuit breaker; wherein, The calculation formula of comprehensive health index CHI is as follows: ； The calculation formula of aging coefficient α is as follows: ； in, α 基准 is the basic aging rate factor, γ is the environmental sensitivity coefficient; The remaining service life (RUL) is calculated as follows: ； Among them, L 初始 is the design life of the circuit breaker.

14. The circuit breaker life prediction system based on multi-source heterogeneous data fusion and dynamic weight correction according to claim 13, characterized in that: The data acquisition module is electrically connected to a plurality of Hall sensors, a displacement sensor, a pressure sensor, and a voltage transformer; wherein the data collected by the plurality of Hall sensors include the opening coil current, the closing coil current, the energy storage current, and the main circuit current; the data collected by the displacement sensor is the displacement mechanical waveform data; the data collected by the pressure sensor is the pressure mechanical waveform data; and the data collected by the voltage transformer is the main circuit voltage; The analysis module, the analysis result storage module, the weight correction module, the fault type judgment module and the life calculation module together constitute a data processing module; the data processing module is electrically connected to a contact temperature acquisition module, a partial discharge detection module and an ambient temperature and humidity detection module.

Images