Switching device performance aging intelligent diagnosis method and device based on multi-parameter fusion

Abstract

The application discloses a switch device performance aging intelligent diagnosis method and device based on multi-parameter fusion, relates to the technical field of power equipment abnormal defect diagnosis, and solves the problem of low precision in the prior art that switch devices rely on single temperature and humidity threshold or single gas parameter for aging judgment. The method comprises the following steps: collecting the temperature and humidity inside and outside the equipment and the internal gas characteristic parameters; constructing multi-channel standardized monitoring data through time synchronization, filtering and denoising and abnormal elimination processing; calculating the absolute humidity difference inside and outside the cabinet, constructing a humidity expectation model and obtaining standardized residual error; performing offset detection and trend smoothing processing on the standardized residual error to obtain cumulative quantity and smoothed trend quantity, fusing the two to calculate an offset index; and outputting a moisture penetration risk score, and triggering an early warning when the moisture penetration risk score and the characteristic gas concentration double indexes exceed the threshold. The application realizes real-time high-precision evaluation of the aging state of the equipment, effectively identifies early hidden dangers, and improves the safety and reliability of the power system.

Description

Intelligent Diagnosis Method and Device for Performance Aging of Switchgear Based on Multi-parameter Fusion Technical Field

[0001] This invention relates to the field of power equipment abnormality and defect diagnosis technology, specifically to a method and device for intelligent diagnosis of switchgear performance aging based on multi-parameter fusion. Background Technology

[0002] With the continuous growth of urban power load and the increasing demands on power grid reliability, the safe operation of switchgear, as a key control and protection unit in the power system, has become a core aspect of operation and maintenance. During long-term operation, switchgear is susceptible to condensation, gas leakage, and insulation degradation due to various factors such as changes in environmental humidity, aging of insulation materials, and decreased sealing performance of the casing. This can lead to safety hazards such as discharge and short circuits, directly threatening the stability of the power system.

[0003] Currently, the diagnosis of aging performance in switchgear mainly relies on three technical methods: ultra-high frequency partial discharge detection, ultrasonic detection, and transient ground voltage detection. All of these methods involve capturing instantaneous electrical / acoustic signals, assessing the equipment's condition based on the discharge intensity at a specific moment. However, a long-standing misconception in this field exists: the assumption that the detected discharge signal intensity is directly equivalent to the severity of aging, while ignoring the dynamic modulation effect of environmental parameters on the discharge physical process.

[0004] In fact, humidity is a direct factor inducing surface discharge in insulation, but its nature is temperature-regulated. Inside relatively enclosed switchgear, temperature changes not only alter the saturated moisture content of the air but also affect the condensation rate and the adsorption / desorption kinetics of moisture on the insulating material surface. Especially when switchgear is placed in a computer room with significant temperature fluctuations, the coupling relationship between temperature and humidity inside and outside the cabinet exhibits a highly nonlinearity, leading to a situation where the apparent "excessive humidity" is actually a temperature-driven imbalance in moisture migration. Traditional monitoring methods fail to reveal this causal chain, merely judging humidity thresholds or capturing instantaneous discharges in isolation, resulting in the following technical difficulties:

[0005] (1) Transient detection cannot reflect the cumulative effect of the environment. The intensity of the discharge signal itself is strongly correlated with real-time temperature and humidity: in high humidity environments, even if the amplitude of the discharge signal is low, its erosion rate and failure risk to the insulating material will increase significantly; conversely, high-intensity discharge signals in low humidity environments may only be occasional interference. Because the existing technology has not established a correlation model between temperature and humidity and discharge intensity, it is difficult to distinguish between the real aging process and short-term environmental disturbances, resulting in a high misjudgment rate and an inability to accurately quantify the severity of aging.

[0006] (2) The single threshold method lags behind early degradation. Existing diagnostic devices usually rely on the characteristic gas concentration or temperature and humidity change range exceeding the set threshold to trigger an alarm. However, when the gas concentration or humidity change range exceeds the set threshold, insulation aging or sealing failure has often entered the middle and late stages. Moreover, there is a lack of a collaborative analysis mechanism between temperature and humidity and gas parameters, which increases the probability of missing hidden defects and results in insufficient diagnostic sensitivity and accuracy.

[0007] (3) Difficulty in identifying microenvironmental imbalance in closed systems. As a relatively closed system, the change in internal gas composition of switchgear (such as increased humidity) is essentially a combined result of external moisture infiltration and internal material release after seal deterioration. Those skilled in the art are accustomed to treating temperature, humidity and gas as two independent physical quantities, without realizing that the two can form a causal chain of evidence, thus making it difficult to accurately identify the precursory characteristics of early insulation aging.

[0008] Therefore, current technology urgently needs an intelligent diagnostic method for the performance aging of switchgear to achieve early and accurate identification of insulation aging and sealing deterioration of switchgear. Summary of the Invention

[0009] The technical solution of the present invention is used to solve the problem that the aging judgment of switchgear in the prior art relies on a single temperature and humidity threshold or a single gas parameter, which results in low diagnostic accuracy.

[0010] This invention solves the above-mentioned technical problems through the following technical means: a smart diagnostic method for the performance aging of switching equipment based on multi-parameter fusion, comprising:

[0011] S1. Collect internal environmental parameters, external environmental parameters, and internal gas characteristic parameters of the switchgear;

[0012] S2. Perform data preprocessing operations on the collected internal environmental parameters, external environmental parameters and internal gas characteristic parameters to obtain multi-channel standardized monitoring data. The data preprocessing includes: time synchronization processing, filtering and noise reduction processing, abnormal data removal processing and format standardization processing.

[0013] S3. Calculate the absolute humidity difference between inside and outside the cabinet based on multi-channel standardized monitoring data, calculate the expected absolute humidity inside the cabinet, calculate the residual between the actual absolute humidity and the expected value, standardize the residual to obtain the standardized residual, perform offset detection on the standardized residual to obtain the cumulative amount, perform trend smoothing on the standardized residual to obtain the smoothed trend amount, and perform feature fusion on the obtained cumulative amount and smoothed trend amount to generate the offset index.

[0014] S4. Calculate the moisture leakage risk score based on the offset index, and combine it with the characteristic gas concentration corresponding to the internal gas characteristic parameters to generate a comprehensive diagnostic result of the aging degree of the switchgear.

[0015] Furthermore, the internal environmental parameters include: cabinet temperature and cabinet humidity; the external environmental parameters include: cabinet temperature and cabinet humidity; the internal gas characteristic parameters include: water vapor content, characteristic gas concentration and its changing trend; and the aging type of the switchgear includes at least one of: insulation aging, decreased sealing performance, condensation risk and environmental anomalies.

[0016] Furthermore, the formula for calculating the absolute humidity difference between the inside and outside of the cabinet is as follows:

[0017]

[0018]

[0019] in, This refers to absolute humidity. Temperature in Celsius Relative humidity, The absolute humidity inside the cabinet. The absolute humidity outside the cabinet. This represents the absolute humidity difference.

[0020] Furthermore, regarding the absolute humidity difference value The positive portion is normalized to obtain the normalized absolute humidity difference index. The calculation formula is as follows:

[0021]

[0022] in, This is the historical 95th percentile of the absolute humidity difference.

[0023] Furthermore, the formula for calculating the expected absolute humidity value inside the cabinet is as follows:

[0024]

[0025] in, These are the regression coefficients obtained through training with historical data. The heater is in working condition. The temperature inside the cabinet. The outside temperature of the cabinet. The absolute humidity outside the cabinet.

[0026] Furthermore, the step of detecting the offset of the standardized residual to obtain the cumulative amount specifically involves:

[0027] Robust standardization is performed on the residuals between the actual absolute humidity inside the cabinet and the expected value to obtain the standardized residuals. Build a positive CUSUM cumulative amount The update formula for the positive CUSUM cumulative amount is:

[0028]

[0029] in, For the first Standardized residuals corresponding to each monitoring session For the number of monitoring sessions, The initial conditions for the positive CUSUM accumulation. This is the offset tolerance constant.

[0030] Furthermore, the formula for calculating the offset index is as follows:

[0031]

[0032] in, For the first The offset index corresponding to the next monitoring. To smooth out the trend, This is a reference threshold obtained based on historical training periods. It is a Sigmoid type mapping function. These are the weighting coefficients.

[0033] Furthermore, the formula for calculating the seepage risk score is as follows:

[0034]

[0035] in, and These are the weighting coefficients for the normalized absolute humidity difference index and the offset index, respectively. + =1.

[0036] Furthermore, the comprehensive diagnostic results for the aging degree of the generated switching equipment are specifically as follows:

[0037] Based on the moisture leakage risk score and the concentration of characteristic gases, the status of switchgear is divided into different levels, and the determination rules for each level are as follows:

[0038] when and characteristic gas concentration When this occurs, it is considered a normal state;

[0039] when and characteristic gas concentration When this occurs, it is determined to be in a state of concern or inspection;

[0040] when and characteristic gas concentration At that time, it was determined to be in a state requiring planned maintenance;

[0041] when and characteristic gas concentration When the condition is determined to be severely aged or in a strongly abnormal state, an audible and visual alarm is triggered and uploaded to the monitoring system;

[0042] in, , and These are the preset first, second, and third threshold values ​​for moisture leakage risk scoring. This is a preset characteristic gas concentration threshold.

[0043] This invention also provides an intelligent diagnostic device for the performance aging of switching equipment based on multi-parameter fusion, comprising:

[0044] The temperature and humidity detection module is installed inside and outside the switch cabinet to collect the temperature and humidity inside and outside the cabinet in real time and output the collected data.

[0045] The gas detection module is installed inside the switchgear and is used to detect the characteristic parameters of the gas inside the equipment. The characteristic parameters of the gas inside the equipment include: water vapor content, characteristic gas concentration and its changing trend.

[0046] The data preprocessing module, connected to the temperature and humidity detection module and the gas detection module, is used to receive multi-source sensor data, perform time synchronization processing, filtering and noise reduction processing, abnormal data removal processing and format standardization processing, and output the processed feature data.

[0047] The algorithm calculation module, connected to the data preprocessing module, is used to receive the processed feature data and perform calculations on the absolute humidity inside and outside the cabinet, the difference in absolute humidity, the normalized absolute humidity difference index, the expected value of absolute humidity inside the cabinet, the residual value between the actual absolute humidity and the expected value, the standardized residual value, the cumulative amount, the smoothing trend amount, the offset index, and the moisture leakage risk score. Combined with the moisture leakage risk score and the characteristic gas concentration, a comprehensive diagnostic result of the aging degree of the switchgear is generated.

[0048] The output and early warning module is connected to the algorithm calculation module and is used to display real-time monitoring values ​​and diagnostic results. When the moisture infiltration risk score and characteristic gas concentration exceed the corresponding preset threshold, it outputs audible and visual alarm information and uploads the audible and visual alarm information to the external monitoring system.

[0049] The communication module is used to realize wired or wireless data communication between the intelligent diagnostic device and the external monitoring system, and to synchronize the diagnostic results, audible and visual alarm information and historical data to the monitoring backend.

[0050] The storage module, connected to the algorithm operation module, is used to store historical monitoring data, operation results, model parameters and operation logs, and is used for subsequent offline analysis and model self-learning updates.

[0051] The advantages of this invention are:

[0052] This invention constructs multi-source data by synchronously collecting internal and external temperature and humidity parameters and internal gas characteristic parameters of switchgear to distinguish between environmental interference and physical anomalies. Through time synchronization, filtering and noise reduction, anomaly removal, and format standardization, multi-channel standardized monitoring data is formed, eliminating sampling time differences and electromagnetic interference, and enhancing the model's anti-interference and robustness. A dynamic expectation model is constructed based on absolute humidity differences to achieve adaptive benchmark assessment of environmental changes. Residual analysis amplifies small deviations and integrates cumulative and smoothing trend values, accurately identifying gradual aging while suppressing instantaneous fluctuations. This significantly improves the sensitivity and trend extraction capability of slow-changing anomalies. Combining moisture infiltration risk scores and characteristic gas concentrations as dual indicators to classify equipment status levels, it achieves collaborative diagnosis of external moisture infiltration and internal insulation degradation. Early warning is triggered based on linkage thresholds, greatly reducing the false alarm rate caused by single factors. Ultimately, it achieves real-time, high-precision identification of early-stage hazards in switchgear, improving the economy and safety of power system operation and maintenance. Attached Figure Description

[0053] Figure 1 is a flowchart of the intelligent diagnostic method for performance aging of switching equipment based on multi-parameter fusion according to Embodiment 1 of the present invention;

[0054] Figure 2 is a trend diagram of characteristic gas concentration detected by the T1 sensor in Embodiment 1 of the present invention;

[0055] Figure 3 is a device module diagram of Embodiment 2 of the present invention;

[0056] Figure 4 is a schematic diagram of the host, sensor and device in Embodiment 2 of the present invention;

[0057] Figure 5 is an installation diagram of the monitoring device host and sensor in Embodiment 2 of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Example 1

[0060] As shown in Figure 1, the intelligent diagnostic method for the performance aging of switchgear based on multi-parameter fusion includes:

[0061] S1. Collect internal environmental parameters, external environmental parameters, and internal gas characteristic parameters of the switchgear.

[0062] Specifically, the internal environmental parameters include: cabinet temperature and humidity; the external environmental parameters include: cabinet temperature and humidity; and the internal gas characteristic parameters include: water vapor content, characteristic gas concentration, and their changing trends. Figure 2 shows the characteristic gas concentration trend detected by the T1 sensor.

[0063] S2. Perform data preprocessing operations on the collected internal environmental parameters, external environmental parameters, and internal gas characteristic parameters to obtain multi-channel standardized monitoring data. Data preprocessing includes: time synchronization processing, filtering and noise reduction processing, and outlier data removal processing.

[0064] Specifically, the temperature and humidity data inside and outside the cabinet, along with the internal gas characteristic parameters, are synchronized in time, filtered and denoised, and outlier data is removed. Valid data is cached in chronological order and uniformly converted into a standardized format for subsequent algorithm analysis. Multi-channel data is aligned to a unified time axis, and filtering algorithms are used to smooth short-term spikes. Outliers that clearly exceed physically reasonable ranges are removed or corrected.

[0065] S3. Calculate the absolute humidity difference between inside and outside the cabinet based on multi-channel standardized monitoring data, calculate the expected absolute humidity inside the cabinet, calculate the residual between the actual absolute humidity and the expected value, standardize the residual to obtain the standardized residual, perform offset detection on the standardized residual to obtain the cumulative amount, perform trend smoothing on the standardized residual to obtain the smoothed trend amount, and perform feature fusion on the obtained cumulative amount and smoothed trend amount to generate the offset index.

[0066] Specifically, 1) Absolute humidity and absolute humidity difference

[0067] To more accurately characterize the stability of water vapor content, this embodiment uses absolute humidity as the core indicator. Based on the collected temperature... and relative humidity Calculate the absolute humidity of the air inside and outside the cabinet. The formula for calculating absolute humidity is:

[0068]

[0069] in, This is absolute humidity, in units of... , Temperature in Celsius This represents the relative humidity. The absolute humidity inside the cabinet is then obtained accordingly. relative humidity outside the cabinet And calculate the absolute humidity difference:

[0070]

[0071] This difference is used to characterize the degree of moisture accumulation inside the cabinet relative to the outside.

[0072] 2) Normalization of absolute humidity difference

[0073] To characterize the absolute level of moisture content inside the cabinet relative to the outside at a given moment, the absolute humidity difference is used. The positive portion is normalized to obtain the normalized absolute humidity difference index. :

[0074]

[0075] in, This is the historical 95th percentile of the absolute humidity difference.

[0076] 3) Expected absolute humidity inside the cabinet

[0077] During the period when the cabinet door is closed and the heater is operating stably, this embodiment uses a multiple regression model to establish the expected relationship of the absolute humidity inside the cabinet. The absolute humidity outside the cabinet, the heater status, and the temperature difference between inside and outside the cabinet are used as inputs to estimate the expected value of the absolute humidity inside the cabinet. The calculation formula is as follows:

[0078]

[0079] in, These are the regression coefficients obtained through training with historical data. The heater is in working condition. The temperature inside the cabinet. The outside temperature of the cabinet. The absolute humidity outside the cabinet.

[0080] 4) Residual Standardization and Positive CUSUM Cumulative Offset

[0081] To enhance sensitivity to small deviations and suppress outlier interference, the residuals between the actual absolute humidity inside the cabinet and the expected value are robustly standardized to obtain standardized residuals. Build a positive CUSUM cumulative amount This is used to amplify anomalies of "continuous small deviations". The update formula for positive CUSUM is:

[0082]

[0083] in, For the first Standardized residuals corresponding to each monitoring session The number of monitoring times, t≥1 and rounded to the nearest integer. The initial conditions for the positive CUSUM accumulation. This is the offset tolerance constant.

[0084] 5) Construction of Trend Smoothing and Offset Index

[0085] To improve noise resistance, this embodiment introduces an exponentially weighted moving average (EWMA) to smooth the standardized residuals, resulting in a smoothed trend quantity. The CUSUM cumulative amount and the smoothed trend amount are fused to construct the offset index, and its calculation formula is as follows:

[0086]

[0087] in, To smooth out the trend, This is a reference threshold obtained based on historical training periods. It is a Sigmoid type mapping function. This is a weighting coefficient. OAI comprehensively reflects the cumulative shift of humidity inside the cabinet relative to the outside environment and the strength of its trend.

[0088] S4. Calculate the moisture leakage risk score based on the offset index, and combine it with the characteristic gas concentration corresponding to the internal gas characteristic parameters to generate a comprehensive diagnostic result of the aging degree of the switchgear.

[0089] Specifically, the formula for calculating the seepage risk score is as follows:

[0090]

[0091] in, and These are the weighting coefficients for the normalized absolute humidity difference index and the offset index, respectively. + =1.

[0092] Based on the moisture leakage risk score and the concentration of characteristic gases, the status of switchgear is divided into different levels, and the determination rules for each level are as follows:

[0093] when and characteristic gas concentration When this occurs, it is considered a normal state.

[0094] when and characteristic gas concentration When this occurs, it is determined to be in a state of concern or inspection.

[0095] when and characteristic gas concentration At that time, it was determined to be in a state requiring planned maintenance.

[0096] when and characteristic gas concentration When the condition is determined to be severely aged or in a highly abnormal state, an audible and visual alarm is triggered and uploaded to the monitoring system.

[0097] in, , and These are the preset first, second, and third threshold values ​​for moisture leakage risk scoring. This is a preset characteristic gas concentration threshold.

[0098] The aging types of switchgear include at least one of insulation aging, deterioration of sealing performance, condensation risk, and environmental anomalies.

[0099] 1) When the concentration of the characteristic gas reaches the characteristic gas concentration threshold ( When the insulation aging occurs, it is determined that insulation aging has occurred.

[0100] 2) When the moisture seepage risk score reaches the set warning range ( When this occurs, it can be determined that the sealing performance has deteriorated due to aging.

[0101] 3) When there is a large temperature difference between the inside and outside of the cabinet, the temperature inside the cabinet is lower than the temperature outside the cabinet, and there are significant changes in the humidity inside the cabinet, it is considered a risk of condensation.

[0102] 4) When the monitoring equipment detects a sudden change in the value, it is determined that the switching equipment environment is abnormal.

[0103] Example 2

[0104] Based on Embodiment 1, Embodiment 2 also provides an intelligent diagnostic device for the performance aging of switching equipment based on multi-parameter fusion, as shown in Figure 3, including:

[0105] The temperature and humidity detection module is installed inside and outside the switch cabinet to collect the temperature and humidity inside and outside the cabinet in real time and output the collected data.

[0106] The gas detection module is installed inside the switchgear to detect the characteristic parameters of the gas inside the equipment. These internal gas characteristic parameters include: water vapor content, characteristic gas concentration and their changing trends.

[0107] The data preprocessing module, connected to the temperature and humidity detection module and the gas detection module, is used to receive data from multiple sources of sensors, perform time synchronization processing, filtering and noise reduction processing, abnormal data removal processing, and format standardization processing, and output the processed feature data.

[0108] The algorithm calculation module, connected to the data preprocessing module, is used to receive the processed feature data and perform calculations on the absolute humidity inside and outside the cabinet, the absolute humidity difference, the normalized absolute humidity difference index, the expected absolute humidity inside the cabinet, the residual value between the actual absolute humidity and the expected value, the standardized residual value, the cumulative amount, the smoothing trend amount, the offset index, and the moisture leakage risk score. Combined with the moisture leakage risk score and the characteristic gas concentration, it generates a comprehensive diagnostic result of the aging degree of the switchgear.

[0109] Specifically, the formula for calculating the absolute humidity difference is:

[0110]

[0111]

[0112] in, This refers to absolute humidity. Temperature in Celsius Relative humidity, The absolute humidity inside the cabinet. The absolute humidity outside the cabinet. This represents the absolute humidity difference.

[0113] For absolute humidity difference The positive portion is normalized to obtain the normalized absolute humidity difference index. :

[0114]

[0115] in, This is the historical 95th percentile of the absolute humidity difference.

[0116] The formula for calculating the expected absolute humidity inside the cabinet is:

[0117]

[0118] in, These are the regression coefficients obtained through training with historical data. The heater is in working condition. The temperature inside the cabinet. The outside temperature of the cabinet. The absolute humidity outside the cabinet.

[0119] The standardized residuals are obtained by robustly standardizing the residuals between the actual absolute humidity and the expected value. Build a positive CUSUM cumulative amount The positive CUSUM update formula is:

[0120]

[0121] in, For the first Standardized residuals corresponding to each monitoring session For the number of monitoring sessions, The initial conditions for the positive CUSUM accumulation. This is the offset tolerance constant.

[0122] The CUSUM cumulative amount and the smoothing trend amount are fused using multiple parameters to construct the offset index, and its calculation formula is changed to:

[0123]

[0124] in, To smooth out the trend, This is a reference threshold obtained based on historical training periods. It is a Sigmoid type mapping function. These are the weighting coefficients.

[0125] The formula for calculating the moisture leakage risk score is:

[0126]

[0127] in, and These are the weighting coefficients for the normalized absolute humidity difference index and the offset index, respectively. + =1.

[0128] The output and early warning module, connected to the algorithm calculation module, is used to display real-time monitoring values ​​and diagnostic results. When the moisture infiltration risk score and characteristic gas concentration exceed the corresponding preset threshold, it outputs audible and visual alarm information and uploads the audible and visual alarm information to the external monitoring system.

[0129] Specifically, based on the moisture leakage risk score and the trend of characteristic gas concentration changes, the status of switchgear is divided into different levels, and the determination rules for each level are as follows:

[0130] when and characteristic gas concentration When this occurs, it is considered a normal state.

[0131] when and characteristic gas concentration When this occurs, it is determined to be in a state of concern or inspection.

[0132] when and characteristic gas concentration At that time, it was determined to be in a state requiring planned maintenance.

[0133] when and characteristic gas concentration When the condition is determined to be severely aged or in a highly abnormal state, an audible and visual alarm is triggered and uploaded to the monitoring system.

[0134] in, , and These are the preset first, second, and third threshold values ​​for moisture leakage risk scoring. This is a preset characteristic gas concentration threshold.

[0135] The communication module is used to enable wired or wireless data communication between the intelligent diagnostic device and the external monitoring system, and to synchronize diagnostic results, audible and visual alarm information and historical data to the monitoring backend.

[0136] The storage module, connected to the algorithm computation module, is used to store historical monitoring data, computation results, model parameters, and operation logs, and is used for subsequent offline analysis and model self-learning updates. Figure 4 shows an overall schematic diagram of the host, sensors, and device, and Figure 5 shows the installation diagram of the monitoring equipment host and sensors.

[0137] Example 3

[0138] An electronic device includes a memory and a processor, the memory being used to store a program that supports the processor in executing the intelligent diagnostic method for switching device performance aging based on multi-parameter fusion as described in Embodiment 1, the processor being configured to execute the program stored in the memory.

[0139] Example 4

[0140] A storage medium storing a computer program, which, when run by a processor, executes the steps of the intelligent diagnostic method for performance aging of switching equipment based on multi-parameter fusion in Embodiment 1.

[0141] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligent diagnosis of performance aging of switchgear based on multi-parameter fusion, characterized in that, include: S1. Collect internal environmental parameters, external environmental parameters, and internal gas characteristic parameters of the switchgear; S2. Perform data preprocessing operations on the collected internal environmental parameters, external environmental parameters, and internal gas characteristic parameters to obtain multi-channel standardized monitoring data. The data preprocessing includes: time synchronization processing, filtering and noise reduction processing, abnormal data removal processing, and format standardization processing. S3. Calculate the absolute humidity difference between the inside and outside of the cabinet based on the multi-channel standardized monitoring data, calculate the expected absolute humidity value inside the cabinet, calculate the residual value between the actual absolute humidity and the expected value, and standardize the residual value to obtain standardized residuals. Perform offset detection on the standardized residuals to obtain a cumulative amount, perform trend smoothing processing on the standardized residuals to obtain a smoothing trend amount, and perform feature fusion on the obtained cumulative amount and smoothing trend amount to generate an offset index. S4. Calculate the moisture leakage risk score based on the offset index, and combine it with the characteristic gas concentrations corresponding to the internal gas characteristic parameters to generate a comprehensive diagnostic result of the aging degree of the switchgear. The calculation formula for the moisture leakage risk score is: in, This is a normalized absolute humidity difference index. For the first The offset index corresponding to the next monitoring. and These are the weighting coefficients for the normalized absolute humidity difference index and the offset index, respectively. + =1; The comprehensive diagnostic result of the aging degree of the switchgear is specifically as follows: combining the moisture leakage risk score and the characteristic gas concentration, the condition of the switchgear is divided into different levels, and the judgment rules for each level are as follows: when and characteristic gas concentration When, it is judged as a normal state; when and characteristic gas concentration When, it is determined to be in a state of concern or inspection; when and characteristic gas concentration When, it is determined to be in a state requiring planned maintenance; when and characteristic gas concentration When the condition is determined to be severely aging or in a highly abnormal state, an audible and visual alarm is triggered and uploaded to the monitoring system; among which, 、 and These are the preset first, second, and third threshold values ​​for moisture leakage risk scoring. This is a preset characteristic gas concentration threshold.

2. The intelligent diagnostic method for performance aging of switchgear based on multi-parameter fusion according to claim 1, characterized in that, The internal environmental parameters include: cabinet temperature and cabinet humidity; the external environmental parameters include: cabinet temperature and cabinet humidity; the internal gas characteristic parameters include: water vapor content, characteristic gas concentration and its changing trend; the aging type of the switchgear includes at least one of: insulation aging, decreased sealing performance, condensation risk and environmental anomalies.

3. The intelligent diagnostic method for performance aging of switchgear based on multi-parameter fusion according to claim 1, characterized in that, The formula for calculating the absolute humidity difference between the inside and outside of the cabinet is: in, This refers to absolute humidity. Temperature in Celsius Relative humidity, The absolute humidity inside the cabinet. The absolute humidity outside the cabinet. This represents the absolute humidity difference.

4. The intelligent diagnostic method for performance aging of switchgear based on multi-parameter fusion according to claim 3, characterized in that, For the absolute humidity difference The positive portion is normalized to obtain the normalized absolute humidity difference index. The calculation formula is as follows: in, This is the historical 95th percentile of the absolute humidity difference.

5. The intelligent diagnostic method for performance aging of switching equipment based on multi-parameter fusion according to claim 1, characterized in that, The formula for calculating the expected absolute humidity inside the cabinet is: in, These are the regression coefficients obtained through training with historical data. The heater is in working condition. The temperature inside the cabinet. The outside temperature of the cabinet. The absolute humidity outside the cabinet.

6. The intelligent diagnostic method for performance aging of switchgear based on multi-parameter fusion according to claim 1, characterized in that, The process of obtaining the cumulative amount by offset detection of the standardized residuals specifically involves: performing robust standardization on the residuals between the actual absolute humidity inside the cabinet and the expected value to obtain the standardized residuals. Build a positive CUSUM cumulative amount The update formula for the positive CUSUM cumulative amount is: in, For the first Standardized residuals corresponding to each monitoring session For the number of monitoring sessions, The initial conditions for the positive CUSUM accumulation. This is the offset tolerance constant.

7. The intelligent diagnostic method for performance aging of switching equipment based on multi-parameter fusion according to claim 1, characterized in that, The formula for calculating the offset index is: in, For the first The offset index corresponding to the next monitoring. To smooth out the trend, This is a reference threshold obtained based on historical training periods. It is a Sigmoid type mapping function. These are the weighting coefficients. For the first The cumulative positive CUSUM of each monitoring session.

8. A smart diagnostic device for the performance aging of switchgear based on multi-parameter fusion, characterized in that, The device integrates the intelligent diagnostic method for switchgear performance aging based on multi-parameter fusion as described in any one of claims 1 to 7. The device includes: a temperature and humidity detection module, installed inside and outside the switchgear, for real-time acquisition of internal temperature, internal humidity, external temperature, and external humidity, and outputting the acquired data; a gas detection module, installed inside the switchgear, for detecting internal gas characteristic parameters, including water vapor content, characteristic gas concentration, and their changing trends; a data preprocessing module, connected to the temperature and humidity detection module and the gas detection module, for receiving multi-source sensor data, performing time synchronization processing, filtering and noise reduction processing, abnormal data removal processing, and format standardization processing, and outputting the processed characteristic data; and an algorithm calculation module, connected to the data preprocessing module, for receiving the processed characteristic data and performing calculations on the absolute humidity inside and outside the switchgear, the absolute humidity difference, and normalization. The system calculates the humidity difference index, expected absolute humidity inside the cabinet, residual value between actual absolute humidity and expected value, standardized residual value, cumulative value, smoothing trend value, offset index, and moisture leakage risk score. Combined with the moisture leakage risk score and characteristic gas concentration, it generates a comprehensive diagnostic result for the aging degree of the switchgear. An output and early warning module, connected to the algorithm calculation module, displays real-time monitoring values ​​and diagnostic results. When the moisture leakage risk score and characteristic gas concentration exceed the corresponding preset thresholds, it outputs audible and visual alarm information and uploads it to an external monitoring system. A communication module enables wired or wireless data communication between the intelligent diagnostic device and the external monitoring system, synchronizing diagnostic results, audible and visual alarm information, and historical data to the monitoring backend. A storage module, connected to the algorithm calculation module, stores historical monitoring data, calculation results, model parameters, and operation logs, and is used for subsequent offline analysis and model self-learning updates.

9. A processing device, characterized in that, The method includes at least one processor and at least one memory communicatively connected to the processor, wherein the memory stores program instructions executable by the processor, and the processor can execute the method as described in any one of claims 1 to 7 by invoking the program instructions.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause the computer to perform the method as described in any one of claims 1 to 7.

Images