Insulation state evaluation method and system for online monitoring device of voltage transformer

Abstract

The invention discloses an insulation state evaluation method for a voltage transformer online monitoring device, and the method comprises the following steps: obtaining an original dielectric loss angle tangent value through a dielectric loss tester, and collecting the temperature and relative humidity through an environment sensor; performing temperature and humidity compensation on the collected dielectric loss angle tangent value to obtain a compensated dielectric loss angle tangent value; performing trend prediction on the compensated dielectric loss angle tangent value based on an autoregressive integral moving average model to obtain a predicted value; judging an abnormal state by comparing a predicted value with a dynamic threshold value; performing risk assessment by integrating the equipment state parameters, and outputting a corresponding maintenance decision according to an assessment result; according to the invention, the interference of environmental factors on the measurement of the dielectric loss angle tangent value is eliminated, and the measurement accuracy is improved; the abnormal change trend of the dielectric loss value is found in advance through trend prediction, and early warning is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power monitoring, and particularly relates to a method and system for evaluating the insulation state of an on-line monitoring device for a voltage transformer. Background Art

[0002] A voltage transformer is an electrical device used to measure high voltages. It reduces the high voltage to a safe level through the principle of electromagnetic induction for measurement by a conventional voltmeter or data acquisition system. An on-line monitoring device for a voltage transformer is a device used to monitor the working state of the voltage transformer in real time. Since the on-line monitoring device for a voltage transformer is exposed to a complex electromagnetic environment and dynamic changes in temperature and humidity in an outdoor scene for a long time, it is easy to cause the gradual aging of insulating materials and abnormal fluctuations in the dielectric loss value, thereby leading to equipment failure problems.

[0003] Existing insulation state evaluation methods mainly rely on fixed threshold determination or single parameter monitoring, and have the following defects:

[0004] First, the tangent value of the dielectric loss angle (tanδ) is significantly affected by environmental factors such as temperature and humidity. Traditional methods do not consider the dynamic compensation of environmental parameters, resulting in large measurement errors. For example, an increase in temperature may cause an overestimation of tanδ, and humidity changes may introduce non-linear interference. Fixed thresholds cannot adapt to such fluctuations and are prone to false alarms or missed alarms.

[0005] Second, insulation deterioration is a gradual process. Traditional methods only evaluate based on instantaneous values or short-term average values, and cannot capture the long-term trend changes of the dielectric loss value, making it difficult to early warn of potential faults.

[0006] Third, fixed thresholds are usually based on empirical values or equipment factory standards, and do not combine the operating history data of the equipment and environmental characteristics, resulting in the disconnection between the threshold and the actual working conditions and being unable to reflect the cumulative effect of equipment aging or environmental changes. Summary of the Invention

[0007] The purpose of the present invention is to provide a method and system for evaluating the insulation state of an on-line monitoring device for a voltage transformer in view of the problems existing in the prior art. It eliminates the interference of environmental factors on the measurement of the tangent value of the dielectric loss angle and improves the measurement accuracy; it can early detect the abnormal change trend of the dielectric loss value through trend prediction to achieve early warning.

[0008] To achieve the above purpose, the technical solution adopted by the present invention is:

[0009] A method for evaluating the insulation state of an on-line monitoring device for a voltage transformer, comprising the following steps:

[0010] S1. Obtain the original tangent value of the dielectric loss angle tanδ through a dielectric loss tester raw, the temperature T and relative humidity H are collected through environmental sensors;

[0011] S2. Compensate the collected tangent value of the dielectric loss angle for temperature and humidity to obtain the compensated tangent value of the dielectric loss angle tanδ final ;

[0012] S3. Based on the autoregressive integrated moving average model, predict the trend of the compensated tangent value of the dielectric loss angle to obtain the predicted value

[0013] S4. Determine the abnormal state by comparing the predicted value with the dynamic threshold;

[0014] S5. Conduct a risk assessment based on the comprehensive equipment status parameters and output the corresponding maintenance decision according to the assessment result.

[0015] Further, step S2 includes:

[0016] S2.1. Compensate the collected tangent value of the dielectric loss angle tanδ raw for temperature through the following formula:

[0017]

[0018] where tanδ comp is the tangent value of the dielectric loss angle after temperature compensation; tanδ raw is the original collected tangent value of the dielectric loss angle; α i is the temperature compensation coefficient determined by fitting experimental data; T is the collected ambient temperature; T ref is the preset reference temperature; n is the polynomial order, taking positive integers.

[0019] Further, step S2 includes:

[0020] S2.2. Correct the tangent value of the dielectric loss angle after temperature compensation tanδ comp for humidity through the following formula:

[0021] tanδ final = tanδ comp ·(1 + β(H - H ref ));

[0022] where tanδ final is the final tangent value of the dielectric loss angle after temperature and humidity compensation; β is the humidity influence coefficient determined by experiment; H is the collected ambient humidity; H ref is the preset reference humidity.

[0023] Further, step S3 includes:

[0024] S3.1. Construct an autoregressive integrated moving average model ARIMA(1,0,1) with the formula as follows:

[0025] tanδ t = φ1tanδ t-1 + θ1∈ t-1 + ∈ t ;

[0026] Among them, φ1 is the autoregressive coefficient, representing the linear relationship between the current value and the previous moment value; θ1 is the moving average coefficient, representing the linear relationship between the current error and the previous moment error; ε t is the white noise at the current moment, following a normal distribution with a mean of 0 and a variance of σ 2 ;

[0027] S3.2. Take the dielectric loss tangent values tanδ final after temperature and humidity compensation for the most recent N ones, denoted as tanδ t-1 , tanδ t-2 ,..., tanδ t-N , and confirm the data stationarity through the Augmented Dickey-Fuller test;

[0028] S3.3. Construct the likelihood function:

[0029]

[0030] Among them, ∈ t = tanδ t - φ1tanδ t-1 - θ1∈ t-1 ;

[0031] Maximize lnL through the Newton iteration method to solve the estimated values of φ1 and θ1 and

[0032] S3.4. The prediction value recurrence formula is:

[0033] When k = 1,

[0034]

[0035] When k ≥ 2,

[0036]

[0037] Furthermore, step S4 includes:

[0038] S4.1. Take the dielectric loss tangent values tanδ final after temperature and humidity compensation for the historical N ones, denoted as tanδ t-1 , tanδt-2 ,..., tanδ t-N ;

[0039] S4.2. Calculate the average insulation state level μ of the historical N tangent values of the dielectric loss angle t , and the formula is as follows:

[0040]

[0041] S4.3. Calculate the standard deviation σ of the historical N tangent values of the dielectric loss angle t , and the formula is as follows:

[0042]

[0043] S4.4. Determine the abnormal state by comparing the predicted value with the dynamic threshold range, and the rules are as follows:

[0044] When the predicted value satisfies

[0045]

[0046] it means that the insulation state is within the historical normal fluctuation range and no intervention is required;

[0047] When the predicted value satisfies

[0048]

[0049] or

[0050]

[0051] it means that the insulation state significantly deviates from the historical mean, triggering a secondary alarm, and manual inspection needs to be arranged and the monitoring period needs to be shortened;

[0052] When the predicted value satisfies

[0053]

[0054] or

[0055]

[0056] it means that the insulation state is extremely abnormal, and the equipment should be shut down for maintenance immediately to avoid accidents caused by equipment failures.

[0057] Furthermore, step S5 includes:

[0058] S5.1. Establish a risk assessment model, and the formula is as follows:

[0059]

[0060] where It represents the ratio of the predicted dielectric loss value to the allowable maximum value; It represents the ratio of the operating years to the designed life; It represents the ratio of the historical number of failures to the total operating cycle; w1, w2, and w3 are weighting factors and satisfy w1 + w2 + w3 = 1;

[0061] S5.2. Output the maintenance decision according to the risk assessment value R:

[0062] When R ≤ 0.3, the insulation state is good, the predicted value does not reach the warning threshold, and the equipment can continue to operate with regular monitoring;

[0063] When 0.3 < R ≤ 0.6, there is a potential deterioration trend in the insulation state, manual inspection needs to be arranged, the monitoring period is shortened to 50% of the original interval, and key parameters are recorded;

[0064] When R > 0.6, the insulation state of the equipment is seriously abnormal, immediately stop the machine and carry out a comprehensive overhaul, check the cause of the failure, and replace the aging components.

[0065] Furthermore, it also includes:

[0066] S6. Store the collected original data, the compensated tangent value of the dielectric loss angle, the predicted value, the abnormal determination result, and the maintenance decision in the database, and display the following information in real time through the human - machine interface:

[0067] The real - time monitoring curves of the current temperature T and humidity H;

[0068] The original tangent value of the dielectric loss angle tanδ raw and the compensated value tanδ final of the historical comparison trend chart;

[0069] The future change curve and confidence interval of the tangent value of the dielectric loss angle predicted by the ARIMA model;

[0070] The real - time update status of the dynamic threshold range;

[0071] The grade identification of the risk assessment value R and the corresponding maintenance suggestions.

[0072] An insulation state assessment system for an on - line monitoring device of a voltage transformer, including:

[0073] A data acquisition module, which acquires the voltage U2 and current I2 through a secondary - side sensor, obtains the original tangent value of the dielectric loss angle tanδ raw , and acquires the temperature T and relative humidity H through an environmental sensor;

[0074] A compensation calculation module, which performs temperature and humidity compensation on the acquired tangent value of the dielectric loss angle to obtain the compensated tangent value of the dielectric loss angle tanδfinal ;

[0075] A trend prediction module that performs trend prediction on the compensated tangent value of the dielectric loss angle based on an autoregressive integrated moving average model to obtain a predicted value

[0076] An anomaly determination module that determines the anomaly status by comparing the predicted value with a dynamic threshold;

[0077] A risk assessment module that comprehensively assesses the risk based on the device status parameters and outputs corresponding maintenance decisions according to the assessment results.

[0078] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the above method steps.

[0079] An electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above method steps are implemented.

[0080] Compared with the prior art, the beneficial effects of the present invention are:

[0081] 1. Through the above temperature and humidity compensation measures, it is possible to accurately eliminate the interference of environmental factors on the measurement of the tangent value of the dielectric loss angle, improve the measurement accuracy, and provide a reliable data basis for the insulation status assessment;

[0082] 2. Through trend prediction, it is possible to discover the abnormal change trend of the dielectric loss value in advance, achieve early warning, provide strong support for the preventive maintenance of the device, avoid the sudden occurrence of device failures, and improve the reliability and service life of the device;

[0083] 3. The dynamic threshold mechanism avoids the false alarm or missed alarm problems caused by a fixed threshold, improves the accuracy of determination; the hierarchical early warning mechanism realizes a differentiated operation and maintenance strategy, which not only reduces unnecessary maintenance costs but also can quickly respond to serious anomalies and effectively prevent device failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0085] Figure 1 It is a schematic flow chart of a method for evaluating the insulation status of an on-line monitoring device for a voltage transformer in this application. Detailed implementation manners

[0086] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0087] The sequence numbers of the steps in the description of this application do not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of this application.

[0088] In the description of this application specification and the appended claims, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and should not be construed as indicating or implying relative importance. It should also be understood that although the terms "first", "second", etc. are used in the text to describe various elements in some embodiments of this application, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, the first table can be named the second table, and similarly, the second table can be named the first table without departing from the scope of the various described embodiments. The first table and the second table are both tables, but they are not the same table.

[0089] The reference to "one embodiment" or "some embodiments" etc. in the description of this application specification means that specific features, structures, or characteristics described in connection with the embodiment are included in one or more embodiments of this application. Thus, the statements "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "comprise", "include", "have" and their variants all mean "include but not limited to", unless otherwise specifically emphasized in other ways.

[0090] A voltage transformer is an electrical device used to measure high voltages. It reduces the high voltage to a safe level through the principle of electromagnetic induction for measurement by a conventional voltmeter or data acquisition system. An on-line monitoring device for a voltage transformer is a device used to monitor the working state of the voltage transformer in real time. Since the on-line monitoring device for the voltage transformer is long-term exposed to a complex electromagnetic environment and an outdoor scene with dynamic changes in temperature and humidity, it is likely to cause the gradual aging of insulating materials and abnormal fluctuations in dielectric loss values, thereby leading to problems of equipment failures.

[0091] The existing insulation status assessment methods mainly rely on fixed threshold judgment or single parameter monitoring, and have the following defects:

[0092] First, the tangent value of the dielectric loss angle (tanδ) is significantly affected by environmental factors such as temperature and humidity. Traditional methods do not consider the dynamic compensation of environmental parameters, resulting in large measurement errors. For example, an increase in temperature may cause an overestimation of tanδ, and humidity changes may introduce non-linear interference. Fixed thresholds cannot adapt to such fluctuations and are prone to false alarms or missed alarms.

[0093] Second, insulation deterioration is a gradual process. Traditional methods only evaluate based on instantaneous values or short-term averages, unable to capture the long-term trend changes of the dielectric loss value, and it is difficult to early warn of potential faults.

[0094] Third, fixed thresholds are usually based on empirical values or equipment factory standards, without combining the equipment operation historical data and environmental characteristics, resulting in the disconnection between the threshold and the actual working conditions, and unable to reflect the cumulative effect of equipment aging or environmental changes.

[0095] To address the above technical problems, as Figure 1 shown, in the first aspect of this application, an insulation status assessment method for an on-line monitoring device of a voltage transformer is provided, including the following steps S1 - S6.

[0096] S1. Obtain the original tangent value of the dielectric loss angle tanδ through a dielectric loss tester raw , and collect the temperature T and relative humidity H through an environmental sensor.

[0097] Specifically, a digital dielectric loss tester is used, which supports real-time on-line measurement and has a resolution of not less than 0.001% to obtain the original tangent value of the dielectric loss angle tanδ raw . The measurement electrode of the dielectric loss tester needs to be closely attached to the insulation component of the on-line monitoring device of the voltage transformer to avoid external electromagnetic interference. The environmental sensor is installed in a nearby ventilated place to avoid the influence of direct sunlight or local heat sources.

[0098] Synchronously collect the dielectric loss value and environmental parameters through a multi-channel data acquisition card, with a sampling frequency ≥1 kHz. Filter the original signal to eliminate high-frequency noise. The collected data is transmitted to the industrial control computer through RS485 or Ethernet and stored in a structured database, including timestamp, device ID, and parameter value.

[0099] S2. Perform temperature and humidity compensation on the collected tangent value of the dielectric loss angle to obtain the compensated tangent value of the dielectric loss angle tanδ final .

[0100] In some embodiments, step S2 includes the following steps S2.1 - S2.2.

[0101] S2.1. Perform temperature compensation on the collected tangent value of the dielectric loss angle tanδ raw through the following formula:

[0102]

[0103] where tanδ comp is the tangent value of the dielectric loss angle after temperature compensation; tanδ raw is the original tangent value of the dielectric loss angle collected; α i is the temperature compensation coefficient determined by fitting experimental data; T is the ambient temperature collected; T ref is the preset reference temperature; n is the polynomial order, taking a positive integer.

[0104] Specifically, in some embodiments, at a constant humidity, measure the dielectric loss values corresponding to different temperatures T, and solve the temperature compensation coefficient α i through regression analysis. Select the typical temperature during normal operation of the device as the preset reference temperature T ref .

[0105] The above steps fit the non-linear relationship between temperature and dielectric loss value through a polynomial model, measure the dielectric loss values corresponding to different temperatures at a constant humidity, solve the temperature compensation coefficient, and correct the original dielectric loss value according to the polynomial expansion of the temperature difference, effectively eliminating the interference of ambient temperature on the measurement of the tangent value of the dielectric loss angle.

[0106] S2.2. Perform humidity correction on the tangent value of the dielectric loss angle tanδ comp after temperature compensation through the following formula:

[0107] tanδ final = tanδ comp ·(1 + β(H - H ref ));

[0108] where tanδ final is the final tangent value of the dielectric loss angle after temperature and humidity compensation; β is the humidity influence coefficient determined through experiments; H is the ambient humidity collected; H ref is the preset reference humidity.

[0109] Specifically, in some embodiments, at a constant temperature, measure the change rate of the dielectric loss value corresponding to different humidities H, and the percentage of the relative change in the dielectric loss value is the humidity influence coefficient β.

[0110] Adopt a linear correction model, and the humidity difference ΔH = H - H ref is used as the scaling factor. When the humidity is higher than the reference value H ref , the dielectric loss value is scaled up proportionally; otherwise, it is scaled down, thereby eliminating the influence of ambient humidity on the dielectric loss value tanδ after temperature compensationcomp Residual influence

[0111] Exemplarily, the operating ambient temperature of a certain on-line monitoring device for voltage transformers is 35°C, T ref is 25°C, and the humidity is 70% RH (H ref = 50%).

[0112] Temperature compensation:

[0113] If α1 = 0.002 and α2 = 0.0001, then:

[0114]

[0115] If the measured tanδ raw = 0.05, then tanδ comp = 0.05 - 0.03 = 0.02.

[0116] Humidity correction:

[0117] If β = 0.005, then:

[0118] tanδ final = 0.02 × (1 + 0.005 × (70 - 50)) = 0.02 × 1.1 = 0.022.

[0119] The above steps adopt a linear correction model, measure the change rate of the dielectric loss value corresponding to different humidities at a constant temperature, determine the humidity influence coefficient, and use the humidity difference as a proportional factor to correct the dielectric loss value after temperature compensation, further eliminating the residual influence of the ambient humidity on the dielectric loss value and improving the accuracy of the dielectric loss value measurement.

[0120] S3. Based on the autoregressive integrated moving average model, perform trend prediction on the compensated tangent value of the dielectric loss angle to obtain the predicted value

[0121] In some embodiments, step S3 includes the following steps S3.1 - S3.4.

[0122] S3.1. Construct an autoregressive integrated moving average model ARIMA(1,0,1), and the formula is as follows:

[0123] tanδ t = φ1tanδ t-1 + θ1∈ t-1 + ∈ t ;

[0124] Among them, φ1 is the autoregressive coefficient, indicating the linear relationship between the current value and the previous moment value; θ1 is the moving average coefficient, indicating the linear relationship between the current error and the previous moment error; ε tis white noise at the current moment, following a normal distribution with a mean of 0 and a variance of σ 2 ;

[0125] S3.2. Take the tangent values of the dielectric loss angle tanδ after temperature and humidity compensation for the last N values final , denoted as tanδ t-1 , tanδ t-2 ,..., tanδ t-N , and confirm the stationarity of the data through the Augmented Dickey-Fuller test;

[0126] Specifically, assume that the sequence has a unit root, calculate the ADF statistic, and compare it with the critical value. If the ADF statistic < critical value, reject the original hypothesis and the data is stationary. If the data is non-stationary, the compensation model needs to be readjusted or differencing processing is required.

[0127] S3.3. Construct the likelihood function:

[0128]

[0129] where the error term calculation ∈ t = tanδ t - φ1tanδ t-1 - θ1∈ t-1 ;

[0130] Maximize lnL through the Newton-Raphson method to solve the estimated values of φ1 and θ1 and

[0131] Specifically, set φ1 (0) = 0.1, θ1 (0) = 0.1. Input data: a sequence of tangent values of the dielectric loss angle with a length of N.

[0132] Calculate the gradient of the likelihood function lnL:

[0133]

[0134] Calculate the Hessian matrix H:

[0135]

[0136] Parameter update, Newton iteration:

[0137]

[0138] Set the convergence condition:

[0139] |lnL (k+1) - ln L (k) |< 10 -6

[0140] Or, set the maximum number of iterations.

[0141] After multiple iterations, output and

[0142] S3.4. The recurrence formula for the predicted value is:

[0143] When k = 1,

[0144]

[0145] When k ≥ 2,

[0146]

[0147] The above steps perform trend prediction on the tangent value of the dielectric loss angle after temperature and humidity compensation by constructing an ARIMA(1,0,1) model. First, the ADF test is used to ensure the stationarity of the data, and then the autoregressive coefficient and moving average coefficient are solved by the Newton iteration method, and the likelihood function is maximized to optimize the model parameters. When predicting, a recurrence formula is used, combined with the observed value and error term at the previous moment, to generate the predicted value at the future moment. This method can effectively capture the temporal characteristics of the dielectric loss value and provide a reliable basis for subsequent anomaly determination.

[0148] On the one hand, through strict stationarity tests and parameter optimization, the accuracy of the prediction is improved, avoiding the misjudgment problem of the traditional fixed threshold method; on the other hand, the dynamic recurrence mechanism can update the prediction results in real time, adapt to the changes in the equipment state, provide timely and reliable data support for the insulation state assessment, and thus achieve early warning and precise maintenance.

[0149] S4. Determine the abnormal state by comparing the predicted value with the dynamic threshold.

[0150] In some embodiments, step S4 includes the following steps S4.1 - S4.4.

[0151] S4.1. Take the historical N tangent values of the dielectric loss angle tanδ after temperature and humidity compensation final , denoted as tanδ t-1 , tanδ t-2 ,..., tanδ t-N ;

[0152] S4.2. Calculate the average insulation state level μ of the historical N tangent values of the dielectric loss angle t , and the formula is as follows:

[0153]

[0154] S4.3. Calculate the standard deviation σ of the historical N tanδ values t , and the formula is as follows:

[0155]

[0156] S4.4. Determine the abnormal state by comparing the predicted value with the dynamic threshold range, and the rules are as follows:

[0157] When the predicted value satisfies

[0158]

[0159] it means that the insulation state is within the historical normal fluctuation range and no intervention is required;

[0160] When the predicted value satisfies

[0161]

[0162] or

[0163]

[0164] it means that the insulation state significantly deviates from the historical mean, triggering a secondary alarm, and manual inspection needs to be arranged and the monitoring period should be shortened;

[0165] When the predicted value satisfies

[0166]

[0167] or

[0168]

[0169] it means that the insulation state is extremely abnormal, and the machine should be stopped for maintenance immediately to avoid accidents caused by equipment failures.

[0170] In the traditional method, the fixed threshold cannot reflect the influence of environmental fluctuations such as temperature / humidity on the dielectric loss value, which is prone to false alarms. The above steps use a dynamic threshold mechanism to determine abnormal states. First, calculate the mean and standard deviation of the tanδ values based on historical data to establish a dynamic threshold range. Subsequently, compare the predicted value of the ARIMA model with the threshold, and classify it into three levels according to the deviation degree: normal fluctuation, significant deviation, and extreme abnormality, and trigger different levels of maintenance responses respectively. This method quantifies the degree of abnormality through statistical principles and realizes refined classification and early warning of the insulation state.

[0171] On the one hand, the dynamic threshold can adapt to environmental changes and equipment aging, avoid false alarms or missed alarms caused by fixed thresholds, and improve the accuracy of judgment; on the other hand, the hierarchical early warning mechanism realizes differentiated operation and maintenance strategies, which not only reduces unnecessary maintenance costs, but also can quickly respond to serious anomalies, thus effectively preventing equipment failures and ensuring the safe and stable operation of the power system.

[0172] S5. Conduct a risk assessment based on comprehensive equipment status parameters and output corresponding maintenance decisions according to the assessment results.

[0173] In some embodiments, step S5 includes the following steps S5.1 - S5.2.

[0174] S5.1. Establish a risk assessment model with the following formula:

[0175]

[0176] Where represents the ratio of the predicted dielectric loss value to the allowable maximum value; represents the ratio of the operating years to the designed life; represents the ratio of the historical failure times to the total operating cycle; w1, w2, w3 are weight factors and satisfy w1 + w2 + w3 = 1;

[0177] S5.2. Output maintenance decisions according to the risk assessment value R:

[0178] When R ≤ 0.3, the insulation state is good, the predicted value does not reach the early warning threshold, and the equipment can continue to operate and only requires regular monitoring;

[0179] When 0.3 < R ≤ 0.6, there is a potential deterioration trend in the insulation state. It is necessary to arrange manual inspections, shorten the monitoring cycle to 50% of the original interval, and record key parameters;

[0180] When R > 0.6, the insulation state of the equipment is severely abnormal. Immediately stop the machine and carry out a comprehensive overhaul, check the cause of the failure, and replace the aging components.

[0181] The above steps establish a risk assessment model through multi - dimensional parameter fusion, comprehensively consider key indicators such as dielectric loss value, operating years, and historical failure rate, and assign different weight factors to quantify the overall risk level of the equipment insulation state. According to the range of R values, the system automatically matches differentiated maintenance decisions, forming a closed - loop management from regular monitoring to emergency shutdown. The model in this embodiment not only covers the current dielectric loss value state, but also introduces long - term factors (aging, historical failures) to achieve a comprehensive assessment of risks.

[0182] On the one hand, by weighted fusion of multi-source parameters, the limitations of a single index are avoided, making the evaluation results more objective and reliable. On the other hand, the hierarchical response mechanism is linked with the dynamic threshold, which not only optimizes the allocation of operation and maintenance resources but also can timely block high-risk states, significantly improving the intelligent level and economy of equipment management. For example, when the proportion of operation years is relatively high, even if the dielectric loss value is normal, an inspection will be triggered, reflecting the design concept of preventive maintenance.

[0183] S6. Store the collected original data, compensated tangent value of dielectric loss angle, predicted value, abnormal determination result, and maintenance decision in the database, and display the following information in real time through the man-machine interface:

[0184] Real-time monitoring curves of the current temperature T and humidity H;

[0185] Original tangent value of dielectric loss angle tanδ raw And the historical comparison trend chart with the compensated value tanδ final ;

[0186] Future change curve and confidence interval of the tangent value of dielectric loss angle predicted by the ARIMA model;

[0187] Real-time update status of the dynamic threshold range;

[0188] Grade identification of the risk assessment value R and corresponding maintenance suggestions.

[0189] In the second aspect of the present application, an insulation status evaluation system for an on-line monitoring device of a voltage transformer is provided, including:

[0190] A data acquisition module, which obtains the original tangent value of dielectric loss angle tanδ raw through a dielectric loss tester and acquires the temperature T and relative humidity H through an environmental sensor;

[0191] A compensation calculation module, which performs temperature and humidity compensation on the collected tangent value of dielectric loss angle to obtain the compensated tangent value of dielectric loss angle tanδ final ;

[0192] A trend prediction module, which performs trend prediction on the compensated tangent value of dielectric loss angle based on the autoregressive integrated moving average model to obtain a predicted value

[0193] An abnormal determination module, which determines the abnormal state by comparing the predicted value with the dynamic threshold;

[0194] A risk assessment module, which comprehensively evaluates the risk based on the equipment status parameters and outputs corresponding maintenance decisions according to the evaluation results.

[0195] In a third aspect of the present application, there is provided a computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the above-mentioned method steps are implemented.

[0196] In a fourth aspect of the present application, there is provided an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the above-mentioned method steps are implemented.

[0197] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An insulation state evaluation method for an on-line monitoring device of a voltage transformer, characterized in that It includes the following steps: S1. Obtain the original dielectric loss tangent value tanδ through a dielectric loss tester raw , and collect the temperature T and relative humidity H through an environmental sensor; S2. Perform temperature and humidity compensation on the collected tangent value of the dielectric loss angle to obtain the compensated tangent value of the dielectric loss angle tanδ final ; S3. Based on the autoregressive integrated moving average model, perform trend prediction on the compensated tangent value of the dielectric loss angle to obtain the predicted value S4. Determine the abnormal state by comparing the predicted value with the dynamic threshold; S5. Conduct a risk assessment based on the comprehensive equipment status parameters, and output corresponding maintenance decisions according to the assessment results.

2. The insulation status evaluation method of an on-line monitoring device for a voltage transformer according to claim 1, characterized in that Step S2 includes: S2.

1. Perform temperature compensation on the collected tangent value of the dielectric loss angle tanδ raw which is achieved through the following formula: Among them, tanδ comp is the tangent value of the dielectric loss angle after temperature compensation; tanδ raw is the originally collected tangent value of the dielectric loss angle; α i is the temperature compensation coefficient determined by fitting experimental data; T is the collected ambient temperature; T ref is the preset reference temperature; n is the polynomial order, taking positive integers.

3. The insulation status evaluation method of an on-line monitoring device for a voltage transformer according to claim 2, characterized in that Step S2 includes: S2.

2. Humidity correction is performed on the temperature-compensated tangent value of the dielectric loss angle tanδ comp and is achieved through the following formula: tanδ final = tanδ comp ·(1 + β(H - H ref )); Among them, tanδ final is the final tangent value of the dielectric loss angle after temperature and humidity compensation; β is the humidity influence coefficient determined through experiments; H is the collected ambient humidity; H ref is the preset reference humidity.

4. The insulation status evaluation method of an on-line monitoring device for a voltage transformer according to claim 1, characterized in that, Step S3 includes: S3.

1. Construct an autoregressive integrated moving average model ARIMA(1,0,1), and the formula is as follows: tanδ t = φ1tanδ t-1 + θ1∈ t-1 + ∈ t ; Among them, φ1 is the autoregressive coefficient, representing the linear relationship between the current value and the value at the previous moment; θ1 is the moving average coefficient, representing the linear relationship between the current error and the error at the previous moment; ε t is the white noise at the current moment, following a normal distribution with a mean of 0 and a variance of σ 2 ; S3.

2. Take the most recent N tanδ values after temperature and humidity compensation final , denoted as tanδ t-1 , tanδ t-2 ,..., tanδ t-N . Confirm the data stationarity through the Augmented Dickey-Fuller test; S3.

3. Construct the likelihood function: where, ∈ t = tanδ t -φ1tanδ t-1 -θ1∈ t-1 ; Maximize lnL by the Newton-Raphson method to solve for the estimated values of φ1 and θ1 and The recurrence formula for the predicted value is: When k = 1, When k ≥ 2, 5. The insulation status evaluation method of an on-line monitoring device for a voltage transformer according to claim 1, characterized in that, Step S4 includes: S4.

1. Take the historical N tanδ values after temperature and humidity compensation, final , denoted as tanδ t-1 , tanδ t-2 ,..., tanδ t-N ; S4.

2. Calculate the average insulation state level μ of the historical N tangent values of the dielectric loss angle t , and the formula is as follows: S4.

3. Calculate the standard deviation σ of the historical N tanδ values t , and the formula is as follows: S4.

4. Determine the abnormal state by comparing the predicted value with the dynamic threshold range, and the rules are as follows: When the predicted value satisfies It indicates that the insulation state is within the normal historical fluctuation range and no intervention is required; When the predicted value satisfies Or It indicates that the insulation state significantly deviates from the historical mean, triggers a secondary alarm, and manual inspection needs to be arranged and the monitoring period should be shortened; When the predicted value satisfies Or It indicates that the insulation state is extremely abnormal, and the equipment should be shut down immediately for maintenance to avoid accidents caused by equipment failures.

6. The insulation state evaluation method of an on-line monitoring device for a voltage transformer according to claim 1, characterized in that, Step S5 includes: S5.

1. Establish a risk assessment model, and the formula is as follows: Among them, represents the ratio of the predicted dielectric loss value to the allowable maximum value; represents the ratio of the operating years to the design life; represents the ratio of the number of historical faults to the total operating cycle; w1, w2, and w3 are weighting factors, and satisfy w1 + w2 + w3 = 1; S5.

2. Output maintenance decisions according to the risk assessment value R: When R ≤ 0.3, the insulation state is good, the predicted value does not reach the warning threshold, and the equipment can continue to operate and only needs regular monitoring; When 0.3 < R ≤ 0.6, there is a potential deterioration trend in the insulation state, manual inspection needs to be arranged, the monitoring period should be shortened to 50% of the original interval, and key parameters should be recorded; When R > 0.6, the insulation state of the equipment is seriously abnormal, and the equipment should be shut down immediately and a comprehensive maintenance should be carried out to check the cause of the failure and replace the aging components.

7. The insulation status evaluation method of an on-line monitoring device for a voltage transformer according to claim 1, characterized in that, It also includes: S6. Store the collected original data, the compensated tangent value of the dielectric loss angle, the predicted value, the abnormal determination result, and the maintenance decision in the database, and display the following information in real time through the human-computer interaction interface: Real-time monitoring curves of the current voltage U2, current I2, temperature T, and humidity H; Original dielectric loss tangent value tanδ raw and the compensated value tanδ final historical comparison trend chart; Future change curves and confidence intervals of the tangent value of the dielectric loss angle predicted by the ARIMA model; Real-time update status of the dynamic threshold range; Level identification of the risk assessment value R and corresponding maintenance suggestions.

8. An insulation state evaluation system for an on-line monitoring device of a voltage transformer, characterized in that, It includes: Data acquisition module, which obtains the original tangent value of dielectric loss angle tanδ through a dielectric loss tester raw , and collects temperature T and relative humidity H through an environmental sensor; The compensation calculation module performs temperature and humidity compensation on the collected tangent value of the dielectric loss angle to obtain the compensated tangent value of the dielectric loss angle tanδ final ; The trend prediction module performs trend prediction on the compensated tangent value of the dielectric loss angle based on the autoregressive integrated moving average model to obtain a predicted value An abnormal determination module that determines the abnormal state by comparing the predicted value with the dynamic threshold; A risk assessment module that conducts a risk assessment based on the comprehensive equipment status parameters and outputs corresponding maintenance decisions according to the assessment results.

9. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, the method steps described in any one of claims 1 to 7 are implemented.

10. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the computer program, the method steps described in any one of claims 1 to 7 are implemented.

Images