Determining the condition of electrical equipment using diagnostic parameter prediction error variances.

Abstract

An embodiment is disclosed for determining the condition of electrical equipment using diagnostic parameter prediction errors. A prediction error value is determined for a plurality of predicted diagnostic parameter values ​​over a predetermined time period for at least one component of the electrical equipment. The prediction error value constrains observed variations in the behavior of the at least one component. The determined prediction error value is compared to an expected prediction error value. An indication of the condition of the at least one component is selectively generated based on the comparison.

Description

[Technical field]

[0001] background The present disclosure relates to the analysis of electrical equipment such as high voltage transformers. In particular, the present disclosure relates to determining the condition of electrical equipment using diagnostic parameter prediction errors. [Background technology]

[0002] Many diagnostic parameters of electrical equipment components exhibit variations due to ambient and other field conditions, making it difficult or slow to detect underlying problems with the components. For example, variations in insulation parameters of insulating bushings for high voltage transformers, such as capacitance or power factor, may indicate deterioration or failure of the bushing. However, these insulation parameters may be highly influenced by ambient conditions, such as temperature, humidity, overvoltage, or other changing environmental, electrical, and / or thermal conditions in and around the electrical equipment. As a result, conventional diagnostic techniques based on such sensitive diagnostic parameters may not be able to proactively detect ongoing failures or may not accurately detect the condition of the components. Thus, such techniques may require taking the transformer offline to accurately detect the condition of the components. Summary of the Invention [Means for solving the problem]

[0003] overview According to some embodiments, a method includes determining, by a processor circuit, a prediction error value of a plurality of predictive diagnostic parameter values ​​over a predetermined time period for at least one component of an electrical equipment, the prediction error value constraining observed ambient variations in behavior of the at least one component. The method further comprises: Expectationsand comparing the predicted error value to the predicted error value. The method further includes selectively generating, by the processor circuitry, an indication of the condition of the at least one component based on the comparison.

[0004] According to some embodiments, the at least one component includes an insulation component of the electrical equipment. The plurality of predictive diagnostic parameter values ​​includes a plurality of predictive insulation diagnostic parameter values.

[0005] According to some embodiments, the plurality of predicted insulation diagnostic parameter values ​​includes a plurality of at least one of a predicted capacitance value, a predicted capacitive current value, a predicted dissipation factor value, and a predicted power factor value of the at least one insulation component.

[0006] According to some embodiments, the electrical equipment comprises a transformer and the at least one component comprises a high voltage bushing of the transformer.

[0007] According to some embodiments, the observed variation in the behavior of the at least one component that is suppressed includes variation due to ambient conditions.

[0008] According to some embodiments, the variation due to ambient conditions includes variation due to at least one of environmental conditions, noise, vibration, and special cause variation.

[0009] According to some embodiments, determining the predicted error value further includes at least one of predicting, by a processor circuit, at least one error value for the plurality of predicted diagnostic parameter values; determining a variation in the at least one error value due to ambient conditions observed in the behavior of the at least one component; and generating the predicted error value based on the at least one error value and the determined variation.

[0010] According to some embodiments, determining the predicted error value further comprises predicting a plurality of predicted diagnostic parameter values ​​for a plurality of respective time points of the predetermined time period based on the obtained diagnostic parameter values, determining the predicted error value further comprises determining a plurality of error values ​​based on a comparison of the plurality of predicted diagnostic parameter values ​​for the respective time points and a plurality of actual diagnostic parameter values ​​obtained at the respective time points, the predicted error value comprising an average error value of the plurality of error values.

[0011] According to some embodiments, the plurality of actual diagnostic parameter values ​​are obtained from a parameter value data stream generated from a device associated with the at least one component.

[0012] According to some embodiments, the plurality of time points comprises at least 100 time points within a predetermined period of time.

[0013] According to some embodiments, the plurality of predicted diagnostic parameter values ​​are related to a predicted behavior of the at least one component, and the prediction error value represents a deviation of an observed behavior of the at least one component from a predicted behavior of the at least one component.

[0014] According to some embodiments, Expectations A prediction error value is determined based on a comparison of a plurality of previously predicted diagnostic parameter values ​​to a corresponding plurality of previously obtained diagnostic parameter values.

[0015] According to some embodiments, the plurality of predicted diagnostic parameter values ​​are determined based on a plurality of determined relationships between a predetermined number of diagnostic parameter values ​​of the plurality of previously acquired diagnostic parameter values ​​and at least one subsequent parameter value of the plurality of previously acquired diagnostic parameter values.

[0016] According to some embodiments, the plurality of previously obtained diagnostic parameter values ​​are obtained from components different from the at least one component.

[0017] According to some embodiments, the plurality of predicted diagnostic parameter values ​​are determined based on at least one of a machine learning model and a statistical model.

[0018] According to some embodiments, Expectations The prediction error value is determined based on at least one of a machine learning model and a statistical model.

[0019] According to some embodiments, selectively generating the indication comprises determining whether the predicted error value is: Expectations and determining, by the processor circuit, whether a predetermined prediction error threshold is met based on the prediction error value. Selectively generating an indication further includes generating a first alert indication in response to the prediction error value meeting the predetermined prediction error threshold.

[0020] According to some embodiments, selectively generating an indication further includes generating a second alert indication in response to the prediction error value not meeting a predetermined prediction error threshold.

[0021] According to some embodiments, an insulation diagnostic system comprises a processor circuit and a memory including machine-readable instructions. When executed by the processor circuit, the instructions cause the processor circuit to determine a plurality of predicted diagnostic parameter values ​​over a predetermined time period for at least one component of an electrical equipment. Further, the instructions cause the processor circuit to obtain a plurality of actual diagnostic parameter values ​​over the predetermined time period from the at least one component. Further, the instructions cause the processor circuit to determine a predicted error value based on the plurality of predicted diagnostic parameter values ​​and the plurality of actual parameter values, the predicted error value constraining observed ambient variations in behavior of the at least one component. Further, the instructions cause the processor circuit to calculate the determined predicted error value. Expectations The instructions further cause the processor circuit to selectively transmit an indication of the condition of the at least one component to the electrical equipment based on the comparison.

[0022] According to some embodiments, the at least one component includes an insulation component of the electrical equipment. The plurality of predictive diagnostic parameter values ​​includes a plurality of predictive insulation diagnostic parameter values. The plurality of actual diagnostic parameter values ​​includes a plurality of actual insulation diagnostic parameter values.

[0023] According to some embodiments, the plurality of predicted insulation diagnostic parameter values ​​include a plurality of at least one of a predicted capacitance value, a predicted capacitive current value, a predicted dissipation factor value, and a predicted power factor value of the at least one insulating component, and the plurality of actual insulation diagnostic parameter values ​​represent a plurality of at least one of an actual capacitance value, an actual capacitive current value, an actual dissipation factor value, and an actual power factor of the at least one insulating component.

[0024] According to some embodiments, the observed variation in the behavior of the at least one component that is suppressed includes variation due to ambient conditions.

[0025] BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are included to provide a further understanding of the disclosure and constitute a part of this application, illustrate certain non-limiting embodiments of the inventive concepts. [Brief description of the drawings]

[0026] [Figure 1A] 1 illustrates a technique for obtaining diagnostic parameter values ​​of an isolation component according to some embodiments. [Figure 1B] 1 illustrates a technique for obtaining diagnostic parameter values ​​of an isolation component according to some embodiments. [Figure 1C] 1 illustrates a technique for obtaining diagnostic parameter values ​​of an isolation component according to some embodiments. [Figure 2A] 1 illustrates operations according to some embodiments for determining a condition of electrical equipment based on a prediction error of a predicted diagnostic parameter value of the electrical equipment. [Figure 2B]1 illustrates operations according to some embodiments for determining a condition of electrical equipment based on a prediction error of a predicted diagnostic parameter value of the electrical equipment. [Diagram 3] 2C illustrates operations according to some embodiments for determining an expected prediction error value as part of the operations of FIG. 2B. [Figure 4A] 1 is a graphical plot of historical power factor data for a transformer bushing in accordance with some embodiments. [Figure 4B] 4 illustrates conversion of the time series data stream of historical power factor data of FIG. 3 into a regression model flat table for use by a machine learning model according to some embodiments. [Figure 5A] 1 is a graphical plot illustrating a comparison of predicted and actual power factor over a period of time for a properly functioning transformer bushing in accordance with some embodiments. [Figure 5B] 5B is a graphical plot illustrating the power factor prediction error of the comparison of FIG. 5A over the period of time for a properly functioning transformer bushing in accordance with some embodiments. [Figure 6A] 1 is a graphical plot illustrating a comparison of predicted capacitance and actual capacitance of a transformer bushing over time in accordance with some embodiments. [Figure 6B] 1 is a graphical plot illustrating capacitance prediction error over time for a properly functioning transformer bushing in accordance with some embodiments. [Figure 7] 2C illustrates operations according to some embodiments for determining a prediction error value as part of the operations of FIG. 2B. [Figure 8A] 1 is a graphical plot illustrating a comparison of predicted capacitance and actual capacitance of a transformer bushing over time according to some embodiments, showing a steep increase in actual capacitance. [Figure 8B] 1 is a graphical plot illustrating capacitance prediction error over time for a transformer bushing according to some embodiments. [Figure 9A]1 is a graphical plot illustrating a comparison of predicted capacitance and actual capacitance of a transformer bushing over time in accordance with some embodiments, with the actual capacitance showing a linear increase over time. [Figure 9B] 1 is a graphical plot illustrating capacitance prediction error over time for a transformer bushing according to some embodiments. [Figure 10A] 1 is a graphical plot illustrating a comparison of predicted and actual power factor of a transformer bushing over time in accordance with some embodiments, where the actual power factor shows a linear increase over time. [Figure 10B] 1 is a graphical plot illustrating power factor prediction error over time for a transformer bushing in accordance with some embodiments. [Figure 11] 1 illustrates operations according to some embodiments for determining a prediction error based on an average of predicted diagnostic parameter values ​​and an average of obtained prediction error values. [Figure 12A] 1 illustrates the calculation of average values ​​from a time series data stream of diagnostic parameter values ​​according to some embodiments. [Figure 12B] 1 illustrates the calculation of average values ​​from a time series data stream of diagnostic parameter values ​​according to some embodiments. [Figure 13A] 1 is a graphical plot illustrating a comparison of multiple averages of predicted and actual power factors of a transformer bushing over time in accordance with some embodiments. [Figure 13B] 1 is a graphical plot illustrating mean power factor prediction error values ​​over time for a properly functioning transformer bushing in accordance with some embodiments. [Figure 14A] 1 is a graphical plot illustrating a comparison of average predicted power factor and actual average power factor of a transformer bushing over time in accordance with some embodiments, where the actual power factor shows an exponential increase over time. [Figure 14B] 1 is a graphical plot illustrating mean power factor prediction error values ​​over time for a transformer bushing in accordance with some embodiments. [Figure 15] FIG. 1 is a block diagram illustrating a transformer monitoring system for performing operations according to some embodiments. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0027] Detailed Description The inventive concept will now be described more fully hereinafter with reference to the accompanying drawings, which show examples of embodiments of the inventive concept. However, the inventive concept can be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the inventive concept to those skilled in the art. It should also be noted that these embodiments are not mutually exclusive. Elements from one embodiment can be inferred to be present / used in another embodiment even if not specifically shown.

[0028] The following description presents various embodiments of the subject matter of the present disclosure. These embodiments are presented as teaching examples and should not be construed as limiting the scope of the subject matter of the present disclosure. For example, certain details of the described embodiments can be modified, omitted, or expanded without departing from the scope of the described subject matter.

[0029] Embodiments include a method of determining the condition of a component of electrical equipment by detecting a change in a prediction error of a diagnostic parameter value of the component. For example, a prediction error value can be determined for a plurality of predicted diagnostic parameter values ​​over a predetermined time period for at least one component of the electrical equipment. Furthermore, the prediction error value can suppress ambient variations observed in the behavior of the at least one component, thus resulting in a more stable and / or reliable determination. As used herein, the term "ambient variations" refers to variations due to ambient conditions, such as, for example, environmental temperature, noise, vibration, humidity, space / surface charge effects, component temperature, fluid pressure (e.g., gas leakage in seal components or transformer housing), vibration, electrical load, and / or special cause variations.

[0030] The determined prediction error value is Expectations A comparison can be made to the predicted error value and an indication of the state of the component can be selectively generated based on the comparison.

[0031] For purposes of illustration, many of the examples described herein relate to determining the condition of bushings or other insulating components for high voltage transformers using the features disclosed herein. However, it should be understood that the disclosure and claims are not so limited and have broad applicability beyond the specific examples provided herein. As used herein, the term "diagnostic parameter value" can refer to any parameter of electrical equipment.

[0032] Before describing the features of embodiments of the present disclosure, Figures 1A-1C show some examples of diagnostic parameters for an insulating component. Figure 1A shows an insulator 100 separating a pair of metal plates 102. As shown by the circuit diagram in Figure 1B, when a voltage V is applied across the insulator 100, a total current i flows through the insulator 100, which is a natural capacitive current i C and the resistive loss current i R For example, in the case of an undamaged and functioning insulator 100, C is the loss current i R As a result, the ratio between the two parameters can be a reliable indicator of the actual condition or quality of the insulator 100. Several parameters can be measured or derived in this way, as shown by the vector diagram in FIG. 1C. For example, the capacitive current i C The angle δ between the current and the total current i defines the dielectric tangent (tan δ), tan δ = i R / i C The loss current i R The complementary angle φ between the current and the total current i determines the power factor (cosφ), and cosφ=i R For a high quality insulator 100 with small δ, the dissipation factor and power factor become very small and are numerically very close.

[0033] However, if the insulator 100 contains defects such as plate shorts, plate punctures, voids, moisture, and / or particulate contamination, for example, the capacitive current and the ratio of loss current to total current will be significantly higher. As a result, capacitance, capacitive current, dissipation factor, and power factor are all useful diagnostic parameters for determining the condition of the insulator 100.

[0034] Many conventional diagnostic techniques for insulating components, such as transformer bushings, require isolating the transformer and measuring capacitance, power factor, and / or other diagnostic parameters offline. While it is possible to measure these diagnostic parameters while the transformer is online, this typically introduces a number of variations in the measured parameter values, such as variations due to ambient conditions (e.g., environmental conditions, temperature, noise, vibration, special cause variations, etc.), making it difficult to obtain accurate readings and therefore difficult to detect problems with bushings or other insulating components while the transformer is online.

[0035] To address this issue, according to some embodiments, a plurality of predicted diagnostic parameter values ​​are obtained over a predetermined period of time, and a corresponding plurality of actual diagnostic parameter values ​​are obtained over the same period of time. The predicted diagnostic parameter values ​​are compared to the actual diagnostic parameter values ​​to obtain a prediction error value for the predicted diagnostic parameter values. The prediction error value is then: Expectations The predicted error values ​​are compared to accurately determine the condition of the insulation components without having to take transformers or other electrical equipment offline.

[0036] The predicted diagnostic parameter values, the predicted error values, and Expectations The prediction error value can be obtained in a number of ways. For example, in some embodiments: ExpectationsThe prediction error value may be obtained by training a machine learning model to predict the diagnostic parameter value based on historical data. For example, the training may be based on determining a plurality of relationships between a predetermined number of the plurality of previously obtained diagnostic parameter values ​​and at least one subsequent parameter value of the plurality of previously obtained diagnostic parameter values. The previously obtained diagnostic parameter values ​​may be obtained from the same component or from different components, as desired.

[0037] The trained machine learning model can then, for example, predict diagnostic parameter values ​​for the insulation components known to be functioning properly based on the determined relationships and compare the predicted values ​​to corresponding actual diagnostic parameter values ​​for the insulation components that are functioning properly. The resulting prediction error values ​​can then be used to estimate future measurements of the insulation components in the field. Expectations It can be used as a prediction error value.

[0038] The machine learning model can similarly obtain predicted diagnostic parameter values ​​over a period of time for a field insulating component, such as a high voltage bushing of a transformer that is connected and online. The predicted diagnostic parameter values ​​are compared to the corresponding actual diagnostic parameter values ​​to obtain predicted error values, which are then calculated as Expectations The predicted error value is compared to determine the actual state of the isolation component. For a properly functioning component, the predicted error value is: Expectations The predicted error value should be very close to the predicted error value, but in the case of damaged or malfunctioning components, the predicted error may be Expectations This can be an order of magnitude increase compared to predicted error values, making it possible to detect problems very quickly and reliably without taking electrical equipment offline.

[0039] These and other embodiments can suppress observed variations in the behavior of the component in several ways. For example, a machine learning model can consider variations in the data as part of its training process and suppress these variations when predicting the predicted diagnostic parameter values. Alternatively or additionally, a moving average of multiple data points can be used to suppress these variations. For example, a predicted error value can be obtained by comparing an average value of predicted diagnostic parameter values ​​for a plurality of predicted diagnostic parameter values ​​(e.g., 100 diagnostic parameter values) with an average value of the corresponding actual diagnostic parameter values. In another example, a plurality of error values ​​(e.g., 100 error values) can be obtained for each of the plurality of predicted diagnostic parameter values, and then an average predicted error value can be calculated for the plurality of error values.

[0040] Reference is now made to Figure 2A, which illustrates operations 200 according to some embodiments for determining a condition of electrical equipment based on a prediction error of a predicted diagnostic parameter value of the electrical equipment. In this example, operations 200 include determining, by a processor circuit, a prediction error value of a plurality of predicted diagnostic parameter values ​​over a predetermined time period for at least one component of the electrical equipment, the prediction error value constraining observed ambient variations in the behavior of the at least one component (block 208). Operations 200 include: Expectations and comparing the predicted error value to a predicted error value (block 210). The operations 200 further include selectively generating, by the processor circuitry, an indication of the state of the at least one component based on the comparison (block 212). It should also be appreciated that any or all of these operations 200 may be used in other embodiments disclosed herein, such as, for example, operation 200' of FIG. 2B, which is described in further detail below. Additionally, any or all of these operations 200 may be used in conjunction with other operations disclosed herein, such as, for example, the additional operations for determining a predicted error value described in FIG. 7, below.

[0041]

[0023] Referring now to Figure 2B, a more detailed example of operations 200' according to some embodiments for determining a condition of electrical equipment based on a prediction error of a predicted diagnostic parameter value of the electrical equipment is shown. In describing the individual operations 200' of Figure 2B, reference is also made to Figures 3-12B which show additional operations and examples of these and other features.

[0042] The operation 200' of FIG. 2B is Expectations This may include determining a prediction error value (Block 202'). Expectations Determining the predicted error value can be accomplished in a number of ways. For example, in some embodiments, a machine learning model can be trained to predict diagnostic parameter values ​​for normally functioning components and these compared to the actual diagnostic parameter values ​​to determine, e.g., a "baseline" predicted error value. Expectations A prediction error value can be determined.

[0043] In this regard, FIG. 3 illustrates, as part of operation 200′ of FIG. 2B, Expectations 3 illustrates additional operations for determining a prediction error value, which may include obtaining time series data of historical diagnostic parameter values ​​(block 302) and converting the time series data into a flat file (block 304).

[0044] For example, Figure 4A is a graphical plot 400 of historical power factor data 402 for a high voltage transformer bushing. As shown by Figure 4A, a time series data stream 404 of the historical power factor data 402 is converted into a flat file 406, such as a regression model flat table in this example, having a number of rows 408 and columns 410, where each row 408 corresponds to a sequence (e.g., a "moving window") of successive diagnostic parameter values ​​in the historical power factor data 402.

[0045] Referring again to FIG. 3, additional operations may further include training a machine learning model to predict diagnostic parameter values ​​based on historical data (block 306). For example, the flat file 406 of FIG. 4B may be used by a machine learning model to repeatedly apply a multivariate regression algorithm to each row 408 of the flat file 406, with the final column 412 as the target output of the input based on the previous matrix to determine and refine the algorithm over time. However, it should be understood that the selection of the number of variables (i.e., predictors) depends on each individual problem. The number of variables in each row 408 may be determined and optimized based on additional tests for, for example, sensitivity, model accuracy, hardware and software constraints, and other parameters.

[0046] One advantage of this data transformation technique in Figure 4B is the transformation of a single variable data set (e.g., tan δ versus time) into a multivariate problem, facilitating the use of a large number of machine learning models suitable for regression or classification applications. The strength of such machine learning models lies in the fact that they can "learn" from large data sets that contain a large number of cases (or examples) and also contain a large number of features (or predictors, or independent variables).

[0047] One advantage of using these and other predictive techniques in conjunction with diagnostic parameter data is that these techniques can provide extremely accurate predictions of future diagnostic parameter values ​​based on a relatively small, single-variable data set of historical diagnostic parameter values ​​over time, without the need for other external parameters such as temperature, holidays, events, etc. Furthermore, much of the contribution of variability introduced by external ambient conditions is suppressed through the application of these and other predictive techniques, thereby providing a more accurate indication of the actual condition of the electrical equipment.

[0048] In some embodiments, a number of different machine learning models (e.g., linear and non-linear algorithms) are trained using the flattened data and the results are compared to determine the most accurate machine learning model. A number of different criteria can be used to judge accuracy, such as root mean square error (RMSE), mean absolute error (MAE), etc. Examples of suitable linear machine learning models can include general linear regression, logistic regression (e.g., for classification), linear discriminant analysis, etc. Examples of suitable non-linear machine learning models can include classification and regression trees, naive Bayesian, K-nearest neighbors, support vector machines, etc. Examples of suitable ensemble machine learning models can include random forests, tree bagging, extreme gradient boosting machines, artificial neural networks, etc.

[0049] The predicted diagnostic parameter values ​​can be predicted using machine learning models, statistical models, or any other suitable techniques. For example, supervised or unsupervised machine learning models, such as neural networks, can be used to recognize underlying relationships in a data set to more accurately predict future values. In another example, a statistical model, such as an autoregressive integrated moving average ("ARIMA"), can take into account past values ​​in the time series and learn from them, thus resulting in a more accurate prediction of future values. However, it should be understood that any number of prediction techniques may be used and the disclosed embodiments are not limited to the above examples. In many embodiments, improved accuracy in predicting diagnostic parameter values ​​can result in more reliable prediction of future values. Expectations A prediction error value may be provided, and therefore the diagnostic value of an unexpected increase in prediction error may be increased. However, it should be understood that any technique that allows prediction of a diagnostic prediction value for electrical equipment may be used in the embodiments described herein.

[0050] 3, the predicted diagnostic values ​​are then compared (block 308) with corresponding actual diagnostic parameter values ​​of the historical data to obtain a plurality of error values ​​for the electrical equipment. In this example, the predicted diagnostic values ​​correspond to particular points in time during a predetermined time period, and the corresponding actual diagnostic parameter values ​​correspond to the same respective points in time. In this example, an average error value is determined for the plurality of error values ​​(block 310), which is used as a mean error value for subsequent measurement and comparison. Expectations It can be used as a prediction error value.

[0051] In this way, machine learning models or other suitable predictive techniques can be used to predict the predictive value of some diagnostic parameters. Expectations A predicted error value can be determined where, for the historical power factor data (FIGS. 5A and 5B) and the historical capacitance data (FIGS. 6A and 6B), Expectations An example of determining the prediction error value is described below.

[0052] 5A, a graphical plot 500 illustrates multiple predicted power factors 502 and actual power factors 504 over a period of time for a properly functioning transformer bushing. As discussed above, the predicted power factor 502 may be predicted using a supervised or unsupervised machine learning model such as a neural network, a statistical model such as an autoregressive integrated moving average ("ARIMA"), or other suitable technique for predicting power factor or other diagnostic parameter values. The actual power factor 504 is measured or derived from measurements of the transformer bushing for a corresponding period of time.

[0053] 5B, a comparison of the predicted power factor 502 and the actual power factor 504 produces a graphical plot 506 of a number of error values ​​508 for a properly functioning transformer bushing. The average predicted error 510 of the number of error values ​​508 in this example is 0.98%, which can be used to compare against future power factor prediction error determinations for a transformer bushing in live service. ExpectationsThe predicted error value 512 may be used as the predicted error value. In some embodiments, another value may be used, such as a 95th percentile value 514 (i.e., the maximum of the error values ​​508 excluding the highest 5% of the error values ​​508), a 99th percentile value 516, or the like. Expectations It may be used as a prediction error value.

[0054] These techniques can also be used to determine other diagnostic parameters such as capacitive current, dissipation factor, and / or power factor. Expectations A predicted error value can be determined. In this regard, Figure 6A is a graphical plot 600 illustrating a comparison of predicted capacitance 602 and actual capacitance 604 of a transformer bushing over a period of time. Figure 6B is a graphical plot 606 of a plurality of error values ​​608 generated by comparing predicted capacitance 602 and actual capacitance 604 for a properly functioning transformer bushing. The average predicted error 610 of the plurality of error values ​​608 in this example is 0.0043%, which can be used to compare against a future capacitance prediction error determination for a transformer bushing in live service. Expectations This can be used as the prediction error value 612. As mentioned above, other values ​​such as the 95th percentile value 614, the 99th percentile value 616, etc. can be used if desired.

[0055] 2B, the operations 200' may further include predicting (block 204') a plurality of predicted diagnostic parameter values ​​over a predetermined time period, e.g., using the trained machine learning model described above. The operations 200' may further include obtaining (block 206') a plurality of actual diagnostic parameter values ​​for the predetermined time period, which may be measured or derived, e.g., from measurements of a transformer bushing for the corresponding time period.

[0056] The operations 200' may further include determining a prediction error value for the plurality of predicted diagnostic parameter values ​​(block 208'), e.g., by comparing the predicted diagnostic parameter values ​​to the actual diagnostic parameter values. In this regard, Figure 7 illustrates additional operations according to some embodiments for determining a prediction error value as part of the operations of Figure 2B.

[0057] The additional operations of FIG. 7 may further include predicting a plurality of diagnostic parameter values ​​using a machine learning model (block 702) or other suitable prediction technique. A corresponding plurality of actual diagnostic parameter values ​​are also obtained (block 704). In this regard, FIG. 8A is a graphical plot 800 illustrating predicted capacitance 802 and actual capacitance 804 over time for a transformer bushing in the field. In this example, the predicted diagnostic values ​​correspond to specific points in time during a predetermined period of time, and the corresponding actual diagnostic parameter values ​​correspond to the same respective points in time. In this example, the trained machine learning model used to determine the predicted capacitance 602 of the normally functioning transformer bushing of FIG. 6A is also used to predict the predicted capacitance 802 of the transformer bushing in the field of FIG. 8A. For this example, a relatively small but sudden increase 818 (e.g., 1 pF) in the actual capacitance 804 occurs at time T.

[0058] The additional operations of FIG. 7 may further include comparing the predicted diagnostic parameter values ​​to the obtained actual diagnostic parameter values ​​to determine a plurality of error values. In this regard, FIG. 8B is a graphical plot 806 of a plurality of error values ​​808 generated by comparing the predicted capacitance 802 and the actual capacitance 804 shown in FIG. 8A for a transformer bushing in the field. Due to the sudden increase 818 in the actual capacitance 804 shown in FIG. 8A, the error value 808 also exhibits a large and sustained increase 820 at time T.

[0059] The additional operations of FIG. 7 may further include determining an average error value of the plurality of error values ​​(block 708). For example, as shown in FIG. 8B, the average predicted error 810 of the plurality of error values ​​808 in this example is 0.0435% as a result of the sudden increase 820 of the error values ​​808. In this manner, the average predicted error 810 represents the deviation of the observed behavior, i.e., the sudden increase 818 of the actual capacitance 804, from the expected behavior, i.e., the predicted capacitance 802. As discussed above, other values, such as the 95th percentile value 814, the 99th percentile value 816, etc., may also be used, if desired.

[0060] Referring again to FIG. 2B, operation 200′ converts the determined prediction error value into Expectations 8B for a transformer bushing that is functioning normally, as determined in FIG. 6B. Expectations This can be compared to the predicted error value 612. In this example, the average predicted error 810 of 0.0435% is larger than the increase in capacitance 818 of 0.0043%, even though the increase in capacitance 818 is relatively small in magnitude in absolute terms. Expectations This is approximately 10 times the predicted error value 612. This represents a clear and easily detected indication of abnormal behavior (i.e., a sudden increase in capacitance) by the transformer bushing in the field. In this and other examples, a significant deviation in the actual diagnostic parameter value may result in a corresponding increase in the predicted error that can be detected and monitored.

[0061] In this regard, and referring again to FIG. 2B, the operations 200′ may further include selectively generating an indication of the condition of the at least one component based on the comparison (block 212). For example, in some embodiments, the prediction error threshold may be set to: ExpectationsThe determination may be based on the prediction error value, and a warning indication may be selectively generated when the prediction error value is determined to meet a predetermined prediction error threshold. Alternatively, or in addition, an indication may be selectively generated when the prediction error value does not meet a predetermined prediction error threshold.

[0062] In some examples, the prediction error threshold may be a particular value or a range of values. The indication may include an indication of a particular value or range of values, an indication of a classification type such as "good or bad," "yes or no," level 1, 2, 3, etc., or any other suitable indication, as desired.

[0063] The embodiments disclosed herein can also detect and indicate other types of anomalous behavior. For example, Figure 9A is a graphical plot 900 showing a comparison of predicted capacitance 902 and actual capacitance 904 for a transformer bushing where the actual capacitance exhibits a linear increase 918 over time. As shown by Figure 9B, this relatively small linear increase 918 in capacitance (e.g., 3 pF) also results in a corresponding measurable increase 920 in the error value 908, resulting in an average predicted error 910 of 0.0043%. Expectations This is 0.0408%, which is more than nine times the prediction error value 612 (shown in FIG. 6B). As noted above, other values ​​such as the 95th percentile value 914, the 99th percentile value 916, etc., could be used if desired.

[0064] In another example, Figure 10A is a graphical plot 1000 illustrating a comparison of predicted power factor 1002 and actual power factor 1004 over time for a transformer bushing in the field, where the actual power factor 1004 shows a linear increase 1018 over time. Again, as shown by Figure 10B, this linear increase 1018 in capacitance also results in a corresponding measurable increase 1020 in the error value 1008, which is 0.98% (shown in Figure 5B). ExpectationsThis results in an average prediction error 1010 of 2.36%, which is much larger than the prediction error value 512. As noted above, other values ​​such as the 95th percentile value 1014, the 99th percentile value 1016, etc. could be used if desired.

[0065] As discussed above, variations in observed behavior of transformer bushings (or other components) can be detected and / or suppressed in a number of ways. For example, as discussed above with respect to the example machine learning models disclosed herein, the predictive techniques themselves can suppress much of the variation introduced by external ambient conditions. For example, the predictive techniques can be trained or configured to distinguish between variations due to ambient conditions and normal aging of the component, i.e., "healthy" variations, and variations due to underlying issues with the component, such as damage, excessive wear, or other undesirable variations. Alternatively or additionally, variations can also be suppressed by taking average values ​​of a set of diagnostic parameter values ​​over time.

[0066] 11, operations 1100 according to some embodiments for determining a prediction error based on an average of predicted diagnostic parameter values ​​and an average of obtained prediction error values. These operations 1100 can be performed, for example, by: Expectations 2B, such as determining a prediction error value (block 202') and / or determining a prediction error value for a plurality of predicted diagnostic parameter values ​​(block 208').

[0067] The operation 1100 of FIG. 11 may include determining a plurality of averaged acquired diagnostic parameter values ​​(block 1102) based on a plurality of acquired diagnostic parameter values. For example, FIGS. 12A and 12B illustrate the calculation of a plurality of averaged values ​​(i.e., running averages) from a time series data stream 1204 of diagnostic parameter values ​​according to some embodiments. In FIG. 12A, a power factor value 1202 is obtained from the time series data stream 1204 that may exhibit fluctuations over time due to ambient conditions such as environmental factors, temperature, etc. The power factor value 1202 is converted to a plurality of average power factor values ​​1206 that further reduce the impact of ambient fluctuations on the measurement. In this example, each set of 20 acquired power factor values ​​1202 is averaged to generate a single average power factor value 1206.

[0068] In some examples, a sufficiently large set of acquired diagnostic parameter values ​​can yield a set of usable average diagnostic parameter values ​​(e.g., 100 values ​​for 100 time points) using the technique of FIG. 12A. Alternatively, or in addition, it may be desirable to determine an average value for each acquired diagnostic parameter value to obtain a larger set of data points for use by a machine learning model, statistical model, or other predictive technique. In this regard, FIG. 12B illustrates the conversion of acquired power factor values ​​1202 from a time series data stream 1204 into a number of average power factor values ​​1206′, each based on the acquired power factor value and the preceding 19 acquired power factor values ​​in the series. In this way, each average power factor value 1206′ can still dampen the variance of the acquired power factor value 1202, but there is a much larger number of average power factor values ​​1206′ for use by the predictive technique, thus improving the overall accuracy of the predictive technique.

[0069] The operation 1100 of FIG. 11 may further include predicting a plurality of average predicted diagnostic parameter values ​​based on the plurality of obtained diagnostic parameter values ​​(block 1104). For example, the plurality of predicted diagnostic parameter values ​​may be determined using the machine learning models described above or other prediction techniques. A plurality of average predicted diagnostic parameter values ​​may then be determined based on the plurality of predicted diagnostic parameter values, for example, using the same or similar processes of FIG. 12A and / or FIG. 12B.

[0070] Operation 1100 of FIG. 11 may further include comparing an average predicted diagnostic parameter value with an average historical diagnostic parameter value to obtain a plurality of average error values ​​(block 1106) and determining an average predicted error value of the plurality of average error values ​​(block 1108), similar to the techniques described above with respect to FIGS. 3 and 7, etc.

[0071] Further application of the operation 1100 of FIG. 11 is illustrated by way of example in FIGS. 13A-14B. FIG. 13A is a graph plot 1300 illustrating a comparison of a plurality of average predicted power factors 1302 and a plurality of average values ​​1304 of actual power factors over time for a properly functioning transformer bushing. In this embodiment, each average predicted power factor data point 1318 represents an average of 100 predicted power factor samples (see block 1104 of FIG. 11), and each average data point 1320 of actual power factor represents an average of 100 corresponding actual power factor samples (see block 1102 of FIG. 11). As illustrated by the graph plot 1306 of FIG. 13B, an average error value 1308 is obtained from a comparison of the plurality of average predicted power factors 1302 and the plurality of average values ​​1304 of actual power factors (see block 1106 of FIG. 11). For example, as in the above example, the average prediction error 1310 of FIG. 13B is calculated for a number of average error values ​​1308 (see block 1108 of FIG. 11). In this example, the average prediction error 1310 is 0.03%, which is used to compare against future prediction error determinations for transformer bushings in live service. Expectations3 , may be used as the predicted error value 1312. As noted above, other values ​​may be used, such as a 95th percentile value 1314, a 99th percentile value 1316, etc., if desired. In this example, averaging groups of predicted and actual power factor samples prior to determining the average error value 1308 further reduces variations in the observed behavior of the transformer bushing, and in this example Expectations The forecast error is reduced from 0.98% in this example (see FIG. 5B, which does not use the moving average technique in this example) to 0.03%. Expectations This reduction in the prediction error value 1312, i.e., the baseline value, increases the likelihood of detecting an increase in prediction error when monitoring the equipment in the field, and therefore increases the detectability of anomalous behavior of the equipment.

[0072] Similar moving average techniques can be used to predict diagnostic parameter values ​​and determine predicted error values ​​for field installations (see, e.g., FIG. 7). In this regard, FIG. 14A is a graphical plot 1400 illustrating a comparison of a plurality of average predicted power factors 1402 and a plurality of averages of actual power factors 1404 over time for a field transformer bushing. In this example, the trained machine learning model used to determine the average predicted power factor 1302 for a properly functioning transformer bushing in FIG. 13A is also used to predict the plurality of average predicted power factors 1402 for a field transformer bushing in FIG. 14A. In this example, the plurality of averages of actual power factors 1404 exhibit a linear increase 1414 over time. As shown by the graphical plot 1406 in FIG. 14B, this linear increase 1414 in the average actual power factor 1404 also results in a measurable increase 1416 in a corresponding error value 1408, resulting in an average predicted error value 1410 of 0.9% compared to a measurable increase 1416 of 0.03%. Expectations This is approximately 30 times the prediction error value 1312 (shown in FIG. 14B). As noted above, other values ​​such as the 95th percentile value 1414, the 99th percentile value 1416, etc. may be used if desired.

[0073] As noted above, many of the above embodiments relate to determining the condition of insulating components (e.g., high voltage bushings) based on diagnostic parameters related to capacitance, power factor, etc., however, it should be understood that the embodiments disclosed herein have a wide range of applications. For example, many of the same ambient conditions that affect diagnostic parameters based on capacitance may also affect diagnostic parameters for detecting and measuring other aspects of transformers and other electrical equipment, such as partial discharge (PD), oil temperature, and / or dissolved gas analysis (DGA). Other types of electrical equipment that can benefit from the embodiments disclosed herein can include circuit breakers for monitoring the condition of contacts (i.e., physical wear), gas leaks, operating mechanisms (e.g., travel time), etc.

[0074] For example, diagnostic parameters relevant to monitoring circuit breaker travel times may include forces experienced by the circuit breaker contacts that may be affected by a number of ambient conditions, such as arcing, insulating gas properties (e.g., gas electronegativity, gas mixture), load current, switching instant, temperature around the contacts, space charge in sulfur hexafluoride or other cooling gas, instantaneous potential difference across the contacts, load current, type of load (e.g., impedance), etc. With a sufficient amount of historical data for these various diagnostic parameters, these and other predictive techniques can be trained or configured to detect component conditions and deviations from expected conditions, regardless of the degree or effect of ambient conditions on the measured data.

[0075] Referring now to FIG. 15, a block diagram of a transformer monitoring system 1500 is shown. The transformer monitoring system 1500 in this example is configured to perform operations according to some embodiments, such as those of FIG. 2, FIG. 3, FIG. 7, and / or FIG. 11, etc. The transformer monitoring device 30 of the transformer monitoring system 1500 can monitor one or more transformers 10A, 10B. In some embodiments, the transformer monitoring device 30 can be integrated within the transformer 10A provided as a device, allowing only the transformer 10A to be monitored, while in other embodiments, the transformer monitoring device 30 may be integrated with the transformer 10A to monitor the transformer 10A and optionally also monitor or receive data from one or more nearby electrical equipment (e.g., the transformer 10B or another power or current transformer or circuit breaker) or connected transmission / distribution lines. In yet another embodiment, the transformer monitoring device 30 is separate from the monitored transformers 10A, 10B.

[0076] The transformer monitoring device 30 includes a processor circuit 34, a communications interface 32 coupled to the processor circuit 34, and a memory 36 coupled to the processor circuit 34. The memory 36 includes machine-readable computer program instructions that, when executed by the processor circuit 34, cause the processor circuit 34 to perform some of the operations illustrated and described herein, such as, for example, the operations of Figures 2, 3, 7, and / or 11.

[0077] As shown, the transformer monitoring system 1500 includes a communication interface 32 (also referred to as a network interface) configured to provide communication with other devices, such as sensors 20 in the transformers 10A, 10B, via a wired or wireless communication channel 14. The transformer monitoring device 30 can receive signals from the sensors 20 representative of diagnostic parameters of the transformers 10A, 10B, such as voltages, currents, oil temperature, ambient temperature, etc. associated with the transformers 10A, 10B.

[0078] In this example, the transformer monitoring device 30 is shown as a separate monitoring device communicating with the circuits of the transformers 10A, 10B over the communication channel 14, for example, in a server-client model, a cloud-based platform, a substation automation system used in a substation, a distribution management system used in power system management, or other network configuration. One advantage of the client-server configuration is that monitoring and predictions of diagnostic parameters can be obtained for multiple individual pieces of equipment, such as the transformers 10A, 10B. For example, diagnosing a problem with one piece of electrical equipment in a power system can include redistributing loads among different pieces of electrical equipment based on a determination of the status of the different pieces of electrical equipment. However, it should be understood that in other embodiments, the transformer monitoring device 30 may be part of the transformers 10A, 10B or other pieces of electrical equipment, as desired.

[0079] In another embodiment of the server-client model, a transformer monitoring system may have devices (e.g., clients) associated with the monitored transformers, the devices comprising machine learning models, statistical models, or other predictive tools, and a central system (e.g., server) configured to monitor the multiple electrical equipment / transformers. Additionally, the server may include instances of the machine learning models or other predictive tools included in the devices associated with the transformers. The machine learning models or other predictive tools in the server may be continuously trained, tuned, adapted, etc. with data received from the transformers and / or multiple electrical equipment, and the server provides information / data for tuning / adapting the predictive tools in the server. Additionally, the server may be capable of performing simulations or advanced processing to predict / simulate conditions in the transformers (e.g., failure or degradation of a transformer bushing based on capacitance and / or power factor data made available by devices or sensors connected to the transformers) and provide information regarding such determinations to devices (e.g., clients) connected to the transformers to cause the devices to modify at least one parameter (e.g., cooling, output, online status) associated with the transformers (or other electrical equipment). According to various embodiments, the transformer monitoring device 30 may include electronic, computing, and communications hardware and software for measuring and predicting diagnostic parameter values ​​and performing at least one activity related to the transformer.

[0080] The transformer monitoring device 30 also includes a processor circuit 34 (also referred to as a processor) and a memory circuit 36 ​​(also referred to as a memory) coupled to the processor circuit 34. According to other embodiments, the processor circuit 34 may be defined to include a memory such that a separate memory circuit is not required.

[0081] As described herein, operations of the transformer monitoring device 30 and other aspects of the transformer monitoring system 1500 may be performed by the processor circuitry 34 and / or the communication interface 32. For example, the processor circuitry 34 may control the communication interface 32 to send communications to one or more other devices via the communication interface 32 and / or receive communications from one or more other devices via the network interface. Additionally, modules may be stored in the memory 36, and these modules may provide instructions such that, when the instructions of the modules are executed by the processor circuitry 34, the processor circuitry 34 performs respective operations (e.g., operations described herein with respect to the exemplary embodiments). For example, the modules may be further configured to obtain diagnostic parameter values, predict diagnostic parameter values, determine prediction error values, and determine a state and / or status of components of the electrical equipment.

[0082] The transformer 10, which may be, for example, a high voltage transformer, includes sensors 20 that measure various quantities related to the transformers 10A, 10B, such as voltage, current, operating load, ambient temperature, moisture, and / or oxygen content of various components of the transformer 10, and transmit the measurements to a transformer monitoring device 30 via a communication channel 14. For example, the sensors 30 may be configured to obtain measurements related to, in this example, a bushing 22 or other insulating components of the transformer 10. Additionally, the transformer 10 may include subsystems, such as a live section 24 coupled to a power line 28 (e.g., an overhead power line), a cooling system 26 (e.g., for a transformer or reactor), which may operate by or in response to instructions from, for example, a processor circuit 34.

[0083] In this and other examples, for ease of explanation, the embodiments are described in the context of a transformer, however, it should be understood that many other types of electrical equipment and components thereof, such as reactors, transmission lines, potential transformers, generators, etc., can benefit from the embodiments described herein, and all such electrical equipment should be considered to be within the scope of the present disclosure.

[0084] These measurements may be used by the transformer monitoring device 30 to detect and / or determine the presence of faults in various components or subsystems of the transformers 10A, 10B and / or general fault conditions of the transformer 10. The communication channel 14 may include a wired or wireless link, and in some embodiments may include a wireless local area network (WLAN) or a cellular communication network, such as a 4G or 5G communication network.

[0085] The transformer monitoring system 1500 can receive online or offline measurements of voltage, current, operating load, temperature, moisture, oxygen content, etc. from the transformers 10A, 10B and process the measurements to perform the operations described herein. The transformer monitoring system 1500 can be implemented on a server, a server cluster, a cloud-based remote server system, and / or a standalone device. Sensor data can be obtained by the transformer monitoring system 1500 from a single transformer and / or multiple transformers.

[0086] The transformer monitoring system 1500 described herein can be implemented in many different ways. For example, the transformer monitoring system 1500 according to some embodiments can receive online / offline data that is used by machine learning or other predictive techniques described in various embodiments. Devices may be connectable to one or more transformers 10 to receive diagnostic parameter values ​​and / or other types of measurement data.

[0087] In the above description of various embodiments of the inventive concept, it should be understood that the terms used herein are merely for the purpose of describing specific embodiments, and are not intended to limit the inventive concept. Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by those skilled in the art to which the inventive concept belongs. It should be further understood that terms such as those defined in commonly used dictionaries should be interpreted to have a meaning that is not inconsistent with the meaning of those terms in the context of this specification and related art.

[0088] When an element is referred to as being "connected," "coupled," "responsive" to another element, or variations thereof, it may be a direct connection, coupling, or response, or there may be intervening elements. In contrast, when an element is referred to as being "directly connected," "directly coupled," "directly responsive" to another element, or variations thereof, there are no intervening elements. Like numbers refer to like elements throughout. Furthermore, as used herein, "coupled," "connected," "responsive," or variations thereof may include wireless coupling, connection, or response. As used herein, the singular forms "a," "an," and "the" are intended to include the plural, unless the context is otherwise clear. Well-known features or configurations may not be described in detail for purposes of brevity and / or clarity. The term "and / or" includes any and all combinations of one or more of the items listed in association with this term. The phrase "at least one of A and B" means "A or B" or "A and B."

[0089] Although terms such as "first", "second", "third", etc. may be used herein to describe various elements / operations, it is understood that these elements / operations should not be limited by these terms. These terms are merely used to distinguish one element / operation from another. Thus, a first element / operation in some embodiments can be referred to as a second element / operation in other embodiments without departing from the teachings of the inventive concept. The same reference numbers or characters refer to the same or similar elements throughout this specification.

[0090] As used herein, the terms "comprise", "comprising", "comprises", "include", "including", "includes", "have", "has" or "having" or variations thereof are open-ended and include one or more stated features, wholes, elements, steps, components or functions but do not exclude the presence or addition of one or more other features, wholes, elements, steps, components, functions or groups thereof.

[0091] Exemplary embodiments are described herein with reference to block diagrams and / or flowchart illustrations of computer-implemented methods, apparatus (systems and / or devices), and / or computer program products. It will be understood that blocks of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions executed by one or more computer circuits. These computer program instructions can be provided to general-purpose computer circuits, special-purpose computer circuits, and / or processor circuits of other programmable data processing circuits, and the instructions can be executed via the processor of the computer and / or other programmable data processing apparatus to produce a machine such that transistors, values ​​stored in memory locations, and other hardware components in such circuits are transformed and controlled to perform the functions / operations specified in one or more blocks of the block diagrams and / or flowcharts to produce means (functions) and / or structure for implementing the functions / operations specified in the blocks of the block diagrams and / or flowcharts.

[0092] Furthermore, these computer program instructions may be stored on a tangible computer-readable medium that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored on the computer-readable medium result in an article of manufacture that includes instructions that implement the functions / acts specified in one or more blocks of the block diagrams and / or flowcharts. Thus, embodiments of the inventive concepts may be implemented in hardware and / or software (including firmware, resident software, microcode, etc.) executing on a processor, such as a digital signal processor, which may collectively be referred to as a "circuit," "module," or variants thereof.

[0093] It should also be noted that in some alternative implementations, the functions / acts described in the blocks may occur in a different order than that described in the flowcharts. For example, two blocks shown in succession may in fact be executed substantially simultaneously, or the blocks may sometimes be executed in the reverse order, depending on the functions / acts involved. Furthermore, the functionality of a given block of the flowcharts and / or block diagrams may be separated into multiple blocks, and / or the functionality of two or more blocks of the flowcharts and / or block diagrams may be at least partially integrated. Finally, other blocks may be added / inserted between the illustrated blocks, and / or blocks / acts may be omitted, without departing from the scope of the inventive concept. Furthermore, while some of the figures include arrows on communication paths to indicate a primary direction of communication, it should be understood that communication may occur in a direction opposite to that of the depicted arrows.

[0094] Many variations and modifications can be made to the embodiments without substantially departing from the principles of the inventive concept. All such variations and modifications are intended to be included herein within the scope of the inventive concept. Therefore, the subject matter disclosed above should be considered as illustrative rather than limiting, and the example embodiments are intended to cover all such modifications, enhancements, and other embodiments that fall within the spirit and scope of the inventive concept. Therefore, to the maximum extent permitted by law, the scope of the inventive concept should be determined by the broadest permissible interpretation of this disclosure, including the example embodiments and their equivalents, and should not be limited or restricted by the foregoing detailed description.

Claims

1. determining, by a processor circuit, a prediction error value for a plurality of predicted diagnostic parameter values ​​over a predetermined time period for at least one component of electrical equipment, the prediction error value suppressing observed ambient variations in behavior of the at least one component; comparing the determined prediction error value with an expected prediction error value; selectively generating, by the processor circuitry, an indication of a status of the at least one component based on the comparison; and Including, the plurality of predicted diagnostic parameter values ​​relating to expected behavior of the at least one component; The method of claim 1, wherein the prediction error value represents a deviation of an observed behavior of the at least one component from the expected behavior of the at least one component.

2. the at least one component includes an insulating component of the electrical equipment; The method of claim 1 , wherein the plurality of predicted diagnostic parameter values ​​comprises a plurality of predicted insulation diagnostic parameter values.

3. 3. The method of claim 2, wherein the plurality of predicted insulation diagnostic parameter values ​​comprises at least one of a predicted capacitance value, a predicted capacitive current value, a predicted dissipation factor value, and a predicted power factor value of the at least one insulation component.

4. The electrical equipment includes a transformer; The method of claim 2 , wherein the at least one component comprises a high voltage bushing of the transformer.

5. The method of claim 1 , wherein the ambient variation observed in the behavior of the at least one component being inhibited comprises variation due to ambient conditions.

6. Determining the prediction error value comprises: predicting, by the processor circuit, at least one error value related to the plurality of predicted diagnostic parameter values; determining a variation of the at least one error value due to ambient conditions observed in a behavior of the at least one component; and generating the predicted error value based on the at least one error value and the determined variance. The method of claim 1 further comprising:

7. Determining the prediction error value comprises: predicting the plurality of predicted diagnostic parameter values ​​for a plurality of respective time points within the predetermined time period based on the obtained diagnostic parameter values; determining a plurality of error values ​​based on a comparison of the plurality of predicted diagnostic parameter values ​​for the respective time points and a plurality of actual diagnostic parameter values ​​obtained at the respective time points; Further comprising: The method of claim 1 , wherein the predicted error value comprises an average error value of the plurality of error values.

8. The method of claim 7 , wherein the plurality of actual diagnostic parameter values ​​are obtained from a parameter value data stream generated from a device associated with the at least one component.

9. The method of claim 7 , wherein the plurality of time points comprises at least 100 time points within the predetermined period of time.

10. The method of claim 1 , wherein the expected prediction error value is determined based on a comparison of a plurality of previous predicted diagnostic parameter values ​​to a corresponding plurality of previously obtained diagnostic parameter values.

11. 2. The method of claim 1, wherein the plurality of predicted diagnostic parameter values ​​are determined based on a plurality of determined relationships between a predetermined number of diagnostic parameter values ​​of a plurality of previously acquired diagnostic parameter values ​​and at least one subsequent parameter value of the plurality of previously acquired diagnostic parameter values.

12. The method of claim 11 , wherein the plurality of previously obtained diagnostic parameter values ​​are obtained from a different component than the at least one component.

13. The method of claim 1 , wherein the plurality of predicted diagnostic parameter values ​​are determined based on at least one of a machine learning model and a statistical model.

14. The method of claim 1 , wherein the expected prediction error value is determined based on at least one of a machine learning model and a statistical model.

15. Selectively generating the indication includes: determining, by the processor circuit, whether the prediction error value meets a predetermined prediction error threshold based on the expected prediction error value; generating a first warning indication in response to the prediction error value satisfying the predetermined prediction error threshold; The method of claim 1 further comprising:

16. The method of claim 15 , wherein selectively generating the indication further comprises generating a second warning indication in response to the prediction error value not meeting the predetermined prediction error threshold.

17. A processor circuit; A memory containing machine-readable instructions; Equipped with The machine-readable instructions, when executed by the processor circuit, determining a plurality of predictive diagnostic parameter values ​​over a predetermined time period for at least one component of the electrical equipment; obtaining a plurality of actual diagnostic parameter values ​​from the at least one component over a predetermined period of time; determining a prediction error value based on the plurality of predicted diagnostic parameter values ​​and the plurality of actual parameter values, the prediction error value suppressing ambient variations observed in behavior of the at least one component; comparing said determined prediction error value with an expected prediction error value; selectively transmitting an indication of a status of the at least one component to the electrical equipment based on the comparison. causing the processor circuit to execute the steps of: the plurality of predicted diagnostic parameter values ​​relating to expected behavior of the at least one component; the prediction error value represents a deviation of an observed behavior of the at least one component from the expected behavior of the at least one component. Insulation diagnostic system.

18. the at least one component comprises an insulating component of the electrical equipment; the plurality of predicted diagnostic parameter values ​​includes a plurality of predicted insulation diagnostic parameter values; The system of claim 17 , wherein the plurality of actual diagnostic parameter values ​​comprises a plurality of actual insulation diagnostic parameter values.

19. the plurality of predicted insulation diagnostic parameter values ​​comprising at least one of a predicted capacitance value, a predicted capacitive current value, a predicted dissipation factor value, and a predicted power factor value of the at least one insulation component; 20. The system of claim 18, wherein the plurality of actual insulation diagnostic parameter values ​​represent at least one of an actual capacitance value, an actual capacitive current value, an actual dissipation factor value, and an actual power factor of the at least one insulation component.

20. The system of claim 17 , wherein the ambient variation observed in the behavior of the at least one component includes variation due to ambient conditions.

Images