Insulation deterioration diagnosis device for electric motor and insulation deterioration diagnosis method for electric motor

Abstract

To provide an insulation deterioration diagnosis device for an electric motor that does not require high-level specialized knowledge and can obtain quantitative insulation deterioration diagnosis results without individual differences, on the basis of measurement data related to partial discharges.SOLUTION: An insulation deterioration diagnosis device for an electric motor according to the present disclosure includes: a measuring instrument that outputs a total discharge charge amount and a number of discharges under the diagnostic voltage over a plurality of test counts on the basis of a measurement result of a current value in a state in which the diagnostic voltage is applied to an electric motor being a diagnosis target; a feature quantity calculation unit that acquires the total discharge charge amount and the number of discharges over the plurality of test counts from the measuring instrument as measurement data and calculates a feature quantity being an index value of discharge intensity on the basis of the measurement data; and an estimation unit that estimates an insulation deterioration state of the electric motor on the basis of the feature quantity.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an insulation deterioration diagnosis device and an insulation deterioration diagnosis method for an electric motor that can quantitatively estimate the insulation deterioration state of an electric motor. [Background technology]

[0002] High-voltage electric motors (hereafter referred to as electric motors) used in power plants, general industry, etc. use insulating materials to separate the coil, iron core, and wires to form circuits. Such insulating materials are subjected to thermal, electrical, mechanical, and environmental stresses that occur during motor operation, causing minute voids to form and expand, resulting in insulation deterioration. This ultimately leads to the formation of electrical trees, resulting in insulation breakdown (loss of function).

[0003] Furthermore, if insulation breakdown occurs in an electric motor during operation, it can lead to serious accidents such as long-term shutdowns of power generation plants and general industrial production. For this reason, it is important to carry out maintenance such as insulation renewal before insulation breakdown occurs.

[0004] As a conventional technology for diagnosing the state of insulation deterioration, there is a partial discharge monitoring device that can detect signs of insulation deterioration in high-voltage motors by performing partial discharge measurements at the daily inspection level during operation (see, for example, Non-Patent Document 1).

[0005] According to the conventional technology disclosed in Non-Patent Document 1, the following three effects can be obtained. Effect 1: Partial discharge can be measured in conjunction with insulation diagnosis while the motor is running or stopped. Effect 2: The amount of partial discharge charge and the number of discharges in the coil insulation can be obtained as measurement data, and signs of the progression of insulation deterioration can be identified by trend management of the measurement data. Effect 3: Based on the measurement data, it is possible to consider optimal preventive maintenance measures for electric motors.

[0006] Periodic maintenance, which involves inspections and replacements at set intervals, is the mainstream method for diagnosing deterioration. However, if insulation deterioration progresses faster than the set intervals, insulation breakdown in the motor may occur while the plant is in operation, potentially leading to a serious accident. On the other hand, if the set intervals are set too short, inspections may be performed before insulation deterioration has progressed, resulting in over-maintenance.

[0007] In particular, the stator coil insulation of high-voltage electric motors deteriorates due to a combination of thermal, electrical, mechanical, and environmental stresses caused by long-term operation. The device disclosed in Non-Patent Document 1 can be used as a tool to understand the progression of such deterioration by collecting measurement data and to prevent sudden stator coil insulation failures.

[0008] In other words, the device according to Non-Patent Document 1 can be effectively utilized in periodic maintenance, and based on the measurement data collected during periodic maintenance, it is possible to consider optimal preventive maintenance measures for the motor and appropriately set the timing of the next maintenance.

[0009] Furthermore, the device according to Non-Patent Document 1 can perform online trend management of the partial discharge charge amount and the number of discharges even while the plant is in operation, and can grasp signs of the progression of insulation degradation. [Prior art documents] [Non-patent literature]

[0010] [Non-Patent Document 1] Mitsubishi Electric Plant Engineering Corporation website, Portable Motor Guard (Online Partial Discharge Monitoring Device) (URL: https: / / www.mpec.co.jp / service / monitoring / motor.html) Summary of the Invention [Problem to be solved by the invention]

[0011] As described above, the device according to Non-Patent Document 1 can obtain quantitative measurement data on the amount of partial discharge charge and the number of discharges in the coil insulation of the electric motor to be diagnosed, based on the current state. Furthermore, based on the measurement data obtained by the device according to Non-Patent Document 1, it becomes possible to grasp the current state of insulation degradation and to predict signs of future progression.

[0012] However, in reality, diagnosticians with domain knowledge such as knowledge about insulation degradation and knowledge about the on-site environment are used to understand the current state of insulation degradation and predict signs of future progression based on the measurement data obtained by the device described in Non-Patent Document 1.

[0013] That is, although quantitative measurement data can be obtained by utilizing the device according to Non-Patent Document 1, highly specialized knowledge of insulation deterioration is important to improve the accuracy of deterioration diagnosis based on the measurement data, and currently there is a risk that the diagnosis results may vary depending on the person performing the diagnosis.

[0014] Therefore, it is highly desirable to obtain quantitative insulation deterioration diagnosis results that are consistent with individual differences based on quantitative measurement data.

[0015] The present disclosure has been made to solve the above-mentioned problems, and aims to provide an electric motor insulation deterioration diagnosis device and electric motor insulation deterioration diagnosis method that do not require advanced specialized knowledge and that can obtain quantitative insulation deterioration diagnosis results that are independent of individual differences based on measurement data related to partial discharge. [Means for solving the problem]

[0016] The electric motor insulation deterioration diagnosis device according to the present disclosure includes a measuring instrument that outputs the total discharge charge amount and the number of discharges at the diagnostic voltage over a number of tests based on the measurement results of the current value when the diagnostic voltage is applied to the electric motor to be diagnosed, a feature calculation unit that acquires the total discharge charge amount and the number of discharges over the number of tests as measurement data from the measuring instrument and calculates a feature amount that serves as an index value of the discharge strength based on the measurement data, and an estimation unit that estimates the insulation deterioration state of the electric motor based on the feature amount.

[0017] Furthermore, the method for diagnosing insulation deterioration of an electric motor according to the present disclosure is a method for diagnosing insulation deterioration of an electric motor executed by a controller provided in an insulation deterioration diagnosis device for an electric motor that estimates the state of insulation deterioration of the electric motor based on the measurement results of the current value when a diagnostic voltage is applied to the electric motor to be diagnosed, and includes a first step of receiving the total discharge charge amount and number of discharges over a plurality of inspections as measurement data from a measuring instrument that outputs the total discharge charge amount and number of discharges at the diagnostic voltage over a plurality of inspections, a second step of calculating a feature value that serves as an index value of the discharge intensity based on the measurement data received in the first step, and a third step of estimating the state of insulation deterioration of the electric motor based on the feature value calculated in the second step. [Effects of the Invention]

[0018] According to the present disclosure, it is possible to obtain an insulation deterioration diagnosis device and an insulation deterioration diagnosis method for an electric motor that do not require advanced specialized knowledge and that can obtain quantitative insulation deterioration diagnosis results that are independent of individual differences based on measurement data related to partial discharge. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a functional block diagram of an insulation deterioration diagnosis device for an electric motor according to a first embodiment of the present disclosure. [Figure 2] 1 is an explanatory diagram relating to insulation deterioration that occurs in an electric motor that is a target for diagnosis by an insulation deterioration diagnosis device according to a first embodiment of the present disclosure. [Figure 3]FIG. 2 is an explanatory diagram illustrating a discharge measurement circuit using a measuring instrument according to the first embodiment of the present disclosure. [Figure 4] FIG. 10 is an explanatory diagram showing a plot of the relationship between the total discharge charge amount obtained for a plurality of inspections and the number of discharges in the first embodiment of the present disclosure. [Figure 5] 2 is an explanatory diagram illustrating a feature calculation process executed by a feature calculation unit according to the first embodiment of the present disclosure, divided into input, processing, and output. FIG. [Figure 6] 4 is an explanatory diagram relating to a process of estimating an insulation degradation state when a gradient amount is used as a feature amount in an estimation unit according to the first embodiment of the present disclosure. FIG. [Figure 7] FIG. 10 is an explanatory diagram illustrating a method for estimating an insulation degradation state by an estimating unit when the distribution pattern of measurement data changes so as to shift in the first embodiment of the present disclosure. [Figure 8] 4 is a flowchart showing a series of processes executed in the method for diagnosing insulation deterioration of an electric motor according to the first embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0020] Hereinafter, preferred embodiments of the motor insulation deterioration diagnosis device and motor insulation deterioration diagnosis method of the present disclosure will be described with reference to the drawings. The electric motor insulation deterioration diagnosis device and electric motor insulation deterioration diagnosis method disclosed herein have a technical feature in that they use, as measurement data, the change in current when a diagnostic voltage is applied to the electric motor to be diagnosed, and calculate a feature value that serves as an index value for discharge strength based on the measurement data, thereby achieving quantitative insulation deterioration diagnosis that is independent of individual differences.

[0021] Embodiment 1 1 is a functional block diagram of an insulation deterioration diagnosis device for an electric motor according to a first embodiment of the present disclosure. An insulation deterioration diagnosis device 10 according to the first embodiment is configured with a measuring instrument 11, a feature calculation unit 12, and an estimation unit 13, and has a function of diagnosing the insulation deterioration state of an electric motor that is the diagnosis target.

[0022] The measuring instrument 11 has a function of outputting the total discharge charge amount and the number of discharges at the diagnostic voltage for a plurality of tests based on the measurement result of the current value when the diagnostic voltage is applied to the motor.

[0023] The partial discharge monitoring device disclosed in Non-Patent Document 1 can be used as the measuring instrument 11. A specific measurement method using the measuring instrument 11 will be described later with reference to FIGS.

[0024] The feature calculation unit 12 has the function of acquiring the total discharge charge amount and the number of discharges over a number of tests from the measuring instrument 11 as measurement data, and calculating a feature that serves as an index value of the discharge intensity based on the measurement data.

[0025] The insulation degradation diagnosis device 10 according to the first embodiment realizes quantitative insulation degradation diagnosis that is consistent with individual differences by calculating a feature amount as an index value of discharge intensity using the feature amount calculation unit 12 based on measurement data obtained when a diagnostic voltage is applied. A specific method for calculating the feature amount using the feature amount calculation unit 12 will be described later with reference to Figs. 4 and 5.

[0026] The estimation unit 13 has a function of estimating the insulation deterioration state of the motor based on the feature calculated as the index value of the discharge intensity by the feature calculation unit 12. A specific estimation method by the estimation unit 13 will be described later with reference to Figs. 6 and 7.

[0027] Next, insulation deterioration of the electric motor to be diagnosed will be described with reference to Fig. 2. Fig. 2 is an explanatory diagram relating to insulation deterioration that occurs in the electric motor to be diagnosed by the insulation deterioration diagnosis device according to the first embodiment of the present disclosure. Fig. 2(a) is an explanatory diagram summarizing the process leading up to insulation breakdown in chronological order. Fig. 2(b) is an explanatory diagram showing an equivalent circuit based on a void model in the insulation layer.

[0028] Figure 2(a) shows the progression of insulation deterioration in a motor equipped with an iron core, main insulation, wire insulation, and wires (conductors) as a time series transition between states 1, 2, and 3. As the motor operates, the main insulation, which corresponds to the insulating material provided inside the motor, undergoes the state transitions shown in Figure 2(a) and eventually reaches insulation breakdown.

[0029] <State 1: Defects caused by heat> Thermal stresses associated with motor operation cause tiny air gaps, or voids, to form in the main insulation.

[0030] <State 2: Progression due to mechanical stress> As the operating time progresses, mechanical stresses cause further voids to form and expand.

[0031] <State 3: Electrical tree growth> When voltage is applied to the motor, discharges occur inside and outside the main insulation, and the main insulation is gradually eroded by heat and ion collisions caused by the discharge in the voids. This ultimately results in a ground fault current, which leads to breakdown of the main insulation and material insulation.

[0032] As shown in Figure 2(b), the discharge charge q increases as the spatial distance dg of the void and the area S of the void increase. Therefore, the discharge charge q serves as a barometer that reflects the progression of defects in the insulation. Therefore, by measuring the changes in the current flowing through the motor using measuring instrument 11, it is possible to monitor the progression of insulation deterioration.

[0033] 3 is an explanatory diagram showing a discharge measurement circuit using a measuring instrument 11 according to the first embodiment of the present disclosure. As shown in FIG. 3(a), the measuring instrument 11 can measure the value of a current flowing in a path of a discharge current when a diagnostic voltage is applied to an electric motor to be diagnosed.

[0034] If a void occurs in the main insulation, a discharge current occurs in the void as shown in Fig. 3(b), and this discharge current is measured by the measuring instrument 11. The discharge current generated in the void occurs twice in one cycle of the diagnostic voltage, as a positive discharge and a negative discharge.

[0035] Therefore, the measuring instrument 11 can apply a certain diagnostic voltage and output the total discharge charge amount and the number of discharges as two inspections per cycle based on the discharge current generated at that time. The partial discharge monitoring device disclosed in Non-Patent Document 1 can be used as such a measuring instrument 11.

[0036] Next, the characteristic amount calculation unit 12, which is a technical feature of the insulation deterioration diagnosis device 10 for an electric motor according to the first embodiment, will be described in detail with reference to FIGS.

[0037] 4 is an explanatory diagram showing a plot of the relationship between the total discharge charge amount obtained for a plurality of inspections and the number of discharges in the first embodiment of the present disclosure, based on the assumption that there is a correlation between the number of discharges and the total discharge charge amount.

[0038] Here, the total discharge charge is the sum of the positive and negative discharge charge amounts in one cycle of the diagnostic voltage, and two measurement data can be obtained. The graph shown in Figure 4 shows the number of discharges on the horizontal axis and the total discharge charge on the vertical axis, with 2N measurement data corresponding to N cycles plotted.

[0039] The feature calculation unit 12 calculates the slope of an approximate line based on the least squares method on the basis of the graph of the measurement data shown in FIG. 4, thereby obtaining the slope as a feature that serves as an index value of the discharge intensity.

[0040] Note that the feature calculation unit 12 can also calculate the following values ​​based on the measurement data as feature values ​​that serve as index values ​​of the discharge intensity, in addition to the amount of slope. Based on the measurement data, statistical quantities such as the maximum value, the value obtained by subtracting the minimum value from the maximum value, the average value, the root mean square value, the variance, or the standard deviation are calculated and used as feature quantities that serve as index values ​​for the discharge intensity. From the two-dimensional plane shown in Figure 4, the area is calculated using a graph construction algorithm such as a convex hull, and this is used as a feature value that serves as an index value for the discharge intensity.

[0041] In other words, the feature calculation unit 12 can calculate various values ​​as described above as feature values ​​that serve as index values ​​of the discharge intensity based on the measurement data consisting of multiple test numbers obtained by the measuring instrument 11 according to the magnitude of the applied diagnostic voltage.

[0042] Furthermore, the feature amount calculation unit 12 can calculate appropriate feature amounts in advance based on an expert's viewpoint in accordance with the installation environment of the electric motor to be diagnosed, etc. Furthermore, the feature amount calculation unit 12 may output not one but multiple feature amounts to the estimation unit as feature amounts suited to the installation environment, etc., and may automatically determine, as the feature amount, an index value that maximizes the feature amount using an optimization algorithm, etc.

[0043] In the following, a specific example will be described in which the feature calculation unit 12 calculates the amount of slope when the measurement data is mapped onto a two-dimensional plane with the number of discharges on the X axis and the total discharge charge amount on the Y axis as a feature that serves as an index value for the discharge intensity.

[0044] 5 is an explanatory diagram showing the feature calculation process executed by the feature calculation unit 12 according to the first embodiment of the present disclosure, divided into input, processing, and output. Each of the input, processing, and output stages will be described below.

[0045] <input> In the input stage, the feature calculation unit 12 acquires a plurality of measurement data individually obtained from the measuring instrument 11 when each of a plurality of different diagnostic voltages is applied as data for the number of tests corresponding to the diagnostic voltages. Here, an example of specific voltage values ​​to be applied as the plurality of different diagnostic voltages is five voltages from 1 kV to 5 kV in 1 kV increments as the diagnostic voltages.

[0046] For example, the feature calculation unit 12 can obtain measurement data consisting of the number of discharges and the total discharge charge amount for 2N tests from the measurement results for N cycles by the measuring instrument 11 for each of a plurality of diagnostic voltages.

[0047] <Processing> In the processing step, the feature calculation unit 12 calculates a slope as a feature for each of the plurality of diagnostic voltages. In calculating an individual feature for each diagnostic voltage, the feature calculation unit 12 can perform the following two-stage processing.

[0048] In the first stage of processing, the feature calculation unit 12 performs outlier removal processing on the measurement data for 2N tests. Specifically, as shown in Fig. 5, the feature calculation unit 12 creates a frequency distribution of the total discharge charge amount for 2N tests, with the horizontal axis representing the total discharge charge amount classified into certain ranges and the vertical axis representing the cumulative total of tests.

[0049] Then, the feature calculation unit 12 performs outlier removal processing by, for example, removing data on the total discharge charge amount outside ±2σ from the created frequency distribution. By performing such removal processing, measurement data related to discharges that occur unrelated to insulation degradation can be distinguished from measurement data related to discharges due to voids, and can be removed from the data used to calculate feature amounts.

[0050] Next, as the second stage of processing, the feature calculation unit 12 performs a process of calculating the amount of gradient for the measurement data after the outlier removal process. Specifically, the feature calculation unit 12 maps the measurement data after the outlier removal process onto a two-dimensional plane with the X axis representing the number of discharges and the Y axis representing the total discharge charge amount, and calculates the amount of gradient.

[0051] For example, the feature calculation unit 12 calculates the slope of an approximate line based on the least squares method for the measurement data distributed on a two-dimensional plane, thereby calculating the slope as a feature that serves as an index value of the discharge intensity.

[0052] The feature calculation unit 12 performs a series of processes for calculating the amount of gradient after the process for removing outliers, individually for each measurement data acquired for each diagnostic voltage.

[0053] <output> In the output stage, the feature calculation unit 12 outputs data associating each diagnostic voltage with a feature calculated and processed individually for each diagnostic voltage. Therefore, the estimation unit 13 estimates the insulation deterioration state of the electric motor to be diagnosed based on the individual gradients corresponding to the multiple diagnostic voltages output from the feature calculation unit 12.

[0054] Next, a description will be given of the process of estimating the insulation degradation state executed by the estimation unit 13. Fig. 6 is an explanatory diagram relating to the process of estimating the insulation degradation state when the slope amount is used as the feature amount in the estimation unit 13 according to the first embodiment of the present disclosure.

[0055] Figure 6 shows two diagrams, (a) and (b), as follows: FIG. 6(a): A diagram showing the first feature value FV1 calculated by the feature value calculation unit 12 at the time of the previous diagnosis of the insulation deterioration state, together with a plot of the measurement data at the time of the previous diagnosis.

[0056] Figure 6(b): The second feature FV2 calculated by the feature calculation unit 12 during the current diagnosis of the insulation deterioration state is shown together with a plot of the measurement data during the current diagnosis and the first feature FV1 calculated during the previous diagnosis.

[0057] As shown in Figure 6(b), if the second feature value FV2 corresponding to the slope calculated during the current diagnosis changes in an increasing direction compared to the first feature value FV1 corresponding to the slope calculated during the previous diagnosis, the estimation unit 13 can determine that the insulation deterioration state is progressing.

[0058] If the actual on-site diagnosis is the first time, the previous diagnosis result cannot be referred to. In such a case, the estimation unit 13 substitutes a reference feature quantity set in advance based on domain knowledge as the first feature quantity FV1, and can determine whether the insulation deterioration state is progressing.

[0059] Note that the change in the feature amount may not only be a change in the slope as shown in Fig. 6(b) but also a shift in the distribution pattern of the plotted measurement data. A supplementary explanation will be given below about the method for diagnosing the insulation deterioration state by the estimation unit 13 in such a case.

[0060] FIG. 7 is an explanatory diagram of a method for estimating the insulation degradation state by the estimating unit 13 when the distribution pattern of the measurement data changes so as to shift, according to the first embodiment of the present disclosure.

[0061] Figure 7 shows two diagrams, (a) and (b), as follows: Figure 7(a): The distribution of the measurement data during this diagnosis has shifted from the distribution of the measurement data during the previous diagnosis in the Y-axis direction, where the total discharge charge increases. Figure 7(b): This shows that the distribution of measurement data during this diagnosis has shifted from the distribution of measurement data during the previous diagnosis in the X-axis direction, where the number of discharges increases.

[0062] 7(a) shows a case where the distribution of the measurement data has shifted to region R1, which indicates a pattern where the number of discharges remains unchanged and the total discharge charge amount increases. When such a shift tendency is observed, the estimation unit 13 can estimate that void expansion has occurred since the previous diagnosis, since the total discharge charge amount is on the increase.

[0063] 7(b) shows a case where the distribution of the measurement data has shifted to region R2, which indicates a pattern where the total discharge charge amount remains unchanged and the number of discharges increases. When such a shift tendency is observed, the estimation unit 13 can estimate that an increase in voids has occurred since the previous diagnosis, since the number of discharges is on the rise.

[0064] The estimation unit 13 can determine that the insulation deterioration state is progressing in both the cases of Figure 7(a) and Figure 7(b). In other words, by having the feature calculation unit 12 obtain information on the shift in the distribution state of the measurement data together with the slope as one of the feature amounts, the estimation unit 13 can estimate that the insulation deterioration state is progressing due to the expansion or increase of voids.

[0065] The following provides additional information about a method in which the estimation unit 13 estimates the insulation deterioration state from the difference between the first feature value FV1 and the second feature value FV2 based on the first feature value FV1 from the previous diagnosis and the second feature value FV2 from the current diagnosis.

[0066] For the same motor, by performing multiple diagnoses over a period of time, it is possible to obtain the results of the previous diagnosis and the current diagnosis, as described above, and by comparing the two diagnostic results, it is expected that a more accurate estimation of the insulation deterioration state of an individual motor will be possible.

[0067] Here, as a difference index for estimating the state of insulation deterioration from the difference in the feature quantities between the two diagnostic results, the difference in statistical quantities can be used, as well as common analytical methods such as F1 score, Mahalanobis distance, and cosine distance.

[0068] Furthermore, by appropriately changing the difference index according to the characteristics of the feature value that serves as the index value of the discharge intensity, it is expected that the accuracy of estimating the insulation deterioration state according to the feature value can be improved.

[0069] Furthermore, when obtaining an estimation result of the insulation deterioration state, the estimation unit 13 may display it quantitatively, but if the degree of variation in the feature quantity and the degree of the insulation deterioration state are related, the estimation result may be displayed qualitatively, such as "large / medium / small insulation deterioration," based on some threshold value, or may be displayed in a color-coded manner, such as "red / yellow / blue."

[0070] Next, a supplementary explanation will be given of the advantages of individually calculating a plurality of feature quantities corresponding to a plurality of diagnostic voltages and estimating the insulation deterioration state of the motor using the plurality of feature quantities.

[0071] When measurement data is acquired as online measurement data by measuring instrument 11 while the motor is operating, the applied voltage value during actual operation is used as the diagnostic voltage, and only measurement data corresponding to one diagnostic voltage is obtained.

[0072] On the other hand, during diagnosis, such as before starting operation of the motor or during periodic inspection, a plurality of different diagnostic voltages can be applied to the motor, and a plurality of feature quantities corresponding to each diagnostic voltage can be calculated individually. In other words, during diagnosis, the state of insulation deterioration can be estimated using not only the feature quantity obtained from one type of applied voltage, but also the individual feature quantities obtained from multiple applied voltages.

[0073] Basically, the higher the applied voltage, the higher the possibility of detecting a discharge. Also, if insulation deterioration is advanced, the possibility of detecting a discharge increases even with a relatively low applied voltage.

[0074] However, depending on the installation environment, individual differences, or model characteristics of the electric motor being diagnosed, discharges may occur on the surface of the insulation material, rather than inside the insulation material, unrelated to insulation deterioration.Therefore, the advantages of estimating the state of insulation deterioration using individual feature quantities obtained from multiple applied voltages are supplemented below with three points.

[0075] <Point 1: Perspectives on utilization for condition estimation and cause diagnosis> It is difficult to estimate the state of insulation degradation from a single feature value for a single applied voltage, but by using multiple feature values ​​for multiple types of applied voltage, it may be possible to verify the relationship between discharge and insulation degradation for each applied voltage.

[0076] Specifically, if the deviation of the feature value between low and high applied voltages is too large, there is a high possibility that discharge unrelated to deterioration is occurring. As an example, if five voltages from 1 kV to 5 kV in 1 kV increments are used as diagnostic voltages, 1 kV, 2 kV, and 3 kV can be classified as low applied voltages, and 4 kV and 5 kV can be classified as high applied voltages.

[0077] Therefore, when a situation occurs in which the degree of deviation is too large, the estimation unit 13 does not need to compare the feature values ​​at the time of the previous diagnosis with the feature values ​​at the time of the current diagnosis, but can estimate that there is a high possibility of an unrelated discharge from the feature values ​​corresponding to the multiple diagnostic voltages at the time of the current diagnosis.

[0078] Here, the deviation may be any value that highlights the difference between feature quantities. For example, the deviation may be the difference between feature quantities or its percentage. Alternatively, a value that represents the feature quantities may be found, and the difference from the representative value may be used as the deviation. Examples of representative values ​​include the mean, median, and various percentiles in simple statistics.

[0079] After determining the representative value, the values ​​may be aggregated using statistics and the difference may be determined as the deviation. Alternatively, the feature quantities may be ranked relatively and the difference in the ranking may be used as the deviation. In either case, the deviation is a value obtained by evaluating the difference based on the magnitude relationship of the feature quantities.

[0080] <Point 2: Perspectives for monitoring the start of discharge> By calculating the characteristic amount using a plurality of applied voltages, it is possible to monitor the applied voltage at which discharge can be measured, that is, the voltage at which discharge starts.

[0081] Consider a specific example in which five applied voltages are applied in 1 kV increments from 1 kV to 5 kV in a situation where the insulation deterioration state is estimated using the feature amount at the time of the previous diagnosis and the feature amount at the time of the current diagnosis. In this case, if the following changes are observed, the estimation unit 13 can estimate that the insulation deterioration state has progressed.

[0082] <Inference process when a change occurs> During the previous diagnosis, the estimation unit 13 estimated the state of insulation deterioration by obtaining features from the measurement data. As a result, discharge was confirmed only at 4 kV and 5 kV, but no discharge was confirmed at 1, 2, or 3 kV, and therefore it was estimated that there was a high possibility of an unrelated discharge.

[0083] Then, during this diagnosis, the estimation unit 13 obtains features from the measurement data to estimate the state of insulation deterioration. If discharge is confirmed at 2 kV and 3 kV in addition to 4 kV and 5 kV, it can estimate that a new discharge has started at a lower applied voltage and that the state of insulation deterioration may have progressed.

[0084] <Point 3: Perspectives on determining multifaceted changes> During this diagnosis, the estimation unit 13 can estimate that the discharge occurring at a high applied voltage is unrelated to insulation deterioration by taking into account the degree of deviation explained in point 1. However, even if such an estimation is possible, it cannot be concluded that deterioration of the motor itself is not progressing.

[0085] Therefore, if it is estimated at the time of the previous diagnosis that the deterioration of the motor itself has not progressed, the estimation unit 13 compares the feature values ​​at the time of the previous diagnosis with the feature values ​​at the time of the current diagnosis for multiple applied voltages, thereby selecting the applied voltage that truly requires attention, and can estimate the state of insulation deterioration of the motor based on the amount of change in the feature value at the selected applied voltage.

[0086] For example, as explained in point 2, if it can be estimated that the feature quantities of 4 kV and 5 kV in the previous diagnosis are likely to be unrelated discharges, the estimation unit 13 can select only 1 kV and 2 kV as the applied voltages to pay attention to. Furthermore, the estimation unit can estimate the state of insulation degradation based on the amount of change in the feature quantities calculated corresponding to 1 kV and 2 kV between the previous diagnosis and the current diagnosis.

[0087] In addition, a threshold based on know-how or a threshold based on statistics may be used to estimate the state of insulation deterioration based on the amount of change, and a ranking based on relative ranking may be presented.

[0088] Finally, a series of processes executed by the above-described electric motor insulation deterioration diagnosis device 10 according to the first embodiment will be described using a flowchart. Fig. 8 is a flowchart showing a series of processes executed as the electric motor insulation deterioration diagnosis method according to the first embodiment of the present disclosure.

[0089] Steps S801 to S807 shown in FIG. 8 correspond to the execution of the functions of the measuring instrument 11, the feature calculation unit 12, and the estimation unit 13 by the controller of the insulation degradation diagnosis device 10 shown in FIG. 1.

[0090] In step S801, the measuring instrument 11 measures the total discharge charge amount and the number of discharges when the diagnostic voltage is applied.

[0091] Next, in step S802, the feature calculation unit 12 receives the measurement data from the measuring instrument 11 as the measurement result, and calculates a feature that serves as an index of the discharge intensity based on the measurement result.

[0092] Next, in step S803, the estimation unit 13 determines whether or not there is a feature amount calculated at the time of the previous diagnosis. If the estimation unit 13 determines YES, the process proceeds to step S804, and if the estimation unit 13 determines NO, the process proceeds to step S806.

[0093] When the process proceeds to step S804, the estimation unit 13 sets the feature calculated by the feature calculation unit 12 during the previous diagnosis as the first feature FV1, and sets the feature calculated by the feature calculation unit 12 during the current diagnosis as the second feature FV2.

[0094] Next, in step S805, the estimation unit 13 performs an estimation process of the insulation deterioration state based on the first feature FV1 and the second feature FV2, more specifically, based on the transition state from the first feature FV1 to the second feature FV2, and then proceeds to processing in step S807.

[0095] On the other hand, if the process proceeds to step S806, the estimation unit 13 executes a process of estimating the insulation deterioration state based on the feature calculated by the feature calculation unit 12 during the current diagnosis, and then proceeds to the process of step S807.

[0096] Finally, when the process proceeds to step S807, the estimation unit 13 stores the feature calculated by the feature calculation unit 12 during this diagnosis as the first feature to be utilized during the next diagnosis, and ends the series of processes.

[0097] As described above, according to the first embodiment, the insulation degradation diagnosis device is configured by including a measuring instrument, a feature calculation unit, and an estimation unit, and realizes the following three functions. Function 1: A function that outputs the total discharge charge amount and number of discharges at the diagnostic voltage for multiple tests based on the measurement results of the current value when the diagnostic voltage is applied to the electric motor being diagnosed.

[0098] Function 2: A function that acquires the total discharge charge and number of discharges across multiple inspections as measurement data, and calculates feature values ​​that serve as indicators of discharge strength based on the measurement data. Function 3: A function to estimate the state of insulation deterioration of a motor based on feature quantities.

[0099] By having such functions, it is possible to realize an electric motor insulation deterioration diagnosis device and an electric motor insulation deterioration diagnosis method that do not require advanced specialized knowledge and can obtain quantitative insulation deterioration diagnosis results that are independent of individual differences based on measurement data related to partial discharge.

[0100] Furthermore, by storing the feature values ​​calculated each time a diagnosis is performed, it is possible to perform an insulation degradation diagnosis taking into account time-series changes in the feature values. Furthermore, it is possible to perform an insulation degradation diagnosis taking into account each feature value corresponding to a plurality of applied voltages. Furthermore, it is possible to perform an insulation degradation diagnosis based on the feature values ​​online as needed.

[0101] As the feature amount, a value that serves as an index value of the discharge intensity can be used, and in particular, a statistical amount relating to measurement data from a plurality of inspections can be used as the feature amount.

[0102] As a result, it is possible to improve the accuracy of estimating the state of insulation deterioration for each motor, taking into account the installation environment, operating conditions, etc., and to provide signs of progressing insulation deterioration and optimal preventive maintenance measures for the motor based on quantitative estimation results that are consistent with individual differences. [Explanation of symbols]

[0103] 10 insulation deterioration diagnosis device, 11 measuring instrument, 12 feature calculation unit, 13 estimation unit.

Claims

1. a measuring instrument that outputs a total discharge charge amount and a discharge count at a diagnostic voltage based on a measurement result of a current value when a diagnostic voltage is applied to an electric motor to be diagnosed, over a plurality of tests; a feature calculation unit that acquires the total discharge charge amount and the number of discharges over the plurality of inspections as measurement data from the measuring device, and calculates a feature that serves as an index value of discharge intensity based on the measurement data; an estimation unit that estimates an insulation deterioration state of the electric motor based on the feature amount; An insulation deterioration diagnosis device for an electric motor comprising:

2. The estimation unit storing the feature amount calculated by the feature amount calculation unit at the time of the previous diagnosis as a first feature amount; storing the feature amount calculated by the feature amount calculation unit during the current diagnosis as a second feature amount; An insulation deterioration state of the electric motor is estimated based on the second feature amount and based on a transition state from the first feature amount to the second feature amount. The motor insulation deterioration diagnosis device according to claim 1.

3. the feature calculation unit acquires, as the measurement data, a plurality of pieces of measurement data when a plurality of different diagnostic voltages are applied, and calculates, as the feature data, a plurality of feature amounts corresponding to each of the plurality of pieces of measurement data; The estimation unit estimates an insulation deterioration state of the electric motor based on the plurality of feature amounts. The motor insulation deterioration diagnosis device according to claim 1.

4. the feature calculation unit acquires, as the measurement data, a plurality of pieces of measurement data when a plurality of different diagnostic voltages are applied, and calculates, as the feature data, a plurality of feature amounts corresponding to each of the plurality of pieces of measurement data; The estimation unit estimates an insulation deterioration state of the electric motor based on the plurality of feature amounts. The motor insulation deterioration diagnosis device according to claim 2.

5. the feature calculation unit acquires the measurement data from the measuring instrument as online measurement data while the electric motor is in operation, and calculates the feature based on the online measurement data; The estimation unit estimates an insulation deterioration state of the electric motor online based on the feature amount. The motor insulation deterioration diagnosis device according to claim 1.

6. The feature calculation unit calculates, as the feature, a slope when plotting data corresponding to the plurality of inspection numbers with the horizontal axis representing the number of discharges and the vertical axis representing the total discharge charge amount, for the measurement data acquired from the measuring instrument. The insulation deterioration diagnosis device for an electric motor according to any one of claims 1 to 5.

7. 1. A method for diagnosing insulation degradation of an electric motor, the method being executed by a controller provided in an insulation degradation diagnosis device for an electric motor, which estimates an insulation degradation state of the electric motor based on a measurement result of a current value when a diagnostic voltage is applied to the electric motor, the method comprising: a first step of receiving, as measurement data, the total discharge charge amount and the number of discharges at the diagnostic voltage over a plurality of test times from a measuring instrument that outputs the total discharge charge amount and the number of discharges over the plurality of test times; a second step of calculating a feature value serving as an index value of discharge intensity based on the measurement data received in the first step; a third step of estimating an insulation deterioration state of the motor based on the feature amount calculated in the second step; A method for diagnosing insulation deterioration in an electric motor having the above structure.

Images