JP3693227B2 - Insulation life estimation method for rotating machine and insulation life estimation system for rotating machine - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method for estimating an insulation life of a rotating machine used in an elevator or the like and a system for estimating an insulation life.
[0002]
[Prior art]
There are various patterns of performance deterioration of rotating machines used for elevators and the like. Among them, for example, deterioration of insulating performance of an insulating material such as a winding coating resin is a particularly important problem.
[0003]
As an index indicating the insulation deterioration of the insulating substance, a deterioration index (for example, dielectric breakdown voltage BDV, etc., hereinafter referred to as a breakdown index) accompanied by a breakdown of parts or materials, and a deterioration index (for example, a breakdown index (for example, breakdown voltage BDV)) Insulation resistance R, dielectric loss tangent tan δ, weak point ratio WR, polarization index PI, difference Δtan δ between dielectric loss tangent at rated voltage and dielectric loss tangent at 100 V, or product CR of capacitance C and insulation resistance R, etc. It is described as an indicator). Since the deterioration diagnosis based on the destruction index literally involves destruction, the deterioration state can be known in more detail and more accurately than the deterioration diagnosis based on the non-destructive index, but the inspected parts cannot be used continuously. Therefore, conventionally, such insulation deterioration diagnosis has been performed using a non-destructive index. FIG. 6 shows correlation data between the insulation resistance R and the elapsed operation time T. In the deterioration diagnosis based on these non-destructive indices, for example, in the insulation resistance R, an estimated value of 10 MΩ is set as the limit index value Ye, and the estimated insulation life value Te is calculated as a value corresponding to the limit index value based on this correlation. It was.
[0004]
[Problems to be solved by the invention]
However, in such deterioration diagnosis by the conventional nondestructive inspection, the limit index value by the nondestructive inspection is an empirically determined value with poor accuracy, or as shown in FIG. Due to the large variation of the index with respect to time, it was difficult to determine the insulation lifetime with high accuracy for various samples. Therefore, there has been a problem of waste of resources and costs, such as forced replacement of parts even though the lifetime is still sufficient.
[0005]
[Means for Solving the Problems]
The present invention is a method for estimating the insulation performance life of an insulator of a rotating machine, and is a first of an insulator deterioration index (X) by destructive inspection and an insulator deterioration index (Y) by nondestructive inspection. A step of calculating a limit deterioration index value (Ye) by non-destructive inspection from a limit deterioration index value (Xe) by destructive inspection based on the statistical reliability of the correlation (Fxy), and an elapsed time (T) Based on a plurality of different second correlations (Fty) according to the classification of the deterioration progress rate with the deterioration index (Y) of the insulator by the nondestructive inspection, from the limit deterioration index value (Ye) by the nondestructive inspection And a step of calculating an estimated insulation lifetime value (Te). A method for estimating an insulation lifetime of a rotating machine.
In view of the above problems, the present onset Ming, the insulator according to the nondestructive inspection degradation index (Y) and further detailed and accurate insulator according prehensible destructive testing the deterioration condition deterioration index (X) and the first By determining the critical deterioration index value (Ye) based on the non-destructive inspection based on the correlation (Fxy), it is possible to calculate the estimated insulation life value (Te) with more certainty, thereby suppressing waste of resources and costs. can do.
[0006]
The present onset Ming, estimated insulation lifetime value based on a plurality of different second correlation (Fty) between the elapsed time (T) and the degradation index of the insulation by non-destructive inspection (Y) (Te) By calculating the above, it is possible to estimate the insulation life with higher accuracy on the basis of the correlation with little variation and high correlation.
[0007]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings.
[0008]
First, a description will be given of a first embodiment in which the present invention is applied to an insulation life estimation system for a winding coating of an elevator motor. First, a schematic configuration of the present system will be described with reference to FIG.
[0009]
The system 1 includes an input unit 100 for inputting data necessary for estimation of insulation life, for example, a keyboard 101, a calculation unit 200 for performing all calculations associated with calculation of insulation life, for example, a CPU 201, and data necessary for calculation of insulation life. A storage unit 300 for storing, for example, a memory 301, and an output unit 400 for outputting a result of the calculated insulation lifetime, for example, a display 401 are provided.
[0010]
FIG. 2 shows data stored in the storage unit 300, for example, the memory 301. First non-destructive index data Yda (for example, insulation resistance R, dielectric loss tangent tan δ, etc.) and corresponding breakdown index data Xd (for example, dielectric breakdown voltage BDV) are stored in the memory 301 as the first data set Da. ing. This is shown in the table as the first table 310 in FIG. The second non-destructive index data Ydb and the elapsed time data Td corresponding to the second non-destructive index data Ydb are stored in the memory 301 as the second data set Db. This is shown in the table in the second table 320 of FIG. The second data set Db is classified by individual, for example, by each motor or by individual part, and also by the progression speed of deterioration. These divisions are performed by setting the individual flag Fu and the traveling speed flag Fs. The system 1 is configured such that the values and types of these data stored in the memory 301 can be added or changed from the input unit 100 as necessary.
[0011]
Next, the method for estimating the insulation lifetime according to the present embodiment will be described with reference to FIGS. FIG. 3 is a flowchart of the insulation life estimation method, FIG. 4 is a diagram showing a first correlation Fxy between the failure index X and the non-destructive non-destructive index Yn, and FIG. 5 is made dimensionless with the elapsed time T. It is a figure which shows the 2nd correlation Fty with the nondestructive parameter | index Yn.
[0012]
First, data necessary for estimating the insulation life is input from the input unit 100, for example, the keyboard 101 (data input step 10). In this step 10, the sample nondestructive index value Ys for which the insulation life is to be estimated and the elapsed time Ts from the start of use until the measurement of the sample nondestructive index value Ys is input. In addition, a plurality of types of sample non-destructive index values Ys (for example, the insulation resistance value Rs as the first sample non-destructive index value Ysa and the second sample non-destructive index value depending on the type of the non-destructive inspection method performed here) The dielectric loss tangent value tan δs) can be input as Ysb. Further, in this step 10, it is possible to input, set, change, select a default value, add or delete all parameter values or types stored in the memory 301 or used for estimation of the insulation life.
[0013]
Next, a first correlation Fxy between the destructive index X and the non-destructive index Y and a second correlation Fty between the non-destructive index Y and the elapsed time T are calculated (correlation Fxy and Fty calculation step 20). .
[0014]
In this step 20, first, data necessary for the calculation stored in the memory 301 is called according to the type of the sample non-destructive index value Ys input in the step 10 (data calling step 21). Here, the destruction index data Xd and the corresponding first non-destructive index data Yda are called according to the type of the sample non-destructive index value Ys. For example, for all the first non-destructive index data Yda of the same type as the sample non-destructive index value Ys for one destructive index data Xd, the destructive index data Xd and the first non-destructive index data Yda is called. More specifically, when there are two types of non-destructive index values Ys of the sample, for example, an insulation resistance value Ysa (Rs) and a dielectric loss tangent value Ysb (tan δs), these are obtained with respect to the destructive index data Xd. The same type of non-destructive index data, that is, two types of data, ie, insulation resistance data Ydaa and dielectric loss tangent data Ydab, are input in these data (Xd, Ydaa, Ydab; Call the enclosed data). Similarly, the elapsed time data Td and the corresponding second non-destructive index data Ydb are called according to the type of the sample non-destructive index value Ys. More specifically, the second non-destructive index data Ydb in which all the second non-destructive index data Ydb of the same type as the sample non-destructive index value Ys is stored for one elapsed time data Td. And elapsed time data Td are called.
[0015]
Next, for example, a method developed by the inventors (reference: Kinko, Oshima, Yamada, Iijima: Insulation diagnosis of a rotating machine for an elevator using an expert system incorporating fuzzy reasoning, Journal of the Institute of Electrical Engineers of Japan, Vol. 112, No. 11, Based on one or more first non-destructive index data Yda called in the previous step 21, the first non-destructive index data Ydan made non-dimensional is calculated by pp 931-937) (Non-destructive degradation index data non-dimensionalization step 22). For example, the called first non-destructive index data Yda includes insulation resistance data (Ydaa (i), i = 1, 2, 3,...) And dielectric loss tangent data (Ydab (i), i = 1, 2, 3, 2, and 2), according to this method, all these two types of data are integrated for each correspondence i to make each dimensionless one non-destructive first type Index data (Ydan (i), i = 1, 2, 3,...) Is calculated. Similarly, second non-destructive index data Ydbn that has been made dimensionless is also calculated for the called second non-destructive index data Ydb.
[0016]
The first non-destructive index data Ydan and the corresponding destructive index data Xd are used to obtain the first correlation Fxy, the second non-dimensional nondestructive index data Ydbn and the corresponding elapsed time data Td. The second correlation Fty is calculated from the above (correlation calculation step 23). In this step, for example, a regression coefficient, an average value, a standard deviation, and the like are calculated as the correlation. At this time, the second correlation Fty is calculated according to a flag F set in the second data set Db, for example, a flag Fs for each traveling speed. For example, when a plurality of flags Fs, for example, the third class (Fs = 1, 2, 3) is set, the same number, that is, the second correlation Fty (Fty1, Fty2, Fty3) of the third class is calculated accordingly. Is done. In the present system 1, the flag Fs for each traveling speed can be appropriately rewritten, for example, in the data input step 10, and therefore, the change of the second data set Db applied to the calculation of the second correlation Fty, or It is possible to easily increase the number of classes of the second correlation Fty.
[0017]
Next, a non-destructive sample non-destructive index value Ysn is calculated based on the input one or more sample non-destructive index values Ys by the same method as in step 22 (the non-destructive degradation index value of the sample Non-dimensionalization process 30). For example, when an insulation resistance value Ysa (Rs) and a dielectric loss tangent value Ysb (tan δs) are input as sample non-destructive index values, these are used according to the same method as in the previous step 22 based on these values. A dimensionless sample non-destructive index value Ysn in which all are integrated is calculated.
[0018]
Next, based on the first correlation Fxy, the limit non-destructive index value Ye is obtained from the limit destructive index value Xe (non-destructive limit deterioration index value calculating step 40). This step 40 will be described with reference to FIG. Here, for example, the non-dimensional non-destructive index Yn is made non-dimensional based on one or a plurality of types of non-destructive indices Y as an index having a range of 0 (poor) to 5 (good) according to the deterioration state. In addition, the fracture index X is made dimensionless by dividing the measured value Xd by the initial value X0. FIG. 4 shows a regression line 501 and a line of reliability Rxy of the first correlation Fxy, for example, a −3σ line 502 (σ: standard deviation of correlation) as the first correlation Fxy. Further, the critical fracture index value Xe (for example, Xe = 0.15) is a critical index value for deterioration of insulation performance obtained by a fracture test. Here, the dimensionless limit nondestructive index value Yen (= 1) as the value of the dimensionless nondestructive index Yn corresponding to the limit fracture index value Xe (= 0.15) in the −3σ line 502 of the first correlation Fxy. .8). As described above, by determining the limit non-destructive index value Ye based on the limit destructive index value Xe obtained by the destructive test, the estimation of the insulation life based on the non-destructive index is performed with higher accuracy from a practical viewpoint. Can do. In addition, by performing statistically based on the reliability from the first correlation as described above, it is possible to more reliably estimate the insulation life.
[0019]
Next, based on the second correlation Fty, the estimated insulation life value Te is estimated from the limit nondestructive index value Ye obtained in the previous step 40 (estimated insulation life value calculation step 50). This step 50 will be described with reference to FIG. 5 corresponding to FIG. As described above, the second correlation Fty between the elapsed time T from the start of use of the motor or the part and the dimensionless non-destructive index Yn is, for example, Fty1, Fty2, Fty3 according to the classification according to the deterioration progress rate. It is divided into three correlations. First, the dimensionless sample nondestructive index value Ysn calculated in the previous step 30, the sample elapsed time Ts, and their correlation (Fty1, Fty2, Fty3) are compared, and the degree of deterioration of the sample, for example, the deterioration line 602 is obtained. Estimate (degradation progress estimation step 51). More specifically, for example, referring to the example shown in FIG. 5, the points (603, 604, 605) of Yn = Ysn (= 4.1) and the sample 601 in each first correlation Fxy are, for example, T The values are compared and the relative position of the sample 601 with respect to each second correlation Fty is recognized. Then, from the two first correlations Fty closest to the sample 601 (Fty1 and Fty2 in the example shown in FIG. 5), the degree of deterioration of the sample, for example, a deterioration line 602 is determined. For example, in FIG. 5, the deterioration line 602 has the same ratio of the difference (606, 607) in T between the two first correlations Fty (Fty1, Fty2) closest to the sample 601 and the deterioration line 602 in each Yn. Determine as a line. Next, an estimated insulation life value Te (= 7300 days) is obtained from the degradation line 602 based on the dimensionless limit nondestructive index value Yen (= 1.8) obtained in the previous step 40 (calculation of the insulation life value). Step 52). That is, the estimated insulation life value Te is obtained as the T value when Yn = Ye in the deterioration line 602. As described above, the degree of progress of deterioration for each sample is estimated from the elapsed time value Ts of the sample and the corresponding non-destructive index value Ys based on the second correlation Fty. Highly accurate life estimation is possible. In addition, by performing based on the plurality of second correlations Fty having different deterioration progress states, the deterioration progress state of each sample can be more accurately determined based on the highly accurate correlation subdivided according to the deterioration progress state. Since it can be estimated, the insulation life can be estimated with higher accuracy.
[0020]
Then, it is determined whether or not the estimated insulation life value Te is calculated again by changing the reliability Rxy (determination step 60 for recalculating the estimated insulation life value). For example, when a plurality of reliability levels Rxy are input or set in advance in the previous data input step 10, or when it is necessary to estimate the lifetime with a different reliability level Rxy, the nondestructive limit deterioration is performed. The process returns to the index value calculation step 40. As shown in FIG. 4, for example, when the −2σ line 503 having low reliability is used as the reliability Rxy, unlike the −3σ line 502, the dimensionless limit nondestructive index value Yen2 ( = 0.8) is a low value different from the dimensionless limit non-destructive index value Yen (= 1.8) obtained from the −3σ line 502, and is calculated based on the value as shown in FIG. The insulation life value Te2 (= 7700 days) is a high value different from the previous insulation life value Te (= 7300 days). By ensuring the degree of freedom of setting the reliability Rxy in this way, for example, when estimating the life of a motor or component that is not permitted to be damaged, for example, the reliability can be set to be higher, or Estimate life, such as setting the reliability high when the progress of deterioration is fast or the margin for life is small, while setting the reliability low when the progress of deterioration is slow or the margin for the life is large The optimal reliability can be set for various situations while ensuring the accuracy of life estimation by correlation with the fracture index Xd. It can be carried out.
[0021]
The estimated insulation life value Te or remaining life value Tr (= Te−Ts) calculated in this way is output by the output unit 400 such as the display 401 together with other parameters such as Ys, Ysn, Ts, and Rxy (estimation). Insulation life output step 70). A plurality of estimated insulation lifetime values Te calculated with different reliability Rxy are displayed and graphed for each reliability Rxy.
[0022]
Then, the input non-destructive index value Ys and elapsed time value Ts of the sample are added as values of the second data set Db to the storage unit 300, for example, the memory 301 (data addition step 80). Since these added data are taken into account when estimating the lifetime of the next sample, the lifetime estimation is always performed based on the latest data.
[0023]
Note that the present embodiment is not limited to the above configuration and method.
[0024]
In the above-described embodiment, the measurement value itself of the nondestructive inspection is stored in the storage unit, but this may be a non-destructive nondestructive index value, or the first correlation Fxy or The second correlation Fty or the limit fracture index value Xe may be stored in the storage unit. According to these configurations, the number of processes for estimating the insulation life can be reduced, and the life can be estimated more quickly.
[0025]
In the above-described embodiment, the data set stored in the storage unit is divided into two, but may be stored as the same data set. In this case, when the first correlation Fxy is calculated, the data is called from the one that stores the destruction index data X, and when the second correlation Fty is calculated, the elapsed time data T is stored. Data is called.
[0026]
Further, in the above-described embodiment, the non-destructive non-destructive index data Ydn and the non-dimensional sample non-destructive index value Ysn are used to calculate the non-destructive non-destructive index Yn (Ydn, Ysn). The types of non-destructive indicators Y to be integrated as a group are the same. However, for example, regarding the nondestructive index data Yd of the same type as the sample nondestructive index value Ys, when the total number of the same type of data Yd is small, or the first or second correlation based on the same type of data Yd When the degree of correlation between Fxy and Fty is low, for example, when the degree of correlation cannot be sufficiently secured, the type of the non-destructive index Y used as the base is different from each other, for example, a higher degree of correlation is applied. You may make it do. For this purpose, a step of selecting or setting the type of the non-destructive index Y as the base may be provided. Thereby, life estimation with higher accuracy can be performed.
[0027]
【The invention's effect】
As described above, according to the present invention, since the insulation life can be estimated more reliably, it can be used for a long period of time that is close to the insulation life of the rotating machine, which wastes resources and costs. In addition to being able to be suppressed, it is possible to prevent the function from being stopped due to the end of life during use.
[Brief description of the drawings]
FIG. 1 is an explanatory diagram showing a schematic configuration of an insulation life estimation system for an elevator motor according to a first embodiment.
FIG. 2 is an explanatory diagram showing data stored in a storage unit of the insulation life estimation system for an elevator motor according to the first embodiment;
FIG. 3 is a flowchart of an insulation life estimation method performed by the insulation life estimation system according to the first embodiment.
FIG. 4 shows a first correlation between an insulator deterioration index (X) by a destructive inspection and an insulator deterioration index (Y) by a nondestructive inspection in the insulation life estimation system according to the first embodiment ( It is explanatory drawing which shows Fxy).
FIG. 5 is an explanatory diagram showing a second correlation (Fty) between an elapsed time (T) and a deterioration index (Y) of an insulator by nondestructive inspection in the insulation life estimation system according to the first embodiment; It is.
FIG. 6 is an explanatory diagram showing a correlation between an insulation resistance R and an elapsed operation time T in conventional insulation life estimation.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Insulation life estimation system, 20 Calculation process of correlation Fxy, Fty, Calculation process of limit deterioration index value Ye by 40 nondestructive inspection, Calculation process of 50 estimated insulation life value Te, 100 input part, 200 calculating part, 300 memory | storage Part, 400 output part, Xe limit deterioration index value by destructive inspection, Yn non-destructive non-destructive inspection deterioration index, Yen non-dimensional non-destructive non-destructive inspection index value, Ysn sample is made non-dimensional Degradation index value by non-destructive inspection, elapsed time value from start of use of Ts sample.

Claims (4)

A method of estimating the life of insulation performance of an insulator of a rotating machine,
Based on the statistical reliability of the first correlation (Fxy) between the deterioration index (X) of the insulator by the destructive inspection and the deterioration index (Y) of the insulator by the non-destructive inspection, the limit deterioration index by the destructive inspection Calculating a limit deterioration index value (Ye) by nondestructive inspection from the value (Xe);
Based on a plurality of different second correlations (Fty) according to the classification of the degradation progress rate between the elapsed time (T) and the degradation index (Y) of the insulator by the nondestructive inspection, the limit degradation by the nondestructive inspection is performed. Calculating an estimated insulation lifetime value (Te) from the index value (Ye);
A method for estimating an insulation life of a rotating machine.   The estimated insulation lifetime value (Te) is based on the second correlation (Fty), the elapsed time value (Ts), the corresponding deterioration indicator value (Ys) of the insulator by nondestructive inspection, and the nondestructive value. 2. The method for estimating an insulation life of a rotating machine according to claim 1, wherein the method is calculated from a limit deterioration index value (Ye) by inspection. The estimated insulation life value (Te) is calculated using a non-destructive deterioration index (Yn) based on a non-destructive deterioration index (Yn) based on a non-destructive deterioration index (Y). The estimation method of the insulation lifetime of the rotary machine as described in thru | or 2. A system for estimating the life of insulation performance of an insulator of a rotating machine,
An input unit for inputting a deterioration index value (Ys) of an insulator by nondestructive inspection;
Insulator first degradation index data (Yda) by non-destructive inspection and insulator degradation index data (Xd) by destructive inspection corresponding thereto, and second degradation index data (Ydb) of insulator by non-destructive inspection ) And an elapsed time value (Td) according to this,
The first correlation obtained from the first deterioration index data (Yda) by the nondestructive inspection and the deterioration index data (Xd) by the destructive inspection according to the type of the deterioration index value (Ys) by the nondestructive inspection (Fxy) and a plurality of different second correlations (Fty) corresponding to the categories of deterioration progress rates determined from the second deterioration index data (Ydb) and elapsed time data (Td) by the nondestructive inspection The deterioration index value (Ys) by the nondestructive inspection based on a plurality of different second correlations (Fty) according to the statistical reliability of the first correlation (Fxy) and the deterioration progress rate classification ) An estimated insulation life value (Te) obtained from the calculation unit,
A system for estimating an insulation life of a rotating machine.

Images