JP6801144B2 - Diagnostic device and diagnostic method - Google Patents

Description

The present invention relates to a technique for diagnosing a device or equipment operating on a three-phase AC power supply.

If a rotating machine such as a motor (electric motor) or a generator built into a production facility suddenly breaks down, unplanned repair work or replacement work of the rotating machine becomes necessary, resulting in a decrease in the operating rate of the production equipment and a review of the production plan. Is required. Similarly, if a power converter or cable connected to a rotating machine breaks down, unplanned repair work or replacement work is required, and it is necessary to reduce the operating rate of production equipment and review the production plan.

In order to prevent sudden failures of the rotating machine system (rotating machine and its ancillary equipment (cable, power converter)), the rotating machine system is stopped appropriately and diagnosed offline to grasp the degree of deterioration. Sudden failure can be prevented to some extent. However, since the diagnosis is performed offline, it is necessary to stop the rotating machine system, which causes a decrease in the operating rate of the production equipment. Further, depending on the type of deterioration, there are some that become apparent only when a voltage is applied. Therefore, there is a need for diagnosing the state of the rotating machine based on the current information of the rotating machine system.

Motor Current Signature Analysis (MCSA) is a method for diagnosing equipment and facilities based on current information of a rotating machine system. According to the MCSA, on the frequency spectrum of the current, specific frequency components according to each factor such as rotor bar damage, rotor eccentricity, stator core damage, winding short circuit, bearing deterioration, etc. By detecting, failure or deterioration can be detected.

Further, in Patent Document 1, especially in bearing diagnosis, vibration sensors are installed at two places, vibration sensor data is acquired from these vibration sensors, and the instantaneous value of each vibration sensor data is drawn on each axis. A method for determining an abnormality from the locus inclination of a figure and the temporal change of the radius is disclosed.

Japanese Unexamined Patent Publication No. 2000-258305

However, the techniques disclosed in MCSA and Patent Document 1 have the following problems.

MCSA is required to detect specific frequency components with high accuracy. For that purpose, it is necessary to measure the current value at a high sampling rate for a long period of time. Expensive data loggers are required to perform long-term measurements at high sampling rates, increasing diagnostic costs.

In addition, there are some motor failures in which the phenomenon appears only for a short time. In order to detect the phenomenon without missing it, it is necessary to measure the current for a long time at a high sampling rate and accumulate the data. Therefore, the amount of data to be stored becomes enormous, and the terminal for storing the data requires a large-capacity storage device, which increases the cost. In addition, when the amount of data to be stored becomes enormous, when the data is transferred via a communication path to be sent to a cloud or a local terminal, it occupies a large band of the communication path and can interfere with the communication of other devices. There is sex.

Further, in the technique disclosed in Patent Document 1, since the failure diagnosis of the motor is performed based on the data obtained by the vibration sensor, the position where the vibration changes sensitively to the failure of the motor, for example, the upper part of the motor. It is necessary to install a vibration sensor, and the installation position of the vibration sensor is limited. In addition, a vibration sensor and an expensive data logger having a sufficient sampling rate to obtain a locus of a Lissajous figure are required, which increases the diagnostic cost.

An object of the present invention is to provide a technique for efficiently diagnosing a diagnostic target operating in three-phase alternating current by suppressing the sampling rate and the amount of data for current measurement.

The diagnostic device according to one aspect of the present invention is a diagnostic device that diagnoses the state of a diagnostic object operating on a three-phase AC power supply, and acquires instantaneous values of physical data at the same time in a plurality of phases of three-phase AC. From the part, the conversion part that discretizes the instantaneous value acquired by the acquisition part, and the discrete value due to the discrimination of the conversion part, the appearance density of each combination of the discrete values of the instantaneous values of the physical data of each phase is calculated. Diagnosis that diagnoses the state of the diagnosis target based on the integration unit to be obtained, the storage unit that stores the appearance density data indicating the appearance density for each combination obtained by the integration unit, and the appearance density data accumulated in the storage unit. It has a part and.

According to one aspect of the present invention, the sampling rate and the amount of data for current measurement can be suppressed, and the state of the diagnostic object operating in three-phase alternating current can be efficiently diagnosed.

It is a block diagram of the comparative example of a diagnostic apparatus. It is a block diagram of the diagnostic apparatus according to Example 1. It is a figure which shows the amount of data which is stored in the storage part by discretization in the conversion part shown in FIG. It is a figure which shows the current waveform when the spectrum appears at the frequency 1Hz away from the fundamental wave frequency 50Hz of the U-phase current. It is a figure which shows the swell waveform generated by the current waveform shown in FIG. It is a figure which shows the current waveform of the U phase in a normal state. It is a figure which shows the current waveform of the W phase in a normal state. It is a figure which shows the current waveform of the U phase when the side band wave is generated. It is a figure which shows the current waveform of W phase when the side band wave is generated. It is a figure which shows the result of having analyzed the degree of abnormality with respect to the matrix C defined as a normal state of the matrix A and the matrix B by the current waveform shown in FIGS. 6-9. It is a block diagram of the diagnostic apparatus according to Example 2. It is a figure which shows the abnormality degree of the matrix A obtained from the current waveform of the rotating machine in a normal state acquired by the control command A. It is a block diagram of the diagnostic apparatus according to Example 3. It is a block diagram of the diagnostic apparatus according to Example 4. It is a block diagram of the diagnostic apparatus according to Example 5.

Hereinafter, the present invention will be described with reference to the drawings with reference to a plurality of examples of the embodiments. It should be noted that the following embodiments are merely examples for explanation, and the present invention is not limited to the following embodiments.

In the failure of a rotating machine such as an electric motor or a generator and a rotating machine system including a cable and a power conversion device attached to the rotating machine, the location of the failure and the cause of the failure are diverse. For example, deterioration of insulation, deterioration of bearings, short circuit, disconnection, flooding, etc. can be considered. In addition, electric motors are often installed for a long period of time in a harsh environment, and diagnostic technology according to the installation conditions is required.

FIG. 1 is a diagram showing a comparative example of a diagnostic device. As shown in FIG. 1, this comparative example diagnoses the state of the rotating machine 3 connected to the power supply 1 via the cable 2. In the diagnostic apparatus of this comparative example, the current measuring unit 4 acquires two-phase current data via the current sensors 10a and 10b attached to the cable 2, and the acquired current data is obtained by Fourier transform. After accumulating in the accumulator 8 as a value of the specific frequency spectrum, the diagnostic unit 9 diagnoses the state of the rotating machine 3 based on the current data accumulated in the accumulator 8.

In the diagnostic apparatus configured in this way, since the change in the specific frequency spectrum is measured by the Fourier transform, it is necessary to carry out continuous measurement at a constant sampling rate. Therefore, it is necessary to increase the capacity of the memory for temporarily storing the measured data or to increase the communication speed with the device for storing the data, which requires an expensive device.

Further, since there is a phenomenon that appears only for a short time depending on the failure of the rotating machine 3, if the measurement is performed for a long time at a high sampling speed in order not to miss the state due to the failure, the amount of accumulated data becomes enormous and expensive. A data storage terminal was needed. In addition, when the amount of data to be stored becomes enormous, an expensive communication device compatible with large-capacity communication is required in order to transmit the data to the cloud or a local terminal.

The inventors discretize the intermittent sensor values for two phases of the load current values of the rotating machine by the number of measurement bits for each phase, and use the combination of the instantaneous values of each phase and the appearance density. We considered making a diagnosis.

Since the current sensor data acquired from the three-phase motor, which is a three-phase rotating machine, has a deviation of 120 degrees from each other, when the two-phase current sensor data is combined, the Lissajous figure has an inclined elliptical shape. The inventors noted that the elliptical shapes of the first and second cycles completely overlap when the two-phase current is ideal, perfectly continuous sinusoidal data. In the measured waveform, there is a deviation from the ideal sine wave, and since the sampling interval is a finite value, the probability of measuring points with exactly the same phase is low, but many can be obtained intermittently in a predetermined time. By superimposing the data for the period and accumulating the combination of instantaneous values and their appearance density, it is possible to accumulate the data in a form in which the amount of data is reduced compared to accumulating the two-phase waveform as time series data. In addition, by superimposing data for multiple cycles, even if the sampling interval is not constant or the sampling speed is slow, the combination of instantaneous values and their appearance density will change the operating conditions to be diagnosed. If not, they match, and equipment such as expensive data loggers and expensive data storage devices can be omitted. The important point here is the time synchronism between the first phase and the second phase, and the first phase and the second phase must be at the same time or at a fixed interval of arbitrary design. Fluctuations in the measurement interval between the second phases are directly linked to a decrease in diagnostic accuracy. In order to perform measurement at the same time or at regular intervals, it can be realized by using a device having excellent real-time performance such as a microcomputer.

The diagnostic device of the present embodiment is a diagnostic device that diagnoses the state of a diagnosis target that operates with a three-phase AC power supply, and includes an acquisition unit that acquires instantaneous values of physical data at the same time in a plurality of phases of three-phase AC. A conversion unit that discretizes the instantaneous values acquired by the acquisition unit, and an integration unit that obtains the appearance density of the combination for each combination of the discrete values of the instantaneous values of the physical data of each phase from the discrete values obtained by the discretization of the conversion unit. And the storage unit that stores the appearance density data indicating the appearance density for each combination obtained by the integration unit, and the diagnostic unit that diagnoses the state of the diagnosis target based on the appearance density data accumulated in the storage unit. To prepare.

In addition, the diagnostic device configured as described above can also be incorporated into a rotating machine system. In particular, it is preferable to share the current sensor, the current measuring unit, and the like provided for controlling the rotating machine in the power converter because the number of parts can be reduced. Further, a plurality of rotating machines may be connected to the power converter.

Furthermore, by classifying and evaluating the current data for each frequency condition of the current value measured by the current sensor and the current measuring unit, even if the driving conditions of the rotating machine change, the rotation is appropriate. It is possible to diagnose the state of the machine system.

More specific examples of the above-mentioned diagnostic apparatus will be described below.

FIG. 2 is a block diagram of the diagnostic apparatus according to the first embodiment.

As shown in FIG. 2, the diagnostic device in this embodiment diagnoses the state of the rotating machine 3 to be diagnosed, which is electrically connected to the power supply 1 via the cable 2, and is connected to the current measuring unit 4. , A conversion unit 5, a measurement bit number input unit 6, an integration unit 7, a storage unit 8, and a diagnosis unit 9. The diagnostic device has a processor and a memory, and the processor may use the memory to execute a software program that defines the operation of each of the above parts.

A three-phase AC voltage is output from the power supply 1. There are two types of three-phase AC voltage output: one is controlled by adjusting the timing of operating the switching element of the inverter so that the rotation speed and torque of the motor are at the desired values, and the other is when a commercial power supply is directly connected. Can be considered.

The current measuring unit 4 may be called an acquisition unit. The current measuring unit 4 acquires the instantaneous value of the current data which is the physical data at the same time in the plurality of phases of the three-phase AC flowing through the cable 2 via the current sensors 10a and 10b attached to the cable 2. At this time, the current measuring unit 4 acquires the instantaneous value of the current data while maintaining the time synchronization between the plurality of phases at an arbitrary interval for an arbitrary period.

The conversion unit 5 discretizes the instantaneous value acquired by the current measurement unit 4 by the set number of bits A set by the measurement bit number input unit 6.

The integrating unit 7 has a size of 2 ^A (2 to the A power) × 2 ^A (2 to the A power) in which the rows and columns correspond to the phases from the discrete values obtained by the discretization of the conversion unit 5. Each component of the column generates a matrix that represents the appearance density of the discrete value combinations of the instantaneous values of the current data of each phase, so that the appearance density of the combination of the discrete values of the instantaneous values of the current data of each phase is generated. Ask for.

The storage unit 8 stores the appearance density data indicating the appearance density for each combination obtained by the integration unit 7.

The diagnosis unit 9 diagnoses the state of the rotating machine 3 and peripheral devices such as a power conversion device connected to the rotating machine 3 based on the appearance density data stored in the storage unit 8.

Hereinafter, a diagnostic method for diagnosing the state of the rotating machine 3 by the diagnostic device configured as described above will be described.

First, the current measuring unit 4 acquires current data in two phases of the three-phase alternating current flowing through the cable 2 via the current sensors 10a and 10b attached to the cable 2. At that time, the current measuring unit 4 acquires the instantaneous value of the current data while maintaining the time synchronization between the plurality of phases at an arbitrary interval for an arbitrary period, but the current sensor acquired by the current measuring unit 4 The current data of 10a and 10b do not necessarily have to have a constant sampling interval, and do not necessarily have to be continuous measurements. Further, it is preferable that the interval between the acquisition of the current data via the current sensor 10a and the acquisition of the current data via the current sensor 10b is constant to allow some variation. By applying a measuring device excellent in real-time processing, it is possible to set a constant data acquisition interval that allows a certain variation. As such a measuring device, for example, a measuring device using a microcomputer can be used.

Further, by keeping the interval between the acquisition of the current data via the current sensor 10a and the acquisition of the current data via the current sensor 10b constant, it is not necessary to perform continuous measurement. For example, it is possible to design the current measurement unit 4 in which a certain amount of data is stored in the memory of the microcomputer, the stored data is stored in the storage device, and then the memory of the microcomputer is cleared and the data storage is restarted. It is possible to make a system using a general-purpose current sensor and memory.

Next, in the conversion unit 5, the instantaneous value acquired by the current measurement unit 4 is discretized by the set number of bits A set in the measurement bit number input unit 6, and in the integration unit 7, the conversion unit 5 is discretized. From the discrete values due to the conversion, the appearance density of the combination for each combination of the discrete values of the instantaneous values of the two-phase current data is obtained, and the appearance density data indicating the appearance density for each combination is accumulated in the storage unit 8. Become. At this time, in the conversion unit 5, the instantaneous value of the two-phase current data is discreteized by a predetermined number of bits A, and in the integration unit 7, 2 ^A (2 to the A power) × 2 in which the rows and columns correspond to the phases, respectively. ^{If the} application density is obtained by generating a matrix that has a size of ^A (2 to the A power) and each component of the row and column represents the appearance density of a combination of discrete values of the instantaneous values of the current data of each phase. , The amount of current data for each phase can be compressed to a size of 2 ^A x 2 ^A.

Here, the amount of data due to discretization when an ideal sine wave current is flowing through the rotating machine 3 when the control pattern for the rotating machine 3 does not change and the rotating machine 3 is operating at a constant fundamental frequency. Will be described. The number of measurement bits of the data logger was set to 8 bits.

FIG. 3 is a diagram showing the amount of data stored in the storage unit 8 due to the discretization in the conversion unit 5 shown in FIG.

Let 1 be the amount of data when the two-phase current is measured at a sampling rate of 200 Hz for 200 seconds.

As method 1, the waveform that becomes the current data acquired by the current measuring unit 4 is discreteized by the number of bits 8 set by the measurement bit number input unit 6 in the changing unit 5, and 28 × 28 in the integrating unit 7. The appearance density of 8 bits, that is, the current values discreteed from 1 to 257 was calculated for each matrix element of the size matrix. For example, in a measurement time of 200 seconds, the number of conditions under which the discretized current value of the current sensor 10a is 257 and the discretized current value of the current sensor 10b is 257 is divided by the number of measurement points of 40,000. Asked. The measurement for 200 seconds does not have to be continuous, and if the control pattern for the rotating machine 3 does not change and the rotating machine 3 is operating at a constant fundamental frequency, the measurement should be divided into several parts. Can be done. As a result, it is possible to acquire the data necessary for diagnosis even at a low-speed measurement terminal having a small amount of memory.

In this case, the amount of data was 0.81, and a reduction of 19% in the amount of data could be realized. As a result, it is possible to acquire the data necessary for diagnosis even at a low-speed measurement terminal having a small amount of memory. In this embodiment, the value divided by the number of measurement points is stored in each matrix element of the matrix, but the number of occurrences is saved, and separately, the number of measurement points is saved in a form linked to the file in which the number of occurrences is saved. Is also good. Further, especially when the measurement is divided into several times, the error accumulation due to the integration can be reduced by storing the number of appearances and the number of measurement points separately.

Further, as the method 2, instead of preparing a matrix having a size of 28 × 28 in advance, (the discrete current value 1 of the current sensor 10a, the discrete current value 1 of the current sensor 10a, the number of occurrences), ( The data was stored in the format of the discrete current value 1 of the current sensor 10a, the discrete current value 2 of the current sensor 10a, the number of occurrences), and so on. In the method 2, the data finally obtained is the data excluding the component whose matrix element value of the method 1 is 0, and it is not necessary to prepare a blank cell. Therefore, the storage unit 8 is filled with the matrix. Only the value of the component in which the numerical value other than zero is stored is accumulated, and the amount of accumulated data can be further reduced. The timing of removing the component 0 is not particularly limited.

As the data storage method, (the discrete current value 1 of the current sensor 10a, the discrete current value 1 of the current sensor 10a, the number of occurrences), (the discrete current value 1 of the current sensor 10a, the current). Data may be accumulated in the form of the discrete current value 2 of the sensor 10a, the number of occurrences), ..., Or, once the matrix of the method 1 is constructed, (the current sensor 10a is discreteized). Current value 1, discrete current value 1 of the current sensor 10a, number of appearances), (discrete current value 1 of the current sensor 10a, discrete current value 2 of the current sensor 10a, number of appearances), ...・ ・ You may convert to the format of.

After that, the diagnosis unit 9 diagnoses the state of the rotating machine 3 based on the appearance density data accumulated in the storage unit 8.

Here, a diagnostic method of the diagnostic unit 9 will be described when the control pattern for the rotating machine 3 does not change and the rotating machine 3 is operating at a constant fundamental frequency and a spectrum is generated at a specific frequency for some reason. To do.

FIG. 4 is a diagram showing a current waveform when a spectrum appears at a frequency 1 Hz away from the fundamental wave frequency of the U-phase current of 50 Hz, and FIG. 5 is a diagram showing a swell waveform generated by the current waveform shown in FIG. Is.

For example, when a sideband wave of the fundamental wave frequency is generated due to deterioration in bearing deterioration, and as a result, as shown in FIG. 4, a spectrum appears at a frequency 1 Hz away from the fundamental wave frequency of 50 Hz of the U-phase current for some reason. As shown in FIG. 5, the U-phase current appears as a waveform having a swell with a 1 Hz period.

In the diagnostic apparatus shown in FIG. 1, in order to accurately separate the components of 50 Hz and 51 Hz from this waveform, when the current flowing through the cable 2 is measured at a frequency of 200 Hz, 20000 points of data are continuously collected for at least 200 seconds. Need to measure. Further, the frequency of appearance of sideband waves increases as the deterioration progresses, and the frequency is low in the initial stage of deterioration, so that long-term data measurement is required. Therefore, it is necessary to apply a special and expensive measuring device for accurate separation.

On the other hand, in the diagnostic apparatus shown in FIG. 2, the current measuring unit 4 acquires the current values of two or more phases at an arbitrary interval for an arbitrary period while maintaining the time synchronization between the phases, and the conversion unit 5 obtains the current values. The current data of two or more phases is separated for each phase by the number of bits set in the measurement bit number input unit 6, and the number of appearances or appearance density of the separated current data is set in each matrix element in the integration unit 7. It is obtained by converting it into a matrix to have, and then the diagnostic unit 9 compares the data stored as application density data in the storage unit 8, and based on the change, the rotating machine 3 or the power conversion device connected to the rotating machine 3 or the like. By diagnosing the state of the peripheral devices of the above, it is possible to diagnose the rotating machine 3 without increasing the sampling speed and the amount of data.

Further, in order to acquire high frequency information, it is desirable to measure asynchronously with the switching timing of the power converter.

When slowing down the sampling rate, it is desirable that the current measuring unit 4 measures the current flowing through the cable 2 asynchronously with the fundamental wave frequency via the current sensors 10a and 10b. Specifically, at a sampling rate having a period that is an integral multiple of the fundamental wave, the value of only a specific phase is always acquired, so that there is a risk of misreporting in the case of deterioration in which a change appears only in a specific phase. Therefore, it is desirable that the sampling rate is a frequency different from an integral multiple of the period of the fundamental wave. As a result, it is possible to acquire a matrix similar to the case where data is acquired at a high sampling rate.

FIG. 6 is a diagram showing a current waveform of the U phase in the normal state, and FIG. 7 is a diagram showing the current waveform of the W phase in the normal state. Further, FIG. 8 is a diagram showing a U-phase current waveform when a sideband wave is generated, and FIG. 9 is a diagram showing a W-phase current waveform when a sideband wave is generated. The sampling speed is 200 Hz, and the data for 1 second out of the data measurement time of 100 seconds is enlarged and shown.

In a normal state of 50 Hz sine wave without swell, the U-phase current waveform is as shown in FIG. 6, the W-phase current waveform is as shown in FIG. 7, and the U-phase and W-phase currents are as shown in FIG. They are 120 degrees out of phase with each other.

When the waveforms shown in FIGS. 6, 7, 8 and 9 are measured and acquired by the current measuring unit 4, the current measured by the converting unit 5 is as shown in the above description of the method 1. The waveform is discreteized by the number of bits 8 set by the measurement bit number input unit 6, and in the integrating unit 7, the current value is 8 bits, that is, the current value discreteized by 1 to 257 for each matrix element of the matrix having a size of 28 × 28. The appearance densities were stored in the matrix A and the matrix B, respectively. The matrix A is based on the waveforms shown in FIGS. 6 and 7, and corresponds to the matrix in the normal state. The matrix B is based on the waveforms shown in FIGS. 8 and 9, and is a matrix in the deteriorated state. Corresponds to.

In the diagnostic unit 9, the matrix A and the matrix B are compared with the matrix C set as the normal state by learning in advance, and the normal state is calculated by calculating how close the matrix A and the matrix B are to the matrix C. It is possible to check the deviation from, that is, the degree of abnormality. As the deterioration progresses, the variation of the current value becomes large, and the appearance density of the region with high appearance density decreases accordingly, and the current value is distributed to the region with low appearance density. As a result, the value of each matrix element in the matrix changes, the number of matrix elements whose value becomes 0 decreases, and the matrix is changed from the matrix defined as the normal state.

After making the above diagnosis, the diagnosis unit 9 outputs the diagnosis result. The means for communicating the diagnosis result to the user can be appropriately selected, and the method for communicating the diagnosis result to the user includes display on a display, lighting of a lamp, notification by e-mail, and the like. The contents are also (1) a method of displaying the matrix to be diagnosed on the screen and letting the user judge whether or not there is a response, (2) a method of quantifying the difference in the matrix to be diagnosed in some way and telling the user, (3) in advance. A method of notifying the user when the set threshold is exceeded can be considered.

Here, as a method of quantifying the difference in the matrix to be diagnosed in (2) above, application of machine learning can be considered. As the machine learning algorithm, one that makes the difference between the matrices to be diagnosed clear may be selected. For example, the Bhattacharyya coefficient (Bhattacharyya distance) Z of the matrix A to be diagnosed with respect to the matrix C defined as the normal state is the matrix A. When the matrix elements of are A _{i and j} and the matrix elements of the matrix C are C _{i and j} , they are defined by the following equations.

上記式において、Bhattacharyya係数Ｚが大きければ、正常状態と離れている、つまり劣化が進行していると診断でき、係数Ｚが小さければ、正常状態と診断することができる。Bhattacharyya係数Ｚは異常度を表していると言える。このように、蓄積部８に行列のデータが保持され、診断部９において、蓄積部に保持された行列と、正常状態を示すものとして予め定められた基準行列とを比較することにより診断対象の状態の正常性を診断することで、正常状態とどの程度異なる状態となっているかを容易に知得し、正常性を診断することができる。

In the above equation, if the Bhattacharyya coefficient Z is large, it can be diagnosed that the state is different from the normal state, that is, the deterioration is progressing, and if the coefficient Z is small, it can be diagnosed as the normal state. It can be said that the Bhattacharyya coefficient Z represents the degree of abnormality. In this way, the matrix data is held in the storage unit 8, and the diagnosis unit 9 compares the matrix held in the storage unit with a reference matrix predetermined as indicating a normal state to be diagnosed. By diagnosing the normality of the state, it is possible to easily know how different the state is from the normal state and diagnose the normality.

Here, when using a matrix in which the number of occurrences is stored in each matrix element, it is necessary to match the measurement points of the matrix to be diagnosed with the normal state. As a diagnostic method in that case, a local subspace method can be mentioned. In the local subspace method, for each matrix element of the matrix to be diagnosed, two points closest to each other in the two-dimensional space defined by the rows and columns of the matrix defined as the normal state are selected and the two points are connected. This is a method of defining the degree of deterioration by the distance between the straight line and the point to be diagnosed. In some cases, it is effective to weight the appearance density of normal values. In particular, when the data can be recognized as an abnormal state, the discriminant is adjusted so that the matrix of the normal state and the abnormal state can be easily distinguished. Is desirable.

In addition, when using the local subspace method, in addition to the method of calculating the distances for all the matrix elements to be diagnosed and quantifying the changes in the matrix, the average value of the distances of all the points to be diagnosed is specified. Any evaluation method can be selected according to the variation error of the current waveform, such as digitizing only the distance of the phase points. If the calculation speed is to be prioritized, a clustering method such as vector quantization clustering or K-means clustering can be used. In addition, a method called a deep neural network, which is a method of automatically finding features based on a large amount of data, can be applied.

FIG. 10 is a diagram showing the results of analyzing the degree of abnormality of the matrix A and the matrix B based on the current waveforms shown in FIGS. 6, 7, 8 and 9 with respect to the matrix C defined as the normal state.

The degree of abnormality of the matrix A based on the current waveforms shown in FIGS. 6 and 7 and the matrix B based on the current waveforms shown in FIGS. 8 and 9 can be analyzed from the Bhattacharyya distance Z according to the above equation. As shown in FIG. 10, since the Bhattacharyya distance Z of the matrix A does not exceed the preset threshold value, the diagnostic unit 9 determines that it is normal. On the other hand, as shown in FIG. 10, the Bhattacharyya distance Z of the matrix B exceeds a preset threshold value, so that the diagnostic unit 9 determines that the distance Z is abnormal.

In this way, depending on whether the Bhattacharyya distance Z exceeds a preset threshold value, it is possible to detect an increase in the degree of abnormality before the rotating machine 3 fails.

As described above, since the appearance density of the combination of the instantaneous values of the current data of each phase is integrated and accumulated, the amount of data to be stored can be suppressed. Further, since the state of the rotating machine 3 to be diagnosed is diagnosed based on the integrated data obtained by integrating the appearance densities of the combination of the instantaneous values of the current data of each phase, it is not necessary to measure the current data at a high sampling speed.

In this embodiment, the case where the spectrum is generated at a specific frequency has been described, but even if the deterioration is of a type in which the spectrum is not generated at a specific frequency, the deterioration causes the load of the rotating machine 3 and the impedance change to cause a current. Since some change appears, the abnormality can be detected by the diagnostic apparatus shown in FIG. 2 and the diagnostic method described above. Deterioration that does not appear as a change in the spectrum of a specific frequency is specifically deterioration other than deterioration in which a change appears as a peak of a specific frequency that can be detected by MCSA, that is, deterioration such as grease deterioration, thermal deterioration of an insulating material, and moisture absorption. Things are supposed.

Further, according to the diagnostic device of this embodiment, it is possible to diagnose a motor system including a device electrically or mechanically connected to the rotary machine 3, such as a cable, a power conversion device, and a load, in addition to the rotary machine 3. is there. In a motor system including peripheral devices connected to the rotating machine 3, even if peripheral devices other than the rotating machine 3 fail or deteriorate, the impedance and load of those devices change, so that the current flowing through the rotating machine 3 flows. Therefore, deterioration can be detected by this method.

Next, a case where the control pattern for the rotating machine 3 changes will be described. When the control pattern for the rotating machine 3 changes, the fundamental wave frequency, the carrier frequency, and the like of the rotating machine 3 change. Therefore, the diagnostic apparatus shown in FIG. 1 may diagnose this as an abnormality. In addition, when learning as a normal state including a state in which the control pattern has changed, the change due to deterioration may be overlooked and a false report may be lost. Therefore, it is desirable to diagnose normality for each control pattern.

FIG. 11 is a block diagram of the diagnostic apparatus according to the second embodiment.

As shown in FIG. 11, the diagnostic apparatus in this embodiment is different from the one shown in FIG. 2 in that it has a command unit 13.

The command unit 13 gives a control command indicating information on the control pattern to the integration unit 7. The control command is arbitrary from the command values that can be output by the power supply 1, such as the voltage command value, the current command value, the exciting current command value, the torque current command value, the speed command value, and the frequency command value, and the values equivalent thereto. You can choose to.

The integrating unit 7 obtains the appearance density of the combination for each combination for each control command to the rotating machine 3, and the storage unit 8 accumulates the appearance density data for each control command.

Hereinafter, a case where the state of the rotating machine 3 is diagnosed under two conditions of the control command A and the control command B in which the voltage command value and the frequency command value are different from each other as the control command will be described. The control command A corresponds to a current fundamental wave frequency of 50 Hz, and the control command B corresponds to a current fundamental wave frequency of 100 Hz.

The integrating unit 7 integrates the appearance density obtained by the conversion unit 5 for each current value distributed to the control command A and the control command B based on the fundamental frequency input to the control unit 13. Then, after data of an arbitrarily determined period or score is accumulated, the data is accumulated in the storage unit 8 as appearance density data.

After that, the diagnosis unit 9 compares the appearance density data accumulated in the storage unit 8 with the data defined as normal acquired in the same state of the control command information to obtain the degree of abnormality of the target data, and the rotation machine 3 is normal. Diagnose sex.

FIG. 12 is a diagram showing the degree of abnormality of the matrix A obtained from the current waveform of the rotating machine 3 in the normal state acquired by the control command A. The matrices α and β are matrices defined by the current waveforms of the rotating machine 3 in the normal state measured by the control commands A and B, respectively.

As shown in FIG. 12, when the rotating machine 3 in the normal state is diagnosed by using the matrix α for the matrix A based on the current data of the rotating machine 3 in the normal state, the Bhattacharyya distance Z of the matrix A is shown in FIG. As described above, since the preset threshold value is not exceeded, the diagnostic unit 9 determines that the value is normal.

On the other hand, when the rotating machine 3 in the normal state is diagnosed by using the matrix β for the matrix A based on the current data of the rotating machine 3 in the normal state, the Bhattacharyya distance Z of the matrix A is set in advance as shown in FIG. Since the threshold value has been exceeded, the diagnostic unit 9 determines that the rotating machine 3 is abnormal even though the rotating machine 3 is in a normal state.

As described above, in this embodiment, the control input to the command unit 13 for diagnosing the state of the rotating machine 3 based on the matrix composed of the U-phase and W-phase currents in which the control commands are classified as the same state. By classifying the current waveforms based on the commands and integrating the appearance density for each classified data, the control pattern and current information were combined to improve the diagnostic accuracy. Thereby, a suitable diagnosis can be made for each control state.

The integrating unit 7 does not necessarily have to use all the command values that can be output by the power supply 1, and can use only the command values that are highly sensitive to deterioration of the detection target. The command value with high sensitivity may be selected so that the difference between the deteriorated state and the normal state can be easily seen by comparing the degree of similarity between the matrix in the deteriorated state and the matrix in the normal state. In addition, if the number of conditions is increased, the amount of data will increase by the amount classified by the conditions even if the current data has the same measurement time. Therefore, the number of command values to be used should be limited by the trade-off between diagnostic accuracy and the amount of data. Can be done.

When the information of the command unit 13 cannot be referred to as in the diagnostic device shown in FIG. 11, the characteristic frequency of the current waveform of the rotating machine 3 can be used instead.

FIG. 13 is a block diagram of the diagnostic apparatus according to the third embodiment.

As shown in FIG. 14, the diagnostic apparatus in this embodiment is different from the one shown in FIG. 2 in that it has a frequency extraction unit 12.

When the control pattern of the rotating machine 3 is recognized by the characteristic frequency of the current waveform of the rotating machine 3, the output of the current measuring unit 4 is branched and input to the frequency extracting unit 12 as shown in FIG. The frequency extraction unit 12 extracts characteristic frequencies such as at least one of the fundamental wave frequency and the carrier frequency, and then the integration unit 7 obtains the appearance density of the combination for each combination for each measurement condition having the same characteristic frequency. , In the storage unit 8, the appearance density data is stored for each feature frequency.

As described above, in this embodiment, the output of the current measuring unit 4 is branched in order to diagnose the state of the rotating machine 3 based on the matrix composed of the U-phase and W-phase currents whose control commands are classified as the same state. Then, the characteristic frequency is extracted by the frequency extraction unit 12, the current waveform is classified based on this characteristic frequency, and the appearance density is integrated for each classified data, thereby combining the control pattern and the current information for diagnostic accuracy. I tried to improve. Thereby, a suitable diagnosis can be made for each control state.

FIG. 14 is a block diagram of the diagnostic device according to the fourth embodiment.

As shown in FIG. 14, the diagnostic apparatus in this embodiment differs from the one shown in FIG. 2 in that a plurality of rotating machines 3-1 to 3-n are connected to one power supply 1. It is a thing.

In this embodiment, the load currents of all the rotating machines 3-1 to 3-n were measured by the current sensors 10a and 10b attached to the cable 2 connecting the power supply 1 and the rotating machines 3-1 to 3-n. Diagnosis is made using the results. The method of diagnosis is the same as when one rotating machine is connected, and the diagnosis is performed using the measured diagnostic data as the distribution of the Lissajous figure while classifying using the control command value and reflecting the command value information.

Since the current data of a plurality of rotating machines 3-1 to 3-n is used as one load current for diagnosis, the change in the current waveform of the deteriorated rotating machine is diminished by the current waveforms of the other plurality of rotating machines. Will end up. Therefore, it is desirable to evaluate the degree of anomaly by machine learning, especially the local subspace method.

FIG. 15 is a block diagram of the diagnostic apparatus according to the 56th embodiment.

As shown in FIG. 15, the diagnostic apparatus in this embodiment measures the three-phase load currents supplied from the power supply 1 to the rotating machine 3 with the current sensors 10a to 10c with respect to the one shown in FIG. The point to do is different.

In this embodiment, the current measuring unit 4 acquires the instantaneous value of the current data at the same time in the three phases of the three-phase alternating current, and the integrating unit 7 obtains the discrete value of the three-phase by the discrete value of the conversion unit 5. The appearance density of the combination for each combination of the discrete values of the instantaneous values of the current data is obtained. In this case, when comparing the distributions, it is possible to obtain and evaluate the distribution of the Lissajous figure with a plurality of combinations of two selected from three or more current sensors. In addition, a multidimensional matrix (tensor) can be defined by three or more current sensors, and motor deterioration can be evaluated from changes in the tensor.

In the case of the three-phase AC rotary machine 3, the current value measured by the current sensors 10a to 10c is zero measured by clamping the three-phase components in addition to the load currents of the U-phase, V-phase, and W-phase. Diagnosis of phase current, two-phase current measured by clamping an arbitrarily selected two-phase, leakage current measured by clamping both the winding start and winding end of the motor winding, current flowing from the motor to the ground, etc. A current with high sensitivity may be arbitrarily selected and a current sensor may be installed at that position.

In this way, since the state of the rotating machine 3 is diagnosed based on the three-phase current data of the three-phase alternating current, more accurate diagnosis becomes possible.

1 ... Power supply, 2 ... Cable, 3 ... Rotator, 4 ... Current measurement unit, 5 ... Conversion unit, 6 ... Measurement bit number input unit, 7 ... Integration unit, 8 ... Accumulation unit, 9 ... Diagnosis unit, 10a, 10b , 10c ... Current sensor, 12 ... Frequency extraction unit, 13 ... Command unit

Claims (10)

A diagnostic device that diagnoses the condition of a diagnostic object that operates on a three-phase AC power supply.
An acquisition unit that acquires the instantaneous value of physical data at the same time in multiple phases of the three-phase alternating current,
A conversion unit that discretizes the instantaneous values acquired by the acquisition unit, and
An integration unit for obtaining the appearance density of the combination for each combination of the discrete values of the instantaneous values of the physical data of each phase from the discrete values obtained by the discretization of the conversion unit.
A storage unit that stores appearance density data indicating the appearance density for each combination obtained by the integration unit, and an accumulation unit.
A diagnostic unit that diagnoses the state of the diagnosis target based on the appearance density data accumulated in the storage unit, and
A diagnostic device equipped with. The conversion unit discretizes the instantaneous values of the two-phase physical data with a predetermined number of bits A.
The integration unit has a size of 2 ^A (2 to the A power) × 2 ^A (2 to the A power) in which the rows and columns correspond to the phases, and each component of the rows and columns is the physical data of each phase. Generates a matrix that represents the appearance density of a combination of discrete values of instantaneous values of
The diagnostic device according to claim 1. The storage unit holds the data of the matrix and holds the data of the matrix.
The diagnostic unit diagnoses the normality of the state to be diagnosed by comparing the matrix held in the storage unit with a reference matrix predetermined as indicating a normal state.
The diagnostic device according to claim 2. The diagnostic unit diagnoses the normality of the state of the diagnosis target based on the difference between the matrix accumulated in the storage unit and the reference matrix.
The diagnostic device according to claim 3. The storage unit stores only the values of the components of the matrix in which non-zero values are stored.
The diagnostic device according to claim 2. The diagnostic object includes a controllable rotating machine.
The integrating unit obtains the appearance density of the combination for each combination for each control command to the rotating machine.
The storage unit stores the appearance density data for each control command.
The diagnostic device according to claim 1. The diagnostic object includes a controllable rotating machine.
The integrating unit acquires the appearance density of the combination for each combination for each feature frequency that changes due to the control to the rotating machine.
The storage unit stores the appearance density data for each feature frequency.
The diagnostic device according to claim 1. The characteristic frequency is at least one of the fundamental frequency and the carrier frequency of the current input to each phase of the rotating machine.
The diagnostic device according to claim 7. The acquisition unit acquires instantaneous values of physical data at the same time in the three phases of the three-phase alternating current.
The integrating unit obtains the appearance density of the combination for each combination of the discrete values of the instantaneous values of the three-phase physical data from the discrete values obtained by the discretization of the conversion unit.
The diagnostic device according to claim 1. It is a diagnostic method for diagnosing the state of the diagnostic target operating on a three-phase AC power supply.
Obtain the instantaneous values of physical data at the same time in the multiple phases of the three-phase alternating current,
Discrete the acquired instantaneous value
From the discrete values obtained by the discretization, the appearance density of the combination for each combination of the discrete values of the instantaneous values of the physical data of each phase is obtained.
The appearance density data indicating the appearance density for each of the obtained combinations is accumulated, and the appearance density data is accumulated.
Based on the accumulated appearance density data, the state of the diagnosis target is diagnosed.
Diagnostic method.

Images