Diagnostic support device, rotating machine system and diagnostic support procedure

Abstract

Diagnostic support device (1) comprising the following: a detection unit that detects a q-axis current in a power conversion device (2) that controls a rotating machine (3); an amplitude calculation unit (11) that calculates an amplitude of each frequency around the q-axis current based on the detected q-axis current; and an output unit (12) that outputs information about the amplitude of each frequency, a storage unit (21) which contains information about natural frequencies of the anomalies, which stores information about a frequency for each of the parts in the rotating machine (3) and an auxiliary device of the rotating machine (3), at which an amplitude increases when an anomaly occurs in at least one of the rotating machine (3) and the auxiliary device; and a unit (13) for locating anomalous parts, which compares the amplitude of each frequency calculated by the amplitude calculation unit (11) with the information about the natural frequencies of the anomalies and thereby locates a part where an anomaly occurs in at least one of the rotating machine (3) and the auxiliary device.

Description

TECHNICAL AREA

[0001] The present invention relates to a diagnostic support device, a rotating machine system and a diagnostic support method. TECHNICAL BACKGROUND

[0002] A sudden failure of a rotating machine, such as a motor, within production equipment necessitates unplanned repair or replacement work. This, in turn, reduces the production equipment's capacity utilization and requires a review of production plans. By monitoring for any signs of failure, spare parts can be prepared or a repair plan developed in advance for equipment with a high probability of failure. This minimizes the reduction in capacity utilization and the need for revising production plans.

[0003] The following procedures are possible to prevent the sudden failure of a rotating machine within a rotating machine system and its auxiliary equipment. It is specified that a rotating machine and its auxiliary equipment include an inverter, gears, a coupling, a load device, and the like. (a1) Attach a vibration sensor to each part of the rotating machine system and check for an increase in the effective vibration value. (a2) Checking for an increase in a natural frequency component of the vibration of each part of the rotating machine.

[0004] By performing the procedures according to (a1), (a2) or the like, the anomalous condition of each part can be diagnosed and sudden failure can be avoided to a certain extent.

[0005] However, the procedures (a1) and (a2) described above have the following problems. (b1) Vibration sensors are expensive. (b2) The sensitivity of a vibration sensor depends to a large extent on its mounting position. (b3) Vibration sensors must be fitted to all parts where an anomaly may occur.

[0006] For this reason, a method has been proposed that uses inexpensive current sensors instead of vibration sensors to reduce the number of installed sensors and enables robust diagnostics without relying on the technician. In this method, a current waveform acquired from the rotating machine system is Fourier-transformed, and a frequency component characteristic of an anomaly (the anomaly's natural frequency) is extracted from the Fourier transform results. A diagnosis is then performed based on the amplitude of the anomaly's natural frequency component. Motor current signature analysis (MCSA) is one such method.

[0007] It is known that in the case of a mechanical anomaly, a natural frequency component of the anomaly is expressed by the following formula (1). fC=|f0±fm|

[0008] In formula (1), fc denotes an anomalous natural frequency of a current, f0 a fundamental frequency of the current, and fm a frequency of a mechanical vibration. This formula enables the diagnosis of bearing deterioration, damage to a gear or coupling, an anomaly in a load device, or the like. For example, if the inner ring of a bearing has a scratch, it is known that, depending on the size of the bearing, the number of bearing balls, and the like, a vibration occurs periodically, with the resulting vibration appearing as a sideband wave at the fundamental frequency of the current.

[0009] However, there is still a need to install an additional current sensor, and it is conceivable that an internal value could be used to control an inverter connected to the motor for diagnostic purposes.

[0010] Patent literature 1 discloses a controller for a rotating machine and a washing machine, in which "a section 67 for detecting anomalous rotation calculates a value corresponding to the input power of a motor 60 based on a value actually detected during a position estimation operation, calculates a value corresponding to the output power of the motor 60 based on the rotational speed ω est estimated by a rotational speed / angle estimation section 66, obtains a rotational anomaly index according to the comparison results, and determines anomalous rotation if the rotational anomaly index is greater than a threshold" (see the abstract). Furthermore, JP 2017-221023A discloses a device for detecting fault signs in an air conditioning compressor. The device is configured to calculate the q-axis current of the motor from the measured three-phase current and the rotational angle of the rotor.By frequency-analyzing this q-axis current, which is less noisy than the three-phase current, and subsequently comparing the resulting value with a reference value, anomalies are detected to improve detection accuracy. KR 2005 063 441 A discloses a device designed to diagnose fault conditions in induction motors. The device converts the three-phase input current into a d-axis and q-axis current after an analog-to-digital conversion using a dq transformation. A frequency analysis of the d-axis current generates a spectrum that is compared with a database of known fault signatures to identify the fault type and location. LIST OF COUNTERPOINTS Patent Literature JP 2009 - 65 764 A JP 2017 - 221 023 A KR 10 2005 0 063 441 A SUMMARY OF THE INVENTION Technical Problem

[0011] The technique described in patent literature 1, however, presents the following problem. Specifically, the technique disclosed in patent literature 1 diagnoses an anomaly of the entire motor system based on the ratio between the estimated values ​​of the input power and the output power. Using this method, an anomalous part cannot be easily located. If the anomalous part is not known, it is difficult to identify a component for which a replacement needs to be prepared. Furthermore, it is difficult to plan a repair that must be based on an anomalous part.

[0012] The present invention was made under such circumstances and has the objective of simplifying the localization of an anomalous part in a rotating machine and its auxiliary devices. Solution to the problem

[0013] To solve the above problems, a diagnostic support device with the features of claim 1, a rotating machine system with the features of claim 7, and a diagnostic support method with the features of claim 8 are provided. Advantageous embodiments are described in the dependent claims.

[0014] Further solvents will be described later in the embodiments. Advantageous effects of the invention

[0015] According to the present invention, an anomalous part in a rotating machine and its auxiliary devices can be easily located. BRIEF DESCRIPTION OF THE DRAWINGS Fig. Figure 1 is a graphical representation showing an exemplary configuration of a rotating machine system 100 according to a first embodiment. Fig. 2A is a graphical representation showing an exemplary frequency spectrum of a q-axis current (when there is no anomaly). Fig. 2B is a graphical representation showing an example frequency spectrum of a q-axis current (if there is an anomaly). Fig. Figure 3A is a graphical representation showing an exemplary vibration frequency spectrum of a rotating machine 3 (when there is no anomaly). Fig. 3B is a graphical representation showing an exemplary vibration frequency spectrum of the rotating machine 3 (if there is an anomaly). Fig. 4A is a graphical representation showing an exemplary frequency spectrum of a load current flowing in each phase of the rotating machine 3 (when there is no anomaly). Fig. Figure 4B shows an exemplary frequency spectrum of a load current flowing in each phase of the rotating machine 3 (when there is an anomaly). Fig. Figure 5 is a graphical representation showing an exemplary configuration of a rotating machine system 100a according to a second embodiment. Fig. Figure 6 is a graphical representation illustrating the definition of a degree of anomaly. Fig. Figure 7 is a graphical representation illustrating an anomaly determination based on a degree of anomaly. Fig. Figure 8 is a graphical representation showing an exemplary configuration of a rotating machine system 100b according to a third embodiment. Fig. Figure 9 is a graphical representation showing an exemplary configuration of a rotating machine system 100c according to a fourth embodiment. Fig. Figure 10 is a graphical representation showing an exemplary configuration of a rotating machine system 100d according to a fifth embodiment. Fig. Figure 11 is a graphical representation showing an exemplary configuration of a rotating machine system 100e according to a sixth embodiment. Fig. Figure 12 is a graphical representation showing an exemplary configuration of a rotating machine system 100f according to a seventh embodiment. Fig. Figure 13 is a graphical representation showing the hardware configuration of the diagnostic support devices 1, 1a to 1d according to the first to fifth embodiments. Fig. Figure 14 is a graphic representation showing an exemplary configuration of a rotating machine system 100g according to a comparative example (patent literature 1). DESCRIPTION OF THE EXECUTION FORMS

[0016] The following describes, where necessary, the ways in which the present invention can be carried out (hereinafter referred to as embodiments) with reference to the drawings. It is stated that the embodiments described below are merely examples and are not intended to limit the scope of protection of the invention to those embodiments. [Comparative example]

[0017] First, a comparative example of the present embodiments is shown.

[0018] Fig. Figure 14 is a graphic representation showing an exemplary configuration of a rotating machine system 100g according to a comparative example (patent literature 1).

[0019] In a rotating machine 3, such as a motor, and its auxiliary equipment, there are various possible parts where an anomaly can occur, and there are several possible reasons for such an anomaly. Examples of reasons for an anomaly include deterioration of the rotating machine's bearing, a damaged tooth in a gear, deterioration of the bearing of a load device 4, a foreign object attacking the load device 4, and a scratch in the inner race of the rotating machine's bearing. The auxiliary equipment of the rotating machine 3, as previously described, includes a power conversion device 2, gears, a clutch, the load device 4, and the like.

[0020] As in Fig. As shown in Figure 14, in the rotating machine system 100g, the power conversion device 2 and the rotating machine 3 are electrically connected. Furthermore, the rotating machine 3 and the load device 4 are mechanically connected. Possible methods for the mechanical connection between the rotating machine 3 and the load device 4 include directly connecting the load device 4 to the rotating shaft of the rotating machine 3 or connecting the rotating machine 3 and the load device 4 via gears.

[0021] Additionally, a diagnostic support device 1g is arranged to diagnose an anomaly in the rotating machine 3 or its auxiliary devices by detecting an internal value from the power conversion device 2.

[0022] In the previously used diagnostic support device 1g, which is in Fig. As shown in Figure 14, a power ratio estimation unit 17 first estimates the ratio between the input power and the output power of the rotating machine 3 based on an internal value from the power conversion device 2. Then, an anomaly detection unit 18 determines, based on a change in the estimated power ratio, whether the rotating machine system 100g is experiencing an anomaly.

[0023] However, the ratio between the input power and the output power of rotating machine 3 alone is insufficient to estimate where deterioration has occurred. Consequently, it is difficult to identify a component that requires replacement. Furthermore, the above procedure is problematic because repairs must be planned based on the specific part that has deteriorated.

[0024] The present embodiments aim to solve such problems. [First embodiment]

[0025] Fig. Figure 1 is a graphical representation showing an exemplary configuration of a rotating machine system 100 according to a first embodiment.

[0026] The configurations in Fig. 1, those in Fig. 14 similar items are designated by the same reference symbols and are not described here.

[0027] The inventors discovered that, among the internal values ​​in the power conversion device 2, the oscillation of a specific frequency component in a q-axis current component occurs in accordance with a part exhibiting an anomaly, which is, for example, associated with mechanical deterioration of the rotating machine system 100. The inventors also discovered that the trend of the condition of each anomalous part can be monitored based on a change in the specific frequency component of the q-axis current over time.

[0028] As in Fig. Figure 1 shows a diagnostic support device 1 according to the first embodiment, characterized in that the diagnostic support device 1 has an amplitude calculation unit 11 and a display unit 12.

[0029] The amplitude calculation unit 11 acquires a q-axis current, which is one of the internal values ​​in the power conversion device 2, from the power conversion device 2. It is specified that a q-axis current, as referred to here, is a time-dependent change in a value of the q-axis current. The amplitude calculation unit 11 then transforms the acquired q-axis current into the amplitudes of the respective frequencies. In other words, the amplitude calculation unit 11 performs a Fourier transform on the q-axis current input to it.

[0030] Display unit 12 shows the calculation results obtained by amplitude calculation unit 11. More specifically, display unit 12 visualizes the chronological change of the frequency spectrum calculated by amplitude calculation unit 11, thus presenting an anomaly state to the user.

[0031] The user diagnoses the condition of the rotating machine system 100 based on the calculation results obtained by the amplitude calculation unit 11 and displayed on the display unit 12, i.e., the frequency spectrum of a q-axis current. (Example frequency spectrum)

[0032] The capabilities of the diagnostic support device 1 of the first embodiment are described below by describing an exemplary diagnostic procedure performed by the diagnostic support device 1, or specifically a procedure for diagnosing the condition in which the inner ring of the bearing of the rotating machine has a scratch.

[0033] The Fig. 2A and Fig. Figure 2B shows an example frequency spectrum of a q-axis current. Fig. 3A and Fig. Figure 3B also shows an exemplary vibration frequency spectrum of the rotating machine 3. Furthermore, the Fig. 4A and Fig. 4B an exemplary frequency spectrum of the load currents flowing in the respective phases of the rotating machine 3.

[0034] Here, the bearing speed is controlled to 164.32 rad / s (26.17 Hz). For the q-axis current, the power conversion device 2 is also modified so that an internal value can be output in text format to a microcomputer for controlling the power conversion device 2. If the power conversion device 2 has the capability to output a log of the q-axis current, this log output capability can be used, or a new log output capability can be added for diagnostic purposes.

[0035] The Fig. 2A, Fig. 3A and Fig. 4A shows a frequency spectrum when there is no anomaly, i.e., before a scratch is created, while the Fig. 2B, Fig. 3B and Fig. 4B shows a frequency spectrum after a scratch has been created (if there is an anomaly). The fundamental frequency f0 of a current fed into the rotating machine 3 is 27.77 Hz. Then, a principal component of a natural frequency fq of a q-axis current that occurs due to an anomaly is 149.00 Hz, as shown in Fig. Figure 2B shows that a principal component of the natural frequency fv of an oscillation occurring due to an anomaly is 149.00 Hz, as shown in Figure 2B. Fig. Figure 3B shows that a major component of the natural frequency fc of a load current is 121.20 Hz or 176.80 Hz, as shown in Figure 3B. Fig. 4B is shown. Fig. 2B, Fig. 3B and Fig. 4B is a natural frequency due to an anomaly (natural frequency of an anomaly) denoted by vertical lines. Furthermore, in the Fig. 2B, Fig. 3B and Fig. 4B is the main component of the natural frequency of the anomaly, marked with an upward-pointing arrow.

[0036] The in the Fig. The frequency spectra shown in Figures 2A to 4B compare the states resulting from driving under the same control conditions. From the standpoint of diagnostic accuracy, it is desirable to extract and diagnose the sections driven under the same control conditions because a change in the control conditions is likely to be falsely detected as an anomaly, especially if the magnitude of the anomaly is small.

[0037] As can be seen from the comparison between the Fig. 2A and Fig. 3A and the Fig. 2B and Fig. As is evident in 3B, a change in the frequency spectrum of a q-axis current due to the generation of a scratch reflects the oscillation frequency spectrum of the rotating machine 3. In other words, from a comparison between the Fig. 2B and Fig. 3B shows that the natural frequencies of the anomalies match.

[0038] In contrast, the load current does not reflect the vibration frequency spectrum of the rotating machine 3, as can be seen from the comparison between the Fig. 3A and Fig. 4A and the Fig. 3B and Fig. 4B is evident. In other words, in a comparison between the Fig. 3B and Fig. 4B The natural frequencies of the anomalies do not match.

[0039] This establishes that the q-axis current is suitable for anomaly analysis of the rotating machine 3 and its auxiliary devices.

[0040] In this way, an increase in a natural frequency component due to the generation of a scratch can be verified using a q-axis current. It can be confirmed that the user can diagnose an anomaly in the rotating machine system 100 by checking for an increase in the amplitude of a characteristic spectrum of a q-axis current. Furthermore, it is confirmed that the presence or absence of an anomaly and its location (in this case, the bearing) can be detected through a diagnosis focused on a specific frequency.

[0041] The previously used diagnostic support device 1g uses a q-axis current as the control information. The present embodiment uses a q-axis current not only for control but also for locating an anomalous part.

[0042] In particular, the first embodiment can provide a user with information for locating an anomalous part without requiring additional sensors, such as current sensors or vibration sensors. [Second embodiment]

[0043] Fig. Figure 5 is a graphical representation showing an exemplary configuration of a rotating machine system 100a according to a second embodiment. The configurations in Fig. 5, those in Fig. 1. Similar to those in the above, they are identified by the same reference symbols as those in the above. Fig. 1 is designated and will not be described here.

[0044] A diagnostic support device 1a of the rotating machine system 100a differs from the one in Fig. 1 Diagnostic support device 1 shown in the following points. (c1) The diagnostic support device 1 includes an anomaly natural frequency storage unit 21 which stores the natural frequencies of mechanical anomalous parts in the rotating machine 3 and the auxiliary devices (natural frequencies of the anomalies; information about the natural frequencies of the anomalies). (c2) The diagnostic support device 1a includes a unit 13 for localizing anomalous parts, which locates an anomalous part by comparing a frequency spectrum output by the amplitude calculation unit 11 with the natural frequencies of the anomalies stored in the anomaly natural frequency storage unit 21. (c3) The diagnostic support device 1a includes a notification unit 14 which notifies a user of a localized anomalous part.

[0045] In an example described in the first embodiment, a frequency spectrum is displayed on the display unit 12, and it is the user's responsibility to determine whether an anomaly exists. In contrast, in the second embodiment, the natural frequencies of the anomalies to be monitored are predetermined. The unit 13 then diagnoses the condition of the rotating machine system 100a for locating anomalous parts in a frequency spectrum, locating an anomalous part based on a change in the amplitude of a q-axis current at the predetermined natural frequencies of the anomalies.

[0046] A frequency input into the natural frequency of an anomaly can be a frequency obtained through an actual measurement, or a vibration frequency of the machine can be determined based on the geometric shape of a machine part.

[0047] Here, the natural frequency of an anomaly corresponding to a scratch on a bearing is expressed by the formulas described in Table 1.

[0048] In Table 1, f is denoted r the rotational frequency of a bearing of the rotating machine, n b the number of balls, b d the ball diameter, p d the flank diameter and β the contact angle. Table 1 shows the main components of the respective natural frequencies of the anomalies.

[0049] Only one natural frequency of an anomaly (e.g., the principal component) can be entered into Unit 13 for locating anomalous parts. Alternatively, two or more conceivable principal frequency components of an anomaly's natural frequency can be entered. A higher-order component of an anomaly's natural frequency can also be entered. [Table 1] Teil Eigenfrequenz der Anomalie Käfig fr2(1−bdpd−cosβ) Außenring nbfr2(1−bdpd−cosβ) Innenring nbfr2(1+bdpdcosβ) Kugeln pdbdfr2{1−(bdpdcosβ)2} ⋮ ⋮

[0050] It is stated that the natural frequencies of the anomalies shown in Table 1 are merely examples and that the natural frequencies of the anomalies can be entered into Unit 13 via other parts for the purpose of locating anomalous parts.

[0051] It is also stated that the frequency components can be entered into unit 13 for locating anomalous parts as a natural frequency of an anomaly, not only via the anomalies that gradually progress, but also via an installation error, such as a wave offset.

[0052] Unit 13, for locating anomalous parts, locates an anomalous part by determining whether the amplitudes of the respective input frequencies, corresponding to the natural frequencies of the anomalies, are larger in a frequency spectrum.

[0053] Unit 13 for locating anomalous parts determines, using any of the following methods, whether the amplitude of a frequency corresponding to a natural frequency of an anomaly is greater. (d1) Whether the amplitude exceeds a predetermined threshold. (d2) The ratio of the amplitude of a measured frequency to the amplitude already known as a normal state. (d3) Machine learning. More specifically, machine learning can be used to distinguish an anomalous state from a normal state. Here, a machine learning algorithm is not restricted to a specific algorithm, as long as a chosen method can clearly communicate the difference between a normal state and an anomalous state. (d4) Determination based on a given index.

[0054] It is desirable that the means for notifying the user of a localization result can be appropriately selected.

[0055] The following methods are conceivable for notifying a user: (e1) Display on the screen. (e2) Light from a lamp. It is better if the color of the light from the lamp is different for each part. (e3) Notify by email.

[0056] Furthermore, the following procedures are conceivable as notification procedures. (f1) Notifying a user with a value obtained by performing some type of processing (such as the machine learning described above) on a change in the amplitude of a natural frequency of an anomaly. (f2) Notify a user when a predefined threshold is exceeded. (Determination based on a degree of anomaly)

[0057] Here, a description of anomaly detection based on a degree of anomaly is given as an example of using machine learning for anomaly detection.

[0058] Fig. Figure 6 is a graphical representation illustrating the definition of a degree of anomaly.

[0059] First, as in Fig. Figure 6 shows the defined coordinate axes x1 and x2. These coordinate axes represent parts within the rotating machine 3 and its auxiliary devices. For example, x1 represents the outer ring, while x2 represents the inner ring.

[0060] Furthermore, µ1 is the average amplitude of the natural frequency of the anomaly of x1. Likewise, v2 is the average amplitude of the natural frequency of the anomaly of x2.

[0061] N1 is a probability density function on the amplitude of the eigenfrequency of the anomaly of x1. Similarly, N2 is a probability density function on the amplitude of the eigenfrequency of the anomaly of x2. N1 and N2 are both normal distributions.

[0062] Furthermore, L is a curved line (curved plane) that connects the mean µ1 of x1 and the mean µ2 of x2. Although L in Fig. If line 6 is curved, then L can be straight.

[0063] Furthermore, M1 is a measured value of the amplitude of the natural frequency of the anomaly x1. Similarly, M2 is a measured value of the amplitude of the natural frequency of the anomaly x2. In this context, it is assumed that M1 is located at a position 3σ1 away from the mean µ1, where σ1 is a distribution of the probability density function N1. Likewise, it is assumed that M2 is located at a position σ2 / 3 away from the mean µ2, where σ2 is a distribution of the probability density function N2. In other words, M1 is an event that occurs with a very low probability.

[0064] Next, a point P with coordinates (M1, M2) is defined in a coordinate system formed by x1 and x2. M1 and M2 are the amplitudes of the natural frequencies of the anomalies of the measured parts and can be easily calculated from the information in Table 1.

[0065] Next, a vector OP is defined that connects the origin O and the point P. Then, the intersection point of vector OP with the curved line L is defined. Finally, the distance of a vector QP (the size of the thick arrow, i.e., the length of line QP) is defined as a degree of anomaly.

[0066] It is stated that, for the sake of simplicity, the illustration in Fig. 6. A two-dimensional coordinate system is formed by x1 (e.g., the outer ring) and x2 (e.g., the inner ring). In reality, however, there are as many coordinate axes as there are parts to be located that exhibit an anomaly, thus forming a multi-dimensional coordinate system.

[0067] Fig. Figure 7 is a graphical representation illustrating an anomaly determination based on a degree of anomaly.

[0068] In a Fig. In the example described in 7, a change in the state of the rotating machine system 100a is diagnosed using a degree of anomaly based on the amplitude of an input natural frequency of the anomaly, which contains a higher-order component, a scratch in the inner ring of the bearing.

[0069] If the degree of anomaly exceeds a predetermined threshold, as in Fig. As shown in Figure 7, Unit 13, used for localizing anomalous parts, determines that an anomaly exists. Such a threshold can be set using machine learning, as described above, or can be empirically determined by a user. Then, when the degree of anomaly exceeds the specified threshold, Unit 13 projects the value shown in Figure 7. Fig. 6. Vector OP shown on each of the coordinate axes (the x1-axis and the x2-axis in Fig. 6) Unit 13 for locating anomalous parts then determines that an anomaly occurs at a part corresponding to the coordinate axis with the largest of the orthogonal projection vectors on the coordinate axes resulting from the projection.

[0070] The notification unit 14 then notifies the user about the localized anomalous part, as previously described.

[0071] According to the in the Fig. 6 and Fig. The procedures shown in section 7 allow for the verification of an increase in the degree of anomaly as it progresses. Furthermore, when an anomaly occurs, the anomalous part can be easily located. The anomaly can then be communicated to the user by providing a suitable threshold value, prompting them to take appropriate action.

[0072] According to the second embodiment, the rotating machine 3 and its auxiliary devices can be diagnosed without technical knowledge. [Third embodiment]

[0073] Fig. Figure 8 is a graphical representation showing an exemplary configuration of a rotating machine system 100b according to a third embodiment. The configurations in Fig. 8, those in Fig. 5 are similar, are distinguished by the same reference symbols as those in Fig. 5 are designated and are not described here.

[0074] A diagnostic support device 1b of the rotating machine system 100b differs from the one in Fig. 5 Diagnostic support device 1a shown in the following points. (g1) It is the amplitude calculation unit 11 that receives the input from the anomaly natural frequency storage unit 21. (g2) The amplitude calculation unit 11 performs a discrete Fourier transform on a q-axis current input to it only on a frequency section corresponding to a natural frequency of the anomaly.

[0075] In the second embodiment, after the amplitude calculation unit 11 has performed a frequency transformation on a q-axis current, a specific frequency component input by the anomaly natural frequency storage unit 21 is extracted. In contrast, in the third embodiment, the amplitude extraction is performed by a discrete Fourier transform or the like only on a predetermined natural frequency component of the anomaly of a q-axis current input into it, as described above. In this way, the amount of memory used can be saved if the frequency decomposition is performed by a microcomputer or the like with limited memory capacity. [Fourth embodiment]

[0076] Fig. Figure 9 is a graphical representation showing an exemplary configuration of a rotating machine system 100c according to a fourth embodiment. The configurations in Fig. 9, those in Fig. 5 are similar, are distinguished by the same reference symbols as those in Fig. 5 are designated and are not described here.

[0077] A diagnostic support device 1c of the rotating machine system 100c differs from the one in Fig. 5 Diagnostic support device 1a shown in the following point: (h1) A filter unit 15 is provided upstream of the amplitude calculation unit 11. The filter unit 15 is, for example, a low-pass filter, a high-pass filter, or the like.

[0078] In the first three embodiments, an input q-axis current is directly decomposed into the amplitude of each frequency. In the fourth embodiment, by contrast, after a specific frequency bandwidth has been removed from a q-axis current, the q-axis current is transformed into the amplitude of each frequency. Removing the DC and noise components beforehand can, for example, improve the decomposition capability for the current amplitude and, through expansion, improve diagnostic accuracy. [Fifth embodiment]

[0079] Fig. Figure 10 is a graphical representation showing an exemplary configuration of a rotating machine system 100d according to a fifth embodiment. The configurations in Fig. 10, those in Fig. 5 are similar, are identified by the same reference symbols as those in Fig. 5 are designated and are not described here.

[0080] A diagnostic support device 1d of the rotating machine system 100d differs from the one in Fig. 5 Diagnostic Support Device 1a shown in the following points: (i1) An anomaly natural frequency storage unit 21d stores the natural frequencies of the anomalies of various parts in the rotating machine 3 and the auxiliary equipment in conjunction with the model numbers (identification information) of the rotating machine 3 and the auxiliary equipment. The anomaly natural frequency storage unit 21d stores, for example, a table in which the natural frequencies of the anomalies of various parts in the rotating machine 3 and the auxiliary equipment are assigned to the model numbers of the rotating machine 3 and the auxiliary equipment. (i2) The diagnostic support device 1d includes a model number input unit (identification information input unit) 16 which inputs the model numbers of the rotating machine 3 and its auxiliary devices into the anomaly natural frequency storage unit 21d. (i3) An anomaly natural frequency associated with the model number entered by the model number input unit 16 is entered by the anomaly natural frequency storage unit 21d into the unit 13 for anomalous part localization.

[0081] This configuration allows an anomalous part to be located according to the model of the rotating machine 3 or its auxiliary device, which can improve the accuracy of the localization of the anomalous part. [Sixth embodiment]

[0082] Fig. Figure 11 is a graphical representation showing an exemplary configuration of a rotating machine system 100e according to a sixth embodiment. The configurations in Fig. 11, those in Fig. 1. Similar to those in 1, are identified by the same reference symbols as those in 1. Fig. 1 is designated and will not be described here.

[0083] In the Fig. The rotating machine system 100e shown in 11 includes a power conversion device 2e which has the capabilities of any of the diagnostic support devices 1, 1a to 1d according to the first to fifth embodiments.

[0084] This configuration enables the diagnosis of an anomaly in the rotating machine 3 or its auxiliary devices within the power conversion device 2e to be supported. [Seventh embodiment]

[0085] Fig. Figure 12 is a graphical representation showing an exemplary configuration of a rotating machine system 100f according to a seventh embodiment. The configurations in Fig. 12, those in Fig. 1. Similar reference symbols are identified by the same reference symbols as in Fig. 1 is designated and will not be described here.

[0086] In the Fig. In the rotating machine system 100f shown in Figure 12, the power conversion device 2 and any of the diagnostic support devices 1, 1a to 1d according to the first to fifth embodiments are connected to each other via the Internet NW.

[0087] This configuration enables diagnostic support for the rotating machine 3 and the auxiliary devices at remote locations. [Hardware configuration]

[0088] Fig. Figure 13 is a graphical representation showing the hardware configuration of the diagnostic support devices 1, 1a to 1d according to the first to fifth embodiments.

[0089] The diagnostic support devices 1, 1a to 1d are configured with a PC (personal computer), a PLC (programmable logic controller) or the like.

[0090] The diagnostic support devices 1, 1a to 1d include a memory 201, a CPU (central processing unit) 202, a storage device 203, an input device 204 and an output device 205.

[0091] Then the programs stored in memory unit 203 are loaded into memory 201, and the loaded programs are executed by CPU 202. This activates the amplitude calculation unit 11, the anomalous part detection unit 13, the notification unit 14, the filter unit 15, and the like, which are located in the Fig. 1, 5 and 8 to 10 are shown, embodied.

[0092] The storage device 203 also stores the natural frequencies of the anomalies (the Fig. 5, Fig. 8 and Fig. 9), the natural frequencies of the anomalies in conjunction with model numbers ( Fig. 10) or the like.

[0093] Furthermore, the output device 205 corresponds to the display unit 12 in Fig. 1 and the notification unit 14 in the Fig. 5 and 8 to 10.

[0094] The diagnostic support devices 1, 1a to 1d according to the present embodiments do not require a sensor and can therefore be provided inexpensively.

[0095] The rotating machine systems 100, 100a to 100f according to the present embodiments are applicable to a high-voltage motor, a medium-voltage motor, and a low-voltage motor. The rotating machine systems 100, 100a to 100f according to the embodiments are also applicable to industrial equipment, such as the power conversion devices 2, 2a, and a transformer. Furthermore, the rotating machine systems 100, 100a to 100f according to the embodiments are applicable to an electrical device that performs control using a q-axis current.

[0096] Furthermore, in the second to fifth embodiments, the calculation results obtained by the amplitude calculation unit 11 can be displayed on the display unit 12.

[0097] The present invention is not limited to the embodiments described above and includes various modifications. The embodiments described above are, for example, described in detail to facilitate understanding of the present invention, which is not necessarily limited to a single mode of operation with all described configurations. Furthermore, a part of a configuration in one embodiment can be replaced by a configuration in another embodiment, and a configuration in one embodiment can be added to a configuration in another embodiment. Additionally, a part of a configuration in each embodiment can include an additional configuration, be omitted, or be replaced.

[0098] Moreover, some or all of the configurations, capabilities, units 11, 13 to 16, storage device 203, and the like described above may be implemented in hardware, for example, by being designed in an integrated circuit. As in Fig. As shown in Figure 14, the configurations, capabilities, and the like described above can also be implemented by software through a processor, such as the CPU 202, which interprets and executes the programs for implementing the capabilities. The information, such as the programs for implementing the capabilities, the tables, and the files, can be stored in memory 201 on a hard disk drive (HDD), in a recording device, such as a solid-state drive (SSD), or on a recording medium, such as an integrated circuit (IC) card, an SD card (secure digital) card, or a DVD (digital versatile disk).

[0099] Furthermore, in the embodiments shown, the control lines and information lines are those deemed necessary for illustration; not all control lines or information lines of a product are necessarily shown. In reality, it is safe to say that almost all configurations are interconnected. List of reference symbols 1, 1a to 1d, 1g Diagnostic support device 2, 2a Power conversion device (auxiliary device) 3 rotating machine 4 Load device (auxiliary device) 11 Amplitude calculation unit 12 Display unit 13 Unit for locating anomalous parts 14 notification units 15 filter units 16 Model number input unit (identification information input unit) 21, 21d Anomaly natural frequency storage unit (contains information about the natural frequency of the anomaly)

Claims

[1] Diagnostic support device (1) comprising the following: a detection unit that detects a q-axis current in a power conversion device (2) that controls a rotating machine (3); an amplitude calculation unit (11) that calculates an amplitude of each frequency around the q-axis current based on the detected q-axis current; and an output unit (12) that outputs information about the amplitude of each frequency, a storage unit (21) which contains information about natural frequencies of the anomalies, which stores information about a frequency for each of the parts in the rotating machine (3) and an auxiliary device of the rotating machine (3), at which an amplitude increases when an anomaly occurs in at least one of the rotating machine (3) and the auxiliary device; and a unit (13) for locating anomalous parts, which compares the amplitude of each frequency calculated by the amplitude calculation unit (11) with the information about the natural frequencies of the anomalies and thereby locates a part where an anomaly occurs in at least one of the rotating machine (3) and the auxiliary device. [2] Diagnostic support device (1) according to claim 1, wherein the output unit (12) outputs the amplitude of each frequency. [3] Diagnostic support device (1) according to claim 1, wherein the amplitude calculation unit (11) calculates the amplitude only at a frequency that is stored in the information about the natural frequencies of the anomalies. [4] Diagnostic support device (1) according to claim 1, comprising a filter unit (15) between the power conversion device (2) and the amplitude calculation unit (11), which removes a predetermined frequency. [5] Diagnostic support device (1) according to claim 1, wherein the storage unit (21) stores the information about the natural frequencies of the anomalies in conjunction with the identification information about at least one of the rotating machine (3) and the auxiliary device, and the diagnostic support device (1) comprises an identification information input unit which sends the information about the natural frequencies of the anomalies, which is associated with the input identification information about at least one of the rotating machine (3) and the auxiliary device, to the unit (13) for localizing anomalous parts. [6] Diagnostic support device (1) according to claim 1, wherein the unit (13) for locating anomalous parts defines a coordinate axis for each of the parts in the rotating machine (3) and the auxiliary device via an amplitude of a frequency, defines a mean amplitude value for each of the parts and defines a plane that connects the mean amplitude values ​​of the respective parts, After the amplitudes assigned to the respective parts have been calculated from the measured q-axis current, a point P is defined by graphically representing the amplitudes assigned to the respective parts at the coordinates defined by the coordinate axes, and a line OP that connects point P to an origin point O of the coordinates, creates an intersection point Q of the line OP with the plane, defines the length of a line QP as a degree of anomaly, if the length of the degree of the anomaly exceeds a predetermined threshold, it is determined that there is an anomaly in at least one of the rotating machine (3) and the auxiliary device, and The part with the anomaly is located based on a result obtained by projecting the line QP onto each of the coordinate axes. [7] Rotating machine system (100) comprising the following: a rotating machine (3); a power conversion device (2) that controls the rotating machine (3); and a diagnostic support device (1) according to claim 1, which supports an anomaly diagnosis on the rotating machine (3) and an accessory device of the rotating machine (3). [8] Diagnostic support method performed by a diagnostic support device (1) that supports anomaly diagnosis on a rotating machine (3) and an accessory device of the rotating machine (3), the method comprising: Detection of a q-axis current in a power conversion device (2) that controls the rotating machine (3); Calculating the amplitude of each frequency around the q-axis current based on the detected q-axis current; and Outputting information about the amplitude of each frequency, Storing information about natural frequencies of anomalies, wherein a frequency is stored for each of the parts of the rotating machine (3) and the auxiliary device, at which an amplitude increases when an anomaly occurs in at least one of the rotating machine (3) and the auxiliary device; and Comparing the calculated amplitudes of each frequency with the stored information about the natural frequencies of the anomalies and locating a part in which an anomaly occurs in at least one of the rotating machine (3) and the auxiliary device.

Images