ANOMALITY DIAGNOSTIC DEVICE, ANOMALITY DIAGNOSTIC SYSTEM, ANOMALITY DIAGNOSTIC PROCEDURES AND PROGRAM

Abstract

An anomaly diagnostic device (101) according to the present invention comprises: a current signal storage unit (21A) that stores a current signal of an electric motor (5); a frequency analysis unit (22A) that performs a frequency analysis of a waveform of the current signal; a feature frequency band extraction unit (22B) that extracts data belonging to a feature frequency band such that a plurality of spectrum peak values ​​are included in a predetermined frequency range in a frequency analysis result;a feature size calculation unit (221C) that extracts the majority of spectrum peaks from the data belonging to the feature frequency band, excludes data that have been detected as the majority of spectrum peaks from the data belonging to the feature frequency band, and calculates a sum of signal intensities contained in the data belonging to the feature frequency band from which the above data have been excluded;and an anomaly diagnostic unit (22D) that diagnoses that the rotating machine equipment is abnormal in a case where the sum is equal to or greater than a first threshold value. Consequently, an anomaly resulting from a malfunction mode involving a minor torque fluctuation can be accurately diagnosed. A diagnostic result can be transmitted to a plurality of sets of rotating machine equipment connected to a network.

Description

Technical field

[0001] The present invention relates to an anomaly diagnostic device, an anomaly diagnostic system, an anomaly diagnostic method and a program. State of the art

[0002] There are various types of conventional equipment for a rotating machine, including an electric motor, and load equipment, such as a pump, fan, and blower, that use the electric motor as a power source. For example, anomaly diagnostics for the electric motor and load equipment are performed by detecting a signal intensity that varies as a result of an anomaly, based on a frequency analysis of the electric motor's drive current.

[0003] A minor torque fluctuation of the electric motor, resulting from a malfunction mode such as cavitation in the pump or the trapping of foreign material or air within it, is unlikely to be periodic and consequently less likely to appear as a specific spectrum peak in a frequency analysis of the drive current waveform. Therefore, diagnosing an anomaly resulting from a malfunction mode exhibiting a minor torque fluctuation proved difficult.

[0004] Accordingly, a method for diagnosing an anomaly resulting from a malfunction mode involving a slight torque fluctuation is considered. For example, in an anomaly diagnostic device described in patent document 1, the drive current of an electric motor is subjected to a frequency analysis, a fundamental component and harmonics are removed from a predetermined frequency range, and then a predetermined number of intensity values ​​are summed, starting from the highest in the frequency range, to calculate a degree of deterioration.

[0005] This allows an anomaly diagnosis to be performed for the electric motor in a case where an anomaly has occurred as a result of cavitation. Bibliography Patent document

[0006] Patent document 1: Japanese patent application publication JP 2020-153965A Summary of the invention Problem to be solved with the invention

[0007] However, the following applies to the method described in patent document 1: In a case where a malfunction mode, distinct from cavitation or disturbances resulting from the influence of inverter operation, occurs simultaneously with cavitation, a spectrum peak value, different from a fundamental component and harmonics, appears in a predetermined frequency range, such that the spectrum peak value is contained within the sum of intensity values. Consequently, the accuracy of detecting minor torque fluctuations is reduced, making it difficult to accurately detect an anomaly resulting from a malfunction mode exhibiting a minor torque fluctuation.

[0008] The present invention was designed to solve the above problem, and it is an object of the present invention to provide an anomaly diagnostic device and the like which can accurately detect an anomaly resulting from a malfunction mode which has a slight torque fluctuation. Ways to solve the problem

[0009] An anomaly diagnostic device according to the present invention is an anomaly diagnostic device for diagnosing an anomaly in equipment for a rotating machine and comprises the following: a current signal storage unit that stores a current signal of an electric motor; a frequency analysis unit that performs a frequency analysis of a waveform of the current signal stored in the current signal storage unit; a feature frequency band extraction unit that extracts data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range into a frequency analysis result, which is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit;a feature quantity calculation unit that extracts the majority of spectrum peaks from the data belonging to the feature frequency band, excludes data detected as the majority of spectrum peaks from the data belonging to the feature frequency band, and calculates a sum of signal intensities contained in the data belonging to the feature frequency band from which the data detected as the majority of spectrum peaks have been excluded; and an anomaly diagnostic unit that diagnoses that the equipment for a rotating machine is abnormal in a case where the sum is equal to or greater than a first threshold value.

[0010] Another anomaly diagnostic device according to the present invention is an anomaly diagnostic device for diagnosing an anomaly in equipment for a rotating machine and comprises the following: a current signal storage unit that stores a current signal of an electric motor; a frequency analysis unit that performs a frequency analysis of a waveform of the current signal stored in the current signal storage unit; a feature frequency band extraction unit that extracts data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit;a feature quantity calculation unit that sorts the data belonging to the feature frequency band into an intensity order of signal intensities, excludes data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band, and calculates a sum of the signal intensities contained in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; and an anomaly diagnostic unit that diagnoses that the equipment for a rotating machine is abnormal in a case where the sum is equal to or greater than a first threshold.

[0011] An anomaly diagnostic system according to the present invention is an anomaly diagnostic system for diagnosing an anomaly in equipment for a rotating machine and comprises the following: a current signal storage unit that stores a current signal of an electric motor; a frequency analysis unit that performs a frequency analysis of a waveform of the current signal stored in the current signal storage unit; a feature frequency band extraction unit that extracts data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range into a frequency analysis result, which is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit;a feature quantity calculation unit that extracts the majority of spectrum peaks from the data belonging to the feature frequency band, excludes data detected as the majority of spectrum peaks from the data belonging to the feature frequency band, and calculates a sum of signal intensities contained in the data belonging to the feature frequency band from which the data detected as the majority of spectrum peaks have been excluded; and an anomaly diagnostic unit that diagnoses that the equipment for a rotating machine is abnormal in a case where the sum is equal to or greater than a first threshold value.

[0012] Another anomaly diagnostic system according to the present invention is an anomaly diagnostic system for diagnosing an anomaly in equipment for a rotating machine and comprises the following: a current signal storage unit that stores a current signal of an electric motor; a frequency analysis unit that performs a frequency analysis of a waveform of the current signal stored in the current signal storage unit; a feature frequency band extraction unit that extracts data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit;a feature quantity calculation unit that sorts the data belonging to the feature frequency band into an intensity order of signal intensities, excludes data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band, and calculates a sum of the signal intensities contained in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; and an anomaly diagnostic unit that diagnoses that the equipment for a rotating machine is abnormal in a case where the sum is equal to or greater than a first threshold.

[0013] An anomaly diagnostic method according to the present invention comprises: a current signal detection step for detecting a current signal of the current flowing through an electric motor; a frequency analysis step for performing a frequency analysis of a waveform of the current signal detected in the current signal detection step; a feature frequency band extraction step for extracting data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result, which is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step;a data exclusion step for detecting the majority of spectrum peaks from the data belonging to the feature frequency band, and excluding data detected as the majority of spectrum peaks from the data belonging to the feature frequency band; a feature quantity calculation step for calculating a sum of signal intensities contained in the data belonging to the feature frequency band from which the data detected as the majority of spectrum peaks have been excluded; a determination step for determining whether or not the sum is equal to or greater than a first threshold; and an anomaly diagnosis step for diagnosing that the equipment for a rotating machine is abnormal in a case where the sum is equal to or greater than the first threshold.

[0014] Another anomaly diagnostic method according to the present invention is an anomaly diagnostic method for diagnosing an anomaly in equipment for a rotating machine and comprises: a current signal detection step for detecting a current signal of the current flowing through an electric motor; a frequency analysis step for performing a frequency analysis of a waveform of the current signal detected in the current signal detection step; a feature frequency band extraction step for extracting data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step;a data exclusion step for sorting the data belonging to the feature frequency band in an intensity order of signal intensities, and excluding data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band; a feature quantity calculation step for calculating a sum of the signal intensities contained in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; a determination step for determining whether or not the sum is equal to or greater than a first threshold; and an anomaly diagnosis step for diagnosing that the equipment for a rotating machine is abnormal in a case where the sum is equal to or greater than the first threshold.

[0015] A program according to the present invention is a program for diagnosing an anomaly in equipment for a rotating machine and causes a computer to perform the following: a current signal detection step for detecting a current signal of the current flowing through an electric motor; a frequency analysis step for performing a frequency analysis of a waveform of the current signal detected in the current signal detection step; a feature frequency band extraction step for extracting data belonging to a feature frequency band that includes a plurality of spectrum peaks, in a predetermined frequency range, in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step;a data exclusion step for detecting the majority of spectrum peaks from the data belonging to the feature frequency band, and excluding data detected as the majority of spectrum peaks from the data belonging to the feature frequency band; a feature quantity calculation step for calculating a sum of signal intensities contained in the data belonging to the feature frequency band from which the data detected as the majority of spectrum peaks have been excluded; a determination step for determining whether or not the sum is equal to or greater than a first threshold; and an anomaly diagnosis step for diagnosing that the equipment for a rotating machine is abnormal in a case where the sum is equal to or greater than the first threshold.

[0016] Another program according to the present invention is a program for diagnosing an anomaly in equipment for a rotating machine and causes a computer to perform the following: a current signal detection step for detecting a current signal of the current flowing through an electric motor; a frequency analysis step for performing a frequency analysis of a waveform of the current signal detected in the current signal detection step; a feature frequency band extraction step for extracting data belonging to a feature frequency band that includes a plurality of spectrum peaks, in a predetermined frequency range, in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step;a data exclusion step for sorting the data belonging to the feature frequency band in an intensity order of signal intensities, and excluding data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band; a feature quantity calculation step for calculating a sum of signal intensities contained in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; a determination step for determining whether or not the sum is equal to or greater than a first threshold; and an anomaly diagnosis step for diagnosing that the equipment for a rotating machine is abnormal in a case where the sum is equal to or greater than the first threshold. Effect of the invention

[0017] According to the present invention, it is possible to accurately detect an anomaly resulting from a malfunction mode that includes a slight torque fluctuation. Brief description of the drawings Fig. Figure 1 shows a schematic configuration of an anomaly diagnostic device according to embodiment 1. Fig. Figure 2 shows a schematic configuration of the anomaly diagnostic device according to embodiment 1. Fig. Figure 3 shows a schematic configuration of a diagnostic result output unit according to embodiment 1. Fig. Figure 4 shows a hardware configuration of a monitoring diagnostic unit according to embodiment 1. Fig. Figure 5 is a flowchart showing a processing sequence in the anomaly diagnostic device according to embodiment 1. Fig. Figure 6 shows an example of the waveform of a current signal measured by a current detection unit according to embodiment 1. Fig. Figure 7 shows an example of a frequency analysis result for a current signal measured by the current detection unit, according to embodiment 1. Fig. Figure 8 shows an example of a frequency analysis result for a current signal measured by the current detection unit, according to embodiment 1. Fig. Figure 9 shows an example of a frequency analysis result for a current signal measured by the current detection unit, according to embodiment 1. Fig. Figure 10 shows another schematic configuration of the anomaly diagnostic device according to embodiment 1. Fig. Figure 11 shows another schematic configuration of the anomaly diagnostic device according to embodiment 1. Fig. Figure 12 is a schematic configuration diagram of a circuit that is in Fig. 11 is shown. Fig. Figure 13 shows a schematic configuration of an anomaly diagnostic system according to modification 1 of embodiment 1. Fig. Figure 14 shows a schematic configuration of an anomaly diagnostic system according to modification 2 of embodiment 1. Fig. Figure 15 shows a schematic configuration of an anomaly diagnostic device according to embodiment 2. Fig. Figure 16 is a flowchart showing a processing sequence in the anomaly diagnostic device according to embodiment 2. Fig. Figure 17 shows an example of a frequency analysis result for a current signal measured by a current detection unit according to embodiment 2. Fig. Figure 18 shows an example of sorted data according to embodiment 2. Fig. Figure 19 shows a schematic configuration of an anomaly diagnostic system according to modification 1 of embodiment 2. Fig. Figure 20 shows a schematic configuration of an anomaly diagnostic system according to modification 2 of embodiment 2. Fig. Figure 21 shows a schematic configuration of an anomaly diagnostic device according to embodiment 3. Fig. Figure 22 is a flowchart showing a processing sequence in the anomaly diagnostic device according to embodiment 3. Fig. Figure 23 shows a schematic configuration of an anomaly diagnostic device according to embodiment 4. Fig. Figure 24 is a flowchart showing a processing sequence in an anomaly diagnostic device according to embodiment 4. Fig. Figure 25 shows a schematic configuration of an anomaly diagnostic device according to embodiment 5. Fig. Figure 26 is a flowchart showing a processing sequence in an anomaly diagnostic device according to embodiment 5. Description of embodiments

[0018] Embodiments of the present invention are described below with reference to the accompanying drawings. The drawings are schematic, and the relationships between the sizes and positions of the parts shown in the various drawings are not necessarily precise and can be modified as appropriate. In the following description and the drawings, identical components are designated with the same reference numerals, and their names and functions are the same or similar. Therefore, a detailed description of such components can be omitted. Design 1

[0019] With reference to Fig. 1 to Fig. 9 describes an anomaly diagnostic device 101 according to the present embodiment.

[0020] In Fig. 1 The anomaly diagnostic device 101 comprises the following: a current detection unit 1 connected to one of the wires 9A, 9B, 9C which are connected to an electric motor 5, a monitoring diagnostic unit 2 and a diagnostic result output unit 3.

[0021] Current detection unit 1 measures the current flowing through wire 9A, 9B, 9C, thus detecting the drive current for powering electric motor 5. Current detection unit 1 outputs the detected drive current as a current signal to monitoring and diagnostic unit 2. Monitoring and diagnostic unit 2 performs an anomaly detection in equipment 4 for a rotating machine. After detecting an anomaly in equipment 4 for a rotating machine, monitoring and diagnostic unit 2 sends the result to diagnostic output unit 3. Diagnostic output unit 3 notifies a monitor whether or not an anomaly exists.

[0022] The electric motor 5 is a three-phase AC motor. The electric motor 5 is connected to a commercial power supply 8 via an inverter 7 and is driven by the inverter 7.

[0023] Inverter 7 is connected to the commercial power supply 8 and is formed by a combination of an AC-DC converter, which converts AC power from the commercial power supply 8 into DC power, and a DC-AC converter, which converts DC power into AC power. Inverter 7 supplies the AC power converted by the DC-AC converter to the electric motor 5.

[0024] The equipment 4 for a rotating machine comprises the electric motor 5 and a load equipment 6, which is connected to the electric motor 5 and uses the electric motor 5 as its energy source. For example, the load equipment 6 is a water pump, a vacuum pump, a fan, a blower, etc., which is driven by the electric motor 5 as its energy source.

[0025] As an example, the case is described in which the anomaly diagnostic device 101 is used in a public plant monitoring control system represented by a water treatment plant, such as a water purification plant or a wastewater treatment plant. The load equipment 6, such as a water inlet pump or a water supply pump used in the water treatment plant, is driven by the electric motor 5. The anomaly diagnostic device 101 performs anomaly diagnostics on the equipment 4 for a rotating machine comprising the load equipment 6 and the electric motor 5 in the monitoring diagnostic unit 2 using the current signal detected by the current detection unit 1, which is connected to one of the wires 9A, 9B, or 9C connected to the electric motor 5.A result of the anomaly determination for equipment 4 for a rotating machine by the monitoring diagnostic unit 2 is sent to the diagnostic result output unit 3 to notify an operations manager for the water treatment plant whether or not there is an anomaly.

[0026] In the present embodiment, the equipment 4 for a rotating machine comprises the electric motor 5, which is operated by the inverter 7, and the water pump, which is the load equipment 6, and an example is described in which a current signal input by the current detection unit 1 is analyzed to diagnose an anomaly due to cavitation of the water pump, which includes or entails a slight torque fluctuation.

[0027] Cavitation is a phenomenon where, when the pressure of a liquid is reduced, the liquid evaporates and produces bubbles. Once bubbles have formed, when the liquid returns to its low-pressure state, the bubbles disappear with a significant shock. If cavitation occurs in a water pump, the bubbles obstruct the flow of liquid, thus reducing the pump's pumping capacity. Furthermore, when cavitation disappears, an anomaly, such as damage to the pump or the formation of a leak, can occur due to the shock caused by the bubbles' disappearance.

[0028] In the present embodiment, an example is shown in which the current detection unit 1 is connected to any wire of the commercial power supply 8 that is connected to the electric motor 5, but the current detection unit 1 can also be provided for each phase of the commercial power supply 8. Even in this case, it is sufficient that any one of the phases is measured.

[0029] In the present embodiment, an example is shown in which an anomaly occurs due to water pump cavitation, which includes a minor torque fluctuation. However, an anomaly due to a malfunction mode other than water pump cavitation, which also includes a minor torque fluctuation, can also be detected. Examples of such malfunction modes other than water pump cavitation include: trapping of foreign material in the water pump, trapping of air in the water pump, deposition of a byproduct in a vacuum pump, bearing wear, and damage to a fan rotor blade.

[0030] In Fig. The monitoring diagnostic unit 2 comprises a storage unit 21 and an analysis unit 22. The storage unit 21 comprises: a current signal storage unit 21A, a determination reference storage unit 21B, and an anomaly determination storage unit 21C. The analysis unit 22 comprises: a frequency analysis unit 22A, a feature frequency band extraction unit 22B, a feature size calculation unit 221C, and an anomaly diagnostic unit 22D. Fig. 3 The diagnostic result output unit 3 comprises the following: a display unit 31A, an alarm unit 31B and an external output communication unit 31C.

[0031] Fig. Figure 4 shows a hardware configuration of the monitoring diagnostic unit 2 in the present embodiment. The monitoring diagnostic unit 2 comprises: a transmitter / receiver 23, a processor (central processing unit (CPU)) 24, a memory (read-only memory (ROM)) 25, and a memory (selective access memory (RAM)) 26. In the monitoring diagnostic unit 2, the processor 24 executes a program pre-stored in memory 25 and thus diagnoses an anomaly in the equipment 4 for a rotating machine, and consequently, a diagnostic result is output.

[0032] In the monitoring and diagnostic unit 2, various function modules are implemented by the processor 24, which executes a predetermined program stored in memory 25. The function modules include the analysis unit 22. The memory unit 21 is contained in memory 25 and memory 26. The transmit / receive unit 23 transmits / receives signals to / from the current detection unit 1, which is connected to the monitoring and diagnostic unit 2, and to / from the diagnostic result output unit 3, which is also connected to the monitoring and diagnostic unit 2.

[0033] Each functional module of the monitoring diagnostic unit 2 can be implemented by the processor 24, which performs software processing according to a predetermined program as described above, or for some of the functional modules, predetermined numerical and logical operational processing can be carried out by hardware, such as an electronic circuit, which has a function corresponding to the respective functional module.

[0034] With reference to Fig. Section 5 describes a processing sequence in the anomaly diagnostic device 101 according to the present embodiment, together with a detailed description of each component included in the anomaly diagnostic device 101. The flowchart formed by the steps shown below is executed repeatedly each time a predetermined condition is met.

[0035] In step S1, the current detection unit 1 measures the current flowing through the electric motor 5 and outputs a current signal to the current signal storage unit 21A. The current signal storage unit 21A stores the current signal.

[0036] Fig. Figure 6 shows the waveform of the current signal detected by the current detection unit 1. The vertical axis represents the current value, and the horizontal axis represents the time. In the waveform of the current signal, a dotted line denotes the U-phase, a dashed line denotes the V-phase, and a solid line denotes the W-phase.

[0037] Then, in step S2, the frequency analysis unit 22A performs a frequency analysis of the waveform of the current signal, which is captured by the current signal storage unit 21A.

[0038] Fig. Figure 7 shows a frequency analysis result of the current signal waveform for the U-phase, shown with the dotted line in Fig. 6. The vertical axis indicates the current power spectrum, and the horizontal axis indicates the frequency. The frequency analysis result, which is displayed in Fig. Figure 7 shows the result in a case where the power supply frequency is 50 Hz, and has spectrum peaks at 50 Hz, which is the power supply frequency, and 150 Hz, which is a third-order component with respect to the power supply frequency.

[0039] The power supply frequency is the frequency of the commercial power supply, and the third-order component with respect to the power supply frequency is a frequency that is three times the power supply frequency. When the frequency is x times the power supply frequency, the frequency is expressed as an x-th-order component with respect to the power supply frequency. Here, x is an integer not less than 0.

[0040] This section describes the reason why the frequency analysis result, which is in Fig. Figure 7 shows spectrum peak values ​​at the power supply frequency and the third-order component with respect to the power supply frequency.

[0041] In an energy conversion process by inverter 7, the following applies: When AC energy from the commercial power supply 8 is converted into DC energy by means of the AC-DC converter, a current with a distorted waveform is generated by the converter circuitry. This current combines a fundamental component with the power supply frequency and harmonics with frequencies that are integer multiples of the power supply frequency. The distorted waveform current affects the voltage waveform, thus distorting the voltage waveform as well. Similarly, a current with a distorted waveform also flows in any device to which the voltage with the distorted waveform is applied.

[0042] For the above reason, the frequency analysis result, which is in Fig. Figure 7 shows spectrum peak values ​​at the power supply frequency and the third-order component with respect to the power supply frequency. In the present embodiment, the example shown is where the waveform of the current signal for the U-phase is subjected to a frequency analysis. However, a current signal for another single phase, current signals for a plurality of phases, or current signals for all phases can also be subjected to a frequency analysis.

[0043] The example where the power supply frequency is 50 Hz is in Fig. Figure 7 is shown. However, the following description of the processing sequence in the anomaly diagnostic device 101 describes the processing sequence in a case where the power supply frequency is 60 Hz.

[0044] Then extracted in step S3 in Fig. 5. The feature frequency band extraction unit 22B extracts data belonging to a feature frequency band, such that a plurality of spectrum peaks are included, within a predetermined frequency range. In the present embodiment, data belonging to a feature frequency band are extracted such that a plurality of spectrum peaks are included, within a frequency range from a zeroth-order component to a second-order component with respect to the power supply frequency.

[0045] More precisely: In the data of the frequency analysis result for the current signal for the U-phase, entered by the frequency analysis unit 22A, a frequency range from 0 Hz, which is the zeroth order component, to 120 Hz, which is the second order component with respect to the power supply frequency, is specified as a feature frequency band, and data belonging to the feature frequency band are extracted.

[0046] Here, the feature frequency band is a frequency range that will be used to calculate a feature quantity, which will be described later.

[0047] Next, the reason why the frequency range from the zeroth-order component to the second-order component with respect to the power supply frequency is defined as the feature frequency band is described, and data belonging to this feature frequency band are extracted. In a case where cavitation occurs in the load equipment 6, the torque of the electric motor 5 required to drive the load equipment 6 varies slightly compared to the case where no cavitation occurs. Since the torque of the electric motor 5 is determined by the current value, a slight variation in the torque of the electric motor 5 affects the driving current for the electric motor 5.

[0048] Consequently, in the frequency analysis result for the current signal, which is affected by the slight torque fluctuation, the signal intensity increases in the frequency range from the zeroth order component to the second order component with respect to the power supply frequency, which is a frequency range in the vicinity of the power supply frequency.

[0049] Fig. Figure 8 shows an example of a graph comparing frequency analysis results for current signals in a case with a slight torque variation in the load equipment 6 and a case without a slight torque variation. The vertical axis represents the current power spectrum, and the horizontal axis represents the frequency. A solid line represents the case with a slight torque variation, and a dotted line represents the case without a slight torque variation. The power supply frequency is 60 Hz, and the rotational frequency of the electric motor 5 is 30 Hz. The rotational frequency is the value obtained when the rotational speed of the electric motor 5 is expressed as a frequency.

[0050] As shown by the solid line, it can be confirmed that in the case where there is a slight torque fluctuation, the signal intensity increases in the frequency range of ±40 Hz from the power supply frequency towards the low frequency side and the high frequency side, centered at 60 Hz, which is the power supply frequency, i.e., the frequency range of 20 Hz to 100 Hz, compared to the case where there is no slight torque fluctuation.

[0051] For the above reason, the frequency range from 0 Hz, which is the zeroth order component, to 120 Hz, which is the second order component with respect to the power supply frequency, is extracted as the feature frequency band.

[0052] While Fig. 8 shows the influence of the slight torque variation in the frequency range from the zeroth order component to the second order component with respect to the power supply frequency, which is a frequency range in the vicinity of the power supply frequency, a slight torque variation also has an influence in a frequency range from a fourth order component to a sixth order component with respect to the power supply frequency, which is a frequency range in the vicinity of a fifth order component with respect to the power supply frequency, and in a frequency range from a sixth order component to an eighth order component, which is a frequency range in the vicinity of a seventh order component with respect to the power supply frequency.

[0053] Furthermore - as in Fig. As shown in Figure 8, in the frequency range from the zeroth-order component to the second-order component with respect to the power supply frequency, a plurality of spectrum peaks appear on both sides of the power supply frequency. Fig. At frequencies defined as (power supply frequency ± rotation frequency), i.e., 30 Hz and 90 Hz, a spectrum 12A of sideband components of a modulation wave, which will be described later, appears as spectrum peaks. Additionally, at frequencies defined as (power supply frequency ± 20 Hz), i.e., 40 Hz and 80 Hz, a spectrum 12B of interference components resulting from switching operations of the inverter 7, which will be described later, appears as spectrum peaks.

[0054] The modulation waveform is one of the elements used in pulse-width modulation (PWM) control, which is an energy control method for a DC-AC inverter. Here, the modulation waveform is a fundamental component, and its frequency corresponds to the power supply frequency.

[0055] The sideband components of the modulation wave depend on the rotational frequency of the electric motor 5 and appear at frequencies shifted by the rotational frequency in both directions toward the modulation wave frequency. Fig. 8 The spectrum 12A of the sideband components of the modulation wave appears at frequencies defined as (power supply frequency ± rotation frequency).

[0056] Regarding the disturbance component resulting from the switching operation of inverter 7, the following applies in an energy conversion process by inverter 7: When DC energy is converted to AC energy by means of the DC-AC inverter, the DC voltage output from the AC-DC converter and the AC voltage output to the electric motor 5 vary slightly at the power supply frequency and at frequencies that are integer multiples of the power supply frequency due to the switching operations or switching operation of the inverter circuit. The slightly varying values ​​appear as a spectrum 12B of disturbance components resulting from the switching operations of inverter 7.

[0057] Although the spectrum 12B has interference components at ± 20 Hz with respect to the power supply frequency in Fig. 8 appears, such interference components can also appear at frequencies that differ from (power supply frequency ± 20 Hz), depending on the control method, the type or the like of the inverter 7 to be used.

[0058] Spectrum 12A of sideband components of the modulation wave and spectrum 12B of interference components resulting from the switching operations of inverter 7 are examples of interference caused by the operation of the inverter.

[0059] The description above shows an example where spectrum 12A of sideband components and spectrum 12B of disturbance components appear in the frequency domain from the zeroth-order component to the second-order component with respect to the power supply frequency. However, in a case where an anomaly other than cavitation also occurs at the same time, spectrum peaks that differ from spectrum 12A of the sideband components and spectrum 12B of the disturbance components may also appear in the frequency domain.

[0060] The description above shows an example where data belonging to the feature frequency band are extracted so that they contain a plurality of spectrum peaks in the frequency range from the zeroth-order component to the second-order component with respect to the power supply frequency. However, without restriction, data belonging to the feature frequency band can also be extracted so that they contain a plurality of spectrum peaks in a frequency range from the fourth-order component to the sixth-order component with respect to the power supply frequency, a frequency range from the sixth-order component to the eighth-order component with respect to the power supply frequency, or the like.

[0061] For example, data belonging to the feature frequency band can be extracted so that they contain a plurality of spectrum peak values ​​in a frequency range of 0 Hz to 120 Hz, a frequency range of 200 Hz to 360 Hz, a frequency range of 300 Hz to 480 Hz, or the like, so that the case where the power supply frequency is 50 Hz and the case where the power supply frequency is 60 Hz are both covered.

[0062] Fig. Figure 9 shows a case in which data belonging to the feature frequency band, i.e., the frequency range from 0 to 120 Hz, are extracted from the comparison graph of the frequency analysis results for the current signals, as in Fig. 8 shown.

[0063] In step S3 in Fig. 5 extracts the feature frequency band extraction unit 22B data belonging to the feature frequency band, as shown in Fig. Figure 9 shows data from the frequency analysis result for the current signal for the U-phase, entered by the frequency analysis unit 22A.

[0064] In step S31 according to Fig. 5 detects the feature quantity calculation unit 221C - as spectrum peak values ​​- a spectrum of the power supply frequency, a spectrum of the second order component with respect to the power supply frequency, the spectrum 12A of sideband components and the spectrum 12B of disturbance components and excludes them from data belonging to the feature frequency band input by the feature frequency band extraction unit 22B.

[0065] As an example of a spectrum peak detection method, a predetermined number of data points can be identified as spectrum peaks, specifically the one with the highest signal intensity among the data points belonging to the feature frequency band. Alternatively, the feature frequency band can be divided into predetermined frequency intervals, and an average value of the signal intensities in each of the divided intervals is taken. The signal intensities in each of the divided intervals are then compared to the average value, and data points with signal intensities greater than the average value are identified as spectrum peaks.

[0066] In a case where step S31 is applied to data relating to the feature frequency band in Fig. 9, the spectrum of the power supply frequency, the spectrum of the second-order component with respect to the power supply frequency, the spectrum 12A of sideband components and the spectrum 12B of disturbance components are detected and excluded as spectrum peaks from the data belonging to the feature frequency band.

[0067] In the present embodiment, an example is shown in which the spectrum 12A of sideband components and the spectrum 12B of interference components are excluded from data belonging to the feature frequency band. However, in a case where a spectrum peak value that differs from the spectrum 12A of sideband components and the spectrum 12B of interference components appears in the feature frequency band, the spectrum peak value that differs from the spectrum 12A of sideband components and the spectrum 12B of interference components is similarly excluded. That is, spectrum peak values ​​that appear in the feature frequency band are excluded so that they are not used in the diagnosis of an anomaly resulting from a malfunction mode that includes a minor torque fluctuation.

[0068] This section describes the reason why spectrum peaks are excluded from data belonging to the feature frequency band, specifically in step S31. Fig. 5. As in Fig. As shown in Figure 8, the following applies: In a case where there is a slight torque fluctuation, the current signal affected by the slight torque fluctuation has such a frequency characteristic that the signal intensity increases in the frequency range of ±40 Hz from the power supply frequency.

[0069] Since a change in current due to a slight torque fluctuation is only slight, the increase in signal intensity in the feature frequency band is also slight compared to the spectrum peaks. If the spectrum peaks are included in the data belonging to the feature frequency band, the slight change in signal intensity between the spectrum peaks can be masked or hidden. For this reason, spectrum peaks are excluded from the data belonging to the feature frequency band in step S31.

[0070] Then, in step S41, it is calculated in Fig. 5. The feature parameter calculation unit 221C is a feature parameter derived from data belonging to the feature frequency band from which spectrum peaks have been excluded. The feature parameter can be obtained by calculating the sum of all signal intensities contained in the data belonging to the feature frequency band from which spectrum peaks have been excluded. That is, in the present embodiment, the feature parameter is the sum of all signal intensities contained in data belonging to the feature frequency band from which spectrum peaks have been excluded.

[0071] Here, the signal intensities are current values ​​or a current power spectrum. Fig. 9 corresponds to the feature size of the sum of all current power spectra contained in data belonging to the feature frequency band from which spectrum peaks have been excluded.

[0072] In step S4A in Fig. 5 then determines, using a recording period for initial learning, denoted by T, and a predetermined period, denoted by T0, the feature quantity calculation unit 221C, whether or not the recording period T for initial learning is less than the predetermined period T0.

[0073] Here, "initial learning recording" means that—within a predetermined period from the start of operation of equipment 4 for a rotating machine—the feature size is calculated in step S41 and the calculated feature size is accumulated in the determination reference storage unit 21B. The initial learning recording period is a predetermined timeframe during which the initial learning recording is repeatedly executed to generate a determination reference that is subsequently described.

[0074] If the recording period T for initial learning is less than the predetermined period T0 (YES in step S4A), the process continues with step S4B. If the recording period T for initial learning is equal to or greater than the predetermined period T0 (NO in step S4A), the process continues with step S4E.More precisely: The case in which the initial learning recording period T is less than the predetermined period T0 corresponds to a state within the initial learning recording period from the start of operation of the equipment 4 for a rotating machine, and the case in which the initial learning recording period T is equal to or greater than the predetermined period T0 corresponds to a period in which the anomaly diagnostic device 101 performs an anomaly diagnosis on the equipment 4 for a rotating machine after the initial learning recording period has ended.

[0075] In step S4B, the feature size calculation unit 221C then accumulates the feature size in the determination reference storage unit 21B, so that it performs the recording for initial learning.

[0076] Then, in step S4C, the feature calculation unit 221C determines whether the initial learning recording period T is less than the predetermined period T0. If the initial learning recording period T is less than the predetermined period T0 (YES in step S4C), the process continues with step S1. If the initial learning recording period T is equal to or greater than the predetermined period T0 (NO in step S4C), the process continues with step S4D.

[0077] Then, in step S4D, the feature parameter calculation unit 221C performs statistical processing of the feature parameters that have accumulated in the determination reference storage unit 21B during the predetermined period T0 from the start of operation of the equipment 4 for a rotating machine through the initial learning recording. This generates a determination reference, which is a first threshold value, and stores the generated determination reference in the determination reference storage unit 21B.

[0078] As an example of a statistical processing procedure for generating the determination reference, the determination reference can be generated by calculating the mean value, a dispersion or spread σ, 2σ, 3σ, or the like of the feature quantities that are accumulated in the determination reference storage unit 21B.

[0079] Then, in step S4E, the anomaly diagnostic unit 22D determines whether the feature size is equal to or greater than the first threshold. If the feature size is equal to or greater than the first threshold (YES in step S4E), the process continues with step S5. If the feature size is less than the first threshold (NO in step S4E), the process continues with step S7.

[0080] In step S5, the anomaly diagnostic unit 22D then diagnoses that the equipment 4 for a rotating machine is abnormal and outputs the diagnostic result to the anomaly determination storage unit 21C.

[0081] The following applies to the feature size calculation unit 221C: In a case where the number of data points of the determination reference generated by the initial learning and the number of data points of the feature size are different, the comparison between the determination reference and the feature size cannot be carried out exactly, and therefore processing is performed to equalize or balance the number of data points of the determination reference and the number of data points of the feature size.

[0082] If the number of data points in the reference set is greater than the number of data points in the feature variable, data points with higher signal intensities are sequentially excluded from the reference set so that the number of data points in the reference set and the number of data points in the feature variable match. If the number of data points in the reference set is less than the number of data points in the feature variable, data points with higher signal intensities are sequentially excluded from the feature variable so that the number of data points in the feature variable and the number of data points in the reference set match.

[0083] As an example of such a case, where the number of data points of the determining reference and the number of data points of the feature size are different, there is the case where the resolution of the frequency analysis is the same, but the frequency range of the feature frequency band for extracting data belonging to the feature frequency band is different.

[0084] In step S6, the diagnostic result output unit 3 then receives a diagnostic result from the anomaly determination storage unit 21C. Subsequently, the diagnostic result output unit 3 displays the diagnostic result on the display unit 31A, such as a screen; the alarm unit 31B triggers an alarm, for example by emitting an alarm tone or by activating or flashing an anomaly light, in a case where the equipment 4 for a rotating machine is diagnosed as abnormal; and the external output communication unit 31C transmits the diagnostic result to an external device, such as a control panel, a personal computer (PC), or a cloud server.

[0085] In step S7, the anomaly diagnostic unit 22D determines whether or not the anomaly diagnosis should continue. If the anomaly diagnosis is continued (YES in step S7), the process proceeds to step S1. If the anomaly diagnosis is not continued (NO in step S7), the process terminates.

[0086] As an example of a procedure for determining whether or not to continue anomaly diagnosis, a period during which the anomaly diagnosis will continue is predetermined for the anomaly diagnosis unit 22D, and if this period is exceeded, it can be determined that the anomaly diagnosis will not continue. The period during which the anomaly diagnosis will continue is, for example, the period during which equipment 4 is operating on a rotating machine.

[0087] As described above, the anomaly in equipment 4 for a rotating machine is addressed by steps S1 to S7 in Fig. 5 diagnosed.

[0088] As described above, in the anomaly diagnostic device 101 of the present embodiment, the sum of all signal intensities contained in data belonging to the feature frequency band, excluding data detected as spectrum peaks, is compared with the first threshold value, thus performing anomaly diagnostics for the equipment 4 of a rotating machine. Consequently, a minor change in signal intensities resulting from slight torque fluctuations between spectrum peaks is prevented from being missed or masked, thereby improving the detection accuracy for minor torque fluctuations. As a result, an anomaly resulting from a malfunction mode involving minor torque fluctuations can be accurately diagnosed.

[0089] As described above, the anomaly diagnostic device 101 of the present embodiment is an anomaly diagnostic device 101 for diagnosing an anomaly in the equipment 4 for a rotating machine, and it comprises the following: a current signal storage unit 21A that stores a current signal of an electric motor 5; a frequency analysis unit 22A that performs a frequency analysis of a waveform of the current signal stored in the current signal storage unit 21A; a feature frequency band extraction unit 22B that extracts data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range into a frequency analysis result, which is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit 22A;a feature size calculation unit 221C, which extracts the majority of spectrum peaks from the data belonging to the feature frequency band, excludes data that have been detected as the majority of spectrum peaks from the data belonging to the feature frequency band, and calculates a sum of signal intensities contained in the data belonging to the feature frequency band from which the data that have been detected as the majority of spectrum peaks have been excluded;and an anomaly diagnostic unit 22D that diagnoses that the equipment 4 for a rotating machine is abnormal in a case where the sum is equal to or greater than a first threshold value. Consequently, a minor change in signal intensities resulting from the minor torque variation between the spectrum peaks is prevented from being missed or hidden, thus improving the detection accuracy for the minor torque variation, and as a result, an anomaly resulting from a malfunction mode that includes a minor torque variation can be accurately diagnosed.

[0090] As described above, the anomaly diagnostic method of the present embodiment comprises: a current signal detection step S1 for detecting a current signal of the current flowing through an electric motor 5; a frequency analysis step S2 for performing a frequency analysis of a waveform of the current signal detected in the current signal detection step S1; a feature frequency band extraction step S3 for extracting data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result, which is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step S2;a data exclusion step S31 for detecting the majority of spectrum peaks from the data belonging to the feature frequency band and excluding data detected as the majority of spectrum peaks from the data belonging to the feature frequency band; a feature quantity calculation step S41 for calculating a sum of signal intensities contained in the data belonging to the feature frequency band from which the data detected as the majority of spectrum peaks have been excluded; a determination step S4E for determining whether or not the sum is equal to or greater than a first threshold; and an anomaly diagnosis step S5 for diagnosing that the equipment 4 is abnormal for a rotating machine in a case where the sum is equal to or greater than the first threshold.

[0091] Consequently, a slight change in signal intensity resulting from the slight torque fluctuation between spectrum peaks is prevented from being missed or hidden, thus improving the detection accuracy for the slight torque fluctuation, and as a result, an anomaly resulting from a malfunction mode that includes a slight torque fluctuation can be accurately diagnosed.

[0092] As described above, the program of the present embodiment is a program for diagnosing an anomaly in the equipment 4 for a rotating machine, and it causes a computer to perform the following: a current signal detection step S1 to detect a current signal of the current flowing through an electric motor 5; a frequency analysis step S2 to perform a frequency analysis of a waveform of the current signal detected in the current signal detection step S1; a feature frequency band extraction step S3 to extract data belonging to a feature frequency band that includes a plurality of spectrum peaks, in a predetermined frequency range, in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step S2;a data exclusion step S31 for detecting the majority of spectrum peaks from the data belonging to the feature frequency band and excluding data detected as the majority of spectrum peaks from the data belonging to the feature frequency band; a feature quantity calculation step S41 for calculating a sum of signal intensities contained in the data belonging to the feature frequency band from which the data detected as the majority of spectrum peaks have been excluded; a determination step S4E for determining whether the sum is equal to or greater than a first threshold value;and an anomaly diagnosis step S5 to diagnose that the equipment 4 is abnormal for a rotating machine in a case where the sum is equal to or greater than the first threshold. Consequently, a minor change in signal intensities resulting from the minor torque variation between the spectrum peaks is prevented from being missed or hidden, thus improving the detection accuracy for the minor torque variation, and as a result, an anomaly resulting from a malfunction mode that includes a minor torque variation can be accurately diagnosed.

[0093] In the present embodiment, the example shown is that the electric motor 5 is operated with the inverter 7. As in Fig. As shown in Figure 10, the electric motor 5 can also be connected to the commercial power supply 8 via a plurality of circuit breakers or disconnectors 10A, 10B, 10C and a plurality of electromagnetic switches or contactors 11A, 11B, 11C, and it can be operated with the commercial power supply 8.

[0094] In the present embodiment, an example is shown in which the anomaly diagnostic device 101 comprises the current detection unit 1, the monitoring diagnostic unit 2, and the diagnostic result output unit 3. However, the current detection unit 1 and the diagnostic result output unit 3 can also be provided with an external device that is different from the anomaly diagnostic device 101.

[0095] As in Fig. As shown in Figure 11, in the present embodiment the anomaly diagnostic device 101 can be configured such that at least one of the current detection unit 1, the monitoring diagnostic unit 2 and the diagnostic result output unit 3 is assigned to the inverter 7 by using a current detection unit 71 to detect the current flowing through an AC bus of the electric motor 5 for the purpose of inverter control, and an inverter control device 72, which is formed from a processor (CPU) and a memory, which the inverter 7 has. Fig. Figure 12 is a schematic diagram of a circuit representing the functional configuration that is in Fig. 11 is shown. Fig. Figure 12 shows an example where all the current detection unit 1, the monitoring diagnostic unit 2 and the diagnostic result output unit 3 are provided to the inverter 7.

[0096] The hardware configuration of the inverter control unit 72 is the same as the hardware configuration of the monitoring diagnostic unit 2, which is described in Fig. Figure 4 shows the inverter control unit 72. It comprises the transmit / receive unit 23, the processor (central processing unit (CPU)) 24, the memory (read-only memory (ROM)) 25, and the memory (selective access memory (RAM)) 26. Within the inverter control unit 72, the processor 24, executing a program pre-stored in memory 25, performs a three-phase / dq conversion. This conversion is performed using information about the current flowing through the electric motor 5 and the angle of the electric motor 5, detected by a rotation angle sensor (not shown) associated with the electric motor 5. The output of the current detection unit 71 is a d-axis current detection value and a q-axis current detection value, which are detected and stored in memory 26.

[0097] These detection values ​​are compared with a d-axis current command value and a q-axis current command value specified by a host control unit, so that a d-axis voltage command value and a q-axis voltage command value are calculated. The calculated d-axis and q-axis voltage command values ​​and the rotational position information about the electric motor 5 undergo a two-phase / three-phase conversion, and then the transmit / receive unit outputs 23 voltage command values ​​to the windings of the electric motor 5.

[0098] Using the processor 24 and the memory units 25 and 26 of the inverter control unit 72, as described above, the monitoring diagnostic unit 2 and the diagnostic result output unit 3 of the inverter control unit 72 can be provided. The current detection unit 71 for one phase is used as the current detection unit 1, a predetermined program is stored in memory 25, and the processor 24 executes the program, thus implementing various functional modules of the monitoring diagnostic unit 2.

[0099] In the present embodiment, an example is shown in which the data belonging to the feature frequency band are extracted so that they contain a plurality of spectrum peaks in the frequency range from the zeroth-order component to the second-order component with respect to the power supply frequency, and an anomaly diagnosis for the equipment of a rotating machine is performed using the extracted data belonging to the feature frequency band. However, in a plurality of frequency ranges, respective sets of data belonging to the feature frequency bands can be extracted so that they each contain a plurality of spectrum peaks, and anomaly diagnosis for the equipment of a rotating machine can be performed using the extracted respective sets of data belonging to the feature frequency bands.

[0100] For example, in the frequency range from the zeroth-order component to the second-order component with respect to the power supply frequency and in the frequency range from the fourth-order component to the sixth-order component with respect to the power supply frequency, respective sets of data belonging to the feature frequencies can be extracted, such that they each contain a plurality of spectrum peaks, and an anomaly diagnosis for the equipment for a rotating machine can be performed using the data belonging to the feature frequency band extracted in the frequency range from the zeroth-order component to the second-order component with respect to the power supply frequency, and the data belonging to the feature frequency band extracted in the frequency range from the fourth-order component to the sixth-order component with respect to the power supply frequency.

[0101] In the present embodiment, an example is shown in which the determination reference to be used in anomaly diagnosis is generated by repeatedly performing an initial learning recording to accumulate calculated feature values ​​in the determination reference storage unit 21B during a predetermined period from the start of operation of the equipment for a rotating machine, and then performing statistical processing of the accumulated feature values. However, a predetermined determination reference can also be specified and stored in advance in the determination reference storage unit 21B. Modification 1

[0102] Fig. Figure 13 shows an anomaly diagnostic system 201 according to modification 1 of embodiment 1. In the configuration example shown in Fig. As shown in Figure 1, the anomaly diagnostic device 101 is present, in which the current detection unit 1, the monitoring diagnostic unit 2 and the diagnostic result output unit 3 are integrated, and the anomaly diagnostic device 101 performs an anomaly diagnosis for the equipment 4 for a rotating machine, whereas the anomaly diagnostic system 201, which is in Fig. Figure 13 shows a server 30 containing the monitoring diagnostic unit 2 and the diagnostic result output unit 3, as well as current detection units 1-1 to 1-n connected to the equipment 4-1 to 4-n for a rotating machine, including electric motors 5-1 to 5-n and load equipment 6-1 to 6-n, and the current detection units 1-1 to 1-n and the server 30 are connected via a network.

[0103] Here, n is the number of sets of equipment 4-1 to 4-n for a rotating machine, and it is an integer not less than 1. The current detection units 1-1 to 1-n measure current signals from the respective sets of equipment 4-1 to 4-n for a rotating machine. In this case, the monitoring diagnostic unit 2 of the anomaly diagnostic system 201 acquires the current signals from the current detection units 1-1 to 1-n, corresponding to the respective sets of equipment 4-1 to 4-n for a rotating machine, via the network. The remaining operations and configurations of the anomaly diagnostic system 201 are the same as those shown in embodiment 1.

[0104] Consequently, the anomaly diagnostic system 201, which is in Fig. Figure 13 shows that, according to the above modification, the effect is that it becomes unnecessary to provide the anomaly diagnostic device 101 for each set of equipment 4-1 to 4-n for a rotating machine.

[0105] The electric motors 5-1 to 5-n can be of the same type, or the type of at least one of them can be different from the others of the electric motors 5-1 to 5-n. The load equipment sets 6-1 to 6-n can be of the same type, or the type of at least one of them can be different from the others.

[0106] As described above, the anomaly diagnostic system 201 of the present embodiment is an anomaly diagnostic system 201 for diagnosing an anomaly in the equipment 4 for a rotating machine and comprises the following: a current signal storage unit 21A that stores a current signal of an electric motor 5; a frequency analysis unit 22A that performs a frequency analysis of a waveform of the current signal stored in the current signal storage unit 21A; a feature frequency band extraction unit 22B that extracts data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range into a frequency analysis result, which is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit 22A;a feature size calculation unit 221C, which extracts the majority of spectrum peaks from the data belonging to the feature frequency band, excludes data that have been detected as the majority of spectrum peaks from the data belonging to the feature frequency band, and calculates a sum of signal intensities contained in the data belonging to the feature frequency band from which the data that have been detected as the majority of spectrum peaks have been excluded;and an anomaly diagnostic unit 22D, which diagnoses that the equipment 4 for a rotating machine is abnormal in a case where the sum is equal to or greater than a first threshold value. Consequently, a minor change in signal intensities resulting from the minor torque variation between the spectrum peaks is prevented from being missed or hidden, thus improving the detection accuracy for the minor torque variation, and as a result, an anomaly resulting from a malfunction mode that includes a minor torque variation can be accurately diagnosed. Furthermore, the effect is that it becomes unnecessary to provide the anomaly diagnostic device 101 for each set of equipment 4-1 to 4-n for a rotating machine. Modification 2

[0107] Fig. Figure 14 shows an anomaly diagnostic system 202 according to modification 2 of embodiment 1. In the configuration example shown in Fig. As shown in Figure 1, the anomaly diagnostic device 101 is present, in which the current detection unit 1, the monitoring diagnostic unit 2 and the diagnostic result output unit 3 are integrated, and the anomaly diagnostic device 101 performs an anomaly diagnosis for the equipment 4 for a rotating machine, whereas the anomaly diagnostic system 202, which is in Fig. Figure 14 shows anomaly diagnostic equipment 202-1 to 202-n, which includes monitoring diagnostic units 2-1 to 2-n, as well as current detection units 1-1 to 1-n, which are connected to the equipment 4-1 to 4-n for a rotating machine, including electric motors 5-1 to 5-n and load equipment 6-1 to 6-n, and a server 40, which includes a data acquisition unit 42 and the diagnostic result output unit 3, and the anomaly diagnostic equipment 202-1 to 202-n and the server 40 are connected via a network. In this case, the anomaly diagnostic devices 202-1 to 202-n transmit diagnostic results over the network, and the diagnostic result output unit 3 of server 40 receives the diagnostic results from the anomaly diagnostic devices 202-1 to 202-n over the network. The remaining operations and configurations of the anomaly diagnostic system 202 are the same as those shown in embodiment 1.

[0108] As described above, the anomaly diagnostic system 202, which is in Fig. Figure 14 shows that in the present modification the diagnostic results for equipment 4-1 to 4-n for a rotating machine are collectively displayed, so that the comparison between sets of equipment 4-1 to 4-n for a rotating machine, their overall management and the like are simplified.

[0109] As described above, the anomaly diagnostic system 202 of the present embodiment is an anomaly diagnostic system 202 for diagnosing an anomaly in the equipment 4 for a rotating machine and comprises the following: a current signal storage unit 21A that stores a current signal of an electric motor 5; a frequency analysis unit 22A that performs a frequency analysis of a waveform of the current signal stored in the current signal storage unit 21A; a feature frequency band extraction unit 22B that extracts data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range into a frequency analysis result, which is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit 22A;a feature size calculation unit 221C, which extracts the majority of spectrum peaks from the data belonging to the feature frequency band, excludes data detected as the majority of spectrum peaks from the data belonging to the feature frequency band, and calculates a sum of signal intensities contained in the data belonging to the feature frequency band from which the data detected as the majority of spectrum peaks have been excluded; and an anomaly diagnostic unit 22D, which diagnoses that the equipment 4 for a rotating machine is abnormal in a case where the sum is equal to or greater than a first threshold value.

[0110] Consequently, minor changes in signal intensity resulting from slight torque fluctuations between spectrum peaks are prevented from being missed or masked, thus improving the detection accuracy for these minor torque fluctuations. As a result, anomalies resulting from a malfunction mode involving minor torque fluctuations can be accurately diagnosed. Furthermore, the anomaly diagnostic system 202 can collectively display the diagnostic results for equipment 4-1 to 4-n for a rotating machine, thereby simplifying comparisons between sets of equipment 4-1 to 4-n for a rotating machine, their overall management, and similar tasks. Design 2

[0111] With reference to Fig. 15 to Fig. 18 describes an anomaly diagnostic device 102 according to the present embodiment.

[0112] In embodiment 1, the configuration is described in which the sum of all signal intensities contained in data belonging to the feature frequency band, after excluding spectrum peaks, is compared with the first threshold value, thus performing an anomaly diagnosis for the equipment 4 of a rotating machine. The present embodiment differs from embodiment 1 in that sorted data is generated, in which data belonging to the feature frequency band are sorted in the intensity order of the signal intensities, and data with signal intensities equal to or greater than a predetermined second threshold value are excluded from the feature frequency band. The remaining configurations are the same as in embodiment 1, and parts that are the same as or correspond to those in embodiment 1 are provided with the same reference numerals.

[0113] In contrast to the anomaly diagnostic device 101 in embodiment 1, the monitoring diagnostic unit 2 in the anomaly diagnostic device 102 of the present embodiment has a feature parameter calculation unit 222C instead of the feature parameter calculation unit 221C, as shown in Fig. 15 shown.

[0114] With reference to Fig. Section 16 describes a processing sequence in the anomaly diagnostic device 102 according to the present embodiment, together with a detailed description of each component included in the anomaly diagnostic device 102. The process is the same as in embodiment 1, except for steps S32 and S42.

[0115] In step S32, the feature size calculation unit 222C generates sorted data in which data belonging to the feature frequency band and extracted by the feature frequency band extraction unit 22B are sorted in the intensity order of the signal intensities, and it excludes data of signal intensities equal to or greater than the predetermined second threshold from the feature frequency band.

[0116] As an example of a procedure for determining the second threshold, the lowest signal intensity among the data belonging to the feature frequency band can be determined, and then a value defined as (lowest signal intensity + 10 dB) can be used as the second threshold.

[0117] With reference to Fig. 17 and Fig. Section 18 describes a specific example of step S32.

[0118] Fig. Figure 17 shows a case in which data contained in a frequency range of ±20 Hz from the power supply frequency, i.e., in a frequency range of 40 Hz to 80 Hz, are extracted as data belonging to the feature frequency band in the graph comparing frequency analysis results for current signals, as in Fig. Figure 7 shows the vertical axis representing the current power spectrum, which is a type of signal intensity, and the horizontal axis represents the frequency. In a case where the frequency analysis unit 22A performs a frequency analysis of the current signal with a resolution of 0.25 Hz in step S2, the number of data points belonging to the feature frequency band is 160, and these 160 data points are stored in signal intensity order.

[0119] For example, in a case where the aforementioned second threshold determination method is used, the following applies: If the lowest signal intensity is less than -60 dB in Fig. The second threshold, determined to be -50 dB, is 17.

[0120] Fig. Figure 18 shows sorted data, in which 160 data points are in the feature frequency band that is in Fig. Figure 17 shows the signal intensities sorted in intensity order from left to right in the graph, i.e., 0, to 160. The vertical axis represents the current power spectrum, which is one type of signal intensity, and the horizontal axis indicates the data order. Fig. Figure 18 shows the following with a solid line: Data order ranks are assigned to signal intensities in intensity order. Furthermore, signal intensities (current power spectrum) greater than -50 dB, which is the predetermined second threshold, are excluded from the sorted data.

[0121] Then, in step S42, it is calculated in Fig. 16. The feature size calculation unit 222C derives a feature size from the data belonging to the feature frequency band from which data with signal intensities greater than the second threshold have been excluded. In the present embodiment, the feature size corresponds to the sum of all signal intensities contained in data belonging to the feature frequency band from which data with signal intensities equal to or greater than the second threshold have been excluded.

[0122] As described above, the anomaly diagnostic device 102 of the present embodiment generates sorted data, wherein data belonging to the feature frequency band are sorted in the intensity order of the signal intensities, and it excludes data with signal intensities equal to or greater than the predetermined second threshold from the data belonging to the feature frequency band. The sum of all signal intensities contained in the data belonging to the feature frequency band from which the data with signal intensities equal to or greater than the second threshold have been excluded is then compared with the first threshold, thus performing an anomaly diagnosis for the equipment 4 for a rotating machine. Consequently, a minor change in signal intensities resulting from the slight torque variation between the spectrum peaks is prevented from being overlooked.is hidden, thus improving the detection accuracy for the slight torque fluctuation, and as a result, an anomaly resulting from a malfunction mode that includes a slight torque fluctuation can be accurately diagnosed.

[0123] As described above, the anomaly diagnostic device 102 of the present embodiment is an anomaly diagnostic device 102 for diagnosing an anomaly in the equipment 4 for a rotating machine, and it comprises the following: a current signal storage unit 21A, which stores a current signal of an electric motor 5; a frequency analysis unit 22A, which performs a frequency analysis of a waveform of the current signal stored in the current signal storage unit 21A; a feature frequency band extraction unit 22B, which extracts data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result, which is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit 22A;a feature quantity calculation unit 222C, which sorts the data belonging to the feature frequency band into an intensity order of signal intensities, excludes data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band, and calculates a sum of the signal intensities contained in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; and an anomaly diagnostic unit 22D, which diagnoses that the equipment 4 for a rotating machine is abnormal in a case where the sum is equal to or greater than a first threshold.

[0124] Consequently, a slight change in signal intensity resulting from the slight torque fluctuation between spectrum peaks is prevented from being missed or hidden, thus improving the detection accuracy for the slight torque fluctuation, and as a result, an anomaly resulting from a malfunction mode that includes a slight torque fluctuation can be accurately diagnosed.

[0125] As described above, the anomaly diagnostic method of the present embodiment is an anomaly diagnostic method for diagnosing an anomaly in the equipment 4 for a rotating machine, and it comprises: a current signal detection step S1 for detecting a current signal of the current flowing through an electric motor 5; a frequency analysis step S2 for performing a frequency analysis of a waveform of the current signal detected in the current signal detection step S1; a feature frequency band extraction step S3 for extracting data belonging to a feature frequency band that includes a plurality of spectrum peaks, within a predetermined frequency range, in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step S2;a data exclusion step S32 for sorting the data belonging to the feature frequency band in an intensity order of signal intensities, and excluding data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band; a feature quantity calculation step S42 for calculating a sum of the signal intensities contained in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; a determination step S4E for determining whether or not the sum is equal to or greater than a first threshold;and an anomaly diagnosis step S5 to diagnose that the equipment 4 is abnormal for a rotating machine in a case where the sum is equal to or greater than the first threshold. Consequently, a minor change in signal intensities resulting from the minor torque variation between the spectrum peaks is prevented from being missed or hidden, thus improving the detection accuracy for the minor torque variation, and as a result, an anomaly resulting from a malfunction mode that includes a minor torque variation can be accurately diagnosed.

[0126] As described above, the program of the present embodiment is a program for diagnosing an anomaly in the equipment 4 for a rotating machine, and it causes a computer to perform the following: a current signal detection step S1 to detect a current signal of the current flowing through an electric motor 5; a frequency analysis step S2 to perform a frequency analysis of a waveform of the current signal detected in the current signal detection step S1; a feature frequency band extraction step S3 to extract data belonging to a feature frequency band that includes a plurality of spectrum peaks, in a predetermined frequency range, in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step S2;a data exclusion step S32 for sorting the data belonging to the feature frequency band in an intensity order of signal intensities, and excluding data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band; a feature quantity calculation step S42 for calculating a sum of signal intensities contained in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; a determination step S4E for determining whether or not the sum is equal to or greater than a first threshold; and an anomaly diagnosis step S5 for diagnosing that the equipment 4 for a rotating machine is abnormal in a case where the sum is equal to or greater than the first threshold.

[0127] Consequently, a slight change in signal intensity resulting from the slight torque fluctuation between spectrum peaks is prevented from being missed or hidden, thus improving the detection accuracy for the slight torque fluctuation, and as a result, an anomaly resulting from a malfunction mode that includes a slight torque fluctuation can be accurately diagnosed. Modification 1

[0128] With reference to Fig. 19 describes an anomaly diagnostic system 203 according to modification 1 of embodiment 2.

[0129] In modification 1 of embodiment 1, the configuration is described in which the server 30, which has the diagnostic result output unit 3 and the monitoring diagnostic unit 2 including the feature size calculation unit 221C, and the current detection units 1-1 to 1-n, which are connected to the equipment 4-1 to 4-n for a rotating machine, which has the electric motors 5-1 to 5-n and the load equipment 6-1 to 6-n, are present, with reference to Fig. 13. The present modification differs from modification 1 in embodiment 1 in that the monitoring diagnostic unit 2 is replaced by the monitoring diagnostic unit 2 according to embodiment 2. The remaining configurations are the same as in modification 1 of embodiment 1, and parts that are the same as or correspond to those in modification 1 of embodiment 1 are provided with the same reference numerals.

[0130] Consequently, the anomaly diagnostic system 203, which is in Fig. As shown in the modification above, the effect is that it becomes unnecessary to provide the anomaly diagnostic device 102 for each set of equipment 4-1 to 4-n for a rotating machine.

[0131] As described above, the anomaly diagnostic system 203 of the present embodiment is an anomaly diagnostic system 203 for diagnosing an anomaly in the equipment 4 for a rotating machine and comprises the following: a current signal storage unit 21A that stores a current signal of an electric motor 5; a frequency analysis unit 22A that performs a frequency analysis of a waveform of the current signal stored in the current signal storage unit 21A; a feature frequency band extraction unit 22B that extracts data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range into a frequency analysis result, which is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit 22A;a feature quantity calculation unit 222C, which sorts the data belonging to the feature frequency band into an intensity order of signal intensities, excludes data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band, and calculates a sum of the signal intensities contained in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; and an anomaly diagnostic unit 22D, which diagnoses that the equipment 4 for a rotating machine is abnormal in a case where the sum is equal to or greater than a first threshold.

[0132] Consequently, minor changes in signal intensity resulting from slight torque fluctuations between spectrum peaks are prevented from being missed or masked, thus improving the detection accuracy for these minor torque fluctuations. As a result, anomalies resulting from a malfunction mode involving minor torque fluctuations can be accurately diagnosed. Furthermore, it becomes unnecessary to provide the anomaly diagnostic device 102 for each set of equipment 4-1 to 4-n for a rotating machine. Additionally, the anomaly diagnostic system 202 can collectively display the diagnostic results for all equipment 4-1 to 4-n for a rotating machine, thus simplifying comparisons between sets of equipment 4-1 to 4-n for a rotating machine, their overall management, and similar tasks. Modification 2

[0133] With reference to Fig. 20 describes an anomaly diagnostic system 204 according to modification 2 of the present embodiment.

[0134] In modification 2 of embodiment 1, the configuration is described in which the anomaly diagnostic devices 202-1 to 202-n, including the monitoring diagnostic units 2-1 to 2-n and the current detection units 1-1 to 1-n, which are connected to the equipment 4-1 to 4-n for a rotating machine, including the electric motors 5-1 to 5-n and the load equipment 6-1 to 6-n, and the server 40, including the data acquisition unit 42 and the diagnostic result output unit 3, are present, with reference to Fig. 14. The present modification differs from modification 2 in embodiment 1 in that the anomaly diagnostic device is replaced by anomaly diagnostic devices 204-1 to 204-n, including the current detection units 1-1 to 1-n and the monitoring diagnostic units 2-1 to 2-n, according to embodiment 2. The remaining configurations are the same as in modification 2 of embodiment 1, and parts that are the same as or correspond to those in modification 2 of embodiment 1 are provided with the same reference numerals.

[0135] As described above, the anomaly diagnostic system 204, which is in Fig. Figure 20 shows that in the present modification the diagnostic results for equipment 4-1 to 4-n for a rotating machine are collectively displayed, so that the comparison between sets of equipment 4-1 to 4-n for a rotating machine, their overall management and the like are simplified.

[0136] As described above, the anomaly diagnostic system 204 of the present embodiment is an anomaly diagnostic system 204 for diagnosing an anomaly in the equipment 4 for a rotating machine and comprises the following: a current signal storage unit 21A that stores a current signal of an electric motor 5; a frequency analysis unit 22A that performs a frequency analysis of a waveform of the current signal stored in the current signal storage unit 21A; a feature frequency band extraction unit 22B that extracts data belonging to a feature frequency band that includes a plurality of spectrum peaks in a predetermined frequency range into a frequency analysis result, which is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit 22A;a feature quantity calculation unit 222C, which sorts the data belonging to the feature frequency band into an intensity order of signal intensities, excludes data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band, and calculates a sum of the signal intensities contained in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; and an anomaly diagnostic unit 22D, which diagnoses that the equipment 4 for a rotating machine is abnormal in a case where the sum is equal to or greater than a first threshold.

[0137] Consequently, minor changes in signal intensity resulting from slight torque fluctuations between spectrum peaks are prevented from being missed or masked, thus improving the detection accuracy for these minor torque fluctuations. As a result, an anomaly caused by a malfunction mode involving minor torque fluctuations can be accurately diagnosed. Furthermore, the 204 anomaly diagnostic system can collectively display the diagnostic results for equipment 4-1 to 4-n for a rotating machine, thereby simplifying comparisons between sets of equipment 4-1 to 4-n for a rotating machine, their overall management, and similar tasks. embodiment 3

[0138] With reference to Fig. 21 and Fig. 22 describes an anomaly diagnostic device 103 according to the present embodiment.

[0139] In embodiment 1 or 2, the configuration is described in which an anomaly resulting from a malfunction mode that includes or entails a minor torque fluctuation is accurately diagnosed. The present embodiment differs from embodiment 1 or 2 in that not only is an anomaly resulting from a malfunction mode that includes a minor torque fluctuation accurately diagnosed, but the total number of operating times during which an anomaly exists in the equipment 4 for a rotating machine is also calculated. The remaining configurations are the same as in embodiment 1 or 2. As an example, the case is described in which the feature parameter calculation unit 222C is present, as in embodiment 2. Parts that are the same as or correspond to those in embodiment 1 or 2 are designated with the same reference numerals.

[0140] In the anomaly diagnostic device 103 of the present embodiment, the following applies, as described in Fig. Figure 21 shows that the monitoring diagnostic unit 2 further comprises an anomaly index storage unit 213D and an anomaly index calculation unit 223E, in contrast to the anomaly diagnostic device 101 in embodiment 1 or the anomaly diagnostic device 102 in embodiment 2.

[0141] With reference to Fig. Reference 22 describes a processing sequence in the anomaly diagnostic device 103 according to the present embodiment, together with a detailed description of each component included in the anomaly diagnostic device 103. The process is the same as in embodiment 1 or 2, except for step S5A, step S5B and step S63, and steps that are the same as or correspond to those in embodiment 1 or 2 are designated with the same reference numerals.

[0142] In step S5A, the anomaly detection storage unit 21C accumulates diagnostic results, which are output from the anomaly diagnostic unit 22D.

[0143] Then, in step S5B, using the anomaly diagnostic results recorded by the anomaly determination storage unit 21C, the following is performed: The anomaly index calculation unit 223E calculates - as a cumulative anomaly time - the total number of operating times during which there is an anomaly in the equipment 4 for a rotating machine, and it outputs the cumulative anomaly time to the anomaly index storage unit 213D.

[0144] Then, in step S63, the diagnostic result output unit 3 records the cumulative anomaly time from the anomaly index storage unit 213D. The diagnostic result output unit 3 displays the cumulative anomaly time on the display unit 31A, such as a display, in addition to the diagnostic result recorded by the anomaly determination storage unit 21C in embodiment 1 or 2, and the external output communication unit 31C outputs the cumulative anomaly time to an external device, such as a control panel, a PC, or a cloud server, for example, to notify a monitor regarding the cumulative anomaly time, so that a trend in the cumulative anomaly time is monitored.

[0145] As described above, the anomaly diagnostic device 103 of the present embodiment produces the effects of embodiment 1 or 2, and furthermore, it calculates the total number of operating times during which an anomaly exists in the equipment 4 for a rotating machine, and it notifies a monitor regarding the cumulative anomaly time during which an anomaly exists in the equipment 4 for a rotating machine, together with the anomaly diagnostic result, so that a trend regarding the cumulative anomaly time can be monitored. In addition, by comparing the cumulative anomaly times of a plurality of sets of equipment 4 for a rotating machine, it is possible to use the cumulative anomaly times as useful indices for determining the maintenance and update times of the equipment 4 for a rotating machine.

[0146] In the present embodiment, an example is shown in which the anomaly index calculation unit 223E calculates the cumulative anomaly time. However, an external device, different from the anomaly diagnostic device 103, can also calculate the cumulative anomaly time. Design 4

[0147] With reference to Fig. 23 and Fig. 24 describes an anomaly diagnostic device 104 according to the present embodiment.

[0148] In embodiment 3, the configuration is described in which an anomaly resulting from a malfunction mode involving a minor torque fluctuation is accurately diagnosed, and the total number of operating times during which an anomaly exists in the equipment 4 for a rotating machine is calculated as the cumulative anomaly time. The present embodiment differs from embodiment 3 in that an operating condition of the equipment 4 for a rotating machine that causes an anomaly is estimated. The remaining configurations are the same as in embodiment 3, and parts that are the same as or correspond to those in embodiment 3 are designated with the same reference numerals.

[0149] In the anomaly diagnostic device 104 of the present embodiment - as in Fig. Figure 23 shows that the monitoring diagnostic unit 2 has an anomaly index storage unit 214D and an anomaly index calculation unit 224E instead of the anomaly index storage unit 213D and the anomaly index calculation unit 223E, and it also has an operating condition storage unit 21E, in contrast to the anomaly diagnostic device 103 in embodiment 3.

[0150] The operating condition storage unit 21E stores time-to-time operating conditions of equipment 4 for a rotating machine. Examples of operating conditions include: the volumetric flow rate of water coming from a water pump, which is the load equipment 6; the pressure of water coming from the water pump; and the opening degree of a valve on an inlet / outlet pipe through which water flows into or out of the water pump, in a case of application to a monitoring and control system for a public facility, such as a water treatment plant.

[0151] With reference to Fig. Reference 24 describes a processing sequence in the anomaly diagnostic device 104 according to the present embodiment, together with a detailed description of each component included in the anomaly diagnostic device 104. The process is the same as in embodiment 3, except for steps S5C, S5D, S5E and S64, and steps that are the same as or correspond to those in embodiment 3 are designated with the same reference numerals. Fig. 24 corresponds to a process up to step S7 in the case where the determination is made as "YES" in step S4E in the processing flow, which is in Fig. Figure 22 is shown, and only this area is shown for the sake of simplicity in the description.

[0152] In step S5C, using the cumulative anomaly time, the anomaly index calculation unit 224E calculates a cumulative anomaly time per operating time unit of equipment 4 for a rotating machine and a cumulative anomaly time per total operating time, which is a third threshold.

[0153] Here, the cumulative anomaly time per operating time unit is the proportion of the time during which there is an anomaly in equipment 4 for a rotating machine per operating unit, and the cumulative anomaly time per total operating time is the proportion of the time during which there is an anomaly in equipment 4 for a rotating machine in the total operating time.

[0154] The cumulative anomaly time per operating time unit is calculated by dividing the total number of operating times during which there is an anomaly in equipment 4 for a rotating machine per operating time unit by the operating time unit.

[0155] The cumulative anomaly time per total operating time is calculated by dividing the total number of operating times during which there is an anomaly in the equipment 4 for a rotating machine, from the start of operation of the equipment 4 for a rotating machine until the present time, by the operating time from the start of operation of the equipment 4 for a rotating machine until the present time.

[0156] Then, in step S5D, the anomaly index calculation unit 224E compares the cumulative anomaly time per operating time unit with the cumulative anomaly time per total operating time. If the cumulative anomaly time per operating time unit is equal to or greater than the cumulative anomaly time per total operating time, the anomaly index calculation unit 224E retrieves the operating condition of equipment 4 for a rotating machine in use at that time from the operating condition storage unit 21E.

[0157] Then, in step S5E, the following occurs: If an anomaly exists in equipment 4 for a rotating machine, the anomaly index calculation unit 224E estimates the operating condition actually prevailing in equipment 4 for a rotating machine at that time as the cause of the anomaly. That is, using the operating condition captured by the operating condition storage unit 21E, the anomaly index calculation unit 224E estimates the operating condition of equipment 4 for a rotating machine that causes the anomaly as an anomaly-inducing operating condition. Furthermore, the anomaly index calculation unit 224E outputs the anomaly-inducing operating condition to the anomaly index storage unit 214D.

[0158] Then, in step S64, the diagnostic result output unit 3 acquires the anomaly-causing operating condition from the anomaly index storage unit 214D. The diagnostic result output unit 3 displays the anomaly-causing operating condition on the display unit 31A, such as a display, in addition to the diagnostic result acquired by the anomaly determination storage unit 21C in embodiment 1 or 2. The external output communication unit 31C outputs the anomaly-causing operating condition to an external device, such as a control panel, a PC, or a cloud server, for example, to notify a monitor regarding the anomaly-causing operating condition, so that a trend in the anomaly-causing operating condition is monitored.

[0159] As described above, the anomaly diagnostic device 104 of the present embodiment displays the effects of embodiment 1 or 2, and furthermore, it estimates the operating condition of the equipment 4 for a rotating machine that causes an anomaly and notifies a monitor regarding the operating condition of the equipment 4 for a rotating machine that causes the anomaly, together with the anomaly diagnostic result, so that a trend in the anomaly-causing operating condition can be monitored. In addition, the estimated anomaly-causing operating condition can be used as a useful index for determining the operating condition of the equipment 4 for a rotating machine.

[0160] In the present embodiment, an example is shown in which the anomaly index calculation unit 224E calculates the anomaly-causing operating condition. However, an external device, different from the anomaly diagnostic device 104, can also calculate the anomaly-causing operating condition. Design 5

[0161] With reference to Fig. 25 and Fig. 26 describes an anomaly diagnostic device 105 according to the present embodiment.

[0162] In embodiment 4, the configuration is described in which the operating condition of the equipment 4 for a rotating machine that causes an anomaly is estimated. The present embodiment differs from embodiment 4 in that a deterioration progression rate of the equipment 4 for a rotating machine is calculated. The remaining configurations are the same as in embodiment 4, and parts that are the same as or correspond to those in embodiment 4 are provided with the same reference numerals.

[0163] In the anomaly diagnostic device 105 of the present embodiment - as in Fig. Figure 25 shows that the monitoring diagnostic unit 2 has an anomaly index storage unit 215D and an anomaly index calculation unit 225E instead of the anomaly index storage unit 214D and the anomaly index calculation unit 224E, in contrast to the anomaly diagnostic device 104 in embodiment 4.

[0164] With reference to Fig. Reference 26 describes a processing sequence in the anomaly diagnostic device 105 according to the present embodiment, together with a detailed description of each component included in the anomaly diagnostic device 105. The process is the same as in embodiment 4, except for steps S5F and S65, and steps that are the same as or correspond to those in embodiment 4 are designated with the same reference numerals.

[0165] Fig. 26 corresponds to a process up to step S7 in the case where the determination is made as "YES" in step S4E in the processing flow, which is in Fig. Figure 22 is shown, and only this area is shown for the sake of simplicity in the description.

[0166] In step S5F, the anomaly index calculation unit 225E calculates the deterioration progression rate of equipment 4 for a rotating machine as the product of the feature size calculated in step S41 or step S42 and the cumulative anomaly occurrence time calculated in step S5B, the product of the feature size calculated in step S41 or step S42 and the cumulative anomaly time per unit of time in equipment 4 for a rotating machine calculated in step S5C, or the product of the feature size calculated in step S41 or step S42 and the cumulative anomaly time per total operating time in equipment 4 for a rotating machine calculated in step S5C.

[0167] Here, the deterioration progress rate is an index that indicates the progress rate of deterioration in equipment 4 for a rotating machine, caused by the anomaly in equipment 4 for a rotating machine.

[0168] In step S65, the diagnostic result output unit 3 then acquires the deterioration progress rate from the anomaly index storage unit 215D. The diagnostic result output unit 3 displays the deterioration progress rate on the display unit 31A, such as a display, in addition to the diagnostic result acquired by the anomaly determination storage unit 21C in embodiment 1 or 2, and the external output communication unit 31C transmits the deterioration progress rate to an external device, such as a control panel, a PC, or a cloud server, for example, to notify a monitor regarding the deterioration progress rate, thus monitoring a trend in the deterioration progress rate.

[0169] As described above, the anomaly diagnostic device 105 of the present embodiment produces the effects according to embodiment 1 or 2, and furthermore, it calculates the degree of deterioration and notifies a monitor regarding the degree of deterioration, together with the anomaly diagnostic result, so that a trend in the degree of deterioration can be monitored. In addition, by comparing the degrees of deterioration of a plurality of sets of equipment 4 for a rotating machine, it is possible to use the degrees of deterioration as useful indices for determining the maintenance and update times of the equipment 4 for a rotating machine.

[0170] In the present embodiment, an example is shown in which the anomaly index calculation unit 225E calculates the deterioration progression rate. However, an external device, different from the anomaly diagnostic device 105, can also calculate the deterioration progression rate. Design 6

[0171] In embodiment 1 or 2, the configuration is described in which an anomaly resulting from a malfunction mode that includes or involves a minor torque fluctuation is accurately diagnosed. In embodiment 3, the configuration is described in which the total number of operating times during which an anomaly exists in the equipment 4 for a rotating machine is calculated as the cumulative anomaly time. In embodiment 4, the configuration is described in which the operating condition of the equipment 4 for a rotating machine that causes an anomaly is estimated. In embodiment 5, the configuration is described in which the deterioration rate of the equipment 4 for a rotating machine is calculated.

[0172] The present embodiment differs from embodiments 1 to 5 in that control is carried out to establish such an operating condition as to inhibit deterioration progress, based on the anomaly detection or cumulative anomaly time, the operating condition of the equipment 4 for a rotating machine that causes the anomaly, or the degree of deterioration progress of the equipment 4 for a rotating machine.

[0173] In one case, for example, where the equipment 4 for a rotating machine is a pump, cavitation occurs as a malfunction mode involving a slight torque fluctuation, as described above. Cavitation occurs as follows: When the flow velocity of a liquid increases locally within the pump, the pressure drops. Then, when the pressure falls below the liquid's saturation vapor pressure, the liquid vaporizes, leading to cavitation. Furthermore, the sudden volume change as the vapor condenses back into liquid causes a shock, which damages the pump. To suppress such cavitation, the rotational speed of the equipment 4 for the rotating machine is reduced below that of the operating condition that induces cavitation, thus decreasing the liquid flow velocity.

[0174] Cavitation suppression is just one example. In embodiments 1 or 2, the following applies: When an anomaly is detected, the rotational speed of equipment 4 for a rotating machine is subjected to feedback control so that the occurrence of the anomaly can be suppressed. Control for suppressing the occurrence of an anomaly based on the cumulative anomaly time, calculated in embodiment 3, can be fed back to equipment 4 for a rotating machine. Control for suppressing the occurrence of an anomaly based on the deterioration progression rate, calculated in embodiment 5, can also be fed back to equipment 4 for a rotating machine.

[0175] A control system for carrying out the operation, such that the operating condition of the equipment 4 for a rotating machine is avoided, which causes an anomaly estimated in embodiment 4, can be implemented. In this case, the operating condition storage unit for storing the operating condition of the equipment for a rotating machine must have an operating condition of the equipment for a rotating machine stored therein before the anomaly is diagnosed.

[0176] Although the invention is described above in the form of various exemplary embodiments and implementations, it is understood that the various features, aspects and functionalities described in one or more of the individual embodiments are not limited in their applicability to the particular embodiment in which they are described, but can instead also be applied - alone or in various combinations - with one or more of the embodiments of the invention.

[0177] It is therefore understood that numerous modifications can be used that have not been described by way of example, without deviating from the scope of the present invention. For example, at least one of the components can be modified, added, or omitted. At least one of the components described in at least one of the preferred embodiments can be selected and combined with the components specified in another preferred embodiment. Reference symbol list 1 current detection unit 2 Monitoring and diagnostic units 3 Diagnostic result output unit 4 Equipment for a rotating machine 5 electric motor 6 Load Equipment 7 inverters 8 commercial energy supply 101, 102, 103, 104, 105 Anomaly Diagnostic Unit QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] JP 2020- 153 965 A

[0006]

Claims

[1] An anomaly diagnosis device for diagnosing an anomaly in equipment for a rotating machine, the anomaly diagnosis device comprising: a current signal storage unit that stores a current signal of an electric motor; a frequency analysis unit that performs frequency analysis of a waveform of the current signal stored in the current signal storage unit; a feature frequency band extraction unit that extracts data corresponding to a feature frequency band including a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit; a feature quantity calculation unit that extracts the plurality of spectrum peaks from the data belonging to the feature frequency band, excludes data detected as the plurality of spectrum peaks from the data belonging to the feature frequency band, and calculates a sum of signal intensities included in the data belonging to the feature frequency band from which the data detected as the plurality of spectrum peaks have been excluded; and an abnormality diagnosis unit that diagnoses that the rotating machine equipment is abnormal in a case where the sum is equal to or greater than a first threshold. [2] An anomaly diagnosis device for diagnosing an anomaly in equipment for a rotating machine, the anomaly diagnosis device comprising: a current signal storage unit that stores a current signal of an electric motor; a frequency analysis unit that performs frequency analysis of a waveform of the current signal stored in the current signal storage unit; a feature frequency band extraction unit that extracts data corresponding to a feature frequency band including a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit; a feature size calculation unit that sorts the data belonging to the feature frequency band in an intensity order of signal intensities, excludes data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band, and calculates a sum of the signal intensities included in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; and an abnormality diagnosis unit that diagnoses that the rotating machine equipment is abnormal in a case where the sum is equal to or greater than a first threshold. [3] An abnormality diagnosis device according to claim 1 or 2, wherein the signal intensities are current values ​​or a current power spectrum of the current signal. [4] An abnormality diagnosis device according to claim 3, wherein the predetermined frequency range is a frequency range from a zero-order component to a second-order component with respect to the power supply frequency. [5] An abnormality diagnosis device according to claim 3, wherein the predetermined frequency range is a frequency range from a fourth-order component to a sixth-order component with respect to the power supply frequency. [6] An abnormality diagnosis device according to claim 3, wherein the predetermined frequency range is a frequency range from a sixth-order component to an eighth-order component with respect to the power supply frequency. [7] An abnormality diagnosis device according to claim 3, wherein the predetermined frequency range is a frequency range of 0 Hz to 120 Hz. [8] An abnormality diagnosis device according to claim 3, wherein the predetermined frequency range is a frequency range of 200 Hz to 360 Hz. [9] An abnormality diagnosis device according to claim 3, wherein the predetermined frequency range is a frequency range of 300 Hz to 480 Hz. [10] The abnormality diagnosis device according to claim 3, further comprising a current detection unit connected to a wire connecting the electric motor and a commercial power supply for supplying power to the electric motor, detecting the current for driving the electric motor, and outputting a current signal of the detected current to the current signal storage unit. [11] An abnormality diagnosis device according to claim 3, wherein the first threshold value is a value obtained by performing statistical processing on the sums accumulated during a predetermined period from the start of operation of the rotating machine equipment. [12] The abnormality diagnosis device according to claim 11, wherein in a case where the number of data of the sum and the number of data of the first threshold are different from each other, the feature quantity calculation unit sequentially excludes data having higher signal intensities from the data of the first threshold or sequentially excludes data having higher signal intensities from the data of the sum so that the number of data of the sum and the number of data of the first threshold coincide with each other. [13] Anomaly diagnostic device according to one of claims 1 to 12, further comprising an operating condition storage unit that stores an operating condition of the rotating machine equipment, wherein the rotating machine equipment is operated so as to avoid the operating condition of the rotating machine equipment which is manifested when the rotating machine equipment is diagnosed as abnormal. [14] The abnormality diagnosis device according to claim 3, further comprising an abnormality determination storage unit that stores a diagnosis result from the abnormality diagnosis unit when the abnormality diagnosis unit has diagnosed that the rotating machine equipment is abnormal. [15] The abnormality diagnosis device according to claim 14, further comprising a diagnosis result output unit including at least one of a display unit that displays the diagnosis result, an alarm unit that issues an alarm when the rotating machine equipment is diagnosed as abnormal, and an external output communication unit that transmits the diagnosis result to an external device. [16] The abnormality diagnosis device according to claim 14, further comprising an abnormality index calculation unit that calculates a cumulative abnormality time, which is a total number of operation times during which there is an abnormality in the rotating machine equipment, using the diagnosis result stored in the abnormality determination storage unit. [17] Anomaly diagnostic device according to claim 16, further comprising an operating condition storage unit that stores an operating condition of the rotating machine equipment, wherein the anomaly index calculation unit calculates the cumulative anomaly time per unit operation time of the rotating machine equipment, and in a case where the cumulative anomaly time per unit operation time of the rotating machine equipment is equal to or greater than a third threshold, the anomaly index calculation unit estimates the operation condition of the rotating machine equipment prevailing at that time as an anomaly-causing operation condition that causes the anomaly in the rotating machine equipment. [18] An abnormality diagnosis device according to claim 17, wherein the rotating machine equipment is operated so as to avoid the abnormality-causing operating condition. [19] The abnormality diagnosis device according to claim 16, wherein the abnormality index calculation unit calculates the cumulative abnormality time per total operating time of the rotating machine equipment and calculates a deterioration progress degree which is the product of the sum and the cumulative abnormality time, the product of the sum and the cumulative abnormality time per unit operating time of the rotating machine equipment, or the product of the sum and the cumulative abnormality time per total operating time of the rotating machine equipment. [20] Anomaly diagnostic device according to claim 19, further comprising an operating condition storage unit that stores an operating condition of the rotating machine equipment, wherein the rotating machine equipment is operated so as to avoid the operating condition which is manifested when the rotating machine equipment is diagnosed as abnormal, according to the deterioration progress degree. [21] An anomaly diagnosis system for diagnosing an anomaly in equipment for a rotating machine, the anomaly diagnosis system comprising: a current signal storage unit that stores a current signal of an electric motor; a frequency analysis unit that performs frequency analysis of a waveform of the current signal stored in the current signal storage unit; a feature frequency band extraction unit that extracts data corresponding to a feature frequency band including a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit; a feature quantity calculation unit that extracts the plurality of spectrum peaks from the data belonging to the feature frequency band, excludes data detected as the plurality of spectrum peaks from the data belonging to the feature frequency band, and calculates a sum of signal intensities included in the data belonging to the feature frequency band from which the data detected as the plurality of spectrum peaks have been excluded; and an abnormality diagnosis unit that diagnoses that the rotating machine equipment is abnormal in a case where the sum is equal to or greater than a first threshold. [22] An anomaly diagnosis system for diagnosing an anomaly in equipment for a rotating machine, the anomaly diagnosis system comprising: a current signal storage unit that stores a current signal of an electric motor; a frequency analysis unit that performs frequency analysis of a waveform of the current signal stored in the current signal storage unit; a feature frequency band extraction unit that extracts data corresponding to a feature frequency band including a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed by the frequency analysis unit; a feature size calculation unit that sorts the data belonging to the feature frequency band in an intensity order of signal intensities, excludes data of signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band, and calculates a sum of the signal intensities included in the data belonging to the feature frequency band from which the data of signal intensities equal to or greater than the second threshold have been excluded; and an abnormality diagnosis unit that diagnoses that the rotating machine equipment is abnormal in a case where the sum is equal to or greater than a first threshold. [23] Anomaly diagnostic system according to claim 21 or 22, further comprising an operating condition storage unit that stores an operating condition of the rotating machine equipment, wherein the rotating machine equipment is operated so as to avoid the operating condition of the rotating machine equipment which is manifested when the rotating machine equipment is diagnosed as abnormal. [24] An anomaly diagnosis method for diagnosing an anomaly in equipment for a rotating machine, the anomaly diagnosis method comprising: a current signal detecting step for detecting a current signal of the current flowing through an electric motor; a frequency analysis step of performing frequency analysis of a waveform of the current signal detected in the current signal detection step; a feature frequency band extraction step of extracting data corresponding to a feature frequency band including a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step; a data exclusion step of detecting the plurality of spectrum peaks from the data belonging to the feature frequency band and excluding data detected as the plurality of spectrum peaks from the data belonging to the feature frequency band; a feature quantity calculating step of calculating a sum of signal intensities included in the data belonging to the feature frequency band from which the data detected as the plurality of spectrum peaks have been excluded; a determining step of determining whether or not the sum is equal to or greater than a first threshold; and an abnormality diagnosis step for diagnosing that the rotating machine equipment is abnormal in a case where the sum is equal to or greater than the first threshold. [25] An anomaly diagnosis method for diagnosing an anomaly in equipment for a rotating machine, the anomaly diagnosis method comprising: a current signal detecting step for detecting a current signal of the current flowing through an electric motor; a frequency analysis step of performing frequency analysis of a waveform of the current signal detected in the current signal detection step; a feature frequency band extraction step of extracting data corresponding to a feature frequency band including a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step; a data exclusion step of sorting the data belonging to the feature frequency band in an intensity order of signal intensities, and excluding data of the signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band; a feature size calculation step of calculating a sum of the signal intensities included in the data belonging to the feature frequency band from which the data of the signal intensities equal to or greater than the second threshold have been excluded; a determining step of determining whether or not the sum is equal to or greater than a first threshold; and an abnormality diagnosis step for diagnosing that the rotating machine equipment is abnormal in a case where the sum is equal to or greater than the first threshold. [26] Anomaly diagnostic method according to claim 24 or 25, further comprising an operating condition storing step in which an operating condition of the equipment for a rotating machine is stored, wherein the rotating machine equipment is operated so as to avoid the operating condition of the rotating machine equipment which is manifested when the rotating machine equipment is diagnosed as abnormal. [27] A program for diagnosing an anomaly in equipment for a rotating machine, the program causing a computer to perform the following: a current signal detecting step for detecting a current signal of the current flowing through an electric motor; a frequency analysis step of performing frequency analysis of a waveform of the current signal detected in the current signal detection step; a feature frequency band extraction step of extracting data corresponding to a feature frequency band including a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step; a data exclusion step of detecting the plurality of spectrum peaks from the data belonging to the feature frequency band and excluding data detected as the plurality of spectrum peaks from the data belonging to the feature frequency band; a feature quantity calculating step of calculating a sum of signal intensities included in the data belonging to the feature frequency band from which the data detected as the plurality of spectrum peaks have been excluded; a determining step of determining whether or not the sum is equal to or greater than a first threshold; and an abnormality diagnosis step for diagnosing that the rotating machine equipment is abnormal in a case where the sum is equal to or greater than the first threshold. [28] A program for diagnosing an anomaly in equipment for a rotating machine, the program causing a computer to perform the following: a current signal detecting step for detecting a current signal of the current flowing through an electric motor; a frequency analysis step of performing frequency analysis of a waveform of the current signal detected in the current signal detection step; a feature frequency band extraction step of extracting data corresponding to a feature frequency band including a plurality of spectrum peaks in a predetermined frequency range in a frequency analysis result that is a result of the frequency analysis of the waveform of the current signal performed in the frequency analysis step; a data exclusion step of sorting the data belonging to the feature frequency band in an intensity order of signal intensities, and excluding data of the signal intensities equal to or greater than a second threshold from the data belonging to the feature frequency band; a feature size calculation step of calculating a sum of the signal intensities included in the data belonging to the feature frequency band from which the data of the signal intensities equal to or greater than the second threshold have been excluded; a determining step of determining whether or not the sum is equal to or greater than a first threshold; and an abnormality diagnosis step for diagnosing that the rotating machine equipment is abnormal in a case where the sum is equal to or greater than the first threshold.

Images