State estimation device, drive system, refrigeration system, fan system, state estimation method, and state estimation program

Abstract

In the estimation process, the control unit (31) estimates the state of the device (70) on the basis of the specific frequency component (C1) of the physical quantity obtained from the device (70). In the limiting process, when the noise frequency component (Cn) and the specific frequency component (C1) included in the physical quantity satisfy a predetermined relationship, the control unit (31) limits either the estimation process or the operation of the motor (50). The carrier frequency is set to be fc, the frequency for sampling the physical quantity is set to be fs, the electrical angle frequency is set to be f0, and when the physical quantity is a DC signal, the frequency of the noise frequency component (Cn) is represented by any one of formulae 1 to 3; when the physical quantity is an AC signal, the frequency of the noise frequency component (Cn) is represented by any one of formulae 4-6. [Mathematical formula 1] fc + / -6Nfo... (1) Mfs-6Nfo... (2) Mfs-fc + / -6Nfo... (3) fo + / -fc + / -6Nfo... (4) fo + / -Mfs-6Nfo... (5) fo + / -Mfs-fc + / -6Nfo... (6) wherein M and N are natural numbers.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a state inference technology. BACKGROUND

[0002] Patent Literature 1 discloses an abnormality diagnosing apparatus for diagnosing an abnormality of a motor driven by pulse width modulation control of a power conversion apparatus. The abnormality diagnosing apparatus includes a detection section, an analysis section, a determination section, and a frequency setting section. The detection section detects a current flowing in the motor. The analysis section performs frequency analysis on the current detected by the detection section and outputs an analysis result. The determination section determines whether the motor is abnormal based on a spectral peak of at least one sideband component of a modulation wave obtained from the analysis result. The frequency setting section sets a noise frequency within the current in advance.

[0003] The frequency setting section calculates a greatest common divisor of two or more frequencies including the modulation wave frequency among three frequencies of the pulse width modulation control, i.e., a modulation wave frequency, a carrier wave frequency, and a sampling frequency at which the modulation wave is sampled, and sets the greatest common divisor and integral multiples thereof as the noise frequency. The determination section determines whether the motor is abnormal based on whether the spectral peak of the sideband component has noise interference according to the frequency of the sideband component and the set noise frequency.

[0004] PRIOR ART DOCUMENTS

[0005] PATENT LITERATURE

[0006] Patent Literature 1: Patent No. 6824494 SUMMARY

[0007] PROBLEMS TO BE SOLVED BY THE INVENTION

[0008] In the state inference apparatus (apparatus that infers a state of a device) like the abnormality diagnosing apparatus of Patent Literature 1, only the noise frequency component of the frequency of the above-described greatest common divisor (greatest common divisor of two or more) and integral multiples thereof can be detected, and thus the influence of the noise frequency component of other frequencies than the frequency of the above-described greatest common divisor and integral multiples thereof cannot be taken into account. Therefore, it is difficult to appropriately suppress false inference of the state of the device.

[0009] TECHNICAL SOLUTION FOR SOLVING THE PROBLEMS

[0010] A state inferring apparatus inferring a state of an apparatus 70 in which a motor 50 driven by a direct AC converter 23 is installed, the state inferring apparatus including a control section 31 that performs an inferring process in which the control section 31 infers the state of the apparatus 70 from a magnitude of a specific frequency component C1 of a physical quantity obtained from the apparatus 70, and a limiting process in which the control section 31 limits the inferring process or an operation of the motor 50 when a noise frequency component Cn included in the physical quantity and the specific frequency component C1 satisfy a prescribed relationship, the limiting process being a first limiting process in which the inferring process is limited so as to output a preset inferring result, or a second limiting process in which the operation of the motor 50 is limited so as not to make a false inference in the inferring process, it being assumed that a carrier frequency of the direct AC converter 23 is f c, it being assumed that a reciprocal of a period in which the physical quantity is sampled, that is, a sampling frequency is f s, and it being assumed that an electrical angle frequency of the motor 50 is f 0, when the physical quantity is a direct current signal, a frequency of the noise frequency component Cn is a frequency represented by any one of the following formulas 1, 2, and 3, and when the physical quantity is an alternating current signal, on the condition that two or more frequencies including the electrical angle frequency among the electrical angle frequency, the carrier frequency, and a frequency at which a modulation wave for controlling the direct AC converter 23 is sampled do not have a greatest common divisor of two or more, a frequency of the noise frequency component Cn is a frequency represented by any one of the following formulas 4, 5, and 6. s

[0011] [Math. 1]

[0012] |f c ±6Nf0|…(1)

[0013] |Mf s -6Nf0|…(2)

[0014] |Mf s -|f c ±6Nf0||…(3)

[0015] |f0±|f c ±6Nf0||…(4)

[0016] |f0±|Mf s -6Nf0||…(5)

[0017] |f0±|Mf s -|f c ±6Nf0|||…(6)

[0018] where M and N are natural numbers ​

[0019] In the first aspect, the noise frequency component Cn of the frequency represented by any one of Formula 1 to Formula 6 is a frequency component that can affect the specific frequency component Cl. Therefore, in the restriction processing, by restricting the inference processing or the operation of the motor 50 according to the relationship between the noise frequency component Cn and the specific frequency component Cl, it is possible to suppress erroneous inference of the state of the device 70.

[0020] Note that, in the first aspect, the influence of the noise frequency component of a frequency other than the "maximum common divisor (maximum common divisor of two or more) and an integer multiple thereof disclosed in Patent Document 1" can be taken into account. In this way, it is possible to appropriately suppress erroneous inference of the state of the device 70.

[0021] The second aspect of the present disclosure is based on the state inferring apparatus of the first aspect, and the prescribed relationship is a relationship in which a frequency difference between the noise frequency component Cn and the specific frequency component Cl is below a prescribed frequency difference threshold value.

[0022] In the second aspect, by restricting the inference processing or the operation of the motor 50 according to the frequency difference between the noise frequency component Cn and the specific frequency component Cl, it is possible to suppress erroneous inference of the state of the device 70.

[0023] The third aspect of the present disclosure is based on the state inferring apparatus of the second aspect, and the frequency difference threshold value is 3 Hz.

[0024] In the third aspect, when the "fluctuation amplitude of the specific frequency component Cl when the noise frequency component Cn has an influence" is greater than 10% of the "amplitude of the specific frequency component Cl when the noise frequency component Cn does not have an influence", it is possible to restrict the inference processing or the operation of the motor 50. In this way, it is possible to appropriately suppress erroneous inference of the state of the device 70.

[0025] The fourth aspect of the present disclosure is based on the state inferring apparatus of any one of the first to third aspects, and the first restriction processing is processing that prohibits implementation of the inference processing.

[0026] In the fourth aspect, by prohibiting implementation of the inference processing, it is possible to avoid obtaining an erroneous inference result. In this way, it is possible to suppress erroneous inference of the state of the device 70.

[0027] The fifth aspect of the present disclosure is based on the state inferring apparatus of any one of the first to third aspects, and in the inference processing, the control section 31 infers the state of the device 70 according to a comparison result of a variable calculated based on the amplitude of the specific frequency component Cl and a preset threshold value, and the first restriction processing is processing that corrects at least one of the threshold value and the amplitude of the specific frequency component Cl.

[0028] In the fifth aspect, by correcting at least one of the threshold value and the amplitude of the specific frequency component C1 used in the inference processing, it is possible to avoid obtaining an erroneous inference result. In this way, it is possible to suppress erroneous inference of the state of the device 70.

[0029] The sixth aspect of the present disclosure is based on the state inferring apparatus of any one of the first to third aspects, and the second restriction processing is processing of changing the operating condition of the motor 50 so as not to make an erroneous inference in the inference processing.

[0030] In the sixth aspect, by changing the operating condition of the motor 50, it is possible to make the noise frequency component Cn and the specific frequency component C1 not satisfy the prescribed relationship. In this way, it is possible to suppress fluctuations in the magnitude of the specific frequency component C1 due to the influence of the noise frequency component Cn, and thus it is possible to suppress erroneous inference of the state of the device 70.

[0031] The seventh aspect of the present disclosure is based on the state inferring apparatus of the sixth aspect, and the processing of changing the operating condition of the motor 50 is processing of changing at least one of the carrier frequency, the sampling frequency, and the mechanical angular frequency of the motor 50.

[0032] In the seventh aspect, by changing at least one of the carrier frequency, the sampling frequency, and the mechanical angular frequency of the motor 50, it is possible to make the noise frequency component Cn and the specific frequency component C1 not satisfy the prescribed relationship. In this way, it is possible to suppress fluctuations in the magnitude of the specific frequency component C1 due to the influence of the noise frequency component Cn, and thus it is possible to suppress erroneous inference of the state of the device 70.

[0033] The eighth aspect of the present disclosure is based on the state inferring apparatus of any one of the first to third aspects, and the second restriction processing is processing of prohibiting the motor 50 from operating in an operating condition in which the noise frequency component Cn and the specific frequency component C1 satisfy the prescribed relationship.

[0034] In the eighth aspect, by prohibiting the motor 50 from operating in an operating condition in which the noise frequency component Cn and the specific frequency component C1 satisfy the prescribed relationship, it is possible to avoid obtaining an erroneous inference result. In this way, it is possible to suppress erroneous inference of the state of the device 70.

[0035] The ninth aspect of the present disclosure is based on the state inferring apparatus of any one of the first to eighth aspects, when the physical quantity is a direct current signal, the frequency of the specific frequency component C1 is at least one of 1 times, 1 / 3 times, and 2 / 3 times of the mechanical angular frequency of the motor 50, when the physical quantity is an alternating current signal, the frequency of the specific frequency component C1 is at least one of: a frequency obtained by adding any one of 1 times, 1 / 3 times, and 2 / 3 times of the mechanical angular frequency to the electrical angular frequency, that is, a prescribed frequency; a frequency obtained by subtracting the prescribed frequency from the electrical angular frequency; and 3 times of the electrical angular frequency.

[0036] In the ninth aspect, it is possible to infer various states of the device 70, particularly, abnormalities.

[0037] The tenth aspect of the present disclosure is based on the state inferring apparatus of any one of the first to ninth aspects, in the restriction processing, the control section 31, after starting the restriction of the inferring processing or the operation of the motor 50, if a prescribed restriction release condition is satisfied, releases the restriction after a prescribed standby release condition is satisfied.

[0038] In the tenth aspect, compared to a case where the restriction (restriction of the inferring processing or restriction of the motor 50) is released after the restriction release condition is satisfied without waiting for the standby release condition to be satisfied, it is possible to release the restriction after the specific frequency component C1 for the inferring processing is stabilized. In this way, it is possible to suppress false inferring of the state of the device 70 after the restriction is released.

[0039] The eleventh aspect of the present disclosure is based on the state inferring apparatus of any one of the first to tenth aspects, in the inferring processing, the control section 31 infers whether the device 70 has an abnormality, and the preset inferring result is not an inferring result indicating that the device 70 has an abnormality.

[0040] In the eleventh aspect, it is possible to avoid false inferring that the device 70 has an abnormality.

[0041] The twelfth aspect of the present disclosure relates to a state inferring apparatus that infers a state of an apparatus 70 in which a motor 50 is installed, the state inferring apparatus including a control section 31 that performs an inferring process in which the control section 31 infers a state of the apparatus 70 from a magnitude of a specific frequency component Cl of a physical quantity obtained from the apparatus 70, and a limiting process in which the control section 31 limits the inferring process or operation of the motor 50 based on a magnitude of a periodic component included in a variable calculated from the magnitude of the specific frequency component Cl, the limiting process being a first limiting process in which the inferring process is limited so as to output a predetermined inference result, or a second limiting process in which the operation of the motor 50 is limited so as not to make a false inference in the inferring process.

[0042] In the twelfth aspect, when a noise component affects the specific frequency component Cl, the magnitude (amplitude) of the specific frequency component Cl fluctuates. Therefore, by limiting the inferring process or operation of the motor 50 based on the magnitude of the periodic component included in the variable calculated from the magnitude of the specific frequency component Cl, it is possible to suppress a false inference of the state of the apparatus 70.

[0043] Note that, in the twelfth aspect, it is possible to take into account the influence of a noise frequency component of a frequency other than the "frequency that is an integer multiple of the greatest common divisor (the greatest common divisor of two or more) disclosed in Patent Literature 1". In this way, it is possible to appropriately suppress a false inference of the state of the apparatus 70.

[0044] The thirteenth aspect of the present disclosure relates to a drive system that includes a motor drive apparatus 20 that drives a motor 50 installed on an apparatus 70, and a state inferring apparatus that infers whether the apparatus 70 is abnormal, the state inferring apparatus being the state inferring apparatus of any one of the first to twelfth aspects.

[0045] The fourteenth aspect of the present disclosure relates to a refrigeration system that includes a refrigerant circuit RR1 including a compressor CC having a motor 50, and a state inferring apparatus that infers a state of the refrigeration system, the state inferring apparatus being the state inferring apparatus of any one of the first to twelfth aspects.

[0046] The fifteenth aspect of the present disclosure relates to a fan system that includes a fan FF1 having a motor 50, and a state inferring apparatus that infers a state of the fan system, the state inferring apparatus being the state inferring apparatus of any one of the first to twelfth aspects.

[0047] The sixteenth aspect of the present disclosure relates to a state inferring method that infers a state of an apparatus 70 that mounts an electric motor 50 driven by a direct AC converter 23, the state inferring method including an inferring step in which a control section 31 infers the state of the apparatus 70 in accordance with a magnitude of a specific frequency component C1 of a physical quantity obtained from the apparatus 70, and a restricting step in which the control section 31 restricts the inferring step or an operation of the electric motor 50 when a noise frequency component Cn included in the physical quantity and the specific frequency component C1 satisfy a prescribed relationship, the restricting step being a first restricting step in which the inferring step is restricted so as to output a preset inferring result, or a second restricting step in which the operation of the electric motor 50 is restricted so as not to make a false inference in the inferring step, provided that a carrier frequency of the direct AC converter 23 is f c , provided that a reciprocal of a period of sampling of the physical quantity, that is, a sampling frequency is f s , provided that an electric angle frequency of the electric motor 50 is f0, when the physical quantity is a direct current signal, a frequency of the noise frequency component Cn is a frequency represented by any one of the following Expression 1, Expression 2, and Expression 3, and when the physical quantity is an alternating current signal, on a condition that two or more frequencies including the electric angle frequency in the electric angle frequency, the carrier frequency, and a frequency of sampling a modulation wave for controlling the direct AC converter 23 do not have a greatest common divisor of two or more, the frequency of the noise frequency component Cn is a frequency represented by any one of the following Expression 4, Expression 5, and Expression 6.

[0048]

Mathematical Expression 2

[0049] |f c ±6Nf0|…(1)

[0050] |Mf s -6Nf0|…(2)

[0051] |Mf s -|f c ±6Nf0||…(3)

[0052] |f0±|f c ±6Nf0||…(4)

[0053] |f0±|Mf s -6Nf0||…(5)

[0054] |f0±|Mf s -|f c ±6Nf0|||…(6)

[0055] where M and N are natural numbers

[0056] In the sixteenth aspect, the noise frequency component Cn of the frequency represented by any one of Formula 1 to Formula 6 is a frequency component that is likely to affect the specific frequency component Cl. Therefore, in the limiting step, by limiting the inference process or the operation of the motor 50 according to the relationship between the noise frequency component Cn and the specific frequency component Cl, it is possible to suppress false inference of the state of the device 70.

[0057] Note that, in the sixteenth aspect, the influence of the noise frequency component of the frequency other than the "greatest common divisor (the greatest common divisor of two or more) and an integral multiple thereof of the frequency disclosed in Patent Document 1" can be taken into account. In this way, it is possible to appropriately suppress false inference of the state of the device 70.

[0058] The seventeenth aspect of the present disclosure relates to a state inference method that infers a state of a device 70 in which a motor 50 is installed, the state inference method including an inference step in which a control section 31 infers the state of the device 70 according to the magnitude of a specific frequency component Cl of a physical quantity obtained from the device 70, and a limiting step in which the control section 31 limits the inference step or the operation of the motor 50 based on the magnitude of a periodic component included in a variable calculated according to the magnitude of the specific frequency component Cl, the limiting step being a first limiting step in which the inference step is limited so as to output a preset inference result, or a second limiting step in which the operation of the motor 50 is limited so as not to make a false inference in the inference step.

[0059] In the seventeenth aspect, when a noise component affects the specific frequency component Cl, the magnitude (amplitude) of the specific frequency component Cl fluctuates. Therefore, by limiting the inference step or the operation of the motor 50 based on the magnitude of the periodic component included in the variable calculated according to the magnitude of the specific frequency component Cl, it is possible to suppress false inference of the state of the device 70.

[0060] Note that, in the seventeenth aspect, the influence of the noise frequency component of the frequency other than the "greatest common divisor (the greatest common divisor of two or more) and an integral multiple thereof of the frequency disclosed in Patent Document 1" can be taken into account. In this way, it is possible to appropriately suppress false inference of the state of the device 70.

[0061] The eighteenth aspect of the present disclosure relates to a state inference program that causes a computer to execute the state inference method of the sixteenth or seventeenth aspect. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 is a circuit diagram that shows a configuration example of the drive system of the first embodiment.

[0063] Figure 2 is a diagram showing an example of a flow of the control processing.

[0064] Figure 3 is a graph showing an example of a specific frequency component and a noise frequency component.

[0065] Figure 4 is a graph showing an example of a waveform of the specific frequency component fluctuating with the noise frequency component.

[0066] Figure 5 is a graph showing an example of a waveform of the specific frequency component when the frequency difference between the noise frequency component and the specific frequency component is relatively large.

[0067] Figure 6 is a graph showing an example of a waveform of the specific frequency component when the frequency difference between the noise frequency component and the specific frequency component is relatively small.

[0068] Figure 7 is a graph showing an example of a relationship between the frequency difference between the noise frequency component and the specific frequency component and the fluctuation amplitude of the specific frequency component.

[0069] Figure 8 is a flowchart showing an example of the processing performed by the control section of the first embodiment.

[0070] Figure 9 is a graph showing a specific example of a specific frequency component and a noise frequency component.

[0071] Figure 10 is a graph showing an example of a specific frequency band with the frequency of the specific frequency component as a reference.

[0072] Figure 11 is a flowchart showing an example of the first inference processing.

[0073] Figure 12 is a flowchart showing an example of the second inference processing.

[0074] Figure 13 is a timing chart showing an example of the second inference processing.

[0075] Figure 14 is a flowchart showing an example of the third inference processing.

[0076] Figure 15 is a timing chart showing an example of the third inference processing.

[0077] Figure 16 is a flowchart showing an example of the fourth inference processing.

[0078] Figure 17 is a timing chart showing an example of the fourth inference processing.

[0079] Figure 18 is a flowchart showing an example of the restriction processing of the first embodiment.

[0080] Figure 19 is a flowchart showing an example of the processing flow of the control section of the second embodiment.

[0081] Figure 20 is a graph for explaining the average value of the amount of change of the specific frequency component in a prescribed time.

[0082] Figure 21 is a graph showing an example of the periodic component extracted from the specific frequency component.

[0083] Figure 22 is a graph for explaining the standard deviation of the specific frequency component in a prescribed time.

[0084] Figure 23 is a graph for explaining the kurtosis of the distribution of the specific frequency component in a prescribed time.

[0085] Figure 24 is a graph for explaining the cumulative value of the absolute value of the difference from the average value of the specific frequency component in a prescribed time.

[0086] Figure 25 is a graph for explaining the energy size of the spectrum of the periodic component of the specific frequency component.

[0087] Figure 26 is a graph for explaining the difference between the maximum value and the minimum value of the specific frequency component in a prescribed time.

[0088] Figure 27 is a diagram showing a configuration example of a refrigeration system.

[0089] Figure 28 is a diagram showing a configuration example of a fan system. DETAILED DESCRIPTION

[0090] Hereinafter, the embodiments will be described in detail with reference to the drawings. Note that the same or equivalent portions will be denoted by the same reference numerals, and repeated explanation will not be given.

[0091] (First Embodiment)

[0092] Figure 1A configuration example of the drive system 10 of the first embodiment is shown. The drive system 10 drives the motor 50 using electric power supplied from the power supply 60. The motor 50 is installed in the device 70. The motor 50 is, for example, an IPM motor (Interior Permanent Magnet Motor). In this example, the power supply 60 is an alternating current power supply, and the motor 50 is a three-phase alternating current motor. The drive system 10 is installed in the device 70. The device 70 is, for example, an outdoor unit of an air conditioner. The drive system 10 includes a motor drive device 20 and a control device 30.

[0093] 〔Motor Drive Device〕

[0094] The motor drive device 20 drives the motor 50. Specifically, the motor drive device 20 converts electric power supplied from the power supply 60 into output alternating current (in this example, three-phase alternating current) having a prescribed frequency and voltage, and supplies the output alternating current to the motor 50. In this example, the motor drive device 20 has an AC / DC converter 21, a DC portion 22, and a DC / AC converter 23.

[0095] The AC / DC converter 21 rectifies electric power supplied from the power supply 60. In this example, the AC / DC converter 21 full-wave rectifies alternating current supplied from the power supply 60. The AC / DC converter 21 is, for example, configured by a diode bridge circuit obtained by connecting a plurality of rectifier diodes in a bridge fashion.

[0096] The DC portion 22 generates direct current corresponding to the electric power supplied from the power supply 60. In this example, the DC portion 22 has a capacitor that smoothes the output of the AC / DC converter 21.

[0097] The DC / AC converter 23 has a plurality of switching elements, and converts the output of the DC portion 22 into alternating current (three-phase alternating current) having a prescribed frequency and voltage by switching operation of the plurality of switching elements.

[0098] In this example, the DC / AC converter 23 has six switching elements connected in a bridge fashion, and six freewheeling diodes connected in reverse parallel to the six switching elements, respectively. Specifically, the DC / AC converter 23 has three switching legs each configured by two switching elements connected in series. The midpoints (specifically, the connection points of the switching elements on the upper arm side and the switching elements on the lower arm side) of the three switching legs are connected to the three windings (windings of the U-phase, the V-phase, and the W-phase) of the motor 50, respectively.

[0099] 〔Various Sensors〕

[0100] In the motor driving device 20, various sensors such as a phase current detecting section 41 and an electric angle frequency detecting section 42 are provided. Various information detected by the various sensors is sent to the control device 30. Specifically, detection signals of the various sensors are sent to a control section 31 described later.

[0101] The phase current detecting section 41 detects three-phase phase currents (U-phase current iu, V-phase current iv, and W-phase current iw) flowing in three windings (omitted from the drawing) of the motor 50, respectively. For example, the phase current detecting section 41 can detect all of the three-phase phase currents iu, iv, and iw, or can detect two of the three-phase phase currents iu, iv, and iw, and derive the remaining one phase current from the two detected phase currents. In addition, the phase current detecting section 41 can derive the three-phase phase currents iu, iv, and iw from a direct current detected by a shunt resistor (omitted from the drawing) provided in the direct current section 22 and a switching pattern.

[0102] The electric angle frequency detecting section 42 is configured to detect an electric angle frequency ω of the motor 50. Note that the electric angle frequency detecting section 42 is not necessarily provided, and the electric angle frequency ω of the motor 50 can be calculated by another method or inferred in a sensorless manner.

[0103] 〔Control device (state inferring device)〕

[0104] The control device 30 infers a state of the device 70. The control device 30 is an example of a state inferring device that infers a state of the device 70 in which the motor 50 is installed. The processing in the control device 30 (processing related to the inference of the state of the device 70) is an example of a state inferring method that infers a state of the device 70 in which the motor 50 is installed.

[0105] In this example, the control device 30 infers whether or not the device 70 has an abnormality, and deals with the abnormality of the device 70. The control device 30 also controls the motor 50. Specifically, the control device 30 controls the motor 50 by controlling the motor driving device 20.

[0106] 〔Control section〕

[0107] The control device 30 includes the control section 31. The control section 31 performs various processing. Specifically, the control section 31 acquires information and data from each part of the device 70, and performs various processing based on the above information and data. The processing of the control section 31 will be described in detail later.

[0108] For example, the control unit 31 includes a processor and a memory (storage medium) electrically connected to the processor and storing programs for instructing the processor to operate. Various functions of the control unit 31 are implemented by the processor executing the programs. It should be noted that the control unit 31 is an example of a computer, and the aforementioned program is an example of a state inference program.

[0109] [The processing carried out by the control department]

[0110] In this example, the control unit 31 performs control processing, calculation processing, inference processing, and response processing. It should be noted that the various processes described above are examples of various steps. For example, inference processing is an example of an inference step.

[0111] [Control Processing]

[0112] In the control process, the control unit 31 controls the motor drive device 20 to control the motor 50. Specifically, the control unit 31 inputs target command values ​​such as the command value of the electrical angle frequency ω of the motor 50, and detection signals from various sensors installed in the motor drive device 20. Then, based on the target command values ​​and the detection signals from various sensors, the control unit 31 controls the switching operation of the DC-AC converter 23 to control the AC power supplied from the DC-AC converter 23 to the motor 50.

[0113] In this example, the control unit 31 inputs a carrier signal representing the carrier wave and a sampling clock. The carrier wave is a wave whose amplitude fluctuates at a predetermined frequency (carrier frequency). The sampling clock is a signal representing the sampling time.

[0114] like Figure 8 As shown, the control unit 31 generates a modulation wave signal representing the modulation wave. The modulation wave is a wave corresponding to the waveform of an output AC voltage having a specified frequency and voltage. The control unit 31 samples the instantaneous value of the modulation wave signal synchronously with a sampling clock, and compares the instantaneous value of the modulation wave signal obtained through the sampling with a carrier signal.

[0115] Then, based on the comparison result, the control unit 31 generates a PWM signal to control the switching operation of the DC-AC converter 23. Specifically, when the instantaneous value of the sampled modulated wave signal is higher than the signal level (amplitude value) of the carrier signal, the signal level of the PWM signal becomes high; when the instantaneous value of the sampled modulated wave signal is lower than the signal level of the carrier signal, the signal level of the PWM signal becomes low (e.g., zero).

[0116] The PWM signal generated by the control unit 31 is supplied to the multiple switching elements included in the DC-AC converter 23. The multiple switching elements included in the DC-AC converter 23 are turned on and off according to the PWM signal. In this way, an output AC current with a specified frequency and voltage is generated.

[0117] Note that the frequency of the carrier signal corresponds to the "carrier frequency of the direct-current alternating-converter 23". The inverse of the period of the sampling of the instantaneous value of the modulated wave signal corresponds to the "frequency of the sampling of the modulated wave for controlling the direct-current alternating-converter 23".

[0118] [Calculation processing]

[0119] In the calculation processing, the control section 31 calculates the specific frequency component Cl from the physical quantity obtained from the device 70. Specifically, the control section 31 samples the instantaneous value of the physical quantity at a prescribed sampling period, and calculates the specific frequency component Cl from the instantaneous value of the physical quantity obtained by the sampling. The sampling period is the "period of the sampling of the physical quantity", and the inverse of the sampling period is the "sampling frequency".

[0120] In this example, the control section 31 derives the specific frequency component Cl of the signal based on the current or voltage of the motor 50. The signal based on the current or voltage of the motor 50 is an example of the physical quantity obtained from the device 70. Details of the signal and the specific frequency component Cl will be described later.

[0121] [Inference processing]

[0122] In the inference processing, the control section 31 infers the state of the device 70 from the specific frequency component Cl calculated by the calculation processing (the specific frequency component Cl of the physical quantity obtained from the device 70). Then, the control section 31 outputs the inference result. In this example, the control section 31 infers in the inference processing whether or not the device 70 is abnormal. Further, the control section 31 infers in the inference processing the state of the device 70 from the specific frequency component Cl of the signal based on the current or voltage of the motor 50. Note that details of the inference processing will be described later.

[0123] [Restriction processing]

[0124] In the restriction processing, when the specific frequency component Cl and the noise frequency component Cn included in the physical quantity satisfy a prescribed relationship, the control section 31 restricts the inference processing or the operation of the motor 50. The restriction processing is the first restriction processing or the second restriction processing, in the first restriction processing, the inference processing is restricted so that a preset inference result is output; in the second restriction processing, the operation of the motor 50 is restricted so that a false inference is not made in the inference processing. Details of the first restriction processing and the second restriction processing will be described later.

[0125] In this example (example in which it is inferred whether the device 70 is abnormal in the inference processing), the preset inference result is not an inference result indicating that the device 70 is abnormal. The preset inference result is, for example, an inference result indicating that the device 70 is normal, an inference result indicating that the state of the device 70 is unknown, an inference result indicating that the inference processing is prohibited (stopped), an inference result output before the inference processing is restricted (for example, immediately before being restricted), or the like.

[0126] [Noise frequency component]

[0127] Here, the noise frequency component C1 is described. Hereinafter, let the carrier frequency of the direct-current-to-alternating-current converter 23 be "f c ", let the reciprocal of the period of sampling of the physical quantity, that is, the sampling frequency be "f s ", and let the electrical angle frequency of the motor 50 be "f0".

[0128] When the physical quantity is a direct-current signal, the noise frequency component C1 is a frequency represented by any one of the following Expression 1, Expression 2, and Expression 3. Further, when the physical quantity is an alternating-current signal, under the condition that two or more of the electrical angle frequency of the motor 50, the carrier frequency of the direct-current-to-alternating-current converter 23, and the frequency at which the modulation wave for controlling the direct-current-to-alternating-current converter 23 is sampled, including the electrical angle frequency of the motor 50, do not have a greatest common divisor of 2 or more, the noise frequency component C1 is a frequency represented by any one of the following Expression 4, Expression 5, and Expression 6.

[0129] [Expression 3]

[0130] |f c ±6Nf0|…(1)

[0131] |Mf s -6Nf0|…(2)

[0132] |Mf s -|f c ±6Nf0||…(3)

[0133] |f0±|f c ±6Nf0||…(4)

[0134] |f0±|Mf s -6Nf0||…(5)

[0135] |f0±|Mf s -|f c ±6Nf0|||…(6)

[0136] where M and N are natural numbers

[0137] [Countermeasure processing]

[0138] In the inference processing, when the control section 31 infers that the state of the device 70 is the "abnormal state", countermeasures are taken. The countermeasures are processing for coping with the abnormality of the device 70. For example, the countermeasures are output processing of outputting first information indicating that the state of the device 70 is the abnormal state.

[0139] The output processing is, for example, the following first output processing, second output processing, third output processing, a combination of the above processing, or the like. The first output processing is processing of causing a display device (omitted from the drawing) provided on a remote controller or the like to display the first information by outputting the first information to the display device. The second output processing is processing of causing a control section (omitted from the drawing) that controls the operation of the device 70 to perform an operation for coping with the abnormal state by outputting the first information to the control section. The third output processing is processing of uploading the first information to a data accumulation section (omitted from the drawing) on the cloud.

[0140] 〔Insight of the inventor〕

[0141] Next, an insight of the present inventor will be described. The present inventor has found, through research, that various noise frequency components can affect the "specific frequency component C1 used in the inference processing". For example, among the frequencies that are integer multiples of the electrical angle frequency of the motor 50, integer multiples of the power supply frequency, integer multiples of the carrier frequency of the rectifier 23, and integer multiples of the sampling frequency at which the physical quantity is sampled in the calculation processing, the absolute value of the sum or difference of at least two frequencies, or a noise frequency component of a frequency equivalent to the absolute value, can affect the specific frequency component C1.

[0142] Furthermore, the present inventor has found, through in-depth research, that among the various noise frequency components that can affect the specific frequency component C1, the following noise frequency components have a particularly significant tendency to affect the specific frequency component C1.

[0143] (1) When the physical quantity is a direct current signal, a noise frequency component of a frequency represented by any one of the following Formula 1, Formula 2, and Formula 3;

[0144] (2) When the physical quantity is an alternating current signal, a noise frequency component of a frequency represented by any one of the following Formula 4, Formula 5, and Formula 6.

[0145]

Mathematical Formula 4

[0146] |f c ±6Nf0|…(1)

[0147] |Mf s -6Nf0|…(2)

[0148] |Mf s -|f c ±6Nf0||…(3)

[0149] |f0±|f c ±6Nf0||…(4)

[0150] |f0±|Mf s -6Nf0||…(5)

[0151] |f0±|Mf s -|f c ±6Nf0|||…(6)

[0152] where M, N are natural numbers

[0153] The " |Mf s -6Nf0| " represented by the formula 2, the formula 5 corresponds to a frequency that is accompanied by aliasing that occurs when a frequency component of a frequency that is 6 times the electrical angle frequency of the motor 50 is sampled. The " |Mf s -|f c ±6Nf0|| " represented by the formula 3, the formula 6 corresponds to a frequency that is accompanied by aliasing that occurs when a frequency component of a frequency that is the difference between the " carrier frequency of the direct AC converter 23 " and the " frequency that is 6N times the electrical angle frequency of the motor 50 " is sampled. Note that 6Nf0 is a spatial higher harmonic of the motor, and has the following characteristics: the larger the value of N, the smaller the amplitude of the noise frequency component. Furthermore, the larger f0, the sparser the frequency interval of the noise frequency component.

[0154] Note that in this specification, the noise frequency component of the frequency represented by any one of the above-described formula 1 to formula 6 is referred to as "noise frequency component Cn".

[0155] Furthermore, the influence of the noise frequency component Cn is reflected in the size (amplitude) of the specific frequency component Cl. In other words, the noise frequency component Cn can cause fluctuations in the size of the specific frequency component Cl. For example, as shown in Figure 3 if the noise frequency component Cn is generated near the specific frequency component Cl, as shown in Figure 4 the size of the calculated noise frequency component Cn fluctuates with a period 1 / X represented by the reciprocal of the frequency difference X between the noise frequency component Cn and the specific frequency component Cl.

[0156] Furthermore, there is a tendency that the smaller the difference between the noise frequency component Cn and the specific frequency component Cl, the greater the amplitude fluctuation of the specific frequency component Cl that is influenced by the noise frequency component Cn. For example, Figure 5 a waveform example of the specific frequency component Cl when the frequency difference between the noise frequency component Cn and the specific frequency component Cl is relatively large is shown in Figure 6An example of the waveform of the specific frequency component C1 when the frequency difference between the noise frequency component Cn and the specific frequency component C1 is relatively small is shown. Note that in the example shown in Figure 5 the frequency difference between the noise frequency component Cn and the specific frequency component C1 is 3 Hz, and in the example shown in Figure 6 the frequency difference between the noise frequency component Cn and the specific frequency component C1 is 0.2 Hz.

[0157] Further, the present inventors have found that when the frequency difference between the noise frequency component Cn and the specific frequency component C1 is 3 Hz, the fluctuation amplitude of the specific frequency component C1 when the noise frequency component Cn has an effect is 10% or less of the amplitude of the specific frequency component C1 when the noise frequency component Cn has no effect. Figure 7 An example of the relationship between the frequency difference between the noise frequency component Cn and the specific frequency component C1 and the fluctuation amplitude of the specific frequency component C1 is shown. In the example shown in Figure 7 the fluctuation amplitude of the specific frequency component C1 is represented by the value obtained by dividing the difference between the maximum value and the minimum value of the amplitude of the specific frequency component C1 by 2. Figure 7 The broken line L10 shown represents 10% of the amplitude of the specific frequency component C1 when the noise frequency component Cn has no effect.

[0158] [Flow of processing by the control section]

[0159] Next, the flow of the processing (specifically, the calculation processing, the inference processing, and the restriction processing) by the control section 31 will be described with reference to Figure 8 The control section 31 performs the calculation processing at each prescribed calculation time. By repeatedly performing the calculation processing, the specific frequency component C1 is calculated at each calculation time. The calculation time is set to a time of or more than the time required to calculate the specific frequency component C1 from the physical quantity obtained from the device 70, for example.

[0160] Then, the control section 31 performs the inference processing at each prescribed inference time, based on the specific frequency component C1 calculated by the calculation processing. By repeatedly performing the inference processing, the inference result is obtained at each inference time. The inference time is set to a time of or more than the calculation time. The inference time is set to a time of or more than the time required to infer the state of the device 70 from the specific frequency component C1 calculated by the calculation processing, for example.

[0161] Further, the control section 31 performs the restriction processing in parallel with the calculation processing and the inference processing. Specifically, the control section 31 repeatedly performs the processing shown in

[0162] Steps S101 to S104. Figure 2

[0163] <Step S101>​

[0164] The control section 31 derives the noise frequency component Cn. For example, the control section 31 calculates the noise frequency component Cn according to a prescribed calculation formula (specifically, any one of Formulas 1 to 6). Further, the control section 31 acquires the specific frequency component Cl calculated by the calculation processing. Then, the control section 31 judges whether the noise frequency component Cn and the specific frequency component Cl satisfy a prescribed relationship.

[0165] In this example, the prescribed relationship is a relationship in which the frequency difference between the noise frequency component Cn and the specific frequency component Cl is below a frequency difference threshold value. For example, the frequency difference threshold value is set to 3 Hz. The control section 31 judges whether the frequency difference between the noise frequency component Cn and the specific frequency component Cl is below the frequency difference threshold value.

[0166] When the noise frequency component Cn and the specific frequency component Cl satisfy the prescribed relationship, the processing of step S102 is performed. For example, the processing of step S101 is performed at every judgment time. In other words, the processing of step S101 is repeatedly performed until the noise frequency component Cn and the specific frequency component Cl satisfy the prescribed relationship.

[0167] <Step S102>

[0168] When the noise frequency component Cn and the specific frequency component Cl satisfy the prescribed relationship, the control section 31 restricts the inference processing or the operation of the motor 50.

[0169] <Step S103>

[0170] Next, the control section 31 judges whether a restriction release condition is satisfied. The restriction release condition has, for example, that the noise frequency component Cn and the specific frequency component Cl do not satisfy the prescribed relationship; that the control section 31 receives a release instruction; and the like. For example, when an operator inputs an operation for releasing the restriction to an operation section (omitted from illustration), a release instruction is sent from the operation section to the control section 31.

[0171] When the restriction release condition is satisfied, the processing of step S104 is performed. For example, the processing of step S103 is performed at every prescribed judgment time. In other words, the processing of step S103 is repeatedly performed until the restriction release condition is satisfied.

[0172] <Step S104>

[0173] When the restriction release condition is satisfied, the control section 31 releases the restriction (restriction on the inference processing or restriction on the operation of the motor 50).

[0174] [Specific Examples of Specific Frequency Component and Noise Frequency Component]

[0175] The following will be described with reference to Figure 9Specific examples of particular frequency components and noise frequency components are explained. Figure 9 In this example, there are two specific frequency components C1 and the mechanical angular frequency f of the motor 50. m The electric angular frequency f0 of the motor is 332.4 Hz, and the carrier frequency f is 83.1 Hz. c It is 5900Hz. It should be noted that the mechanical angular frequency f of motor 50... m This is equivalent to the rotational frequency of an electric motor of 50.

[0176] exist Figure 9 In this example, two specific frequency components C1 appear, symmetrical about the electrical angular frequency f0 of the motor. Noise frequency components Cn appear near each of these two specific frequency components C1. One specific frequency component C1 has a frequency of 249.3 Hz, and a noise frequency component Cn located near this specific frequency component C1 has a frequency of 249.2 Hz. The other specific frequency component C1 has a frequency of 415.5 Hz, and another noise frequency component Cn located near this other specific frequency component C1 has a frequency of 415.6 Hz. In this example, the frequency difference between the noise frequency component Cn and the specific frequency component C1 is 0.1 Hz; therefore, in the limiting process, the inference process or the operation of the motor 50 is limited.

[0177] In this way, the influence of noise frequency components C1 other than "the greatest common divisor (greatest common divisor of 2 or more) and its integer multiples disclosed in Patent Document 1" can be taken into consideration. On the other hand, the anomaly diagnosis device of Patent Document 1 cannot take into consideration the influence of the aforementioned noise frequency component C1.

[0178] [Effects of the first embodiment]

[0179] As described above, in the first embodiment, during the restriction process, when the noise frequency component Cn and the specific frequency component C1 contained in the physical quantity satisfy a predetermined relationship, the control unit 31 restricts the inference process or the operation of the motor 50. When the physical quantity is a DC signal, the noise frequency component Cn is a frequency represented by any one of the following equations 1, 2, and 3. When the physical quantity is an AC signal, under the condition that two or more frequencies including the electrical angle frequency among the electrical angle frequency, the carrier frequency, and the frequency used to sample the modulation wave for controlling the DC-AC converter 23 do not have a greatest common divisor of 2 or more, the noise frequency component Cn is a frequency represented by any one of the following equations 4, 5, and 6.

[0180]

Mathematical Expression 5

[0181] |f c ±6Nf0|…(1)

[0182] | Mf s - 6Nf0|…(2)

[0183] | Mf s | f c ± 6Nf0||…(3)

[0184] | f0± | f c ± 6Nf0||…(4)

[0185] | f0± | Mf s - 6Nf0||…(5)

[0186] | f0± | Mf s | f c ± 6Nf0|||…(6)

[0187] where M, N are natural numbers

[0188] In the above configuration, the noise frequency component Cn of the frequency represented by any one of the above Formulae 1 to 6 is a frequency component that can affect the specific frequency component Cl. Therefore, in the limiting process, by limiting the inference process or the operation of the motor 50 in accordance with the relationship between the noise frequency component Cn and the specific frequency component Cl, it is possible to suppress erroneous inference of the state of the device 70. In this example, since erroneous inference of the abnormality of the device 70 can be avoided, erroneous inference of the abnormality of the device 70 can be avoided.

[0189] Note that, in the first embodiment, the influence of the noise frequency component of the frequency other than the "common multiple of the greatest common divisor (the greatest common divisor of two or more) and an integer multiple thereof disclosed in Patent Literature 1" can be taken into consideration. In this way, erroneous inference of the state of the device 70 can be appropriately suppressed.

[0190] Further, in the first embodiment, the prescribed relationship is the relationship in which the frequency difference between the noise frequency component Cn and the specific frequency component Cl is below a prescribed frequency difference threshold. In this way, by limiting the inference process or the operation of the motor 50 in accordance with the frequency difference between the noise frequency component Cn and the specific frequency component Cl, it is possible to suppress erroneous inference of the state of the device 70.

[0191] Further, in the first embodiment, the frequency difference threshold is 3 Hz. In this way, when the "fluctuation amplitude of the specific frequency component Cl when the noise frequency component Cn has an influence" is greater than 10% of the "amplitude of the specific frequency component Cl when the noise frequency component Cn does not have an influence", it is possible to limit the inference process or the operation of the motor 50. As a result, erroneous inference of the state of the device 70 can be appropriately suppressed.

[0192] (Modified example of the prescribed relationship)

[0193] Note that the prescribed relationship that is satisfied by the noise frequency component Cn and the specific frequency component Cl in step S101 is not limited to the relationship in which the frequency difference between the noise frequency component Cn and the specific frequency component Cl is below the frequency difference threshold value.

[0194] For example, as shown in FIG. 8, the prescribed relationship can also be a relationship in which the noise frequency component Cn is included in a specific frequency band FR that is centered on the frequency of the specific frequency component Cl and has a bandwidth of 6 Hz. In step S101, the control unit 31 can also determine whether the noise frequency component Cn is included in the specific frequency band FR that is centered on the frequency of the specific frequency component Cl and has a bandwidth of 6 Hz. Figure 10

[0195] Note that the specific frequency band FR can also be set to a frequency band that is centered on the frequency of the specific frequency component Cl and has a bandwidth of 6 Hz. With the above setting, when the "fluctuation amplitude of the specific frequency component Cl when the noise frequency component Cn has an effect" is greater than 10% of the "amplitude of the specific frequency component Cl when the noise frequency component Cn has no effect", it is possible to limit the inference processing or the operation of the motor 50.

[0196] (Specific examples of signals)

[0197] A specific example of a physical quantity obtained from the device 70, i.e., a "signal based on the current or voltage of the motor 50", will be described below. This signal is roughly classified into a direct current signal and an alternating current signal.

[0198] (Specific examples of direct current signals)

[0199] Examples of the direct current signal include a "signal related to the phase current iu, iv, iw of the motor 50", a "signal related to the phase voltage Vu, Vv, Vw of the motor 50", and a "signal related to the power of the motor 50".

[0200] Other examples of the direct current signal include a "current iγ, iδ obtained by performing coordinate conversion on the phase current iu, iv, iw of the motor 50 using the phase ωi·t of the phase current iu, iv, iw of the motor 50", a "voltage Vγ, Vδ obtained by performing coordinate conversion on the phase voltage Vu, Vv, Vw of the motor 50 using the phase ωv·t of the phase voltage Vu, Vv, Vw of the motor 50", a "voltage iζ, iη obtained by performing coordinate conversion on the phase current iu, iv, iw of the motor 50 using the phase ωv·t of the phase voltage Vu, Vv, Vw of the motor 50", and a "voltage obtained by performing coordinate conversion on the phase voltage Vu, Vv, Vw of the motor 50 using the phase ωi·t of the phase current iu, iv, iw of the motor 50".

[0201] ​Another example of the DC signal is, for example, "d-q axis magnetic fluxes λd, λq obtained by coordinate conversion of the armature cross-magnetic flux generated by the permanent magnet" and "the magnitude λ0 of the armature cross-magnetic flux vector obtained by combining the armature cross-magnetic flux of the permanent magnet and the armature reaction".

[0202] Note that, in the following description, "the phase currents iu, iv, iw of the motor 50" are the phase currents iu, iv, iw of the motor 50 detected by the phase current detection section 41. "The phase voltages Vu, Vv, Vw of the motor 50" are the phase voltages Vu, Vv, Vw of the motor 50 indicated by the voltage command value used in the inside of the control section 31 or the phase voltages Vu, Vv, Vw of the motor 50 detected by the phase voltage detection section (not shown) provided in the motor drive device 20. "The electric angle frequency ω of the motor 50" is the electric angle frequency ω of the motor 50 detected by the electric angle frequency detection section 42.

[0203] 〔1. Specific examples of signals related to the phase currents of the motor〕

[0204] Specific examples of the signals related to the phase currents iu, iv, iw of the motor 50 are, for example, the current vector amplitude Ia, the square value of the current vector amplitude Ia 2 , the phase current amplitude I, the phase current effective value Irms, and the like.

[0205] Note that the current vector amplitude Ia and the square value of the current vector amplitude Ia 2 are examples of values corresponding to the sum of the square values of the respective phase currents iu, iv, iw of the three phases of the motor 50. The values corresponding to the sum of the square values of the respective phase currents iu, iv, iw of the three phases of the motor 50 are examples of values proportional to the integral power of the magnitudes of the phase currents iu, iv, iw of the motor 50.

[0206] (1) Current vector amplitude

[0207] The current vector amplitude Ia is derived from the phase currents iu, iv, iw of the motor 50. Further, the current vector amplitude Ia can also be derived from the α-phase current iα and the β-phase current iβ obtained by converting the phase currents iu, iv, iw of the motor 50 to the fixed coordinate system. Further, the current vector amplitude Ia can also be derived from the M-axis current iM and the T-axis current iT obtained by coordinate-converting the phase currents iu, iv, iw of the motor 50 according to an angle based on the direction of the primary magnetic flux. Further, the current vector amplitude Ia can also be derived from the d-axis current id and the q-axis current iq obtained by coordinate-converting the phase currents iu, iv, iw of the motor 50 according to an angle based on the direction of the magnetic pole position. Specifically, the current vector amplitude Ia can be expressed by the following equation.

[0208] [Math. 6]

[0209]

[0210] (2) Square value of current vector amplitude

[0211] Square value of current vector amplitude Ia 2 is derived from the phase currents iu, iv, iw of the motor 50. Further, the square value of the current vector amplitude Ia 2 can also be derived from the α-phase current iα and the β-phase current iβ obtained by converting the phase currents iu, iv, iw of the motor 50 to the fixed coordinate system. Further, the square value of the current vector amplitude Ia 2 can also be derived from the M-axis current iM and the T-axis current iT obtained by coordinate-converting the phase currents iu, iv, iw of the motor 50 according to an angle based on the direction of the primary magnetic flux. Further, the square value of the current vector amplitude Ia 2 can also be derived from the d-axis current id and the q-axis current iq obtained by coordinate-converting the phase currents iu, iv, iw of the motor 50 according to an angle based on the direction of the magnetic pole position. Specifically, the square value of the current vector amplitude Ia 2 can be expressed by the following equation.

[0212] [Math. 7]

[0213]

[0214] (3) Phase current amplitude

[0215] The phase current amplitude I is derived from one of the phase currents (e.g., the U-phase current iu) of the motor 50 and the phase of the phase current ωi. Note that the phase of the phase current ωi is derived from the phase currents iu, iv, iw of the motor 50, for example. Specifically, the phase current amplitude I can be expressed by the following equation.

[0216] [Math. 8]

[0217]

[0218] (4) Phase current effective value

[0219] The phase current effective value Irms is derived from the phase current amplitude I. Specifically, the phase current effective value Irms can be expressed by the following equation.

[0220] [Math. 9]

[0221]

[0222] (5) Others

[0223] In the above description, the case where the current vector amplitude Ia is derived from the phase currents iu, iv, iw of the three phases of the motor 50 is exemplified, but the current vector amplitude Ia can also be derived from the phase currents of two of the three phases of the motor 50. Further, the current vector amplitude Ia can also be derived from the direct current of the direct-current inverter 23 detected by a direct-current detection unit (e.g., a shunt resistor, not shown) provided in the motor drive device 20. The square value Ia 2 of the current vector amplitude is also the same.

[0224] 〔2. Specific examples of signals related to phase voltages of the motor〕

[0225] Specific examples of signals related to the phase voltages Vu, Vv, Vw of the motor 50 are the voltage vector amplitude Va, the square value Va 2 of the voltage vector amplitude, the phase voltage amplitude V, the phase voltage effective value Vrms, and the like.

[0226] Note that the voltage vector amplitude Va and the square value Va 2 of the voltage vector amplitude are examples of values corresponding to the sum of the squares of the respective phase voltages Vu, Vv, Vw of the three phases of the motor 50. The values corresponding to the sum of the squares of the respective phase voltages Vu, Vv, Vw of the three phases of the motor 50 are examples of values proportional to an integral power of the magnitudes of the phase voltages Vu, Vv, Vw of the motor 50.

[0227] (1) Voltage vector amplitude

[0228] The voltage vector amplitude Va is derived from the phase voltages Vu, Vv, Vw of the motor 50. Further, the voltage vector amplitude Va can also be derived from the α-phase voltage Vα and the β-phase voltage Vβ obtained by converting the phase voltages Vu, Vv, Vw of the motor 50 to the fixed coordinate system. Further, the voltage vector amplitude Va can also be derived from the M-axis voltage VM and the T-axis voltage VT obtained by coordinate-converting the phase voltages Vu, Vv, Vw of the motor 50 based on the angle of the direction of the primary magnetic flux. Further, the voltage vector amplitude Va can also be derived from the d-axis voltage Vd and the q-axis voltage Vq obtained by coordinate-converting the phase voltages Vu, Vv, Vw of the motor 50 based on the angle of the direction of the magnetic pole position. Specifically, the voltage vector amplitude Va can be expressed by the following equation.

[0229]

Math. 10

[0230]

[0231] (2) Square value of voltage vector amplitude

[0232] Square value of voltage vector amplitude Va 2 is derived from the phase voltages Vu, Vv, Vw of the motor 50. Further, the square value of the voltage vector amplitude Va 2 can also be derived from the α-phase voltage Vα and the β-phase voltage Vβ obtained by converting the phase voltages Vu, Vv, Vw of the motor 50 to the fixed coordinate system. Further, the square value of the voltage vector amplitude Va 2 can also be derived from the M-axis voltage VM and the T-axis voltage VT obtained by coordinate-converting the phase voltages Vu, Vv, Vw of the motor 50 based on the angle of the direction of the primary magnetic flux. Further, the square value of the voltage vector amplitude Va 2 can also be derived from the d-axis voltage Vd and the q-axis voltage Vq obtained by coordinate-converting the phase voltages Vu, Vv, Vw of the motor 50 based on the angle of the direction of the magnetic pole position. Specifically, the square value of the voltage vector amplitude Va 2 can be expressed by the following equation.

[0233]

Math. 11

[0234]

[0235] (3) Phase voltage amplitude

[0236] The phase voltage amplitude V is derived from one of the phase voltages (e.g., the U-phase voltage Vu) of the phase voltages Vu, Vv, Vw of the motor 50 and the phase ωv of the phase voltage. Note that the phase ωv of the phase voltage is derived from the phase voltages Vu, Vv, Vw of the motor 50, for example. Specifically, the phase voltage amplitude V can be expressed by the following equation.

[0237] [Math. 12]

[0238]

[0239] (4) Phase voltage effective value

[0240] The phase voltage effective value Vrms is derived from the phase voltage amplitude V. Specifically, the phase voltage effective value Vrms can be expressed by the following equation.

[0241] [Math. 13]

[0242]

[0244] (5) Others

[0245] In the above description, the case where the voltage vector amplitude Va is derived from the three-phase phase voltages Vu, Vv, Vw of the motor 50 is exemplified, but the voltage vector amplitude Va can also be derived from two-phase phase voltages of the three-phase phase voltages Vu, Vv, Vw of the motor 50. The square value Va2 of the voltage vector amplitude is also the same. 2

[0246] 〔3. Specific examples of signals related to the power of the motor〕

[0247] The signals related to the power of the motor 50 include, for example, the instantaneous power p, the instantaneous reactive power q, the apparent power S, the active power P, the reactive power Q, and the like.

[0248] (1) Instantaneous power

[0249] ​The instantaneous power p is derived from the phase currents iu, iv, iw of the motor 50 and the phase voltages Vu, Vv, Vw of the motor 50. In addition, the instantaneous power p can also be derived from the α-phase current iα and the β-phase current iβ obtained by converting the phase currents iu, iv, iw of the motor 50 to the fixed coordinate system, and the α-phase voltage Vα and the β-phase voltage Vβ obtained by converting the phase voltages Vu, Vv, Vw of the motor 50 to the fixed coordinate system. In addition, the instantaneous power p can also be derived from the M-axis current iM and the T-axis current iT obtained by coordinate-converting the phase currents iu, iv, iw of the motor 50 based on the direction of the primary magnetic flux, and the M-axis voltage VM and the T-axis voltage VT obtained by coordinate-converting the phase voltages Vu, Vv, Vw of the motor 50 based on the direction of the primary magnetic flux. The instantaneous power p can also be derived from the d-axis current id and the q-axis current iq obtained by coordinate-converting the phase currents iu, iv, iw of the motor 50 based on the direction of the magnetic pole position, and the d-axis voltage Vd and the q-axis voltage Vq obtained by coordinate-converting the phase voltages Vu, Vv, Vw of the motor 50 based on the direction of the magnetic pole position. Specifically, the instantaneous power p can be expressed by the following equations.

[0250] [Math. 14]

[0251] p = v u i u + v v i v + v w i w = v α i α + v β i β = v M i M + v T i T = v d i d + v q i q

[0252] (2) Instantaneous reactive power

[0253] The instantaneous reactive power q is derived from the α-phase current iα and the β-phase current iβ obtained by converting the phase currents iu, iv, iw of the motor 50 to the fixed coordinate system, and the α-phase voltage Vα and the β-phase voltage Vβ obtained by converting the phase voltages Vu, Vv, Vw of the motor 50 to the fixed coordinate system. Further, the instantaneous reactive power q can also be derived from the M-axis current iM and the T-axis current iT obtained by coordinate-converting the phase currents iu, iv, iw of the motor 50 based on the direction of the primary magnetic flux, and the M-axis voltage VM and the T-axis voltage VT obtained by coordinate-converting the phase voltages Vu, Vv, Vw of the motor 50 based on the direction of the primary magnetic flux. Further, the instantaneous reactive power q can also be derived from the d-axis current id and the q-axis current iq obtained by coordinate-converting the phase currents iu, iv, iw of the motor 50 based on the direction of the magnetic pole position, and the d-axis voltage Vd and the q-axis voltage Vq obtained by coordinate-converting the phase voltages Vu, Vv, Vw of the motor 50 based on the direction of the magnetic pole position. Specifically, the instantaneous reactive power q can be expressed by the following equation.

[0254] [Math. 15]

[0255] q = v α i β -v β i α = v M i T -v T i M = v d i q -v q i d

[0256] (3) Apparent Power

[0257] The apparent power S is derived from the phase voltage effective value Vrms and the phase current effective value Irms. Specifically, the apparent power S can be expressed by the following equation.

[0258] [Math. 16]

[0259] S = 3V rms I rms

[0260] (4) Active Power

[0261] The active power P is derived from the phase voltage effective value Vrms, the phase current effective value Irms, and the phase difference φ1 of the phase voltage and the phase current. The phase difference φ1 of the phase voltage and the phase current, which is the phase difference of one phase voltage (for example, U-phase voltage Vu) and one phase current (for example, U-phase current iu), is derived from the phase ωi of the phase current and the phase ωv of the phase voltage. Specifically, the active power P can be expressed by the following equation.

[0262] [Equation 17]

[0263]

[0264] (5) Reactive power

[0265] The reactive power Q is derived from the phase voltage effective value Vrms, the phase current effective value Irms, and the phase difference φ1 of the phase voltage and the phase current. The phase difference φ1 of the phase voltage and the phase current, which is the phase difference of the U-phase voltage Vu and the U-phase current iu, for example, is derived from the phase ωi of the phase current and the phase ωv of the phase voltage. Specifically, the reactive power Q can be expressed by the following equation.

[0266] [Equation 18]

[0267]

[0268] 〔4. Current obtained by coordinate conversion of phase current using phase of phase current〕

[0269] The currents iγ, iδ obtained by coordinate conversion of the phase currents iu, iv, iw of the motor 50 using the phases ωi•t of the phase currents iu, iv, iw of the motor 50 can be expressed by the following equation.

[0270] [Equation 19]

[0271]

[0272] 〔5. Voltage obtained by coordinate conversion of phase voltage using phase of phase voltage〕

[0273] The voltages Vγ, Vδ obtained by coordinate conversion of the phase voltages Vu, Vv, Vw of the motor 50 using the phases ωv•t of the phase voltages Vu, Vv, Vw of the motor 50 can be expressed by the following equation.

[0274] [Equation 20]

[0275]

[0276] 〔6. Current obtained by coordinate conversion of phase current using phase of phase voltage〕

[0277] The currents iζ, iη obtained by coordinate-converting the phase currents iu, iv, iw of the motor 50 with the phase ωv-t of the phase voltages Vu, Vv, Vw of the motor 50 can be expressed by the following equations.

[0278] [Math. 21]

[0279]

[0280] 〔7. Voltage obtained by coordinate-converting phase voltage with phase of phase current〕

[0281] The voltages Vζ, Vη obtained by coordinate-converting the phase voltages Vu, Vv, Vw of the motor 50 with the phase ωi-t of the phase currents iu, iv, iw of the motor 50 can be expressed by the following equations.

[0282] [Math. 22]

[0283]

[0284] 〔8. Size of armature cross-magnetic flux vector〕

[0285] The dq-axis magnetic fluxes λd, λq obtained by coordinate-converting the armature cross-magnetic flux generated by the permanent magnet and the size λ0 of the armature cross-magnetic flux vector obtained by combining the armature cross-magnetic flux of the permanent magnet and the armature reaction can be expressed by the following equations. "Ld" in the following equations is the d-axis inductance, and "Lq" is the q-axis inductance.

[0286] [Math. 23]

[0287]

[0288] 〔9. Another example of DC signal〕

[0289] Further, the DC signal can also be a DC signal obtained by three-phase two-phase converting the phase currents, the phase voltages, the line currents, and the line-to-line voltages of the motor 50 and further performing rotational coordinate conversion. For example, the DC signal can also be a DC signal obtained by three-phase two-phase converting the phase currents of the motor 50 to obtain α-axis and β-axis currents and performing rotational coordinate conversion of the α-axis and β-axis currents at an angle based on the orientation of the magnetic poles of the rotor of the motor 50 to obtain d-axis and q-axis currents. Further, the DC signal can also be a DC signal obtained by performing rotational coordinate conversion of the α-axis and β-axis currents at an angle based on the direction of the primary magnetic flux of the rotor of the motor 50 to obtain M-axis and T-axis currents.

[0290] Further, the DC signal can also be the power input to the AC / DC converter 21 of the motor drive device 20, the power output from the AC / DC converter 21, the power output from the DC portion 22, the current flowing between the AC / DC converter 21 and the DC portion 22, the current flowing between the DC portion 22 and the AC / DC converter 23, and the like.

[0291] (Specific examples of AC signals)

[0292] The AC signals are, for example, "phase currents iu, iv, iw of the motor 50", "phase voltages Vu, Vv, Vw of the motor 50", and "AC link fluxes Ψfu, Ψfv, Ψfw of the respective phases".

[0293] The AC link fluxes Ψfu, Ψfv, Ψfw of the respective phases can be expressed by the following expressions.

[0294] [Equation 24]

[0295]

[0296] Another example of the AC signal is, for example, a fixed coordinate current, voltage, AC link flux obtained by performing three-phase / two-phase conversion on the above-described AC signals.

[0297] Further, the AC signal can also be a line current, a line-to-line voltage, and the like of the motor 50. Further, the AC signal can also be a two-phase AC current (for example, an α-axis current and a β-axis current), a two-phase AC voltage obtained by performing three-phase / two-phase conversion on a phase current, a phase voltage, a line current, a line-to-line voltage. Further, the AC current can also be a current flowing between a commercial power system (specifically, the AC power supply 60) and the AC / DC converter 21 of the motor drive device 20.

[0298] (Specific examples of specific frequency components)

[0299] The specific examples of the specific frequency components will be described below.

[0300] For example, when the physical quantity is a DC signal, the frequency of the specific frequency component C1 is at least one of 1 times, 1 / 3 times, and 2 / 3 times the mechanical angular frequency of the motor 50. When the physical quantity is an AC signal, the frequency of the specific frequency component C1 is at least one of a frequency obtained by adding a prescribed frequency to the electrical angular frequency of the motor 50, a frequency obtained by subtracting the prescribed frequency from the electrical angular frequency of the motor 50, and 3 times the electrical angular frequency of the motor 50. The prescribed frequency is any one of 1 times, 1 / 3 times, and 2 / 3 times the mechanical angular frequency of the motor 50.

[0301] As described above, by setting the frequency of the specific frequency component C1, it is possible to infer various states (particularly, abnormalities) of the device 70.

[0302] (Variety of states of the device)

[0303] Here, a variety of states of the device 70 are described. The states of the device 70 are, for example, oil seal failure of the compression chamber, insulation aging of the motor 50, bearing wear, dilution, liquid strike, imbalance, flow path clogging, and the like. The states of the device 70 described above are examples of abnormalities of the device 70.

[0304] 〔1. Oil seal failure of the compression chamber〕

[0305] The oil seal failure of the compression chamber is a state in which the lubricating oil that seals the compression chamber (omitted illustration) within the compressor is insufficient, and is a state that can occur when the motor 50 is a "motor 50 mounted on a compressor".

[0306] When the physical quantity is a direct current signal, there is a tendency that if the oil seal failure of the compression chamber occurs, the frequency component corresponding to 1 times the mechanical angular frequency of the motor 50 decreases. Therefore, when the physical quantity is a direct current signal, by setting the frequency of the specific frequency component C1 to "1 times the mechanical angular frequency of the motor 50", it is possible to infer whether or not the "oil seal failure of the compression chamber" occurs.

[0307] Further, when the physical quantity is an alternating current signal, there is a tendency that if the oil seal failure of the compression chamber occurs, the frequency component corresponding to the frequency obtained by adding "1 times the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50 and the frequency component corresponding to the frequency obtained by subtracting "1 times the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50 decrease. Therefore, when the physical quantity is an alternating current signal, by setting the frequency of the specific frequency component C1 to the frequency obtained by adding "1 times the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50 or the frequency obtained by subtracting "1 times the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50, it is possible to infer whether or not the "oil seal failure of the compression chamber" occurs.

[0308] 〔2. Insulation aging of the motor〕

[0309] The insulation aging of the motor 50 is a state (abnormality) in which the insulation performance of the motor 50 is insufficient. When the insulation performance of the motor 50 is aged to a certain degree, the insulation class of the motor 50 will be lower than the allowable class.

[0310] When the physical quantity is an alternating current signal, there is a tendency that if the insulation aging of the motor 50 occurs, the frequency component corresponding to 3 times the electrical angular frequency of the motor 50 increases. Therefore, when the physical quantity is an alternating current signal, by setting the frequency of the specific frequency component C1 to "3 times the electrical angular frequency of the motor 50", it is possible to infer whether or not the "insulation aging of the motor 50" occurs.

[0311] 〔3. Bearing wear〕

[0312] Bearing wear is a state (anomaly) of the bearing of the rotating shaft or the rotating shaft rotating by the driving of the motor 50.

[0313] When the physical quantity is a direct current signal, there is a tendency that if bearing wear occurs, frequency components corresponding to 1 / 3 and 2 / 3 of the mechanical angular frequency of the motor 50 increase, respectively. Therefore, when the physical quantity is a direct current signal, by setting the frequency of the specific frequency component C1 to "at least one of 1 / 3 and 2 / 3 of the mechanical angular frequency of the motor 50", it is possible to infer whether or not "bearing wear" occurs.

[0314] Furthermore, when the physical quantity is an alternating current signal, there is a tendency that if bearing wear occurs, frequency components corresponding to a frequency obtained by adding "1 / 3 of the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50, a frequency obtained by subtracting "1 / 3 of the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50, a frequency obtained by adding "2 / 3 of the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50, and a frequency obtained by subtracting "2 / 3 of the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50 increase. Therefore, when the physical quantity is an alternating current signal, by setting the frequency of the specific frequency component C1 to a frequency obtained by adding "1 / 3 of the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50, a frequency obtained by subtracting "1 / 3 of the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50, a frequency obtained by adding "2 / 3 of the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50, and a frequency obtained by subtracting "2 / 3 of the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50, it is possible to infer whether or not "bearing wear" occurs.

[0315] 〔4. Dilution〕

[0316] Dilution is a state (anomaly) that can occur when the motor 50 is "a motor 50 mounted on a compressor", and is a state in which the oil concentration of the compressor is insufficient. If dilution occurs, the friction of the bearing portion (omitted from the drawing) in the compressor becomes large, which can cause the bearing portion to wear. Furthermore, if dilution occurs, the sliding portion (omitted from the drawing) in the compressor can generate friction due to insufficient lubricating oil, which can cause the surface roughness of the sliding portion to increase, and the like.

[0317] When the physical quantity is a direct-current signal, there is a tendency that, if dilution occurs, a frequency component corresponding to 1 time the mechanical angular frequency of the motor 50 decreases. Therefore, when the physical quantity is a direct-current signal, by setting the frequency of the specific frequency component C1 to "1 time the mechanical angular frequency of the motor 50", it is possible to infer whether or not "dilution" occurs.

[0318] Further, when the physical quantity is an alternating-current signal, there is a tendency that, if dilution occurs, a frequency component corresponding to a frequency obtained by adding "1 time the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50 and a frequency component corresponding to a frequency obtained by subtracting "1 time the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50 increase. Therefore, when the physical quantity is an alternating-current signal, by setting the frequency of the specific frequency component C1 to the frequency obtained by adding "1 time the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50 or the frequency obtained by subtracting "1 time the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50, it is possible to infer whether or not "dilution" occurs.

[0319] 〔5. Liquid strike〕

[0320] The liquid strike is a state (an abnormality) that can occur when the motor 50 is a "motor 50 mounted on a compressor", and is a state in which a liquid fluid is sucked into a compression chamber (omitted illustration) of the compressor. If the liquid strike occurs, it can cause metal fatigue and breakage of components constituting the compression chamber, and the like.

[0321] When the physical quantity is a direct-current signal, there is a tendency that, if the liquid strike occurs, a frequency component corresponding to 1 time the mechanical angular frequency of the motor 50 increases. Therefore, when the physical quantity is a direct-current signal, by setting the frequency of the specific frequency component C1 to "1 time the mechanical angular frequency of the motor 50", it is possible to infer whether or not "the liquid strike" occurs.

[0322] Further, when the physical quantity is an alternating-current signal, there is a tendency that, if the liquid strike occurs, a frequency component corresponding to a frequency obtained by adding "1 time the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50 and a frequency component corresponding to a frequency obtained by subtracting "1 time the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50 increase. Therefore, when the physical quantity is an alternating-current signal, by setting the frequency of the specific frequency component C1 to the frequency obtained by adding "1 time the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50 or the frequency obtained by subtracting "1 time the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50, it is possible to infer whether or not "the liquid strike" occurs.

[0323] 〔6. Imbalance〕

[0324] The unbalance is a state (an abnormality) that can occur when the motor 50 is "a motor 50 that drives the blades of the fan" and is a state in which the plurality of blades provided on the fan are unbalanced. For example, the unbalance is caused by the respective deformations, breakage, fouling, and frosting degrees of the plurality of blades being different.

[0325] When the physical quantity is an alternating-current signal, there is a tendency that, if the unbalance occurs, a frequency component corresponding to a frequency obtained by adding "1 times the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50 and a frequency component corresponding to a frequency obtained by subtracting "1 times the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50 increase. Therefore, when the physical quantity is an alternating-current signal, by setting the frequency of the specific frequency component C1 to the frequency obtained by adding "1 times the mechanical angular frequency of the motor 50" to the electrical angular frequency of the motor 50 or the frequency obtained by subtracting "1 times the mechanical angular frequency of the motor 50" from the electrical angular frequency of the motor 50, it is possible to infer whether or not the "unbalance" occurs.

[0326] 〔7. Flow path clogging〕

[0327] The flow path clogging is a state (an abnormality) that can occur when the motor 50 is "a motor 50 that drives the blades of the fan" and is a state in which the flow path provided with the fan is clogged. The flow path clogging is caused, for example, by a filter (not shown) provided on the flow path together with the fan being clogged. If the flow path clogging occurs, the reaction force at the time of the air supply can be disturbed by load fluctuations due to turbulence, resulting in a change in the speed.

[0328] When the physical quantity is a direct-current signal, there is a tendency that, if the flow path clogging occurs, a frequency component having a peak value near 1 times the mechanical angular frequency of the motor 50 fluctuates. Therefore, when the physical quantity is a direct-current signal, by setting the frequency of the specific frequency component C1 to "1 times the mechanical angular frequency of the motor 50", it is possible to infer whether or not the "flow path clogging" occurs.

[0329] Further, when the physical quantity is an alternating-current signal, there is a tendency that, if the flow path clogging occurs, the electrical angular frequency of the motor 50 fluctuates. Therefore, when the physical quantity is an alternating-current signal, by setting the frequency of the specific frequency component C1 to the electrical angular frequency of the motor 50, it is possible to infer whether or not the "flow path clogging" occurs.

[0330] (Detailed example of inference processing)

[0331] The following describes specific examples of the inference processing. The inference processing is, for example, the following four kinds of inference processing (first to fourth inference processing). Note that the following describes a case where the inference processing is performed based on the amplitude of the specific frequency component C1. The "specific frequency component C1" in the following description is the "amplitude of the specific frequency component C1".

[0332] 〔First Inference Processing〕

[0333] First, with reference to Figure 11 The first inference processing is described. In the first inference processing, the control section 31 infers whether the state of the device 70 is an abnormal state based on the instantaneous value of the specific frequency component C1. Specifically, the control section 31 performs the following processing (steps S11 to S13) at each predetermined detection period.

[0334] 〈Step S11〉

[0335] The control section 31 acquires the instantaneous value of the specific frequency component C1.

[0336] 〈Step S12〉

[0337] Next, the control section 31 determines whether the instantaneous value of the specific frequency component C1 acquired in step S11 is equal to or greater than a predetermined upper limit value. Note that this upper limit value is an example of a threshold value used in the inference processing. When the instantaneous value of the specific frequency component C1 is equal to or greater than the upper limit value, the processing of step S13 is performed. Otherwise, the processing of step S14 is performed.

[0338] 〈Step S13〉

[0339] When the instantaneous value of the specific frequency component C1 is equal to or greater than the upper limit value in step S12, the control section 31 infers that the state of the device 70 is an abnormal state. Also, the control section 31 outputs an inference result indicating that the state of the device 70 is an abnormal state.

[0340] 〈Step S14〉

[0341] On the other hand, when the instantaneous value of the specific frequency component C1 is not equal to or greater than the upper limit value in step S12, the control section 31 infers that the state of the device 70 is a normal state. Also, the control section 31 outputs an inference result indicating that the state of the device 70 is a normal state.

[0342] 〔Second Inference Processing〕

[0343] With reference to Figure 12 and Figure 13The second inference processing will be described. In the second inference processing, the control section 31 infers whether the state of the device 70 is an abnormal state or not, based on the rate of change of the specific frequency component Cl. Specifically, the control section 31 repeatedly performs the following processing (step S21 to step S23) at each of the preset detection periods.

[0344] <Step S21>

[0345] The control section 31 acquires the rate of change of the specific frequency component Cl. For example, the control section 31 derives the rate of change of the specific frequency component Cl by differentiating the instantaneous value of the specific frequency component Cl with respect to time. In the example of FIG. 6, the rate of change of the specific frequency component Cl changes at times tl, t2, t3, respectively. Figure 13

[0346] <Step S22>

[0347] The control section 31 determines whether the rate of change of the specific frequency component Cl acquired in step S21 is above a preset upper limit value or not. Note that this upper limit value is an example of the threshold value used in the inference processing. When the rate of change of the specific frequency component Cl is above the upper limit value, the processing of step S23 is performed. Otherwise, the processing of step S24 is performed. In the example of FIG. 6, the rate of change of the specific frequency component Cl reaches above the upper limit value at time t2. Figure 13

[0348] <Step S23>

[0349] When the rate of change of the specific frequency component Cl is above the upper limit value in step S22, the control section 31 infers that the state of the device 70 is an abnormal state. Also, the control section 31 outputs the inference result indicating that the state of the device 70 is an abnormal state.

[0350] <Step S24>

[0351] On the other hand, when the rate of change of the specific frequency component Cl is not above the upper limit value in step S22, the control section 31 infers that the state of the device 70 is a normal state. Also, the control section 31 outputs the inference result indicating that the state of the device 70 is a normal state.

[0352] 〔Third Inference Processing〕

[0353] The third inference processing will be described with reference to Figure 14 and Figure 15 In the third inference processing, the control section 31 infers whether the state of the device 70 is an abnormal state or not, based on the cumulative time which is obtained by accumulating the time during which the specific frequency component Cl is above the upper limit value. Specifically, the control section 31 repeatedly performs the following processing (step S31 to step S33) at each of the preset detection periods. ​​

[0354] <Step S31>

[0355] The control section 31 acquires the cumulative value (cumulative time) of the time when the specific frequency component Cl is equal to or greater than the upper limit value. Note that this upper limit value is an example of the threshold value used in the inference processing.

[0356] For example, the control section 31 acquires the instantaneous value of the specific frequency component Cl that becomes a representative value of the detection period, and determines whether the instantaneous value of the specific frequency component Cl is equal to or greater than the upper limit value. Also, when the instantaneous value of the specific frequency component Cl is equal to or greater than the upper limit value, the control section 31 increments the count value, and when the instantaneous value of the specific frequency component Cl is not equal to or greater than the upper limit value, the control section 31 does not increment the count value. This count value corresponds to the cumulative value (cumulative time) of the time when the specific frequency component Cl is equal to or greater than the upper limit value. For example, the value obtained by multiplying the count value by the time corresponding to the detection period is the cumulative time.

[0357] In the example of FIG. 12, the third inference processing is performed at times t1 to t12, and the count value is incremented at times t2 to t5 and t9 to t12 among the times t1 to t12. Figure 15 <Step S32>

[0358] Next, the control section 31 determines whether the cumulative time (count value in this example) acquired in step S31 exceeds a predetermined threshold value. This threshold value is an example of the threshold value used in the inference processing. When the cumulative time exceeds the threshold value, the processing of step S33 is performed, and otherwise, the processing of step S34 is performed. In the example of FIG. 12, the control section 31 determines that the count value exceeds the threshold value at time t11.

[0359] Figure 15 <Step S33>

[0360] If the cumulative time exceeds the threshold value in step S32, the control section 31 infers that the state of the device 70 is an abnormal state. Also, the control section 31 outputs the inference result indicating that the state of the device 70 is an abnormal state.

[0361] <Step S34>

[0362] On the other hand, if the cumulative time does not exceed the threshold value in step S32, the control section 31 infers that the state of the device 70 is a normal state. Also, the control section 31 outputs the inference result indicating that the state of the device 70 is a normal state.

[0363] 〔Fourth Inference Processing〕

[0364]

[0365] The following describes the fourth inference processing with reference to FIGS. 13 and 14. Figure 16 Figure 17 ​​​The fourth inference processing will be described. In the fourth inference processing, the control section 31 infers whether the state of the device 70 is an abnormal state or not, based on a ratio (time ratio) of a time in which the specific frequency component Cl is equal to or greater than a predetermined upper limit value within a predetermined judgment time. Specifically, the control section 31 repeatedly performs the following processing (steps S41 to S43) at each of predetermined detection periods.

[0366] <Step S41>

[0367] The control section 31 acquires a ratio (time ratio) of a time in which the specific frequency component Cl is equal to or greater than the upper limit value within the judgment time. Note that the upper limit value is an example of a threshold value used in the inference processing.

[0368] For example, the control section 31 stores the specific frequency component Cl within a judgment period (a period corresponding to at least the judgment time) ending at the current time. Then, the control section 31 derives a time in which the specific frequency component Cl is equal to or greater than the upper limit value (upper limit value exceeding time) within a judgment period ending at the current time and having a period length corresponding to the judgment time, from the stored specific frequency component Cl, and derives the time ratio by dividing the upper limit value exceeding time by the judgment time. Note that in the example of FIG. 10, the time t2 is the current time, the period from the time tl to the time t2 is the judgment period, the time Tl and the time T2 are the upper limit value exceeding times, and a value obtained by dividing the sum of the time Tl and the time T2 by the judgment time T0 is the time ratio. Figure 17

[0369] <Step S42>

[0370] Next, the control section 31 determines whether the time ratio (ratio of a time in which the specific frequency component Cl is equal to or greater than the upper limit value within the judgment time) acquired in step S41 exceeds a predetermined threshold value. Note that the threshold value is an example of a threshold value used in the inference processing. When the time ratio exceeds the threshold value, the processing of step S43 is performed. Otherwise, the processing of step S44 is performed.

[0371] <Step S43>

[0372] When the time ratio exceeds the threshold value in step S42, the control section 31 infers that the state of the device 70 is an abnormal state. Then, the control section 31 outputs an inference result indicating that the state of the device 70 is an abnormal state.

[0373] <Step S44>

[0374] On the other hand, when the time ratio does not exceed the threshold value in step S42, the control section 31 infers that the state of the device 70 is a normal state. Then, the control section 31 outputs an inference result indicating that the state of the device 70 is a normal state. ​

[0375] Note that, in the explanation of the above inference processing, a case where the specific frequency component C1 is compared with the upper limit value is exemplified, but the present application is not limited to this. For example, the specific frequency component C1 can be compared with a preset lower limit value. That is, the above "upper limit value or more" can be replaced with "lower limit value or less". Further, the specific frequency component C1 can be compared with a preset allowable range. That is, the above "upper limit value or more" can be replaced with "outside the allowable range".

[0376] (Specific example of the first restriction processing)

[0377] A specific example of the first restriction processing will be described below. The "processing for restricting the inference processing" performed in the first restriction processing is exemplified by the following five inference restriction processes (first to fifth inference restriction processes).

[0378] (First inference restriction process)

[0379] First, the first inference restriction process will be described. The first inference restriction process is a process of prohibiting (stopping) the inference processing. In the first inference restriction process, the control section 31 stops the inference processing, and outputs a preset inference result. In this way, the inference processing is directly restricted so as to output the preset inference result. As such, since it is possible to avoid obtaining an erroneous inference result, it is possible to suppress erroneous inference of the state of the device 70.

[0380] (Second inference restriction process)

[0381] Next, the second inference restriction process will be described. The second inference restriction process is a process of correcting the inference result obtained by the inference processing. In the second inference restriction process, the control section 31 continues the inference processing, while correcting the inference result obtained by the inference processing to a "preset inference result". In this way, the inference processing is directly restricted so as to output the preset inference result. As such, since it is possible to avoid obtaining an erroneous inference result, it is possible to suppress erroneous inference of the state of the device 70.

[0382] (Third inference restriction process)

[0383] Next, the third inference restriction process will be described. The third inference restriction process is a process of prohibiting (stopping) the calculation processing. In the third inference restriction process, the control section 31 stops the calculation processing, and sets the specific frequency component C1 used in the inference processing to a "preset specific frequency component C1" so as to obtain a preset inference result in the inference processing. In this way, the inference processing is indirectly restricted so as to output the preset inference result. As such, since it is possible to avoid obtaining an erroneous inference result, it is possible to suppress erroneous inference of the state of the device 70.

[0384] 〔Fourth inference restriction processing〕

[0385] Next, the fourth inference restriction processing will be described. The fourth inference restriction processing is processing of correcting the specific frequency component C1 calculated by the calculation processing. In the fourth inference restriction processing, the control section 31 continues the calculation processing while correcting the specific frequency component C1 calculated by the calculation processing to "a preset specific frequency component C1" so as to obtain a preset inference result in the inference processing. In this way, the inference processing is indirectly restricted so as to output a preset inference result. As such, since it is possible to avoid obtaining an erroneous inference result, it is possible to suppress erroneous inference of the state of the device 70.

[0386] 〔Fifth inference restriction processing〕

[0387] Next, the fifth inference restriction processing will be described. The fifth inference restriction processing is processing performed when the control section 31 infers the state of the device 70 in the inference processing based on a comparison result of the amplitude of the specific frequency component C1 and a preset threshold, and is processing of correcting at least one of the threshold and the amplitude of the specific frequency component C1. In the fifth inference restriction processing, the control section 31 corrects the threshold used in the inference processing to "a preset threshold" so as to obtain a preset inference result in the inference processing. Alternatively, in the fifth inference restriction processing, the control section 31 corrects the specific frequency component C1 used in the inference processing to "a preset specific frequency component C1" so as to obtain a preset inference result in the inference processing. In this way, the inference processing is directly restricted so as to output a preset inference result. As such, since it is possible to avoid obtaining an erroneous inference result, it is possible to suppress erroneous inference of the state of the device 70.

[0388] (Specific example of the second restriction processing)

[0389] Next, a specific example of the second restriction processing will be described. The "processing for restricting the operation of the motor 50" performed in the second restriction processing has, for example, the following two operation restriction processes (first and second operation restriction processes).

[0390] 〔First operation restriction processing〕

[0391] First, the first operation restriction processing will be described. The first operation restriction processing is processing of changing the operation condition of the motor 50 so as not to make a false inference in the inference processing. The processing of changing the operation condition of the motor 50 is processing of changing at least one of the carrier frequency, the sampling frequency at which the physical quantity is sampled in the calculation processing, and the mechanical angular frequency of the motor 50. In the first operation restriction processing, the control section 31 changes the operation condition of the motor 50 including at least one of the carrier frequency, the sampling frequency, and the mechanical angular frequency of the motor 50 so as not to make a false inference in the inference processing.

[0392] As such, by changing the operation condition of the motor 50, it is possible to make the noise frequency component Cn and the specific frequency component Cl not satisfy the prescribed relationship. In this way, it is possible to suppress fluctuation in the magnitude of the specific frequency component Cl due to the influence of the noise frequency component Cn, and thus it is possible to suppress false inferences about the state of the device 70.

[0393] 〔Second operation restriction processing〕

[0394] Next, the second operation restriction processing will be described. The second operation restriction processing is processing of prohibiting the motor 50 from operating in an operation condition in which the noise frequency component Cn and the specific frequency component Cl satisfy the prescribed relationship. In the second operation restriction processing, when the operation condition of the motor 50 is "an operation condition in which the noise frequency component Cn and the specific frequency component Cl satisfy the prescribed relationship", the control section 31 prohibits the motor 50 from operating, and otherwise, the control section 31 allows the motor 50 to operate.

[0395] As such, by prohibiting the motor 50 from operating in an operation condition in which the noise frequency component Cn and the specific frequency component Cl satisfy the prescribed relationship, it is possible to avoid obtaining a false inference result. In this way, it is possible to suppress false inferences about the state of the device 70.

[0396] (First embodiment variant)

[0397] In the drive system 10 of the first embodiment variant, the restriction processing performed by the control section 31 is different from that of the drive system 10 of the first embodiment. The other configurations and processing of the drive system 10 of the first embodiment variant are the same as those of the drive system 10 of the first embodiment.

[0398] In the restriction processing of the modification example of the first embodiment, the control section 31 restricts the inference processing or the operation of the motor 50 when the noise frequency component Cn and the specific frequency component Cl satisfy a prescribed relationship. Also, if a prescribed restriction release condition is satisfied after the control section 31 starts to perform the restriction on the inference processing or the operation of the motor 50, the control section 31 releases the restriction on the inference processing or the operation of the motor 50 after a prescribed standby release condition is satisfied. The standby release condition will be described later in detail.

[0399] As shown in Fig. 10, in the restriction processing of the modification example of the first embodiment, step S110 is performed between step S103 and step S104. Figure 18

[0400] 〈Step S110〉

[0401] In step S103, when the prescribed restriction release condition is satisfied, the control section 31 determines whether the standby release condition is satisfied. When the standby release condition is satisfied, the processing of step S104 is performed. For example, the processing of step S110 is performed at every prescribed determination time. In other words, the processing of step S110 is repeatedly performed until the standby release condition is satisfied.

[0402] 〔Effects of the modification example of the first embodiment〕

[0403] In the modification example of the first embodiment, the same effects as those of the first embodiment can be obtained.

[0404] Further, in the modification example of the first embodiment, compared with a case where the restriction (the restriction on the inference processing or the restriction on the motor 50) is released after the restriction release condition is satisfied without waiting for the standby release condition to be satisfied, it is possible to release the restriction after the specific frequency component Cl used for the inference processing is stabilized. In this way, it is possible to suppress erroneous inference of the state of the device 70 after the restriction is released.

[0405] (Specific examples of the standby release condition)

[0406] Specific examples of the standby release condition will be described below. The standby release condition has, for example, the following three standby release conditions (first to third standby release conditions).

[0407] 〔First standby release condition〕

[0408] ​First, the first standby cancel condition will be described. The first standby cancel condition is a condition that a prescribed time elapses from the time when the restriction cancel condition is established. When the restriction cancel condition is established, the control section 31 cancels the restriction on the inference processing or the operation of the motor 50 after the prescribed time elapses. Specifically, when the restriction cancel condition is established, the control section 31 starts measuring the elapsed time. And when the elapsed time reaches the prescribed time, the control section 31 cancels the restriction on the inference processing or the operation of the motor 50. Note that the prescribed time is set to, for example, a time longer than the time required for the characteristic quantity that infers the state of the device to change from the unstable state to the stable state. The characteristic quantity can be, for example, the specific frequency component Cl as long as it is a characteristic quantity calculated from the physical quantity obtained from the device 70 within a prescribed period.

[0409] 〔Second standby cancel condition〕

[0410] Next, the second standby cancel condition will be described. The second standby cancel condition is a condition that the number of consecutive times that the restriction cancel condition is established reaches a prescribed number. The control section 31 repeatedly judges whether the restriction cancel condition is established, and when the restriction cancel condition is established, cancels the restriction on the inference processing or the operation of the motor 50 after the number of consecutive times that the restriction cancel condition is established reaches the prescribed number. Specifically, when the restriction cancel condition is established, the control section 31 starts measuring the number of consecutive times that the restriction cancel condition is established. And when the number of consecutive times that the restriction cancel condition is established reaches the prescribed number, the control section 31 cancels the restriction on the inference processing or the operation of the motor 50. Note that the prescribed number is set to, for example, a number larger than the number of consecutive times that the restriction cancel condition is established measured until the characteristic quantity changes from the unstable state to the stable state.

[0411] 〔Third standby cancel condition〕

[0412] Next, the third standby cancel condition will be described. The third standby cancel condition is a condition that the specific frequency component Cl calculated by the calculation processing reaches the stable state. The control section 31 monitors the specific frequency component Cl calculated by the calculation processing, and when the restriction cancel condition is established, cancels the restriction on the inference processing or the operation of the motor 50 after the specific frequency component Cl calculated by the calculation processing reaches the stable state. Specifically, when the prescribed restriction cancel condition is established, the control section 31 judges whether the specific frequency component Cl calculated by the calculation processing is in the stable state. And when the control section 31 judges that the specific frequency component Cl calculated by the calculation processing is in the stable state, cancels the restriction on the inference processing or the operation of the motor 50.

[0413] Note that the stable state of the specific frequency component C1 is, for example, a state in which a ratio of a current value to a previous value of a moving average value of the specific frequency component C1 within a prescribed time is within a prescribed ratio. For example, the moving average value within the prescribed time is a moving average value within 1 second, the previous value is a moving average value acquired 1 second before (preceding) the current value, and the prescribed ratio is 1%. The unstable state of the specific frequency component C1 is a state in which the ratio of the current value to the previous value of the moving average value of the specific frequency component C1 within the prescribed time is higher than the prescribed ratio.

[0414] (Second Embodiment)

[0415] In the drive system 10 of the second embodiment, the processing performed by the control section 31 of the control device 30 is different from that of the drive system 10 of the first embodiment. The other configurations and processing of the drive system 10 of the second embodiment are the same as those of the drive system 10 of the first embodiment.

[0416] 〔Processing Performed by Control Section〕

[0417] In the second embodiment, the control section 31 performs control processing, calculation processing, inference processing, determination processing, restriction processing, and response processing. The control processing, the calculation processing, the inference processing, and the response processing are the same as those of the first embodiment.

[0418] 〔Determination Processing〕

[0419] In the determination processing, it is determined whether or not there is noise interference on the basis of the size of a periodic component included in a variable calculated from the size of the specific frequency component C1. The "variable calculated from the size of the specific frequency component C1" is, for example, a value obtained by dividing a short-term moving average value of the specific frequency component C1 by a long-term moving average value. The short-term moving average value is a moving average value of the specific frequency component C1 within a first time. The long-term moving average value is a moving average value of the specific frequency component C1 within a second time. The second time is longer than the first time.

[0420] Other examples of the "variable calculated from the size of the specific frequency component C1" include an instantaneous value of the specific frequency component C1, a square root of a sum of squares of the specific frequency component C1, a rate of change of the specific frequency component C1, a ratio of the short-term moving average value to the long-term moving average value of the specific frequency component C1, arithmetic operations among a plurality of specific frequency components C1, power operations of the specific frequency component C1, a combination of at least two of the above, and the like.

[0421] Note that when there is noise interference (specifically, when a noise component affects the specific frequency component C1), the size (amplitude) of the specific frequency component C1 fluctuates. Furthermore, when there is noise interference, the periodic component included in the variable calculated from the size of the specific frequency component C1 becomes large.

[0422] For example, when the periodic component is contained in the specific frequency component C1, the control section 31 determines that there is noise disturbance; when the periodic component is not contained in the specific frequency component C1, the control section 31 determines that there is no noise disturbance. Note that the specific example of the determination processing will be described later in detail.

[0423] [Limiting Processing]

[0424] In the limiting processing, when the control section 31 determines that there is noise disturbance in the determination processing, the control section 31 limits the inference processing or the operation of the motor 50. The limiting processing is the first limiting processing or the second limiting processing, in the first limiting processing, the inference processing is limited so as to output a predetermined inference result; in the second limiting processing, the operation of the motor 50 is limited so as not to make a false inference in the inference processing.

[0425] [Flow of Processing by Control Section]

[0426] Next, the flow of the processing (specifically, the calculation processing, the inference processing, the determination processing, and the limiting processing) by the control section 31 will be described with reference to Figure 19 The control section 31 performs the calculation processing at each predetermined calculation time. By repeatedly performing the calculation processing, the specific frequency component C1 is calculated at each calculation time.

[0427] Then, the control section 31 performs the inference processing based on the specific frequency component C1 calculated by the calculation processing at each predetermined inference time. By repeatedly performing the inference processing, an inference result is obtained at each inference time.

[0428] Further, the control section 31 performs the determination processing in parallel with the calculation processing and the inference processing. Specifically, the control section 31 performs the determination processing based on the specific frequency component C1 calculated by the calculation processing at each predetermined determination time. By repeatedly performing the determination processing, a determination result is obtained at each determination time. The determination time is set to a time or more required for determining the presence or absence of noise disturbance based on the specific frequency component C1 calculated by the calculation processing.

[0429] Further, the control section 31 performs the limiting processing in parallel with the above-described processing (the calculation processing, the inference processing, and the determination processing). Specifically, the control section 31 repeatedly performs the limiting processing based on the processing of steps S201 to S204 shown in Fig. 6.

[0430] Figure 19 The processing of steps S201 to S204 shown in Fig. 6 is repeatedly performed.

[0431] <Step S201>

[0432] ​The control section 31 acquires the determination result of the determination processing. Then, the control section 31 determines whether there is noise interference, based on the determination result of the determination processing. When there is noise interference, the processing of step S202 is performed. For example, the processing of step S201 is performed every prescribed determination time. In other words, the processing of step S201 is repeatedly performed until a determination result indicating that there is noise interference is obtained.

[0433] <Step S202>

[0434] When there is noise interference, the control section 31 restricts the inference processing or the operation of the motor 50.

[0435] <Step S203>

[0436] Next, the control section 31 determines whether a restriction release condition is satisfied. The restriction release condition includes, for example, that there is no noise interference, that the control section 31 receives a release instruction, and the like. The release instruction is automatically transmitted to the control section 31, for example, after a prescribed time elapses from when the restriction release condition is satisfied.

[0437] When the restriction release condition is satisfied, the processing of step S204 is performed. For example, the processing of step S203 is performed every prescribed determination time. In other words, the processing of step S203 is repeatedly performed until the restriction release condition is satisfied.

[0438] <Step S204>

[0439] When the restriction release condition is satisfied, the control section 31 releases the restriction (restriction on the inference processing or restriction on the operation of the motor 50).

[0440] Effects of the Second Embodiment

[0441] As described above, in the second embodiment, the control section 31 determines whether there is noise interference, based on the size of the periodic component included in the variable calculated from the size of the specific frequency component Cl, in the determination processing. Also, when the control section 31 determines that there is noise interference in the determination processing, the control section 31 restricts the inference processing or the operation of the motor 50 in the restriction processing.

[0442] In the above-described configuration, when a noise component affects the specific frequency component Cl, the size (amplitude) of the specific frequency component Cl fluctuates. Therefore, in the determination processing, it is possible to appropriately determine whether there is noise interference, based on the size of the periodic component included in the variable calculated from the size of the specific frequency component Cl. Also, in the restriction processing, by restricting the inference processing or the operation of the motor 50 based on the determination result of the determination processing, it is possible to suppress erroneous inference of the state of the device 70.

[0443] Note that in the second embodiment, the influence of noise frequency components other than the "maximum common divisor (a maximum common divisor of two or more) and an integer multiple thereof disclosed in Patent Literature 1" can be taken into account. In this way, the erroneous estimation of the state of the device 70 can be appropriately suppressed.

[0444] (Details of the judgment processing)

[0445] Details of the judgment processing will be described below. The judgment processing includes, for example, the following seven kinds of judgment processing (first to seventh judgment processing).

[0446] [First judgment processing]

[0447] First, the judgment processing is described with reference to FIG. 7. Figure 20 The first judgment processing will be described. In the first judgment processing, the control section 31 judges the presence or absence of noise interference on the basis of the average value of the absolute value of the fluctuation amount (fluctuation amount per unit time) of the specific frequency component Cl within a prescribed time.

[0448] Specifically, the control section 31 acquires the instantaneous value of the specific frequency component Cl at each prescribed time t. Then, the control section 31 derives the fluctuation amount per unit time of the specific frequency component Cl within the prescribed time by dividing the difference between the instantaneous value U(k) of the specific frequency component Cl acquired at the kth (k is a natural number) time and the instantaneous value U(k+1) of the specific frequency component Cl acquired at the (k+1)th time by the time t. In this way, the fluctuation amount per unit time of the specific frequency component Cl within the prescribed time can be derived at each prescribed time t. k k+1

[0449] The control section 31 derives the average value of the fluctuation amount (fluctuation amount per unit time) of the specific frequency component Cl within the prescribed time by averaging the absolute value of the "fluctuation amount per unit time of the specific frequency component Cl within the prescribed time" derived at each prescribed time t. The averaging period (period for deriving the average value) is preferably set to a time period equal to or longer than the period of fluctuation of the specific frequency component Cl assumed in the presence of noise interference. For example, the averaging period is set to 3 minutes. Then, when the average value of the fluctuation amount of the specific frequency component Cl within the prescribed time derived exceeds a threshold value, the control section 31 judges that there is noise interference, and otherwise, the control section 31 judges that there is no noise interference. For example, by setting the threshold value to "0.24% of the average value of the amplitude of the specific frequency component Cl within the averaging period", when the specific frequency component Cl fluctuates by 10% or more with a period of 3 minutes, it can be judged that "there is noise interference".

[0450] ​​Note that the average of the fluctuation amount (fluctuation amount per unit time) of the specific frequency component C1 in the prescribed time can be derived using various filters such as a low-pass filter, or can be derived using various processes such as a moving average.

[0451] 〔Second determination process〕

[0452] Next, the second determination process will be described with reference to Figure 21 The second determination process will be described. In the second determination process, the control section 31 determines the presence or absence of noise interference based on the magnitude of the periodic component included in the specific frequency component C1.

[0453] Specifically, the control section 31 frequency-converts the specific frequency component C1, and extracts the "periodic component having the largest amplitude" among the periodic components of the specific frequency component C1. Then, the control section 31 determines the presence or absence of noise interference based on the magnitude (amplitude) of the periodic component of the specific frequency component C1. Specifically, when the amplitude of the "periodic component having the largest amplitude" of the specific frequency component C1 exceeds a threshold value, the control section 31 determines that noise interference is present, and otherwise, the control section 31 determines that noise interference is not present. For example, by setting the threshold value to "10% of the average value of the prescribed time", when there is a periodic component having a fluctuation of 10% or more in the specific frequency component C1, it is possible to determine that "noise interference is present".

[0454] 〔Third determination process〕

[0455] Next, the third determination process will be described with reference to Figure 22 The third determination process will be described. In the third determination process, the control section 31 determines the presence or absence of noise interference based on the magnitude of the standard deviation of the specific frequency component C1 in the prescribed time. Note that the prescribed time is preferably set to a time period longer than the period of fluctuation of the specific frequency component C1 assumed in the case where noise interference is present. For example, the prescribed time is 3 minutes.

[0456] As shown in Figure 22 the distribution of the specific frequency component C1 in the prescribed time has the following tendency: in the case where noise interference is not present (in the case where the influence of noise interference is relatively small), the distribution is a distribution in which the amplitude values, i.e., data values, of the specific frequency component C1 are concentrated around the average value; and in the case where noise interference is present (in the case where the influence of noise interference is relatively large), the distribution is a distribution in which the data values are not concentrated around the average value but are dispersed.

[0457] Specifically, in the third judgment process, the control unit 31 derives the standard deviation of the specific frequency component C1 within a specified time period based on multiple data values ​​(amplitude values ​​of the specific frequency component C1) obtained within that specified time period. Then, when the standard deviation of the specific frequency component C1 within the specified time period exceeds a threshold, the control unit 31 determines that there is noise interference; otherwise, the control unit 31 determines that there is no noise interference. For example, the threshold is 7% of the average amplitude of the specific frequency component C1 within the specified time period. In this case, when there is a periodic component with fluctuations of more than 10% in the specific frequency component, it is possible to determine that "there is noise interference".

[0458] It should be noted that in the third judgment process, the control unit 31 can also determine whether there is noise interference based on the moving average of the standard deviation of the specific frequency component C1 within a specified time. Specifically, the control unit 31 can also determine that there is noise interference when the moving average of the standard deviation of the specific frequency component C1 within a specified time exceeds a threshold; otherwise, the control unit 31 determines that there is no noise interference.

[0459] [Fourth Judgment Processing]

[0460] Below, refer to Figure 23 The fourth judgment process will be explained. In the fourth judgment process, the control unit 31 determines whether there is noise interference based on the kurtosis of the distribution of a specific frequency component C1 within a specified time. It should be noted that the specified time is preferably set to a time longer than the period of the fluctuation of the specific frequency component C1 under the assumption of noise interference.

[0461] like Figure 23 As shown, the distribution of a specific frequency component C1 within a specified time period has the following trends: in the absence of noise interference (when the influence of noise interference is relatively small), the kurtosis of the distribution is relatively high; in the presence of noise interference (when the influence of noise interference is relatively large), the kurtosis of the distribution is relatively low.

[0462] Specifically, in the fourth determination process, the control unit 31 derives the kurtosis of the distribution of the specific frequency component C1 within a specified time period based on multiple data values ​​(amplitude values ​​of the specific frequency component C1) obtained within a specified time period. Then, when the kurtosis of the distribution of the specific frequency component C1 within the specified time period is lower than a specified value, the control unit 31 determines that there is noise interference; otherwise, the control unit 31 determines that there is no noise interference. For example, the specified value (the threshold for kurtosis) is "-1.5".

[0463] [Fifth Judgment and Processing]

[0464] Below, refer to Figure 24The fifth determination processing will be described. In the fifth determination processing, the control section 31 determines the presence or absence of noise disturbance on the basis of the cumulative value of the absolute value of the difference between the specific frequency component Cl and the average value in a prescribed time. The average value is the average value of the specific frequency component Cl in the prescribed time. In the example shown in FIG. 12, the cumulative value of the absolute value of the difference between the specific frequency component Cl and the average value in the prescribed time corresponds to the area of the hatched region. Note that the prescribed time is preferably set to a time period longer than the period of fluctuation of the specific frequency component Cl assumed in the presence of noise disturbance. For example, the prescribed time is 3 minutes. Figure 24

[0465] Specifically, in the fifth determination processing, the control section 31 derives the average value of the specific frequency component Cl in the prescribed time on the basis of a plurality of data values (amplitude values of the specific frequency component Cl) obtained in the prescribed time. Then, the control section 31 derives the cumulative value of the absolute value of the difference between each of the data values (amplitude values of the specific frequency component Cl) obtained in the prescribed time and the average value by summing the absolute values, and thereby derives the cumulative value of the absolute value of the difference between the specific frequency component Cl and the average value in the prescribed time. Then, when the cumulative value of the absolute value of the difference between the specific frequency component Cl and the average value in the prescribed time exceeds a threshold value, the control section 31 determines the presence of noise disturbance, and otherwise, the control section 31 determines the absence of noise disturbance. For example, when the absolute value of the difference between each of the data values (amplitude values of the specific frequency component Cl) obtained in the prescribed time and the average value is calculated every second and summed, the threshold value is set to "11% of the average value of the amplitude of the specific frequency component Cl in the prescribed time". In this case, when there is a periodic component having a fluctuation of 10% or more in the specific frequency component Cl, it is possible to determine "the presence of noise disturbance".

[0466] 〔Sixth determination processing〕

[0467] Next, the sixth determination processing will be described with reference to FIG. 13. Figure 25 In the sixth determination processing, the control section 31 determines the presence or absence of noise disturbance on the basis of the cumulative value of the amplitude of the frequency component other than 0 Hz and within a prescribed bandwidth in the frequency spectrum of the periodic component included in the specific frequency component Cl. In the example shown in FIG. 13, the cumulative value of the amplitude of the frequency component within the prescribed bandwidth in the frequency spectrum of the periodic component included in the specific frequency component Cl corresponds to the area of the hatched region. Figure 25

[0468] ​​Specifically, in the sixth determination processing, the control section 31 extracts a periodic component of the specific frequency component Cl by performing frequency conversion on the specific frequency component Cl. Then, the control section 31 derives a cumulative value of the amplitude values of the frequencies of the periodic component by summing the amplitude values of the frequencies of the periodic component of the specific frequency component Cl within a prescribed bandwidth. Then, when the cumulative value of the amplitude values of the frequencies of the periodic component of the specific frequency component Cl exceeds a threshold value, the control section 31 determines that there is noise interference, and otherwise, the control section 31 determines that there is no noise interference. For example, the threshold value is 10% of the average value of the amplitudes of the specific frequency component Cl within a prescribed time. In this case, when there is a periodic component having a fluctuation of 10% or more in the specific frequency component Cl, it is possible to determine "there is noise interference".

[0469] [Seventh determination processing]

[0470] Next, the seventh determination processing will be described with reference to Figure 26 The seventh determination processing will be described. In the seventh determination processing, the control section 31 determines the presence or absence of noise interference based on the magnitude of the difference between the maximum value and the minimum value of the specific frequency component Cl within a prescribed period TP. Note that the prescribed period TP is preferably set to a time period longer than the period of fluctuation of the specific frequency component Cl assumed in the case where there is noise interference. For example, the prescribed period TP is 3 minutes.

[0471] Specifically, the control section 31 acquires the instantaneous values of the specific frequency component Cl within the prescribed period TP and derives the difference between the maximum value and the minimum value of the specific frequency component Cl within the prescribed period TP. In this way, it is possible to derive the difference between the maximum value and the minimum value of the specific frequency component Cl within each prescribed period TP. Also, when the difference between the maximum value and the minimum value of the specific frequency component Cl within the prescribed period TP exceeds a threshold value, the control section 31 determines that there is noise interference, and otherwise, the control section 31 determines that there is no noise interference. For example, by setting the threshold value to "20% of the average value of the specific frequency component Cl within the prescribed period TP", it is possible to determine "there is noise interference" when there is a periodic component having a fluctuation of 10% or more in the specific frequency component Cl.

[0472] (Modified example of the second embodiment)

[0473] In the drive system 10 of the modified example of the second embodiment, the restriction processing performed by the control section 31 is different from that of the drive system 10 of the second embodiment. The other configurations and processing of the drive system 10 of the modified example of the second embodiment are the same as those of the drive system 10 of the second embodiment.

[0474] In the restriction processing of the modification example of the second embodiment, when the control section 31 determines that there is noise interference in the determination processing, the control section 31 restricts the estimation processing or the operation of the motor 50. Also, if a prescribed restriction release condition is satisfied after the control section 31 starts to restrict the estimation processing or the operation of the motor 50, the control section 31 releases the restriction of the estimation processing or the operation of the motor 50 after the prescribed standby release condition is satisfied. The standby release condition in the modification example of the second embodiment is the same as the standby release condition in the modification example of the first embodiment.

[0475] 〔Effects of the modification example of the second embodiment〕

[0476] In the modification example of the second embodiment, the same effects as those of the second embodiment can be obtained.

[0477] Further, in the modification example of the second embodiment, the restriction (restriction of the estimation processing or restriction of the motor 50) can be released after the specific frequency component Cl for the estimation processing is stabilized, compared with the case where the restriction is released after the prescribed restriction release condition is satisfied without waiting for the standby release condition to be satisfied. In this way, it is possible to suppress erroneous estimation of the state of the device 70 after the restriction is released.

[0478] (Refrigeration system)

[0479] Figure 27 A configuration example of a refrigeration system RR is shown. The refrigeration system RR includes a refrigerant circuit RR1 in which a refrigerant is filled, a motor drive device 20, and a control device 30. Figure 27 The motor drive device 20 and the control device 30 shown are the motor drive device 20 and the control device 30 in the first embodiment or the second embodiment.

[0480] The refrigerant circuit RR1 has a compressor CC, a heat radiator RR5, a pressure reducing mechanism RR6, and an evaporator RR7. In this example, the pressure reducing mechanism RR6 is an expansion valve. The refrigerant circuit RR1 performs a vapor compression type refrigeration cycle.

[0481] The compressor CC has a compression mechanism Cca and a motor 50. The compression mechanism Cca is connected to the motor 50 through a rotating shaft. The motor 50 drives the compression mechanism Cca to rotate by driving the rotating shaft to rotate. The motor drive device 20 drives the motor 50. The control device 30 determines the state of the refrigeration system RR in which the compressor CC having the motor 50 is installed.

[0482] In the refrigeration cycle, the refrigerant discharged from the compressor CC is heat- released in the heat radiator RR5. The refrigerant flowing out of the heat radiator RR5 is depressurized in the pressure-reducing mechanism RR6 and evaporated in the evaporator RR7. And the refrigerant flowing out of the evaporator RR7 is sucked into the compressor CC.

[0483] In this example, the refrigeration system RR is an air conditioner. The air conditioner can be a refrigeration-only machine, or a heating-only machine. In addition, the air conditioner can also be an air conditioner that switches between refrigeration and heating. In this case, the air conditioner has a switching mechanism (for example, a four-way reversing valve) that switches the direction of circulation of the refrigerant. In addition, the refrigeration system RR can also be a water heater, a cooling unit, a cooling device that cools the air in a box, and the like. The cooling device cools the internal air of a refrigeration chamber, a freezer, a container, and the like.

[0484] (Fan system)

[0485] Figure 28 A configuration example of the fan system FF is shown. The fan system FF includes a fan FF1, the motor drive device 20, and the control device 30. Figure 28 The motor drive device 20 and the control device 30 shown are the motor drive device 20 and the control device 30 in the first embodiment or the second embodiment.

[0486] The fan FF1 is provided in a flow path (omitted from the drawing). The fan FF1 has a rotating body FFa having a blade and a motor 50. The rotating body FFa is connected to the motor 50 through a rotating shaft. The motor 50 drives the rotating body FF2a to rotate by driving the rotating shaft to rotate. The motor drive device 20 drives the motor 50. The control device 30 infers the state of the fan system FF in which the motor 50 is installed. The state of the fan system FF inferred by the control device 30 includes a state in which the flow path in which the fan FF1 is provided is clogged.

[0487] (Other embodiments)

[0488] In the above description, the state of the device 70 can be a state of the entire device 70, or a state of a part of the components included in the device 70. For example, when the device 70 is an "outdoor unit in which the compressor CC having the motor 50 is installed", the state of the device 70 can be a state of the outdoor unit, can be a state of the compressor CC, can be a state of the motor 50, or can be a state of other components of the outdoor unit.

[0489] In the above description, the signal based on the current or voltage of the motor 50 is cited as an example of the physical quantity obtained by the device 70, but is not limited thereto. For example, the physical quantity can be a signal indicating the vibration of the device 70, or a signal indicating the sound of the device 70. The signal indicating the vibration of the device 70 can also be acquired by a vibration sensor (omitted from the drawing) provided on the device 70. The signal indicating the sound of the device 70 can also be acquired by a sound sensor (omitted from the drawing) provided on the device 70.

[0490] Further, in the above description, the case where it is inferred in the inference processing whether or not the device 70 has an abnormality is cited as an example, but is not limited thereto.

[0491] For example, the control section 31 can also infer the degree of wear of a rotating shaft (omitted from the drawing) or a bearing (omitted from the drawing) of the rotating shaft that is rotated by the motor 50 in the inference processing. In this case, the control section 31 can also infer the degree of wear from the specific frequency component Cl, for example, the larger the specific frequency component Cl obtained by the calculation processing, the higher the degree of wear of the rotating shaft or the bearing. The "preset inference result" output when the inference processing of inferring the degree of wear is restricted is, for example, an inference result indicating that the degree of wear of the device 70 is unknown, an inference result indicating that the inference processing is prohibited (stopped), or the like. The degree of wear is an example of the degree of deterioration.

[0492] Further, the control section 31 can also infer the remaining life of the motor 50 in the inference processing. In this case, the control section 31 can also infer the remaining life from the characteristic quantity, for example, the larger the characteristic quantity obtained by the calculation processing, the shorter the remaining life of the motor 50. The "preset inference result" output when the inference processing of inferring the remaining life is restricted is, for example, an inference result indicating that the remaining life of the device 70 is unknown, an inference result indicating that the inference processing is prohibited (stopped), or the like. The remaining life is an example of the degree of deterioration.

[0493] Further, in the above description, in the calculation processing, a band-pass filter that extracts a signal of a frequency band including the frequency of the specific frequency component Cl can also be used. Further, in the calculation processing, a band-pass filtering process that extracts a signal of a frequency band including the frequency of the specific frequency component Cl can also be performed.

[0494] Further, in the above description, in extracting the periodic component included in the specific frequency component Cl, various filters such as a Kalman filter can also be used. Further, in extracting the periodic component included in the specific frequency component Cl, various frequency analysis processes such as wavelet analysis can also be used. The same applies to the calculation (extraction) of the specific frequency component Cl.

[0495] Further, in the above description, the instantaneous value of the specific frequency component C1, the square root of the sum of squares of the specific frequency component C1, the rate of change of the specific frequency component C1, the ratio of the short-term moving average value to the long-term moving average value of the specific frequency component C1, the arithmetic operation between a plurality of specific frequency components C1, the power operation of the specific frequency component C1, a combination of at least two of the above, and the like can be used as the index value, and the inference processing can be performed based on the index value. Further, the inference processing can be performed based on the duration for which the index value exceeds the threshold value, the number of times the index value exceeds the threshold value, and the like. Further, the inference processing can be performed based on a combination of the specific frequency component C1 and another frequency component having a characteristic (for example, a frequency component corresponding to an integral multiple of the mechanical angular frequency of the motor 50).

[0496] Further, in the above description, the specific frequency component C1 can be a frequency component corresponding to a single frequency, or can be a frequency component corresponding to a frequency range centered on a single frequency.

[0497] Further, in the above description, when the physical quantity is an alternating current signal, the inference processing can be performed based on either of two specific frequency components C1 symmetrical about the electrical angular frequency of the motor 50, or the inference processing can be performed based on both of the two specific frequency components C1.

[0498] Further, in the above description, the physical quantity can be a measured value, an instructed value, or an inferred value.

[0499] Further, in the above description, the control unit 31 can be configured to perform the judgment processing using an algorithm (an algorithm for judging the state based on the change in the signal) constructed by a neural network or machine learning.

[0500] Further, in the above description, the control unit 31 can be implemented by one processor, or can be implemented by a plurality of processors. Further, the control unit 31 can be implemented by a plurality of arithmetic processing devices (computers) that communicate with each other via a communication network.

[0501] The embodiments and modified examples have been described above, but it should be understood that various changes can be made to the configurations and details thereof without departing from the gist and scope of the claims. The components related to the above-described embodiments, modified examples, and other embodiments can be appropriately combined or replaced.

[0502] -Industrial Applicability-

[0503] As described above, the present disclosure is useful as a state inference technology.

[0504] -Symbol Explanation-

[0505] 10 Drive system

[0506] 20 motor drive device

[0507] 21 AC / DC converter

[0508] 22 DC portion

[0509] 23 AC / DC converter

[0510] 30 control device (state inferring device)

[0511] 31 control portion

[0512] 41 phase current detecting portion

[0513] 42 electric angle frequency detecting portion

[0514] 50 motor

[0515] 60 power supply

[0516] 70 apparatus

[0517] CC compressor

[0518] RR refrigeration system

[0519] RR1 refrigerant circuit

[0520] FF fan system

[0521] FF1 fan

Claims

1. A state inferring apparatus that infers a state of an installation (70) that mounts an electric motor (50) driven by a direct-AC converter (23), characterized by comprising: a control section (31) that infers a state of the installation (70) based on a magnitude of a specific frequency component (Cl) of a physical quantity obtained from the installation (70), and that restricts the inferring processing or an operation of the electric motor (50) when a noise frequency component (Cn) included in the physical quantity and the specific frequency component (Cl) satisfy a prescribed relationship, wherein the restricting processing is first restricting processing in which the inferring processing is restricted so as to output a preset inferring result, or second restricting processing in which the operation of the electric motor (50) is restricted so as not to make a false inference in the inferring processing, wherein when the physical quantity is a direct-current signal, a frequency of the noise frequency component (Cn) is a frequency represented by any one of the following Formula 1, Formula 2, and Formula 3, wherein when the physical quantity is an alternating-current signal, a frequency of the noise frequency component (Cn) is a frequency represented by any one of the following Formula 4, Formula 5, and Formula 6, on the condition that two or more frequencies including the electrical angle frequency among two or more frequencies of a carrier frequency and a frequency at which a modulation wave for controlling the direct-AC converter (23) is sampled do not have a greatest common divisor of 2 or more, [Mathematical Formula 1] Formula 1: M * N wherein M and N are natural numbers, [Mathematical Formula 2] Formula 2: M * N + 1 wherein M and N are natural numbers, [Mathematical Formula 3] Formula 3: M * N - 1 wherein M and N are natural numbers, [Mathematical Formula 4] Formula 4: M * N wherein M and N are natural numbers, [Mathematical Formula 5] Formula 5: M * N + 1 wherein M and N are natural numbers, [Mathematical Formula 6] Formula 6: M * N - 1 wherein M and N are natural numbers.

2. The state inferring apparatus according to claim 1, characterized in that: the prescribed relationship is a relationship in which a frequency difference between the noise frequency component (Cn) and the specific frequency component (Cl) is below a prescribed frequency difference threshold.

3. The state inferring apparatus according to claim 2, characterized in that: the frequency difference threshold is 3 Hz.

4. The state inferring apparatus according to any one of claims 1 to 3, characterized in that: the first restricting processing is processing that prohibits the inferring processing.

5. The state inferring apparatus according to any one of claims 1 to 3, characterized in that: in the inferring processing, the control section (31) infers the state of the installation (70) based on a comparison result of a variable calculated based on an amplitude of the specific frequency component (Cl) and a preset threshold, and the first restricting processing is processing that corrects at least one of the threshold and the amplitude of the specific frequency component (Cl).

6. The state inferring apparatus according to any one of claims 1 to 3, characterized in that: the second restricting processing is processing that changes an operation condition of the electric motor (50) so as not to make a false inference in the inferring processing. Let the carrier frequency of the direct AC converter (23) be f c Let the inverse of the period of sampling the physical quantity, i.e. the sampling frequency, be f s Let the electrical angular frequency of the electric motor (50) be f0, 7. The state inferring apparatus according to claim 6, characterized in that: ​ ​ |f c ±6Nf0|…(1) | Mf s -6Nf0|…(2) | Mf s | f c ± 6Nf o ||…(3) |f0±|f c ±6Nf0||…(4) |f0±|Mf s -6Nf o ||…(5) |f0±|Mf s -|f c ±6Nf o |||…(6) ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ The process of changing the operating condition of the motor (50) is a process of changing at least one of the carrier frequency, the sampling frequency, and the mechanical angular frequency of the motor (50).

8. The state inferring apparatus according to any one of claims 1 to 3, characterized in that: The second restriction process is a process of prohibiting the motor (50) from operating under an operating condition in which the noise frequency component (Cn) and the specific frequency component (Cl) satisfy the prescribed relationship.

9. The state inferring apparatus according to any one of claims 1 to 8, characterized in that: When the physical quantity is a direct current signal, the frequency of the specific frequency component (Cl) is at least one of 1 times, 1 / 3 times, and 2 / 3 times the mechanical angular frequency of the motor (50), When the physical quantity is an alternating current signal, the frequency of the specific frequency component (Cl) is at least one of a frequency obtained by adding any one of 1 times, 1 / 3 times, and 2 / 3 times the prescribed frequency to the electrical angular frequency, a frequency obtained by subtracting the prescribed frequency from the electrical angular frequency, and 3 times the electrical angular frequency.

10. The state inferring apparatus according to any one of claims 1 to 9, characterized in that: In the restriction process, if a prescribed restriction release condition is satisfied after the control section (31) starts to impose the restriction on the inferring process or the restriction on the operation of the motor, the control section (31) releases the restriction after a prescribed standby release condition is satisfied.

11. The state inferring apparatus according to any one of claims 1 to 10, characterized in that: In the inferring process, the control section (31) infers whether the device (70) is abnormal, The preset inferring result is not an inferring result indicating that the device (70) is abnormal.

12. A state inference device for inferring the state of a device (70) on which an electric motor (50) is installed, characterized in that: The state inferring apparatus includes a control section (31) that performs an inferring process and a restriction process, In the inferring process, the control section (31) infers the state of the device (70) based on the magnitude of a specific frequency component (Cl) of a physical quantity obtained from the device (70), In the restriction process, the control section (31) restricts the inferring process or the operation of the motor (50) based on the magnitude of a periodic component included in a variable calculated based on the magnitude of the specific frequency component (Cl), The restriction process is a first restriction process in which the inferring process is restricted so as to output a preset inferring result or a second restriction process in which the operation of the motor (50) is restricted so as not to make a false inference in the inferring process.

13. A drive system including a motor drive apparatus (20) that drives a motor (50) mounted on a device (70) and a state inferring apparatus that infers whether the device (70) is abnormal, characterized in that: The state inferring apparatus is the state inferring apparatus according to any one of claims 1 to 12.

14. A refrigeration system comprising a refrigerant circuit (RR1) including a compressor (CC) having a motor (50) and a state inferring apparatus, characterized in that: The state inferring apparatus is the state inferring apparatus according to any one of claims 1 to 12, inferring a state of the refrigeration system.

15. A fan system comprising a fan (FF1) having a motor (50) and a state inferring apparatus, characterized in that: The state inferring apparatus is the state inferring apparatus according to any one of claims 1 to 12, inferring a state of the fan system.

16. A state inferring method of inferring a state of an apparatus (70) that installs an electric motor (50) driven by a direct current-ac conversion device (23), characterized by: The state inferring method includes an inferring step and a limiting step, In the inferring step, the control section (31) infers a state of the device (70) from a magnitude of a specific frequency component (C1) of a physical quantity obtained from the device (70), In the limiting step, the control section (31) limits the inferring step or the operation of the motor (50) when a noise frequency component (Cn) included in the physical quantity and the specific frequency component (C1) satisfy a prescribed relationship, The limiting step is a first limiting step in which the inferring step is limited so as to output a preset inferring result or a second limiting step in which the operation of the motor (50) is limited so as not to make a false inference in the inferring step, Let the carrier frequency of the direct AC converter (23) be f c Let the reciprocal of the period of sampling the physical quantity, i.e. the sampling frequency, be f s Let the electrical angular frequency of the electric motor (50) be f0, When the physical quantity is a direct current signal, the frequency of the noise frequency component (Cn) is a frequency represented by any one of the following Formula 1, Formula 2, and Formula 3, When the physical quantity is an alternating current signal, the frequency of the noise frequency component (Cn) is a frequency represented by any one of the following Formula 4, Formula 5, and Formula 6, on the condition that two or more frequencies including the electrical angle frequency among two or more frequencies of the carrier frequency, the frequency at which a modulation wave for controlling the direct alternating current converter (23) is sampled, and the electrical angle frequency do not have a greatest common divisor of 2 or more, 【Mathematical Formula 2】 |f c ±6Nf0|…(1) | Mf s -6Nf0|…(2) | Mf s | f c ± 6Nf0||... (3) |f0±|f c ±6Nf0||…(4) |f0±|Mf s -6Nf0||…(5) |f0±|Mf s -|f c ±6Nf o |||…(6) where M and N are natural numbers.

17. A state inferring method of inferring a state of an apparatus (70) in which a motor (50) is installed, characterized by: The state inferring method includes an inferring step and a limiting step, In the inferring step, the control section (31) infers a state of the device (70) from a magnitude of a specific frequency component (C1) of a physical quantity obtained from the device (70), In the limiting step, the control section (31) limits the inferring step or the operation of the motor (50) according to a magnitude of a periodic component included in a variable calculated on the basis of the magnitude of the specific frequency component (C1), The limiting step is a first limiting step in which the inferring step is limited so as to output a preset inferring result or a second limiting step in which the operation of the motor (50) is limited so as not to make a false inference in the inferring step.

18. A state inferring program characterized by comprising: The program causes a computer to execute the state inferring method according to claim 16 or 17.

Images