Diagnostic device for electric motors

Abstract

The present application provides a kind of motor diagnostic device (100) with the current detection circuit (7) of detecting the current of motor (5), the operation processing part (10) of operating processing to the current detected and detecting motor anomaly, and storage part (11), operation processing part (10) has the effective value calculation unit (21) of calculating the effective value of current, the peak value of signal intensity of specific frequency band is extracted from sideband by FFT analysis in advance, and is stored in storage part (11) in association with the effective value of current at this time, and, the threshold value set to the peak value of signal intensity of specific frequency band is stored in storage part (11) in advance, the peak value of signal intensity of specific frequency band based on the current detected during the diagnosis of motor (5) is compared with the peak value of signal intensity of specific frequency band and threshold value of each current effective value stored in storage part (11) in advance, to carry out the abnormality determination of motor.

Description

Technical Field

[0001] This application relates to a diagnostic device for an electric motor. Background Technology

[0002] Previously, load current of induction motors was measured and frequency analysis was performed, focusing on the sidebands generated on both sides of the operating frequency. Based on the disorder of short-period vertical waveforms and the oscillation of long-period vertical waveforms, i.e., fluctuations, equipment fault diagnosis methods for diagnosing faults of induction motors and equipment driven by induction motors were proposed (e.g., Patent Document 1).

[0003] In previous equipment anomaly diagnosis methods, when the load torque of an induction motor changes, the spectral intensity near the power supply frequency (operating frequency) increases and becomes greater than the oscillation intensity of the sidebands that generate peaks on both sides of the power supply frequency, thus making it difficult to detect the sidebands.

[0004] In contrast, the applicant filed an application for a diagnostic device for an electric motor that can diagnose whether there is an abnormality in the motor even in a motor with fluctuating load torque by detecting periods when the load does not change (e.g., Patent Document 2).

[0005] Existing technical documents

[0006] Patent documents

[0007] Patent Document 1: Japanese Patent No. 4782218

[0008] Patent Document 2: Japanese Patent No. 6190841 Summary of the Invention

[0009] The technical problem that the invention aims to solve

[0010] However, there is a recent expectation for further improvements in the precision of motor anomaly diagnosis. This necessitates the requirement for the same level of detection accuracy when load changes occur as when no load changes occur.

[0011] This application discloses a technology for solving the above-mentioned problems, and its purpose is to provide a diagnostic device for motors that can determine the occurrence of motor abnormalities regardless of load variations.

[0012] Technical means for solving technical problems

[0013] The electric motor diagnostic device disclosed herein includes a current detection circuit for detecting the current of the electric motor and a voltage detection circuit for detecting the voltage; an arithmetic processing unit for performing calculations on the current detected by the current detection circuit and the voltage detected by the voltage detection circuit and detecting abnormalities in the electric motor; and a storage unit for storing the calculation results of the arithmetic processing unit. The arithmetic processing unit includes a torque calculation unit for calculating a torque value based on the current detected by the current detection circuit and the voltage detected by the voltage detection circuit; a state determination unit for determining whether the calculated torque value is in a stable state; and a unit for performing FFT analysis on the current detected by the current detection circuit and extracting information from a specific frequency band from the sideband. The analysis unit analyzes the peak value of the signal intensity and determines whether the motor is malfunctioning. The analysis unit stores the peak value of the extracted signal intensity of a specific frequency band and the torque value at that time in the storage unit in advance. It also sets a threshold for the normal range of the motor for the extracted peak value of the signal intensity of the specific frequency band and stores the threshold in advance in the storage unit. The malfunction determination unit compares the peak value of the signal intensity of the specific frequency band obtained by performing FFT analysis on the current detected by the current detection circuit with the peak value of the signal intensity of the specific frequency band for each torque value stored in the storage unit and the threshold stored in the storage unit in advance, thereby determining whether the motor is malfunctioning.

[0014] Invention Effects

[0015] According to the diagnostic device for electric motors disclosed herein, it is possible to determine the occurrence of electric motor malfunctions regardless of load variations. Attached Figure Description

[0016] Figure 1 This is a diagram showing the simplified structure and setup of the diagnostic device for the electric motor according to Embodiment 1.

[0017] Figure 2 This diagram shows the structure of the computational processing unit of the diagnostic device for the electric motor according to Embodiment 1, and is a diagram for explaining the signal flow when learning is performed using current analysis.

[0018] Figure 3 This diagram shows the structure of the computational processing unit of the diagnostic device for an electric motor according to Embodiment 1, and is a diagram for explaining the signal flow when performing diagnostics using current analysis.

[0019] Figure 4A This is a flowchart showing the sequence of diagnostic procedures performed using the diagnostic device for the electric motor according to Embodiment 1.

[0020] Figure 4BThis is a flowchart showing the sequence of diagnostic procedures performed using the diagnostic device for the electric motor according to Embodiment 1.

[0021] Figure 5 This is a diagram illustrating the effect of the diagnostic device for the electric motor according to Embodiment 1.

[0022] Figure 6 This is a diagram showing the simplified structure and setup of the diagnostic device for the electric motor according to Embodiment 2.

[0023] Figure 7 This is a diagram showing the structure of the computational processing unit of the diagnostic device for the electric motor according to Embodiment 2, and a diagram illustrating the signal flow during the learning process of current and voltage analysis.

[0024] Figure 8 This diagram shows the structure of the computational processing unit of the diagnostic device for the electric motor according to Embodiment 2, and is a diagram illustrating the signal flow during diagnosis of current and voltage analysis.

[0025] Figure 9A This is a flowchart showing the sequence of diagnostic procedures performed using the diagnostic device for the electric motor according to Embodiment 2.

[0026] Figure 9B This is a flowchart showing the sequence of diagnostic procedures performed using the diagnostic device for the electric motor according to Embodiment 2.

[0027] Figure 10 This is a diagram illustrating the effect of the diagnostic device for the electric motor according to Embodiment 2.

[0028] Figure 11 This is a hardware structure diagram of the diagnostic device for the electric motor involved in the implementation method. Detailed Implementation

[0029] Hereinafter, this embodiment will be described with reference to the accompanying drawings. Furthermore, in each drawing, the same reference numerals denote the same or equivalent parts.

[0030] Implementation method 1.

[0031] The diagnostic device for the electric motor according to Embodiment 1 will now be described using the accompanying drawings.

[0032] Figure 1This diagram illustrates the simplified structure and setup of the diagnostic device for an electric motor according to Embodiment 1. The diagnostic device for an electric motor according to Embodiment 1 is primarily used in a closed switchboard, i.e., a control center. In the diagram, the main circuit 1, which is supplied from the power system, includes a wiring circuit breaker 2, an electromagnetic contactor 3, and an instrument transformer 4 for detecting the load current of the main circuit 1. The main circuit 1 is connected to an electric motor 5, such as a three-phase induction motor, which serves as the load, and the electric motor 5 drives the operation of mechanical equipment 6. The diagnostic device 100 for the electric motor is connected to the instrument transformer 4 and includes a current detection circuit 7 that detects the load current of the main circuit 1 and converts it into a predetermined signal, and an arithmetic processing unit 10 that performs predetermined calculations based on the output of the current detection circuit 7.

[0033] The storage unit 11 is connected to the setting circuit 12 and the arithmetic processing unit 10, and exchanges data with the arithmetic processing unit 10.

[0034] The setting circuit 12 is a circuit that sets the power supply frequency, the rated output of the motor, the rated voltage, the rated current, the number of poles, the rated rotation frequency, etc., and stores this information in the storage unit 11.

[0035] The display unit 13 is connected to the arithmetic processing unit 10 and displays abnormal status, warnings, etc. when physical quantities such as the detected load current are detected, or when the arithmetic processing unit 10 detects an abnormality in the motor 5.

[0036] The drive circuit 14 is connected to the arithmetic processing unit 10, and outputs a control signal for switching the electromagnetic contactor 3 based on the current detected by the instrument transformer 4 and the output of the arithmetic processing unit 10.

[0037] The external output unit 15 outputs abnormal status and alarms to the outside based on the output from the arithmetic processing unit 10.

[0038] The external monitoring device 200, consisting of a PC (personal computer) or similar device, is connected to one or more motor diagnostic devices 100. It receives the output of the processing unit 10 via a communication loop 16 and monitors the operation of the motor diagnostic devices 100. The connection between the external monitoring device 200 and the communication loop 16 of the motor diagnostic devices 100 can be via cable or wirelessly. Alternatively, a network can be established among multiple motor diagnostic devices 100 and connected via the network.

[0039] Next, the structure of the arithmetic processing unit 10 will be explained. Figure 2 and Figure 3 This is a diagram showing the structure of the arithmetic processing unit 10. Figure 2 This is a diagram used to illustrate the signal flow when learning is performed using current analysis. Figure 3 This is a diagram used to illustrate the signal flow when using current analysis for diagnosis.

[0040] Figure 2 and Figure 3 In the process, the arithmetic processing unit 10 includes a current conversion unit 20, a state determination unit 30, an analysis unit 40, and an anomaly determination unit 50, and works in conjunction with a storage unit 11 which has a current and specific frequency band storage device 60 that stores current and specific frequency, and a threshold storage device 61 that stores threshold values.

[0041] First, use Figure 2 This will illustrate the signal flow when learning is performed using current analysis.

[0042] A predetermined current signal, converted by the current detection circuit 7, is input into the current conversion unit 20, and the effective value of the current is calculated by the effective value calculation unit 21. The stable state determination unit 31 of the state determination unit 30 determines whether the calculated current effective value is in a stable state. Here, a stable state refers to a situation where the effective value of the current remains constant for a certain period of time. Furthermore, the certain period of time is a predetermined time.

[0043] When the effective value of the current is determined to be in a stable state, the effective value of the current is stored in the current and specific frequency band storage device 60 of the storage unit 11, and the frequency analysis unit 41 of the analysis unit 40 performs FFT (Fast Fourier Transform) analysis on the current. The averaging analysis unit 42 averages the FFT analysis results. This averaging process can reduce noise.

[0044] The sideband analysis unit 43 extracts the sidebands near the power supply frequency from the signal that has undergone averaging.

[0045] Next, the specific frequency band detection unit 44 detects specific frequency bands caused by mechanical system malfunctions. Detected specific frequency bands caused by mechanical system malfunctions include, for example, specific frequency bands caused by rotation frequency (rotation frequency band), specific frequency bands caused by rotor guide bar malfunctions, and specific frequency bands caused by belt rotation frequency.

[0046] Then, the signal strength of a specific frequency band and the effective current value used to calculate that signal strength are stored in the current and specific frequency band storage device 60 of the storage unit 11. Here, the signal strength of a specific frequency band is stored for each effective current value. That is, a normal range can be determined for each effective current value.

[0047] Using the signal strength of a specific frequency band for each effective current value stored in the current and specific frequency band storage device 60, the distribution of the normal range for each effective current value is calculated in the normal range analysis unit 45. Here, for example, the standard deviation σ is calculated through statistical processing, and 3σ is determined as the threshold. The threshold value, which serves as the normal range, is stored in the threshold storage device 61 of the storage unit 11 in the normal range analysis unit 45.

[0048] The threshold is not limited to 3σ and can be determined through statistical processing of each current value. Alternatively, a certain current value can be used as a reference, and a correction factor can be applied to it to set the threshold. For example, the signal strength of a specific frequency band at the rated current can be used as a reference, and the signal strength of the same frequency band at currents other than the rated current can be corrected. In the diagnosis described later, the stored threshold is used directly at the rated current, but at currents other than the rated current, the corrected value of the stored threshold is set as the threshold. The threshold for the normal range is stored in the threshold storage device 61 of the storage unit 11 in the normal range analysis unit 45.

[0049] Next, use Figure 3 This illustrates the signal flow when using current analysis for diagnosis.

[0050] Similar to the learning process, a predetermined current signal converted by the current detection circuit 7 is input into the current conversion unit 20, and the effective value of the current is calculated by the effective value calculation unit 21. The stable state determination unit 31 of the state determination unit 30 determines whether the calculated current effective value is in a stable state.

[0051] When the effective value of the current is determined to be in a stable state, the frequency analysis unit 41 of the analysis unit 40 performs current FFT analysis. The averaging analysis unit 42 then averages the FFT analysis results.

[0052] The sideband analysis unit 43 extracts the sidebands near the power supply frequency from the signal that has undergone averaging.

[0053] Next, the specific frequency band detection unit 44 detects specific frequency bands caused by mechanical system malfunctions. Detected specific frequency bands caused by mechanical system malfunctions include, for example, specific frequency bands caused by rotation frequency (rotation frequency band), specific frequency bands caused by rotor guide bar malfunctions, and specific frequency bands caused by belt rotation frequency.

[0054] The specific frequency band detected by the specific frequency band detection unit 44 is input to the anomaly determination unit 50. The anomaly determination unit 51 inputs the specific frequency band detected by the specific frequency band detection unit 44, the signal strength of the specific frequency band for each effective current value stored in the current and specific frequency band storage device 60, and threshold data stored in the threshold storage device. In the anomaly determination unit 51, the signal strength of the specific frequency band detected by the specific frequency band detection unit 44 is compared with the signal strength of the specific frequency band for each effective current value stored in the current and specific frequency band storage device 60. While determining whether the detected specific frequency band is caused by a mechanical system anomaly, the threshold value for each effective current value is used to determine whether it is within the normal range, i.e., whether an anomaly has occurred. The determination result is output from the anomaly determination unit 50.

[0055] Figure 4A , Figure 4B This is a flowchart illustrating the sequence of diagnostic procedures performed using the diagnostic device for the electric motor according to Embodiment 1. Here, the example of detecting the rotational frequency band is used as a specific frequency band. After a predetermined learning period, the diagnostic device for the electric motor according to Embodiment 1 transitions to a diagnostic period where diagnostics can be performed.

[0056] First, let's start with the learning period.

[0057] In step S101, the current waveform is acquired. Specifically, the current detection circuit 7, which is connected to the instrument transformer 4, detects the load current of the main circuit 1 and converts the load current into a specified signal.

[0058] In step S102, the effective value calculation unit 21 calculates the effective value of the current.

[0059] In step S103, the stability determination unit 31 determines whether the effective value of the current is in a stable state. If it is not in a stable state ("No" in step S103), it returns to step S101. If it is in a stable state ("Yes" in step S103), it proceeds to step S104 and stores the calculated effective value of the current in the current and specific frequency band storage device 60.

[0060] Next, in step S105, the frequency analysis unit 41 performs current FFT analysis, and in step S106, the averaging analysis unit 42 averages the current FFT results obtained from the analysis. This averaging process can reduce noise.

[0061] In step S107, the sideband analysis unit 43 extracts the sidebands from the result of the averaged current FFT. In step S108, the specific frequency band detection unit 44 extracts the peak value of the rotating frequency band from the extracted sidebands, and in step S109, stores the peak value of the extracted rotating frequency band signal strength in the current and specific frequency band storage device 60. The effective value of the current is stored in the current and specific frequency band storage device 60 in association with the peak value of the rotating frequency band signal strength (step S110).

[0062] The above up to step S110 is the process during the learning period. During the learning period, steps S101 to S110 are repeated (step S111 is marked "No"). After the learning period ends following multiple repetitions (step S111 is marked "Yes"), the diagnostic period begins.

[0063] During diagnosis, the current waveform is first acquired in step S112. Similarly, during learning, the current detection circuit 7, connected to the instrument transformer 4, detects the load current of the main circuit 1 and converts the load current into a specified signal.

[0064] In step S113, the effective value calculation unit 21 calculates the effective value of the current.

[0065] In step S114, the steady-state determination unit 31 determines whether the effective value of the current is in a stable state. If the effective value of the current is not in a stable state ("No" in step S114), the process returns to step S112 and acquires the current waveform. If the effective value of the current is in a stable state, the process proceeds to step S115, where the frequency analysis unit 41 performs current FFT analysis.

[0066] Next, in step S116, the averaging analysis unit 42 averages the analysis results of the current FFT. This averaging process reduces noise.

[0067] In step S117, the sideband analysis unit 43 extracts the sideband from the analysis results of the averaged current FFT.

[0068] In step S118, the specific frequency band detection unit 44 extracts the peak value of the rotating frequency band from the extracted sideband.

[0069] In step S119, the anomaly determination unit 51 compares the peak value of the extracted rotating frequency band signal intensity with the peak value of the rotating frequency band signal intensity associated with the effective value of the current stored in the current and specific frequency band storage device 60, and determines whether the specific frequency band is caused by the rotating frequency. Further, based on the threshold data stored in the threshold storage device 61, it determines whether it is within the normal range, i.e., whether an anomaly has occurred (step S120).

[0070] If an anomaly is determined to have occurred in step S120, an alarm is output using the alarm (not shown) included in the anomaly determination unit 50, or an alarm is output using the external output unit 15 and the display unit 13 after receiving the result output from the calculation processing unit 10 (step S121). In addition, the alarm output can also be notified to the monitoring device 200 via the communication loop 16 as the result output from the calculation processing unit 10.

[0071] Figure 5This diagram illustrates the effect of Embodiment 1, showing the change in the peak value of the signal strength in the rotating frequency band when the load changes. When the current load rate is a%, the peak value of the signal strength in the specific frequency band caused by the rotation frequency is less than the peak value of the sideband. However, when the current load rate is b%, the peak value of the signal strength in the specific frequency band caused by the rotation frequency is greater than the peak value of the sideband. Therefore, although detecting the sideband is difficult, in this embodiment, since the peak value of the signal strength in the specific frequency band is pre-learned and stored for each current RMS value, the peak value of the signal strength in the specific frequency band can be extracted even when the load changes. Furthermore, even when the load changes, the peak value of the signal strength in the specific frequency band can be extracted based on each current RMS value. Therefore, even when the load changes, the sideband can be detected by current FFT analysis.

[0072] Although examples of current load rates a% and b% are shown, the current load rate can be learned in intervals, such as from 0 to less than 5%, from more than 5% to less than 10%, and so on in 5% intervals.

[0073] While the above example illustrates an anomaly detection method by extracting the peak signal strength of a rotating frequency band that is a specific frequency band caused by a mechanical system anomaly, it is also possible to investigate the cause of the anomaly by extracting the peak signal strength of other specific frequency bands, such as those caused by rotor bar anomalies or belt rotation frequencies, and performing anomaly detection.

[0074] As described above, according to Embodiment 1, since the peak value of the signal strength in a specific frequency band is pre-learned and stored for each current RMS value, anomalies can be detected with high accuracy even when load changes occur. When a load change occurs, the current RMS value also changes; however, in this embodiment, since each current RMS value has a peak value of the signal strength in a specific frequency band and its normal range threshold is stored together, anomalies can be determined regardless of load changes.

[0075] Implementation method 2.

[0076] The diagnostic device for the electric motor according to Embodiment 2 will now be described using the accompanying drawings.

[0077] Figure 6 This diagram illustrates the simplified structure and installation of the motor diagnostic device according to Embodiment 2. Similar to Embodiment 1, the motor diagnostic device according to Embodiment 2 is primarily used in a closed switchboard, i.e., a control center. Figure 6 In, compared with implementation method 1 Figure 1The difference is that a transformer 8 for detecting the voltage of the main circuit 1 is also provided in the main circuit 1. A voltage detection circuit 9 connected to the transformer 8 detects the voltage of the main circuit 1, converts it into a specified signal, and outputs it to the arithmetic processing unit 10. Other structures are the same as in embodiment 1.

[0078] Next, the structure of the arithmetic processing unit 10 will be explained.

[0079] Figure 7 and Figure 8 This is a diagram showing the structure of the arithmetic processing unit 10. Figure 7 This is a diagram used to illustrate the signal flow during learning using current-voltage analysis. Figure 8 This is a diagram used to illustrate the signal flow when using current-voltage analysis for diagnosis.

[0080] exist Figure 7 and Figure 8 In the process, the arithmetic processing unit 10 includes a torque conversion unit 22, a state determination unit 30, an analysis unit 40, and an anomaly determination unit 50, and works in coordination with a storage unit 11 which has a torque and specific frequency storage device 62 that stores torque and specific frequency band, and a threshold storage device 61 that stores threshold values.

[0081] First, use Figure 7 This illustrates the signal flow when learning using current-voltage analysis.

[0082] In the torque conversion unit 22, a predetermined current signal converted by the current detection circuit 7 and a predetermined voltage signal converted by the voltage detection circuit 9 are input, and the torque is calculated by the torque calculation unit 23. The stability determination unit 32 of the state determination unit 30 determines whether the calculated torque value is in a stable state. Here, a stable state refers to a situation where the torque value remains constant for a certain period of time. Furthermore, the certain period of time is a predetermined time.

[0083] When the torque value is determined to be in a stable state, the torque value is stored in the torque and specific frequency band storage device 62 of the storage unit 11, and the frequency analysis unit 41 of the analysis unit 40 performs FFT analysis on the current. The averaging analysis unit 42 performs averaging processing on the FFT analysis results. This averaging processing can reduce noise.

[0084] The sideband analysis unit 43 extracts the sidebands near the power supply frequency from the signal that has undergone averaging.

[0085] Next, the specific frequency band detection unit 44 detects specific frequency bands caused by mechanical system malfunctions. Detected specific frequency bands caused by mechanical system malfunctions include, for example, specific frequency bands caused by rotation frequency (rotation frequency band), specific frequency bands caused by rotor guide bar malfunctions, and specific frequency bands caused by belt rotation frequency.

[0086] Then, the signal strength of a specific frequency band and the torque value when that signal strength is detected are stored in the torque and specific frequency band storage device 62 of the storage unit 11. Here, the signal strength of a specific frequency band is stored for each torque value. That is, a normal range can be determined for each torque value.

[0087] Using the signal strength of a specific frequency band for each torque stored in the torque and specific frequency band storage device 62, the distribution of the normal range for each torque is calculated in the normal range analysis unit 45. Here, for example, the standard deviation σ is calculated through statistical processing, and 3σ is determined as the threshold. The threshold value as the normal range is stored in the threshold storage device 61 of the storage unit 11 in the normal range analysis unit 45.

[0088] The threshold is not limited to 3σ and can be determined through statistical processing of each torque value. Alternatively, a certain torque value can be used as a reference, and a correction coefficient can be applied to it to serve as the threshold. For example, the signal strength of a specific frequency band at rated torque can be used as a reference, and the signal strength of a specific frequency band at torques other than rated torque can be corrected. During the diagnosis described later, at rated torque, the stored threshold is used directly, but at torques other than rated torque, the value after correcting the stored threshold is set as the threshold. In the normal range analysis unit 45, the threshold as the normal range is stored in the threshold storage device 61 of the storage unit 11.

[0089] Next, use Figure 8 This illustrates the signal flow when using current-voltage analysis for diagnosis.

[0090] Similar to the learning process, in the torque conversion unit 22, a predetermined current signal converted by the current detection circuit 7 and a predetermined voltage signal converted by the voltage detection circuit 9 are input, and the torque is calculated by the torque calculation unit 23. The stability determination unit 32 of the state determination unit 30 determines whether the calculated torque value is in a stable state.

[0091] When the torque value is determined to be in a stable state, the frequency analysis unit 41 of the analysis unit 40 performs current FFT analysis. The averaging analysis unit 42 then averages the FFT analysis results.

[0092] The sideband analysis unit 43 extracts the sidebands near the power supply frequency from the signal that has undergone averaging.

[0093] Next, the specific frequency band detection unit 44 detects specific frequency bands caused by mechanical system malfunctions. Detected specific frequency bands caused by mechanical system malfunctions include, for example, specific frequency bands caused by rotation frequency (rotation frequency band), specific frequency bands caused by rotor guide bar malfunctions, and specific frequency bands caused by belt rotation frequency.

[0094] The specific frequency band detected by the specific frequency band detection unit 44 is input to the anomaly determination unit 50. The anomaly determination unit 51 receives the specific frequency band detected by the specific frequency band detection unit 44, the signal strength of the specific frequency band for each torque value stored in the torque and specific frequency band storage device 62, and threshold data stored in the threshold storage device. Using this data, the anomaly determination unit 51 compares with the threshold for each torque value to determine with high precision whether the specific frequency band is caused by a mechanical system anomaly, thereby determining whether a mechanical system anomaly has occurred. The determination result is output from the anomaly determination unit 50.

[0095] Figure 9A and Figure 9B This is a flowchart illustrating the sequence of diagnostic procedures performed using the diagnostic device for the electric motor according to Embodiment 2. Here, the example of detecting the rotational frequency band is used as a specific frequency band. The diagnostic device for the electric motor according to Embodiment 2 transitions to a diagnostic period after a predetermined learning period.

[0096] First, let's start with the learning period.

[0097] In step S121, current waveforms and voltage waveforms are acquired. Specifically, the current detection circuit 7 connected to the instrument transformer 4 detects the load current of the main circuit 1 and converts it into a specified signal, and the voltage detection circuit 9 connected to the instrument transformer 8 detects the voltage of the main circuit 1 and converts it into a specified signal.

[0098] In step S122, the torque calculation unit 23 calculates the torque.

[0099] In step S123, the stability determination unit 32 determines whether the torque value is in a stable state. If it is not in a stable state ("No" in step S123), it returns to step S121. If it is in a stable state ("Yes" in step S123), it proceeds to step S124 and stores the calculated torque value in the torque and specific frequency band storage device 62.

[0100] Next, similar to Embodiment 1, in step S105, the frequency analysis unit 41 performs current FFT analysis, and in step S106, the averaging analysis unit 42 averages the results of the analyzed current FFT. This averaging process can reduce noise.

[0101] In step S107, the sideband analysis unit 43 extracts the sidebands from the result of the averaged current FFT. In step S108, the specific frequency band detection unit 44 extracts the peak value of the rotating frequency band from the extracted sidebands, and in step S125, stores the extracted peak value of the rotating frequency band in the torque and specific frequency band storage device 62. The torque value is stored in the torque and specific frequency band storage device 62 in association with the peak value of the signal strength in the rotating frequency band (step S126).

[0102] The above up to step S126 is the process during the learning period. During the learning period, steps S121 to S126 are repeated (step S111 is marked "No"). After the learning period ends following multiple repetitions (step S111 is marked "Yes"), the diagnostic period begins.

[0103] During diagnosis, the current waveform and voltage waveform are first acquired in step S127. Similarly, during learning, the current detection circuit 7 connected to the instrument transformer 4 detects the load current of the main circuit 1 and converts it into a specified signal, and the voltage detection circuit 9 connected to the instrument transformer 8 detects the voltage of the main circuit 1 and converts it into a specified signal.

[0104] In step S128, the torque calculation unit 23 calculates the torque.

[0105] In step S129, the stability determination unit 32 determines whether the torque value is in a stable state. If the torque value is not in a stable state ("No" in step S129), the process returns to step S127, and the current waveform and voltage waveform are acquired. If the torque value is in a stable state, the process proceeds to step S115, and the frequency analysis unit 41 performs current FFT analysis.

[0106] Next, in step S116, the averaging analysis unit 42 averages the analysis results of the current FFT. This averaging process reduces noise.

[0107] In step S117, the sideband analysis unit 43 extracts the sideband from the analysis results of the averaged current FFT.

[0108] In step S118, the specific frequency band detection unit 44 extracts the peak value of the rotating frequency band from the extracted sideband.

[0109] In step S130, the anomaly determination unit 51 compares the peak value of the extracted rotating frequency band signal intensity with the peak value of the rotating frequency band signal intensity associated with the torque value stored in the torque and specific frequency band storage device 62, and determines whether the specific frequency band is caused by the rotation frequency. Further, based on the threshold data stored in the threshold storage device 61, it determines whether it is within the normal range, i.e., whether an anomaly has occurred (step S120).

[0110] If an anomaly is determined to have occurred in step S120, an alarm is output using the alarm (not shown) included in the anomaly determination unit 50, or an alarm is output using the result output of the calculation processing unit 10 and the external output unit 15 and the display unit 13 (step S121). Alternatively, the alarm output can also be notified to the monitoring device 200 via the communication loop 16 as a result output of the calculation processing unit 10.

[0111] Figure 10 This diagram illustrates the effect of Embodiment 2, showing the change in the peak value of the signal strength in the rotating frequency band when the load changes. When the torque value is 'a', the peak value of the signal strength in the specific frequency band caused by the rotation frequency is less than the peak value of the sideband. However, when the torque value is 'b', the peak value of the signal strength in the specific frequency band caused by the rotation frequency is greater than the peak value of the sideband. Therefore, although detecting the sideband is difficult, in this embodiment, since the peak value of the signal strength in the specific frequency band is pre-learned and stored for each torque value, the peak value of the signal strength in the specific frequency band can be extracted even when the load changes. Furthermore, since the peak value of the signal strength in the specific frequency band can be extracted for each torque value even when the load changes, the sideband can be detected by current FFT analysis even when the load changes.

[0112] Although examples of torque values ​​a and b are shown, the learning of torque values ​​can be done within a certain range, such as torque rate (=torque value / rated torque value × 100) set in 5% intervals from 0 to less than 5%, from more than 5% to less than 10%, and so on.

[0113] While the above example illustrates an anomaly detection method by extracting the peak signal strength of a rotating frequency band that is a specific frequency band caused by a mechanical system anomaly, it is also possible to investigate the cause of the anomaly by extracting the peak signal strength of other specific frequency bands, such as those caused by rotor bar anomalies or belt rotation frequencies, and performing anomaly detection.

[0114] As described above, according to Embodiment 2, since the peak value of the signal strength in a specific frequency band is pre-learned and stored for each torque value, anomalies can be detected with high accuracy even when load changes occur. When a load change occurs, the torque value also changes; however, in this embodiment, since each torque value has a peak value of the signal strength in a specific frequency band and its normal range threshold is stored together, anomalies can be determined regardless of load changes.

[0115] <Modifications of Implementation Method 2>

[0116] In addition, in embodiment 2, the torque is calculated based on the load current of the main circuit 1 detected by the current detection circuit 7 and the voltage of the main circuit 1 detected by the voltage detection circuit 9 for anomaly determination. However, the load rate of the motor 5 can also be calculated based on the torque value, and anomaly determination can be performed by learning data of a specific frequency band corresponding to the load rate.

[0117] Alternatively, the torque value calculated in Implementation Method 2 can be used to detect torque anomalies and determine motor anomalies caused by torque. The torque anomaly detection will be described below.

[0118] The torque Te is expressed by the stator current and linkage flux of motor 5 as shown in the following equation (1).

[0119] [Mathematical Expression 1]

[0120]

[0121] Among them, P p Magnetic pole number

[0122] Φ d Φ q Stator coil linkage flux

[0123] i q i d Stator current

[0124] In addition, the linkage flux can be obtained from the following equations (2) and (3).

[0125] [Mathematical Expression 2]

[0126]

[0127]

[0128] Among them, v d v q Stator voltage

[0129] R s Stator resistance

[0130] By comparing the torque value obtained by the torque calculation unit 23 with the torque Te obtained by equation (1), torque anomalies can be detected.

[0131] Furthermore, the electric motor diagnostic device 100 described in Embodiments 1 and 2 above is as follows: Figure 11As an example of hardware, it consists of a processor 110 and a storage device 120. Although the storage device is not shown, it includes volatile storage devices such as random access memory (RAM) and non-volatile auxiliary storage devices such as flash memory. Alternatively, an auxiliary storage device such as a hard disk can be used instead of flash memory. The processor 110 executes a program input from the storage device 120. In this case, the program is input from the auxiliary storage device to the processor 110 via the volatile storage device. Furthermore, the processor 110 can output data such as calculation results to the volatile storage device of the storage device 120, and can also save data to the auxiliary storage device via the volatile storage device.

[0132] This disclosure describes various exemplary embodiments and examples, but the various features, forms and functions described in one or more embodiments are not limited to the application of a particular embodiment, and can be applied to the embodiment alone or in various combinations.

[0133] Therefore, it can be assumed that numerous variations not illustrated are also included within the scope of the technology disclosed in this application. For example, this includes cases where at least one constituent element is modified, added to, or omitted, and cases where at least one constituent element is extracted and combined with constituent elements of other embodiments.

[0134] Label Explanation

[0135] 1: Main circuit; 2: Circuit breaker for wiring; 3: Electromagnetic contactor; 4: Transformer for instrumentation; 5: Motor; 6: Mechanical equipment; 7: Current detection circuit; 8: Transformer for instrumentation; 9: Voltage detection circuit; 10: Processing unit; 11: Storage unit; 12: Setting circuit; 13: Display unit; 14: Drive circuit; 15: External output unit; 16: Communication circuit; 200: Monitoring device; 20: Current conversion unit; 21: RMS value calculation unit; 22: Torque converter. 23: Torque Calculation Unit; 30: State Determination Unit; 31, 32: Stable State Determination Unit; 40: Analysis Unit; 41: Frequency Analysis Unit; 42: Averaging Analysis Unit; 43: Sideband Analysis Unit; 44: Specific Frequency Band Detection Unit; 45: Normal Range Analysis Unit; 50: Anomaly Determination Unit; 51: Anomaly Determination Unit; 60: Current and Specific Frequency Band Storage Device; 61: Threshold Storage Device; 62: Torque and Specific Frequency Band Storage Device; 100: Motor Diagnostic Device.

Claims

1. A diagnostic device for an electric motor, comprising a current detection circuit for detecting the current of the electric motor and a voltage detection circuit for detecting the voltage, an arithmetic processing unit for performing calculations on the current detected by the current detection circuit and the voltage detected by the voltage detection circuit and detecting abnormalities in the electric motor, and a storage unit for storing the calculation results of the arithmetic processing unit. The diagnostic device for the electric motor is characterized in that, The arithmetic processing unit includes a torque calculation unit that calculates a torque value based on the current detected by the current detection circuit and the voltage detected by the voltage detection circuit; a state determination unit that determines whether the calculated torque value is in a stable state; an analysis unit that performs FFT analysis on the current detected by the current detection circuit and extracts the peak value of the signal strength of a specific frequency band from the sideband; and an anomaly determination unit that determines whether the motor is malfunctioning. The analysis unit pre-stores the peak value of the extracted signal strength in a specific frequency band and the corresponding torque value in the storage unit, and sets a threshold for the normal range of the motor based on the peak value of the extracted signal strength in the specific frequency band, and pre-stores the threshold value in the storage unit. The anomaly determination unit compares the peak value of the signal strength of a specific frequency band obtained by performing FFT analysis on the current detected by the current detection circuit with the peak value of the signal strength of the specific frequency band of each torque value pre-stored in the storage unit and the threshold value pre-stored in the storage unit to determine whether the motor has an anomaly.

2. The diagnostic device for an electric motor as described in claim 1, characterized in that, The processing unit repeatedly acquires torque values ​​that the state determination unit determines are in a stable state, stores the peak values ​​of signal strengths in specific frequency bands extracted by the analysis unit, the torque values ​​at that time, and the threshold values ​​in the storage unit, and then diagnoses the motor.

3. The diagnostic device for an electric motor as described in claim 1, characterized in that, The load rate of the motor is calculated based on the torque value calculated by the torque calculation unit. The analysis unit pre-stores the peak value of the extracted signal strength in a specific frequency band and the current load rate in the storage unit, and sets a threshold for the normal range of the motor based on the peak value of the extracted signal strength in the specific frequency band, and pre-stores the threshold value in the storage unit. The anomaly determination unit compares the peak value of the signal strength of a specific frequency band obtained by performing FFT analysis on the current detected by the current detection circuit with the peak value of the signal strength of the specific frequency band for each load rate pre-stored in the storage unit, and the threshold value pre-stored in the storage unit, to determine whether the motor has an anomaly.

Images