Abnormality assessment device, learning device, power supply device, abnormality assessment system, and abnormality assessment method

Abstract

An abnormality assessment device (41) comprises an electric current acquisition unit (51), a rotation speed acquisition unit (52), a conversion unit (53), an intensity determination unit (54), and an abnormality assessment unit (55). The electric current acquisition unit (51) acquires time-domain electric current data that indicates the electric current flowing from a power conversion device (11) to an electric motor (91). The rotation speed acquisition unit (52) acquires the rotation speed of the electric motor (91). The conversion unit (53) generates, from the current data, frequency-domain spectrum data. The intensity determination unit (54) determines, from the spectrum data, sideband wave intensity, which is the intensity of a sideband wave, being a frequency component resulting from shifting from the power supply frequency of the power conversion device (11) by the rotational frequency of the electric motor (91). The abnormality assessment unit (55) determines the presence or absence of an abnormality in the electric motor (91) in accordance with a determination criterion corresponding to the sideband wave intensity and the rotation speed.

Description

Abnormality discrimination device, learning device, power supply device, abnormality discrimination system, and abnormality discrimination method

[0001] The present disclosure relates to an abnormality determination device, a learning device, a power supply device, an abnormality determination system, and an abnormality determination method.

[0002] Railway vehicles are equipped with electric motors that generate propulsive force when driven by power supplied from a power conversion device. When the motor deteriorates over time, for example, bearing damage or rotor eccentricity occurs, the magnetic relationship between the stator and rotor fluctuates. This makes it difficult to achieve the target acceleration or deceleration of the railway vehicle. Therefore, railway vehicles or ground facilities are equipped with devices that determine whether or not there is an abnormality in the electric motor. An example of this type of device is disclosed in Patent Document 1.

[0003] The anomaly monitoring device disclosed in Patent Document 1 determines the peak value of sideband waves from a current spectrum obtained by performing a fast Fourier transform on the current signal of a three-phase induction motor during operation. This anomaly monitoring device detects an anomaly in a rotating mechanical system by comparing a deterioration parameter determined from the peak value of the sideband waves with a deterioration judgment reference value set based on destructive tests, maintenance management records, etc.

[0004] JP 2017-181437 A

[0005] In the case of an electric motor with a constant rotational speed and torque, the frequency and peak value of the sideband waves are constant under normal conditions. On the other hand, the rotational speed and torque of an electric motor mounted on a railway vehicle change in response to the railway vehicle's operation commands. With a method such as the anomaly monitoring device disclosed in Patent Document 1, which compares a deterioration parameter determined from the peak value of the sideband waves with a fixed deterioration reference value, it is difficult to accurately determine whether or not an abnormality exists in an electric motor whose rotational speed and torque change.

[0006] The present disclosure has been made in consideration of the above-mentioned circumstances, and aims to provide an abnormality detection device, a learning device, a power supply device, an abnormality detection system, and an abnormality detection method that make it possible to accurately determine whether or not an abnormality exists in an electric motor.

[0007] In order to achieve the above object, the abnormality determination device of the present disclosure includes a current acquisition unit, a rotational speed acquisition unit, a conversion unit, an intensity determination unit, and an abnormality determination unit. The current acquisition unit acquires time-domain current data indicating the current flowing from the power conversion device to the electric motor. The rotational speed acquisition unit acquires the rotational speed of the electric motor. The conversion unit generates frequency-domain spectrum data from the current data. The intensity determination unit determines, from the spectrum data, sideband intensity, which is the intensity of a sideband, which is a frequency component shifted from the power supply frequency of the power conversion device by the rotational frequency of the electric motor. The abnormality determination unit determines whether or not an abnormality exists in the electric motor based on a determination criterion corresponding to the sideband intensity and the rotational speed.

[0008] The abnormality detection device according to the present disclosure determines whether or not an abnormality exists in the electric motor based on a detection criterion according to the sideband wave intensity and the rotation speed, and therefore is able to accurately determine whether or not an abnormality exists in the electric motor.

[0009] FIG. 1 is a block diagram of an abnormality determination system according to embodiment 1; FIG. 2 is a diagram showing a hardware configuration of an abnormality determination device according to embodiment 1; FIG. 3 is a flowchart showing an example of the operation of an abnormality determination process performed by the abnormality determination device according to embodiment 1; FIG. 4 is a diagram showing an example of spectrum data determined by the abnormality determination device according to embodiment 1; FIG. 5 is a diagram showing an example of a target range according to a rotation speed used by the abnormality determination device according to embodiment 1; FIG. 6 is a diagram showing an example of a signal intensity difference in embodiment 1;

[0010] An abnormality detection device, a learning device, a power supply device, an abnormality detection system, and an abnormality detection method according to embodiments of the present disclosure will be described in detail below with reference to the drawings. In the drawings, the same or equivalent parts are designated by the same reference numerals.

[0011] (Embodiment 1) Using an electric motor that is mounted on a railway vehicle and receives a supply of electric power to be driven to generate propulsive force for the railway vehicle as an example, an abnormality detection device that detects an abnormality in the electric motor and an abnormality detection system that includes the abnormality detection device will be described in Embodiment 1. An abnormality detection device 41 included in an abnormality detection system 31 shown in Figure 1 determines whether or not an abnormality exists in an electric motor 91.

[0012] The power conversion device 11 that supplies power to the electric motor 91 is, for example, a DC-to-three-phase conversion device that is mounted on a DC-fed railway vehicle and converts DC power supplied from a power source into three-phase AC power and supplies it to the electric motor 91. To avoid complicating the diagram, one electric motor 91 is shown in Fig. 1, but the number of electric motors 91 to which the power conversion device 11 supplies power is arbitrary.

[0013] The power conversion device 11 includes an input terminal 11a connected to a power supply and an input terminal 11b connected to ground. The power conversion device 11 further includes a power conversion circuit 12 that converts DC power supplied from the power supply into three-phase AC power and supplies the three-phase AC power to the electric motor 91. The power conversion device 11 is provided with a current detection circuit 13 that measures the current flowing from the power conversion device 11 to the electric motor 91. The power conversion device 11 further includes a reactor L1 and a capacitor C1 that are connected in series between the input terminals 11a and 11b.

[0014] The input terminal 11a is electrically connected to a power source, specifically, a current collector that acquires power supplied from a substation via a power supply line, via a contactor, circuit breaker, etc. (not shown). For example, the current collector may be a pantograph that acquires power via an overhead line, which is an example of a power supply line, or a current collector shoe that acquires power via a third rail, which is an example of a power supply line. The input terminal 11b is grounded via a ground ring, ground brush, wheel, etc. (not shown).

[0015] The power conversion circuit 12 is, for example, an inverter that outputs AC power with variable effective voltage and frequency. The power conversion circuit 12 has a plurality of switching elements. Each switching element is, for example, an insulated gate bipolar transistor (IGBT), a gate turn-off thyristor (GTO), or a metal-oxide-semiconductor field-effect transistor (MOSFET).

[0016] The current detection circuit 13 has an electric circuit between the power conversion circuit 12 and the electric motor 91, for example, a current transformer (CT) attached to a bus bar electrically connecting the power conversion circuit 12 and the electric motor 91, and measures the phase currents output by the power conversion circuit 12, specifically, the U-phase current, the V-phase current, and the W-phase current. The current detection circuit 13 sends the measured values ​​of each phase current to the power conversion control device 21.

[0017] One end of the reactor L1 is connected to the input terminal 11a. The other end of the reactor L1 is connected to a primary terminal of the power conversion circuit 12. One end of the capacitor C1 is connected to a connection point between the other end of the reactor L1 and the primary terminal of the power conversion circuit 12. The other end of the capacitor C1 is connected to a connection point between the input terminal 11b and the primary terminal of the power conversion circuit 12. The reactor L1 and the capacitor C1 form an LC filter that attenuates harmonic components generated by the switching operation of the power conversion circuit 12.

[0018] The power conversion control device 21 controls the switching operation of each switching element of the power conversion circuit 12 provided in the power conversion device 11. The power conversion control device 21 acquires a driving command S1 from a cab (not shown). The driving command S1 indicates a command corresponding to an operation by a driver to a master controller provided in the cab. Specifically, the driving command S1 indicates any one of a powering command to accelerate the railway vehicle, a braking command to decelerate the railway vehicle, and a coasting command to coast the railway vehicle. A coasting command indicates a state in which neither a powering command nor a braking command is input.

[0019] The power conversion control device 21 generates a power conversion control signal S2 that controls each switching element of the power conversion circuit 12 in response to the operation command S1 and outputs the power conversion control signal S2 to the power conversion circuit 12. The power conversion control signal S2 is, for example, a PWM (Pulse Width Modulation) signal. For example, when the operation command S1 indicates a powering command, the power conversion control device 21 calculates a target torque, which is a target value of the torque of the electric motor 91, from the target acceleration of the railway vehicle indicated by the powering command. The power conversion control device 21 estimates the rotational speed of the electric motor 91 from the U-phase current value, V-phase current value, and W-phase current value acquired from the current detection circuit 13. The power conversion control device 21 calculates a voltage command value from the estimated rotational speed of the electric motor 91 and the target torque, and generates a PWM signal from the voltage command value and a carrier wave obtained from a local oscillator (not shown).

[0020] The abnormality determination system 31 that determines whether or not there is an abnormality in the above-mentioned electric motor 91 includes an abnormality determination device 41 that determines whether or not there is an abnormality in the electric motor 91, and a monitoring device 61 that outputs the determination result obtained from the abnormality determination device 41.

[0021] The abnormality determination device 41 includes a current acquisition unit 51 that acquires time-domain current data indicating the current flowing from the power conversion device 11 to the electric motor 91, a rotational speed acquisition unit 52 that acquires the rotational speed of the electric motor 91, and a conversion unit 53 that generates frequency-domain spectrum data from the current data. The abnormality determination device 41 also includes an intensity determination unit 54 that determines, from the spectrum data, a sideband wave intensity, which is the intensity of a sideband wave that is a frequency component shifted from the power supply frequency of the power conversion device 11 by the rotational frequency of the electric motor 91, and an abnormality determination unit 55 that determines whether or not an abnormality exists in the electric motor 91 based on a determination criterion according to the sideband wave intensity and the rotational speed.

[0022] The abnormality determination device 41 is provided in ground equipment, for example, an operation control center, and communicates with the power conversion control device 21 via a network.

[0023] The current obtaining unit 51 obtains the value of the current flowing through the electric motor 91, specifically, the U-phase current value Iu of the phase currents detected by the current detection circuit 13, from the power conversion control device 21. The current obtaining unit 51 outputs the obtained U-phase current value to the conversion unit 53 as current data.

[0024] The rotation speed acquisition unit 52 acquires the estimated rotation speed N1 of the electric motor 91 from the power conversion control device 21. The rotation speed acquisition unit 52 outputs the acquired rotation speed N1 to the strength determination unit 54 and the abnormality determination unit 55.

[0025] The converter 53 generates spectrum data in the frequency domain by performing FFT (Fast Fourier Transform) on the current data acquired from the current acquirer 51. The converter 53 sends the generated spectrum data to the intensity determiner 54.

[0026] The intensity determination unit 54 determines the rotation frequency of the electric motor 91 from the rotation speed obtained from the rotation speed acquisition unit 52. The intensity determination unit 54 determines, from the spectrum data obtained from the conversion unit 53, sideband wave intensity, which is the intensity of sideband waves that are frequency components shifted from the power supply frequency of the power conversion device 11 by the rotation frequency of the electric motor 91, and the intensity of the frequency component of the power supply frequency. In detail, the intensity determination unit 54 determines an upper sideband wave intensity, which is the intensity of an upper sideband wave that is a frequency component obtained by adding the rotation frequency of the electric motor 91 to the power supply frequency, and a lower sideband wave intensity, which is the intensity of a lower sideband wave that is a frequency component obtained by subtracting the rotation frequency of the electric motor 91 from the power supply frequency. The intensity determination unit 54 sends the determined sideband wave intensity and the intensity of the frequency component of the power supply frequency to the abnormality determination unit 55.

[0027] The abnormality determination unit 55 specifies a determination criterion to be used for determining whether or not there is an abnormality in the electric motor 91, according to the rotation speed acquired from the rotation speed acquisition unit 52, and determines whether or not there is an abnormality in the electric motor 91 from the determination criterion according to the sideband wave intensity and the rotation speed. When the abnormality determination unit 55 determines that there is an abnormality in the electric motor 91, it transmits a determination result indicating that there is an abnormality to the monitoring device 61.

[0028] In the first embodiment, the abnormality determination unit 55 calculates the signal strength difference for each of the upper sideband and the lower sideband by subtracting the sideband strength from the strength of the frequency component of the power supply frequency. The abnormality determination unit 55 specifies a target range that includes a target value for the signal strength difference and that changes depending on the rotation speed acquired from the rotation speed acquisition unit 52. The abnormality determination unit 55 determines whether the signal strength difference is within the specified target range. If the signal strength difference is within the target range, it can be determined that no abnormality has occurred in the electric motor 91. On the other hand, if the signal strength difference is not within the target range, it can be determined that an abnormality has occurred in the electric motor 91.

[0029] When the monitoring device 61 receives a determination result indicating that an abnormality has occurred from the abnormality determination unit 55 included in the abnormality determination device 41, the monitoring device 61 outputs the determination result indicating that an abnormality has occurred in the electric motor 91 by, for example, a screen display, an audio output, or the like. The monitoring device 61 is provided in a ground facility, for example, an operation control center.

[0030] The hardware configuration of the abnormality determination device 41 having the above-described configuration is shown in FIG. 2. The abnormality determination device 41 includes a processor 81, a memory 82, and an interface 83. The processor 81, the memory 82, and the interface 83 are connected to one another via a bus 80. The functions of each unit of the abnormality determination device 41 are realized by software, firmware, or a combination of software and firmware. The software and firmware are written as programs and stored in the memory 82. The processor 81 reads and executes the programs stored in the memory 82, thereby realizing the functions of each unit described above. In other words, the memory 82 stores programs for executing the processing of each unit of the abnormality determination device 41.

[0031] The memory 82 includes, for example, non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read-Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable and Programmable Read-Only Memory), magnetic disk, flexible disk, optical disk, compact disk, mini disk, DVD (Digital Versatile Disc), etc.

[0032] The abnormality determination device 41 is connected to the power conversion control device 21, the monitoring device 61, etc. via an interface 83. The interface 83 has an interface module that complies with one or more standards depending on the connected device.

[0033] The operation of the abnormality determination device 41 will be described with reference to Fig. 3. When the abnormality determination device 41 is activated, it starts the process shown in Fig. 3. The current acquisition unit 51 included in the abnormality determination device 41 acquires, from the power conversion control device 21, current data indicating the value of the current flowing from the power conversion device 11 to the electric motor 91 (step S11).

[0034] The current acquiring unit 51 determines whether the electric motor 91 is operating based on the electric current data (step S12). For example, if the amplitude of the electric current data is equal to or greater than an amplitude threshold determined according to the minimum value of the amplitude of the electric current supplied to the electric motor 91 after a certain time has elapsed since the electric motor 91 was driven, the electric motor 91 can be considered to be operating.

[0035] When the electric motor 91 is not operating (step S12; No), the process of step S11 is repeated. When the electric motor 91 is operating (step S12; Yes), the current acquisition unit 51 outputs the current data to the conversion unit 53, and the conversion unit 53 performs FFT on the current data to generate spectrum data (step S13).

[0036] The rotational speed acquisition unit 52 acquires the rotational speed of the electric motor 91 from the power conversion control device 21 (step S14). The rotational speed acquisition unit 52 sends the acquired rotational speed to the intensity determination unit 54, which determines the rotational frequency from the rotational speed (step S15). The intensity determination unit 54 determines the sideband intensity from the spectrum data (step S16).

[0037] 4, the spectrum data includes peak values ​​located at the power supply frequency and at a frequency shifted from the power supply frequency by the rotation frequency of the electric motor 91. Specifically, if the rotation frequency of the electric motor 91 is fr and the power supply frequency is f1, peaks exist at the frequency (f1-fr), frequency f1, and frequency (f1+fr). In other words, the frequency components at positions shifted by the rotation frequency fr to the left and right of the power supply frequency f1 are sidebands.

[0038] The intensity determination unit 54 is assumed to previously store information about the power supply frequency f1 of the power conversion device 11 that supplies power to the electric motor 91 that is the target of abnormality detection. The power supply frequency f1 is constant at 50 Hz or 60 Hz. The intensity determination unit 54 determines the intensities of the sideband waves of the frequency (f1-fr) and the frequency (f1+fr) and the intensity of the frequency component of the power supply frequency f1 based on the power supply frequency f1 and the rotational frequency fr calculated from the rotational speed of the electric motor 91 acquired from the rotational speed acquisition unit 52. The intensity determination unit 54 outputs the calculated sideband wave intensities and the intensity of the frequency component of the power supply frequency f1 to the abnormality detection unit 55.

[0039] As shown in FIG. 3 , the abnormality determination unit 55 determines the signal strength difference, which is the difference between the strength of the frequency component of the power supply frequency f1 and the strength of the sideband waves (step S17). Specifically, the abnormality determination unit 55 determines the signal strength difference for each of the upper and lower sideband waves by subtracting the sideband wave strength from the strength of the frequency component of the power supply frequency f1. As shown in FIG. 4 , the strength of the frequency component of the power supply frequency f1 is higher than the sideband wave strength. When the sideband wave strength is higher relative to the strength of the frequency component of the power supply frequency f1, the signal strength difference becomes smaller. When the sideband wave strength is lower relative to the strength of the frequency component of the power supply frequency f1, the signal strength difference becomes larger.

[0040] As shown in FIG. 3 , the abnormality determination unit 55 determines whether the signal intensity difference is within the target range (step S18). Specifically, the abnormality determination unit 55 identifies the target range from the rotation speed acquired from the rotation speed acquisition unit 52. In the first embodiment, as shown in FIG. 5 , the abnormality determination unit 55 is assumed to previously store a correspondence between the rotation speed N1 and the upper and lower limits of the target range. In the example record on the first line of FIG. 5 , if the rotation speed N1 is equal to or less than the reference value Nb1, the lower limit of the target range is Sb1, and the upper limit of the target range is Sb5. In other words, if the rotation speed N1 is equal to or less than the reference value Nb1 and the signal intensity difference is equal to or greater than Sb1 and equal to or less than Sb5, the electric motor 91 can be considered normal.

[0041] The abnormality determination unit 55 determines whether or not there is an abnormality in the motor 91 by comparing each signal strength difference with a target range corresponding to the rotation speed. In Fig. 6, the signal strength difference when the motor 91 can be considered normal is indicated by a black circle, and the signal strength difference when the motor 91 can be considered abnormal is indicated by a white circle. The range enclosed by a dotted line in Fig. 6 indicates the target range corresponding to the rotation speed. By using the target range corresponding to the rotation speed of the motor 91, it is possible to accurately determine whether or not there is an abnormality in the motor 91 based on the sideband intensity that changes according to the rotation speed of the motor 91.

[0042] 3, when the abnormality determination unit 55 determines that the signal strength difference is outside the target range (step S18; No), it outputs a determination result indicating that an abnormality exists to the monitoring device 61 (step S19). When the abnormality determination unit 55 determines that the signal strength difference is within the target range (step S18; Yes), it does not perform the process of step S19. When the signal strength difference is within the target range (step S18; Yes), or when the process of step S19 is completed, the above-described process is repeated from step S11.

[0043] As described above, the abnormality detection device 41 according to the first embodiment determines whether or not there is an abnormality in the electric motor 91 based on the sideband wave intensity and a detection criterion according to the rotation speed. In particular, the abnormality detection device 41 determines whether or not there is an abnormality in the electric motor 91 based on whether or not the signal strength difference, which is the difference between the strength of the frequency component of the power supply frequency and the sideband wave intensity, is within a target range according to the rotation speed. By using the target range according to the rotation speed of the electric motor 91, it becomes possible to accurately determine whether or not there is an abnormality in the electric motor 91 based on the sideband wave intensity, which changes according to the rotation speed of the electric motor 91.

[0044] Second Embodiment The discrimination criteria used by the abnormality discrimination device to discriminate whether or not there is an abnormality in the electric motor 91 are not limited to the above-described examples. In the second embodiment, an abnormality discrimination device that discriminates whether or not there is an abnormality in the electric motor 91 based on discrimination criteria according to the rotation speed and torque of the electric motor 91 will be described, focusing on the differences from the first embodiment.

[0045] The abnormality detection system 31 according to the second embodiment shown in FIG. 7 includes an abnormality detection device 42 that detects whether or not there is an abnormality in the electric motor 91, and a monitoring device 61 that outputs the detection result obtained from the abnormality detection device 42.

[0046] In addition to the configuration of the abnormality determination device 41, the abnormality determination device 42 includes a torque acquisition unit 56 that acquires the torque of the electric motor 91. The hardware configuration of the abnormality determination device 42 is similar to the hardware configuration of the abnormality determination device 41 shown in FIG.

[0047] The torque obtaining unit 56 obtains the target torque τ1 used to control the power conversion circuit 12 from the power conversion control device 21. The torque obtaining unit 56 sends the obtained target torque to the abnormality determination unit 55.

[0048] The abnormality determination unit 55 determines a target range of the signal strength difference according to the rotation speed and the target torque, and determines whether or not there is an abnormality in the electric motor 91 by comparing the signal strength difference with the target range according to the rotation speed and the target torque. In the second embodiment, as shown in FIG. 8 , the abnormality determination unit 55 is assumed to previously store correspondence between the rotation speed and the target torque and the upper and lower limits of the target range. In the example of the record in the first row of FIG. 8 , if the rotation speed N1 is equal to or less than the reference value Nb1 and the target torque τ1 is equal to or less than the reference value τb1, the lower limit of the target range is Sb1 and the upper limit of the target range is Sb5. In other words, if the rotation speed N1 is equal to or less than the reference value Nb1 and the target torque τ1 is equal to or less than the reference value τb1, the electric motor 91 can be considered normal.

[0049] As described above, the abnormality determination device 42 according to the second embodiment determines whether or not an abnormality exists in the electric motor 91 based on a comparison between the signal strength difference and a determination criterion corresponding to the rotation speed and the target torque. In particular, the abnormality determination device 42 determines whether or not an abnormality exists in the electric motor 91 based on whether or not the signal strength difference is within a target range corresponding to the rotation speed and the target torque. By using the target range corresponding to the rotation speed and the target torque of the electric motor 91, it becomes possible to accurately determine whether or not an abnormality exists in the electric motor 91 based on the sideband intensity that changes according to the rotation speed and the torque of the electric motor 91.

[0050] Third Embodiment The discrimination criteria used to determine whether or not there is an abnormality in the electric motor 91 may be generated or updated by an abnormality discrimination device. As an example of the discrimination criteria, an abnormality discrimination device that determines a target range according to the rotation speed will be described in the third embodiment, focusing on the differences from the first embodiment.

[0051] The abnormality detection system 31 according to the third embodiment shown in FIG. 9 includes an abnormality detection device 43 that detects whether or not there is an abnormality in the electric motor 91, and a monitoring device 61 that outputs the detection result obtained from the abnormality detection device 43.

[0052] The abnormality determination device 43 includes a criteria determination unit 57 that determines a determination criterion, specifically, a target range according to the rotation speed, in addition to the configuration of the abnormality determination device 41. The hardware configuration of the abnormality determination device 43 is similar to the hardware configuration of the abnormality determination device 41 shown in FIG.

[0053] The reference determination unit 57 acquires the rotation speed of the electric motor 91 from the rotation speed acquisition unit 52 and acquires the signal strength difference from the abnormality determination unit 55. The reference determination unit 57 stores the rotation speed and the signal strength difference in association with each other. The reference determination unit 57 determines a target range from the association between the rotation speed and the signal strength difference during a normal period from when the electric motor 91 starts operating until the period during which the electric motor 91 can be considered normal has elapsed. The reference determination unit 57 sends the determined target range to the abnormality determination unit 55.

[0054] When a plurality of electric motors 91 are mounted on the same railway vehicle, the reference determination unit 57 may store the rotation speeds of the respective electric motors 91 and the corresponding signal strength differences. In this case, the reference determination unit 57 may determine a target range that is common to the plurality of electric motors 91 mounted on the same railway vehicle.

[0055] The abnormality determination unit 55 outputs the signal strength difference to the reference determination unit 57 and acquires the target range from the reference determination unit 57. The abnormality determination unit 55 compares the signal strength difference with the target range acquired from the reference determination unit 57 to determine whether or not an abnormality exists in the electric motor 91.

[0056] The criteria determination process performed by the abnormality determination device 43 having the above configuration will be described with reference to Fig. 10. The processes from step S11 to step S16 are the same as the processes from step S11 to step S16 performed by the abnormality determination device 41 according to the first embodiment shown in Fig. 3.

[0057] The standard determination unit 57 determines whether or not the current state is within a normal period from when the operation of the electric motor 91 starts until the period during which the electric motor 91 can be considered normal has elapsed (step S21). The normal period can be determined by test operation, simulation, or the like of the electric motor 91, and is a period during which the electric motor 91 can be considered to operate normally without any breakdowns.

[0058] If it is within the normal period (step S21; Yes), the reference determination unit 57 stores the rotation speed and the signal strength difference in association with each other in a memory (not shown) (step S22).

[0059] The reference determination unit 57 determines whether the number of stored data items relating to the correspondence between the rotation speed and the signal strength difference is sufficient to determine the target range (step S23). If the number of data items is not sufficient (step S23; No), the above-described process is repeated from step S11.

[0060] If the number of data is sufficient (step S23; Yes), the reference determination unit 57 determines a target range from the stored correspondence between the rotation speed and the signal strength difference (step S24). For example, the reference determination unit 57 determines a target range indicating a normal range of the signal strength difference according to the range of the rotation speed, taking the stored signal strength difference as a normal value.

[0061] When the processing of step S24 is completed, or when a period in which the motor 91 can be considered normal has elapsed since the start of operation of the motor 91, i.e., when it is outside the normal period (step S21; No), the abnormality determination device 43 terminates the criteria determination processing.

[0062] 10 ends, the abnormality determination device 43 performs the abnormality determination process shown in Fig. 3. In the abnormality determination process, the abnormality determination unit 55 determines whether or not an abnormality exists in the electric motor 91 by comparing the signal strength difference with the target range determined by the reference determination unit 57.

[0063] As described above, the abnormality determination device 43 according to the third embodiment determines the determination criteria for determining whether or not there is an abnormality in the electric motor 91. Therefore, it becomes possible to determine whether or not there is an abnormality in the electric motor 91 based on the determination criteria according to the rotation speed of the electric motor 91, in accordance with the individual characteristics of the electric motor 91.

[0064] (Fourth Embodiment) The abnormality determination device may determine the presence or absence of a sign of an abnormality in addition to determining the presence or absence of an abnormality in the electric motor 91. An abnormality determination device that determines the presence or absence of a sign of an abnormality will be described in a fourth embodiment, focusing on the differences from the first embodiment.

[0065] The abnormality detection system 31 shown in Figure 11 includes an abnormality detection device 44 that detects whether or not there is an abnormality or a sign of an abnormality in the electric motor 91, and a monitoring device 61 that outputs the detection results obtained from the abnormality detection device 44.

[0066] The abnormality determination device 44 includes a sign determination unit 58 that determines whether or not there is a sign of an abnormality, in addition to the configuration of the abnormality determination device 41. The hardware configuration of the abnormality determination device 44 is similar to the hardware configuration of the abnormality determination device 41 shown in FIG.

[0067] The sign determination unit 58 obtains the rotation speed of the electric motor 91 from the rotation speed obtaining unit 52, and obtains the signal strength difference from the abnormality determination unit 55. The sign determination unit 58 stores the correspondence between the rotation speed and the signal strength difference in a memory (not shown).

[0068] The sign determination unit 58 determines whether or not there is a sign of an abnormality in the motor 91 based on a time-dependent change in the signal strength difference within the target range over a sign determination period. The sign determination period is, for example, the period from when the motor 91 starts operating to the point in time at which it is determined whether or not there is a sign of an abnormality. For example, when the sign determination unit 58 detects a change in the signal strength difference that rapidly approaches the lower limit of the target range, it determines that there is a sign of an abnormality in the motor 91.

[0069] When the sign determination unit 58 determines that there is a sign of an abnormality in the electric motor 91, it transmits a determination result indicating that there is a sign of an abnormality to the monitoring device 61. When the monitoring device 61 receives the determination result from the abnormality determination unit 55 or the determination result from the sign determination unit 58, it outputs the determination result.

[0070] As described above, the abnormality determination device 44 according to the fourth embodiment makes it possible to determine whether or not there is a sign of an abnormality in the electric motor 91 before an abnormality occurs in the electric motor 91. This makes it possible to prompt an operator to perform maintenance work on the electric motor 91 before an abnormality occurs in the electric motor 91.

[0071] Fifth Embodiment The discrimination criteria used to determine whether or not there is an abnormality in the electric motor 91 may be obtained by a learning device. An abnormality discrimination system including a learning device will be described in a fifth embodiment, focusing on the differences from the first embodiment.

[0072] The abnormality discrimination system 32 shown in Figure 12 includes an abnormality discrimination device 41 that discriminates whether or not there is an abnormality in the electric motor 91, a monitoring device 61 that outputs the discrimination results obtained from the abnormality discrimination device 41, and a learning device 71 that determines the discrimination criteria used by the abnormality discrimination device 41 to discriminate whether or not there is an abnormality in the electric motor 91.

[0073] The hardware configuration of the abnormality determination device 41 according to the fifth embodiment is similar to the hardware configuration of the abnormality determination device 41 according to the first embodiment shown in FIG.

[0074] The learning device 71 includes a learning unit 72 that learns the correspondence between rotation speed and signal strength difference, and a model generation unit 73 that generates an abnormality discrimination model from the correspondence between rotation speed and signal strength difference learned by the learning unit 72.

[0075] The learning unit 72 acquires the rotation speed of the electric motor 91 from the rotation speed acquisition unit 52 included in the abnormality determination device 41. The learning unit 72 acquires the signal strength difference from the abnormality determination unit 55 included in the abnormality determination device 41. The learning unit 72 learns the association between the rotation speed and the signal strength difference acquired as described above during a normal period from when the electric motor 91 starts operating until the period during which the electric motor 91 can be considered normal has elapsed.

[0076] The model generation unit 73 receives the rotation speed and the signal strength difference from the correspondence between the rotation speed and the signal strength difference learned by the learning unit 72, and determines an anomaly discrimination model, which is a neural network model that outputs the presence or absence of an abnormality in the electric motor 91. The model generation unit 73 sends the generated anomaly discrimination model to the anomaly discrimination unit 55 included in the anomaly discrimination device 41.

[0077] In detail, the model generation unit 73 uses the rotation speed and the signal strength difference as input values ​​and the presence or absence of an abnormality as output values ​​to generate a neural network model having an input layer, an intermediate layer, and an output layer. The correspondence between the rotation speed and the signal strength difference during a normal period learned by the learning unit 72 becomes the input value when the output value indicates that there is no abnormality in the electric motor 91. The model generation unit 73 adjusts the weights between the input layer and the intermediate layer, the weights between the intermediate layers, and the weights between the intermediate layer and the output layer based on the learning data consisting of the rotation speed and the signal strength difference.

[0078] The abnormality determination unit 55 included in the abnormality determination device 41 determines whether or not there is an abnormality in the electric motor 91 by applying the rotation speed and the signal strength difference to the abnormality determination model acquired from the learning device 71 .

[0079] As described above, the abnormality detection system 32 according to the fifth embodiment uses the learning device 71 to generate an abnormality detection model that serves as a detection criterion for detecting an abnormality in the electric motor 91. This makes it possible to determine whether or not there is an abnormality in the electric motor 91 based on a detection criterion according to the rotation speed of the electric motor 91, in accordance with the individual characteristics of the electric motor 91.

[0080] The present disclosure is not limited to the above-described examples. Any combination of the above-described embodiments may be used. As an example, the abnormality determination unit 55 included in the abnormality determination device 43, 44 may determine whether or not an abnormality exists in the electric motor 91 by comparing the signal strength difference with a target range corresponding to the rotation speed and target torque of the electric motor 91.

[0081] As another example, learning device 71 may learn the correspondence between rotational speed, target torque, and signal strength difference during a normal period, and may obtain an abnormality determination model that receives the rotational speed, target torque, and signal strength difference as input and outputs whether or not there is an abnormality in electric motor 91. In particular, learning unit 72 may learn the correspondence between rotational speed and target torque and signal strength difference, and model generation unit 73 included in learning device 71 may generate an abnormality determination model from the correspondence between rotational speed and target torque and signal strength difference.

[0082] In order to improve the accuracy of the determination, the abnormality determination devices 41-44 may determine whether or not there is an abnormality in the electric motor 91 only when the disturbance strength is within an allowable range. As an example, the operation of the abnormality determination device 41 that determines whether or not there is an abnormality only when the disturbance strength is within an allowable range is shown in Fig. 13. The processing from steps S11 to S19 in Fig. 13 is the same as the processing from steps S11 to S19 performed by the abnormality determination device 41 according to the first embodiment shown in Fig. 3.

[0083] The intensity determination unit 54 determines the disturbance intensity from the spectrum data (step S31). Specifically, the intensity determination unit 54 determines the disturbance intensity, which is the intensity of the frequency component between the power supply frequency and the frequency of the sideband.

[0084] The intensity determination unit 54 determines whether the disturbance intensity is within an allowable range (step S32). The allowable range is determined according to the value that the disturbance intensity can take when the effect of the disturbance on the accuracy of determining whether or not there is an abnormality in the electric motor 91 based on the sideband intensity is sufficiently small.

[0085] If the disturbance intensity is within the allowable range (step S32; Yes), the intensity determination unit 54 sends the intensity of the sideband waves and the intensity of the frequency component of the power supply frequency to the abnormality determination unit 55. The abnormality determination unit 55 performs the subsequent processing as in embodiment 1. If the disturbance intensity is outside the allowable range (step S32; No), the subsequent processing is not performed, and the processing is repeated from step S11.

[0086] As another example of improving the accuracy of determination, the abnormality determination unit 55 included in the abnormality determination device 41-44 may determine whether or not an abnormality has occurred in the electric motor 91 based on the number of times the signal strength difference exceeds a target range during a period for which abnormality determination is to be performed. The period for which abnormality determination is to be performed is, for example, the period from when operation of the electric motor 91 begins to when abnormality determination is performed. For example, the abnormality determination unit 55 may determine that an abnormality has occurred in the electric motor 91 when the number of times the signal strength difference exceeds the target range is equal to or greater than an abnormality threshold. The abnormality threshold is a value that can prevent erroneous determination that an abnormality has occurred in the electric motor 91 due to temporary fluctuations in sideband wave intensity.

[0087] As one example, the current obtaining unit 51 may obtain a current value directly from the current detection circuit 13. As another example, the current obtaining unit 51 may obtain a V-phase current value or a W-phase current value. In this case, the conversion unit 53 may perform an FFT on the current data indicating the V-phase current value or the W-phase current value to generate spectrum data.

[0088] The method of anomaly determination by the anomaly determination unit 55 is not limited to the above example, and any method may be used as long as it can determine whether or not an anomaly has occurred in the motor 91 based on the sideband wave intensity and a determination criterion corresponding to the rotation speed. As an example, the anomaly determination unit 55 may determine whether or not an anomaly has occurred in the motor 91 by comparing the sideband wave intensity with a target intensity range corresponding to the rotation speed. In more detail, the anomaly determination unit 55 determines that no anomaly has occurred in the motor 91 if the sideband wave intensity is within the target intensity range, and determines that an anomaly has occurred in the motor 91 if the sideband wave intensity is outside the target intensity range. In this case, the criterion determination unit may determine the target intensity range, which is the range of target values ​​for the sideband wave intensity, from the correspondence between the rotation speed and the sideband wave intensity.

[0089] As another example, the abnormality determination unit 55 may determine the presence or absence of an abnormality in the electric motor 91 by comparing the ratio of the sideband wave intensity to the intensity of the frequency component of the power supply frequency with a target ratio range corresponding to the rotation speed. In this case, the reference determination unit may determine the target ratio range, which is the range of target values ​​for the ratio, by associating the rotation speed with the ratio of the sideband wave intensity to the intensity of the frequency component of the power supply frequency.

[0090] The reference determination unit 57 may update the preset target range based on the association between the rotation speed and the sideband intensity.

[0091] 3, the determination of whether the electric motor 91 is operating may be performed based on the rotation speed. In this case, the acquisition of the rotation speed may be performed in parallel with step S11 in step S14.

[0092] The rotation speed acquisition unit 52 may acquire the rotation speed of the electric motor 91 from a speed sensor that measures the rotation speed of the electric motor 91 .

[0093] 14, the power conversion device 11, the power conversion control device 21, and the abnormality determination device 41 may be realized together as a power supply device 74. The same applies to the abnormality determination devices 42-44.

[0094] The power conversion device 11 is not limited to being mounted on a DC-fed railway vehicle, but can be mounted on any vehicle, such as an AC-fed railway vehicle or a railway vehicle equipped with an internal combustion engine. When the power conversion device 11 is mounted on an AC-fed railway vehicle, it is sufficient to provide a transformer that steps down the voltage of AC power supplied from a current collector and a converter that converts the AC power stepped down by the transformer into DC power. The power conversion device 11 converts the DC power supplied from the converter into AC power and supplies the converted AC power to the electric motor 91.

[0095] The abnormality determination devices 41-44 may be implemented as one function of a train information management system. Also, the abnormality determination devices 41-44 may be provided in, for example, a train operation control center, rather than being mounted on a railway vehicle.

[0096] The electric motor 91 may be either a three-phase induction motor or a three-phase synchronous motor. Furthermore, the electric motor 91 is not limited to a three-phase motor, and may be, for example, a single-phase motor or a DC motor. The electric motor 91 may be an inner rotor or an outer rotor.

[0097] The core part of the control processing, which includes the processor 81, memory 82, and interface 83, can be realized using a normal computer system rather than a dedicated system. For example, the abnormality determination devices 41-44 that perform the above-described processing may be realized by storing and distributing a computer program for executing the above-described operations on a computer-readable recording medium (such as a flexible disk, a CD-ROM (Compact Disc-Read Only Memory), or a DVD-ROM (Digital Versatile Disc-Read Only Memory)), and installing the computer program on a computer. Alternatively, the abnormality determination devices 41-44 may be realized by storing the computer program in a storage device of a server device on a communication network and downloading it to a normal computer system.

[0098] When the functions of the abnormality determination devices 41-44 are realized by sharing the functions between an OS (Operating System) and an application program, or by collaboration between the OS and an application program, only the application program portion may be stored on a recording medium, storage device, etc.

[0099] It is also possible to superimpose a computer program on a carrier wave and distribute it via a communication network. For example, the computer program may be posted on a bulletin board system (BBS) on the communication network and distributed via the communication network. The computer program may then be started and executed under the control of an OS in the same way as other application programs, thereby executing the above-described processing.

[0100] 15 , the abnormality determination devices 41-44 may be realized by a processing circuit 84. The processing circuit 84 is connected to the power conversion control device 21, the monitoring device 61, etc. via an interface circuit 85. When the processing circuit 84 is dedicated hardware, the processing circuit 84 is, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof. Each unit of the abnormality determination devices 41-44 may be realized by an individual processing circuit 84, or each unit of the abnormality determination devices 41-44 may be realized by a common processing circuit 84.

[0101] Some of the functions of the abnormality determination devices 41-44 may be realized by dedicated hardware, and other functions may be realized by software or firmware. For example, the current acquisition unit 51 and the rotation speed acquisition unit 52 included in the abnormality determination device 41 may be realized by a processing circuit 84 shown in Fig. 15, and the conversion unit 53, the intensity determination unit 54, and the abnormality determination unit 55 may be realized by the processor 81 shown in Fig. 2 reading and executing programs stored in a memory 82.

[0102] The present disclosure allows various embodiments and modifications without departing from the broad spirit and scope of the present disclosure. Furthermore, the above-described embodiments are intended to illustrate the present disclosure and do not limit the scope of the present disclosure. That is, the scope of the present disclosure is defined by the claims, not the embodiments. Various modifications made within the scope of the claims and the meaning of equivalent disclosures are considered to be within the scope of the present disclosure.

[0103] 11 Power conversion device, 11a, 11b Input terminal, 12 Power conversion circuit, 13 Current detection circuit, 21 Power conversion control device, 31, 32 Abnormality discrimination system, 41, 42, 43, 44 Abnormality discrimination device, 51 Current acquisition unit, 52 Rotational speed acquisition unit, 53 Conversion unit, 54 Strength determination unit, 55 Abnormality discrimination unit, 56 Torque acquisition unit, 57 Reference determination unit, 58 Sign discrimination unit, 61 Monitoring device, 71 Learning device, 72 Learning unit, 73 Model generation unit, 74 Power supply device, 80 Bus, 81 Processor, 82 Memory, 83 Interface, 84 Processing circuit, 85 Interface circuit, 91 Electric motor, C1 Capacitor, L1 Reactor, S1 Operation command, S2 Power conversion control signal, Iu U-phase current value, N1 Rotational speed, τ1 Target torque.

Claims

1. An abnormality detection device comprising: a current acquisition unit that acquires time domain current data indicating a current flowing from a power conversion device to an electric motor; a rotational speed acquisition unit that acquires the rotational speed of the electric motor; a conversion unit that generates frequency domain spectrum data from the current data; an intensity determination unit that determines from the spectrum data a sideband wave intensity, which is the intensity of a sideband wave, which is a frequency component shifted from the power supply frequency of the power conversion device by the rotational frequency of the electric motor; and an abnormality detection unit that determines the presence or absence of an abnormality in the electric motor from the sideband wave intensity and a detection criterion according to the rotational speed.

2. The abnormality detection device according to claim 1, wherein the intensity determination unit further determines the intensity of the frequency component of the power supply frequency from the spectrum data, and the abnormality detection unit includes a target value of a signal intensity difference which is the difference between the intensity of the frequency component of the power supply frequency and the sideband intensity, and uses a target range which changes according to the rotation speed as the detection criterion to determine the presence or absence of an abnormality in the motor by comparing the signal intensity difference with the target range.

3. The abnormality detection device according to claim 2, further comprising a sign detection unit that detects the presence or absence of a sign of an abnormality in the electric motor from a change over time in the signal strength difference within the target range over a period of time subject to sign detection.

4. The abnormality determination device according to claim 2 or 3, wherein the abnormality determination unit determines whether or not an abnormality exists in the electric motor based on the number of times that the sideband wave intensity exceeds the target range during a period subject to abnormality determination.

5. An abnormality detection device as described in any one of claims 1 to 4, further comprising a criterion determination unit that determines the discrimination criterion corresponding to the rotation speed from the sideband wave intensity determined by the intensity determination unit and the rotation speed acquired by the rotation speed acquisition unit during a normal period from the start of operation of the motor to the elapse of a period during which the motor can be considered normal.

6. An abnormality detection device as described in any one of claims 1 to 5, wherein the intensity determination unit determines from the spectrum data a disturbance intensity, which is the intensity of a frequency component between the power supply frequency and the frequency of the sideband wave, and the abnormality detection unit determines the presence or absence of an abnormality in the motor from the sideband wave intensity and the detection criterion when the disturbance intensity is within an allowable range.

7. An abnormality detection device as claimed in any one of claims 1 to 6, further comprising a torque acquisition unit that acquires a torque of the electric motor, wherein the abnormality detection unit determines whether or not there is an abnormality in the electric motor based on the sideband intensity and the detection criterion corresponding to the rotation speed and the torque.

8. The abnormality detection device according to claim 7, further comprising a criterion determination unit that determines the discrimination criterion according to the rotational speed and the torque from the sideband intensity determined by the intensity determination unit, the rotational speed acquired by the rotational speed acquisition unit, and the torque acquired by the torque acquisition unit during a normal period from the start of operation of the motor until the elapse of a period during which the motor can be considered normal.

9. A learning device that generates an abnormality discrimination model that discriminates whether or not an abnormality exists in an electric motor receiving power from a power conversion device, comprising: a learning unit that learns a correspondence between a sideband wave intensity, which is the intensity of a sideband wave that is a frequency component shifted from the power supply frequency of the power conversion device by the rotational frequency of the electric motor, and a signal intensity difference, which is the difference in intensity of the frequency component of the power supply frequency, which is indicated by frequency domain spectrum data based on time domain current data that indicates the current flowing from the power conversion device to the electric motor during a normal period from when the operation of the electric motor begins to the time when the period in which the electric motor can be considered normal has elapsed, and the rotational speed of the electric motor; and a model generation unit that generates an abnormality discrimination model that outputs the presence or absence of an abnormality in the electric motor by inputting the signal intensity difference and the rotational speed of the electric motor, based on the correspondence between the signal intensity difference and the rotational speed learned by the learning unit.

10. A power supply device comprising: a power conversion device that converts power supplied from a power source into power to be supplied to an electric motor, and supplies the converted power to the electric motor; a power conversion control device that acquires a value of a current flowing to the electric motor from the power conversion device and a rotational speed of the electric motor, determines a target torque for the electric motor in accordance with an operation command for the electric motor, and controls the power conversion device based on the value of the current flowing from the power conversion device to the electric motor, the rotational speed, and the target torque; and an abnormality determination device as defined in any one of claims 1 to 6, wherein the current acquisition unit provided in the abnormality determination device acquires, from the power conversion control device, the value of a current flowing to the electric motor used in controlling the power conversion device as the current data, and the rotational speed acquisition unit provided in the abnormality determination device acquires, from the power conversion control device, the rotational speed used in controlling the power conversion device.

11. A power supply device comprising: a power conversion device that converts power supplied from a power source into power to be supplied to an electric motor, and supplies the converted power to the electric motor; a power conversion control device that acquires a value of a current flowing to the electric motor from the power conversion device and a rotational speed of the electric motor from the power conversion device, determines a target torque of the electric motor according to an operation command of the electric motor, and controls the power conversion device based on the value of the current flowing from the power conversion device to the electric motor, the rotational speed, and the target torque; and an abnormality determination device as defined in claim 7 or 8, wherein the current acquisition unit included in the abnormality determination device acquires, as the current data, from the power conversion control device, a value of a current flowing to the electric motor used for controlling the power conversion device; the rotational speed acquisition unit included in the abnormality determination device acquires, from the power conversion control device, the rotational speed used for controlling the power conversion device; and the torque acquisition unit included in the abnormality determination device acquires, from the power conversion control device, the target torque used for controlling the power conversion device as the torque of the electric motor.

12. An abnormality discrimination system comprising: an abnormality discrimination device according to any one of claims 1 to 8; and a monitoring device that acquires a discrimination result of the abnormality discrimination section provided in the abnormality discrimination device and outputs the discrimination result.

13. An abnormality discrimination system comprising: an abnormality discrimination device according to any one of claims 2 to 4; and a learning device according to claim 9, wherein the abnormality discrimination section of the abnormality discrimination device applies the rotation speed acquired by the rotation speed acquisition section and the signal intensity difference, which is the difference between the intensity of the frequency component of the power supply frequency determined by the intensity determination section and the sideband intensity, to the abnormality discrimination model generated by the model generation section of the learning device to determine the presence or absence of an abnormality in the electric motor.

14. A method for discriminating an abnormality, comprising: generating frequency domain spectrum data from time domain current data indicating a current flowing from a power conversion device to an electric motor; determining from the spectrum data a sideband wave intensity, which is the intensity of a sideband wave, which is a frequency component shifted from the power supply frequency of the power conversion device by the rotational frequency of the electric motor; and discriminating whether or not an abnormality exists in the electric motor based on a discrimination criterion according to the sideband wave intensity and the rotational speed of the electric motor.

Images