Elevator abnormality detection device and elevator abnormality detection method

Abstract

To improve the accuracy of detecting abnormalities in an elevator device.SOLUTION: An elevator abnormality detection device according to an embodiment includes a data acquisition unit, a Fourier transform unit, and an abnormality detection unit. The data acquisition unit acquires time series data indicating a voltage or current waveform from a measurement device that measures the waveform of the voltage or current supplied from a power source to a motor. An extraction unit extracts time series data of a specified time length from the time series data acquired by the data acquisition unit. The Fourier transform unit generates a spectral distribution by performing a Fourier transform on the time series data extracted by the extraction unit. The abnormality detection unit detects an abnormality in the items constituting an elevator device based on the generated spectral distribution by AI using supervised data.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] An embodiment of the present invention relates to an elevator abnormality detection device and an elevator abnormality detection method. [Background technology]

[0002] As a method for detecting an abnormality in an elevator system, a method is known in which the voltage waveform and current waveform of the inverter output that supplies power to the motor that drives the elevator car to move up and down are monitored, and whether or not there is an abnormality in the inverter is detected based on the difference from the normal waveform. Another method is known in which the torque ripple of the motor is measured, and based on the difference from the normal torque ripple, an abnormality in the motor or a foreign object being caught is detected.

[0003] In addition, a technology has been disclosed that performs FFT conversion on the voltage waveform of the inverter output and the torque ripple of the motor to obtain a spectral distribution, and then analyzes the spectral distribution using AI to detect abnormalities in the inverter, motor, or equipment driven by the motor. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 5481286 [Patent Document 2] JP 2020-64481 A

[0005] The inverter output voltage waveform and current waveform and motor torque ripple monitored while the elevator car is ascending and descending contain noise such as surrounding mechanical vibrations that accompany the ascending and descending movement of the car. If the presence or absence of an abnormality is judged based on the inverter output voltage waveform and current waveform or motor torque ripple that contain such noise, the probability of erroneous judgment increases. Summary of the Invention [Problem to be solved by the invention]

[0006] The present invention has been made in view of the above-mentioned circumstances, and an object of the present invention is to improve the accuracy of abnormality detection in an elevator system. [Means for solving the problem]

[0007] The elevator abnormality detection device according to the embodiment for solving the above problem is an elevator abnormality detection device that detects an abnormality in an elevator system having a motor that drives the elevator car to ascend and descend, a power source that supplies power to the motor, and a control device that controls the motor. The elevator abnormality detection device according to the embodiment includes a data acquisition unit, a Fourier transformation unit, and an abnormality detection unit. The subtraction processing section The data acquisition unit acquires time series data indicating a voltage or current waveform from a measuring device that measures the waveform of the voltage or current supplied from the power supply to the motor. The extraction unit extracts time series data of a specified time length from the time series data acquired by the data acquisition unit. The Fourier transform unit generates a spectral distribution by performing a Fourier transform on the time series data extracted by the extraction unit. The anomaly detection unit detects anomalies in the components that make up the elevator system based on the generated spectral distribution by AI using supervised data. The subtraction processing unit subtracts the value of the spectral distribution caused by the operation of the equipment when the equipment is determined to be free of abnormality from the value of the spectral distribution generated by the Fourier transform unit. The abnormality detection unit detects an abnormality in the equipment constituting the elevator apparatus based on the spectral distribution after the subtraction process. [Brief description of the drawings]

[0008] [Figure 1] 1 is a perspective view of an elevator apparatus according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a block diagram showing a control system of the elevator apparatus according to the present embodiment. [Diagram 3] FIG. 2 is a block diagram of a drive unit according to the embodiment. [Figure 4] FIG. 2 is a physical block diagram of a control unit according to the present embodiment. [Diagram 5] FIG. 2 is a functional block diagram of a control unit according to the present embodiment. [Figure 6] FIG. 4 is a diagram for explaining an operation pattern according to the present embodiment. [Figure 7] FIG. 2 is a diagram for explaining time-series data according to the embodiment; [Figure 8] 1 is a diagram for explaining the operating cycle of the equipment constituting the elevator apparatus according to the present embodiment. FIG. [Figure 9] 4 is a diagram for explaining time-series data extracted by an extracting unit according to the embodiment; FIG. [Figure 10] 5A to 5C are diagrams for explaining a spectral distribution generated by a Fourier transform section according to the embodiment. [Figure 11] 6 is a diagram for explaining the operation of a subtraction processing unit according to the present embodiment. FIG. [Figure 12] 4 is a flowchart for explaining a data measurement process of the elevator abnormality detection device according to the present embodiment. [Figure 13] 4 is a flowchart for explaining an abnormality detection process of the elevator abnormality detection device according to the present embodiment. [Figure 14] 5 is a flowchart for explaining a process of adding supervised data in the elevator abnormality detection device according to the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, the present embodiment will be described with reference to the drawings. In the description, an XYZ coordinate system consisting of mutually orthogonal X-axis, Y-axis, and Z-axis will be used as appropriate. The drawings and flowcharts used in the description of the present embodiment are merely examples.

[0010] (Embodiment 1) Fig. 1 is a perspective view of an elevator system 10 according to this embodiment. The elevator system 10 is disposed inside a hoistway 100 provided in a building such as a commercial facility or a residential facility. As shown in Fig. 1, the elevator system 10 includes a car 31, a counterweight 50, an elevator motor 40, guide rails 21 to 24, and a control panel 70 (elevator control device).

[0011] Each of the guide rails 21 to 24 is a member whose longitudinal direction is the Z-axis direction. The guide rails 21 and 22 are a pair of members for guiding the car 31 so as to be able to rise and fall freely. The guide rails 23 and 24 are a pair of members for guiding the counterweight 50 so as to be able to rise and fall freely. The guide rails 21 and 22 are arranged spaced apart from each other in the Y-axis direction. Similarly, the guide rails 23 and 24 are arranged spaced apart from each other in the Y-axis direction. In FIG. 1, the guide rails 23 and 24 of the counterweight 50 are arranged spaced apart from each other in the X-axis direction with respect to the guide rails 21 and 22 of the car 31. The arrangement of the guide rails 21 to 24 is not limited to the arrangement shown in FIG. 1.

[0012] The car 31 is a unit that accommodates passengers and moves them up and down the elevator shaft 100. The car 31 is disposed between the guide rails 21 and 22, and is attached to the guide rails 21 and 22 so as to be movable in the up and down direction.

[0013] An opening 31a for entering and exiting the interior is formed on the side surface on the +X side of the car 31. The opening 31a is closed or opened by a pair of doors 32 that move along the side surface of the car 31. The doors 32 are opened and closed by an opening / closing motor (not shown in FIG. 1).

[0014] The counterweight 50 is attached so as to be movable up and down relative to the guide rails 23 and 24. The weight of the counterweight 50 is adjusted to be a predetermined ratio to the weight of the car 31.

[0015] The lift motor 40 is a motor for lifting and lowering the car 31. The lift motor 40 is disposed at the upper part of the elevator shaft 100 such that the rotation axis is parallel to the Y axis. A pulley 42 is fixed to the rotation axis of the lift motor 40.

[0016] A wire 43 is wound around the pulley 42 of the lift motor 40. One end of the wire 43 is fixed to the car 31, and the other end is fixed to the counterweight 50.

[0017] The control panel 70 is disposed in the hoistway 100. The control panel 70 houses a control device for controlling the hoisting motor 40 and the equipment provided in the car 31. Note that, in the following embodiment, a machine room-less elevator in which the control panel 70 is disposed in the hoistway 100 will be described as an example, but this embodiment can also be applied to a case in which a machine room is provided.

[0018] 2 is a block diagram showing a control system of the elevator apparatus 10. The control system includes a control unit 80 and a drive unit 91 housed in a control panel 70, and an operation panel 36 provided in the car 31.

[0019] The operation panel 36 is provided on the inner wall surface of the car 31. The operation panel 36 is an interface for receiving destination floors and the like from users of the car 31. Users can register destination floors and the like of the car 31 and open and close the door 32 by operating the operation panel 36. The operation panel 36 is connected to a control unit 80 housed in the control panel 70 via a cable 44 shown in FIG. 1.

[0020] 2 drives the lift motor 40 and the opening / closing motor 41 (not shown in FIG. 1) that drives the door 32 of the car 31 by supplying power to the lift motor 40 and the opening / closing motor 41. The drive unit 91 drives the lift motor 40 based on instructions from the control unit 80. The drive unit 91 also drives the opening / closing motor 41 based on instructions from the control unit 80.

[0021] 3 is a block diagram of the drive unit 91. The drive unit 91 has an inverter 13 and a measuring device 15. The inverter 13 is a power supply device that supplies power to the lift motor 40 and the opening / closing motor 41. The inverter 13 is configured with a switching regulator. When the lift motor 40 and the opening / closing motor 41 are formed as three-phase AC motors, the inverter 13 outputs a three-phase AC voltage.

[0022] The measurement device 15 has a current measurement device 151 and a voltage measurement device 152. The current measurement device 151 is a device that measures the current supplied by the inverter 13 to the lift motor 40 and the opening / closing motor 41 for each phase. The voltage measurement device 152 is a device that measures the output voltage of the inverter 13 for each phase.

[0023] FIG. 4 is a physical block diagram of the control unit 80. The control unit 80 is a computer having a CPU (Central Processing Unit) 81, a main memory unit 82, an auxiliary memory unit 83, and an interface unit 84, which are interconnected via a bus 85. The CPU 81 executes a process described below according to a program stored in the auxiliary memory unit 83. The main memory unit 82 has a RAM (Random Access Memory) and the like. The main memory unit 82 is used as a working area for the CPU 81. The auxiliary memory unit 83 has a non-volatile memory such as a ROM (Read Only Memory) and a semiconductor memory. The auxiliary memory unit 83 stores the program executed by the CPU 81, various parameters, and the like. The auxiliary memory unit 83 also stores supervised data used for anomaly detection by AI (Artificial Intelligence). The supervised data may be data in which all data is labeled, or may be semi-supervised data in which a portion is labeled. Details will be described later.

[0024] The interface unit 84 has a serial interface, a parallel interface, a wireless LAN interface, etc. The operation panel 36 and the drive unit 91 are connected to the CPU 81 via the interface unit 84. In addition, an input / output device 93 composed of a keyboard, a display, etc. is connected to the interface unit 84.

[0025] 5 is a functional block diagram of the control unit 80. The CPU 81 of the control unit 80 realizes a drive unit control unit 71 and an abnormality detection device 72 by executing a program stored in the auxiliary storage unit 83.

[0026] The drive unit control unit 71 controls the drive unit 91 based on input from the operation panel 36 or the call panel of each floor. For example, when the drive unit control unit 71 rotates the lift motor 40 in the normal direction via the drive unit 91, the car 31 rises and the counterweight 50 descends. When the drive unit control unit 71 rotates the lift motor 40 in the reverse direction via the drive unit 91, the car 31 descends and the counterweight 50 rises. When the drive unit control unit 71 rotates the opening / closing motor 41 in the normal direction via the drive unit 91, the door 32 of the car 31 and the doors provided at the landings on each floor are controlled to be open, and when the opening / closing motor 41 is rotated in the reverse direction, the door 32 of the car 31 and the doors provided at the landings on each floor are controlled to be closed.

[0027] The abnormality detection device 72 detects abnormalities in the equipment constituting the elevator system 10 based on a spectral distribution obtained by Fourier transforming time-series data indicating changes in current or voltage measured by the measuring device 15, using AI with supervised data. The abnormality detection device 72 has an operation pattern setting unit 73, a data acquisition unit 75, an extraction unit 76, a Fourier transform unit 77, a subtraction processing unit 78, and an abnormality detection unit 79.

[0028] The operation pattern setting unit 73 sets an operation pattern that specifies the moving speed and moving acceleration of the car 31 in the drive unit control unit 71. The operation patterns include an operation pattern for normal operation to transport passengers and an operation pattern for abnormality detection to detect an abnormality in the equipment constituting the elevator device 10. The operation patterns are stored in the storage unit in advance. The operation pattern setting unit 73 selects an operation pattern from the storage unit and supplies it to the drive unit control unit 71.

[0029] FIG. 6 shows an example of an operation pattern when an abnormality is detected. FIG. 6(A) shows an operation pattern in which the speed of the car 31 is changed at a lower acceleration than during normal operation. By changing the speed at a low acceleration, it becomes easier to detect noise caused by the vibration of the product or the operating cycle that appears at a specific speed. FIG. 6(B) shows an operation pattern in which the car 31 moves while maintaining a constant speed. FIG. 6(C) shows an operation pattern in which the speed of the car 31 is increased at a higher acceleration than during normal operation, and then the lifting and lowering speed of the car 31 is reduced at a low acceleration. FIG. 6(D) shows an operation pattern in which the acceleration control of FIG. 6(A) is performed in a stepped manner. The operation pattern when an abnormality is detected is set not only for the lifting and lowering control of the car 31, but also for the opening and closing control of the door 32.

[0030] Returning to FIG. 5, the data acquisition unit 75 acquires time series data showing the waveform of the voltage or current from the measuring device 15 that measures the waveform of the voltage or current supplied by the inverter 13 to the lift motor 40 and the opening / closing motor 41. The time series data measured by the measuring device 15 may be temporarily stored in the storage unit, and the data acquisition unit 75 may acquire the time series data from the storage unit. FIG. 7 shows an example of a part of the time series data showing the acquired voltage waveform. The time series data includes frequency components corresponding to voltage changes accompanying motor control, as well as noises of various frequency components caused by the vibration and operation cycle of the equipment. The time length of the time series data acquired in one operation pattern is, for example, 10 seconds, 1 minute, or 10 minutes.

[0031] Meanwhile, the components constituting the elevator device 10 each have their own operating cycle. This operating cycle is also related to the operation pattern. For example, the rotation cycle of the sheave related to the cable connecting the car 31 and the counterweight 50 is determined by the ascending and descending speed of the car 31. Also, the cycle of vibration of the guide rails 21 to 24 is determined by the ascending and descending speed and ascending and descending acceleration of the car 31 and the moving distance. Also, the rotation cycle of the sheave used to open and close the door 32 is determined by the opening and closing speed of the door 32. FIG. 8 shows an example of the unique operating cycle of each of the components constituting the elevator device 10. The information shown in FIG. 8 is stored for each operating speed of the operation pattern.

[0032] Returning to FIG. 5, the extraction unit 76 specifies a time length from the time series data acquired by the data acquisition unit 75 and extracts a part of the time series data. FIG. 9 is a diagram for explaining the time series data. The time length is set based on the operation cycle specific to the product shown in FIG. 8. For example, the time length Wm for detecting an abnormality in the sheave is a time length (e.g., about 0.1 seconds) several times (e.g., two or four times) 1 / fm, and the time length Wn for detecting an abnormality in the guide rail is a time length (e.g., about 100 seconds) several times 1 / fn. When inspecting products related to the door 32, the time length may be, for example, about 2 seconds.

[0033] Returning to Fig. 5, the Fourier transform unit 77 generates a spectral distribution by performing a Fourier transform on the time series data extracted by the extraction unit 76. Fig. 10 shows an example of the spectral distribution generated by the Fourier transform unit 77. The spectral values ​​of the frequencies corresponding to the operation cycles of the respective products shown in Fig. 8 are large.

[0034] The subtraction processing unit 78 performs a process of subtracting the spectral distribution value caused by the operation of the equipment when no abnormality is detected for the equipment that is determined to be free of abnormality from the spectral distribution value generated by the Fourier transform unit 77. For example, when it is determined that no abnormality is detected for the roller guide shown in FIG. 8, the spectral distribution value caused by the operation when no abnormality is detected for the roller guide is subtracted from the spectral distribution value shown in FIG. 10. For the time series data acquired in the operation pattern shown in FIG. 6, the spectral distribution data for the equipment when no abnormality is detected corresponding to the time length specified by the extraction unit 76 is stored in advance in the storage unit. By this subtraction process, the spectrum indicated by the dotted line in FIG. 11 is deleted, or the value of the spectrum indicated by the dotted line is reduced.

[0035] The anomaly detection unit 79 detects anomalies in the products constituting the elevator system 10 based on the spectral distribution generated and output by the subtraction processing unit 78 using an AI that uses supervised data. The anomaly detection unit 79 is composed of an AI with a learning function. The AI ​​is composed of, for example, a neural network or a support vector machine. The AI ​​that constitutes the anomaly detection unit 79 detects whether or not there is an abnormality in the products constituting the elevator system 10 by comparing the feature amount of the spectral distribution output by the subtraction processing unit 78 with the feature amount of the supervised data.

[0036] Here, the supervised data will be described. The supervised data is data in which the data of the spectrum distribution when there is an abnormality is labeled as "abnormal" and the data of the spectrum distribution when there is no abnormality is labeled as "no abnormality". For example, the spectrum distribution when the sheave is deformed may be labeled as "deformed sheave" and the spectrum distribution when the sheave is scratched may be labeled as "scratched sheave". In the case of the guide rail, the spectrum distribution data corresponding to the distortion of the guide rail or the loosening of the mounting screw may be labeled to indicate the cause of the abnormality. This supervised data is created for a time length set by the extraction unit 76 and a large number of data for each operation pattern shown in FIG. 6, and stored in the storage unit. This supervised data is created for a time length set by the extraction unit 76 and a large number of data for each operation pattern shown in FIG. 6, even when all the items are normal. That is, this supervised data can be used as semi-supervised data.

[0037] Next, the measurement process of data used in the abnormality detection process of the elevator device 10 will be described with reference to the flowchart shown in Fig. 12. The following control is performed based on a program stored in the auxiliary storage unit 83, and is mainly controlled by the control unit 80 (CPU 81). The measurement device 15 measures the output current and output voltage of the inverter 13, and notifies the control unit 80 of the measured values. Here, a case will be described in which data is measured by moving the car 31 up and down in the four types of operation patterns shown in Fig. 6.

[0038] The operation pattern setting section 73 first sets the operation pattern shown in FIG. 6(A) in the drive unit control section 71 (step S11).

[0039] The measuring device 15 measures time series data of voltage and current when the car 31 is under control to move or the door 32 is under control to open and close according to the corresponding operation pattern (step S12), and stores the data in the memory unit.

[0040] The driving pattern setting unit 73 judges whether or not all driving patterns have been performed (step S13). If all driving patterns have not been performed (step S13: No), the driving pattern setting unit 73 returns to step S11 and sets the driving pattern that has not been performed. For example, in the second step S13, the driving pattern setting unit 73 sets the driving pattern shown in FIG. 6(B) in the drive unit control unit 71. In addition, in the third step S13, the driving pattern setting unit 73 sets the driving pattern shown in FIG. 6(C) in the drive unit control unit 71. On the other hand, if all driving patterns have been performed (step S13: Yes), the abnormality detection device 72 ends the data measurement process.

[0041] Next, the abnormality detection process (elevator abnormality detection method) of the elevator device 10 will be described with reference to the flowchart shown in Fig. 13. It is assumed that the above-mentioned data acquisition process has been completed and the measured time series data has been stored in the storage unit. The supervised data according to the operation pattern and the time length set by the extraction unit 76 has been stored in the storage unit in advance.

[0042] First, the data acquiring unit 75 extracts time series data for each driving pattern from the storage unit (step S31). For example, the data acquiring unit 75 first extracts time series data measured in the driving pattern shown in FIG. 6(A) from the storage unit.

[0043] Next, the extraction unit 76 specifies a time length from the time series data acquired by the data acquisition unit 75 and extracts a part of the time series data (step S32). The time length is set in advance for each product to be inspected based on the operation cycle of the product to be inspected, etc. The extraction unit 76 first sets the shortest time length. For example, when the roller guide shown in FIG. 8 is to be inspected, the extraction unit 76 sets the time length Wi shown in FIG. 9 to a time length several times (for example, two or four times) 1 / fi. Then, the extraction unit 76 supplies the time series data of the time length Wi to the Fourier transform unit 77.

[0044] Next, the Fourier transform unit 77 generates a spectral distribution by Fourier transforming the time series data of the time length Wi (step S33). The spectral distribution by Fourier transforming the time series data limited to the time length Wi does not include noise caused by products whose operation cycle is longer than fi, such as the sheave and guide rail shown in Fig. 8, or the ratio of the power of the spectrum caused by products whose operation cycle is longer than fi in the converted spectral distribution is low.

[0045] Next, the subtraction processing unit 78 performs a process of subtracting the value of the spectral distribution caused by the operation of the product when no abnormality is present for the product determined to be free of abnormality from the value of the spectral distribution generated by the Fourier transform unit 77 (step S34). Since there is no product determined to be free of abnormality in the initial process of step S34, the process of step S34 is skipped.

[0046] Next, the abnormality detection unit 79 detects whether or not there is an abnormality in the equipment constituting the elevator device 10 by comparing the feature amount of the spectrum distribution output by the subtraction processing unit 78 with the feature amount of the supervised data (step S35). The spectrum distribution obtained by Fourier transforming the time series data limited to the time length Wi does not include noise caused by equipment whose operation cycle is longer than fi, such as the sheave and guide rail shown in FIG. 8, or the ratio of the power of the spectrum caused by the equipment whose operation cycle is longer than fi in the transformed spectrum distribution is low. Therefore, the spectrum distribution obtained by Fourier transforming the time series data limited to the time length Wi has a high ratio of the feature amount of noise caused by the equipment whose operation cycle is shorter than fi. By making the feature amount of the target equipment prominent, the abnormality detection unit 79 can accurately detect whether or not there is an abnormality in the target equipment (the equipment whose operation cycle is shorter than fi).

[0047] The abnormality detection device 72 judges whether or not all the products have been analyzed (step S36). Specifically, it judges whether or not all the settings of the time lengths shown in Fig. 9 corresponding to the products shown in Fig. 8 to be inspected have been executed. If the analysis of all the products has not been completed (step S36: No), the abnormality detection device 72 returns to step S32, changes the time length, and repeats the processes from step S32 to step S36.

[0048] In the subtraction process of step S34 for the second time, for example, if it is determined in the process of step S35 for the first time that there is no abnormality in the roller guide, the value of the spectral distribution caused by the operation of the roller guide when there is no abnormality in the roller guide is subtracted from the value of the spectral distribution generated by the Fourier transform unit 77 in step S33 for the second time. As explained with reference to Fig. 11, the spectral components indicated by the dotted line are removed, so that the spectral components in the frequency band lower than the frequency fi are emphasized.

[0049] Furthermore, in the subtraction process of step S34 for the third time, if it is determined in the process of step S35 for the first time that there is no abnormality in the roller guide and in the process of step S35 for the second time that there is no abnormality in the motor and sheave, the value of the spectral distribution caused by the operation of the roller guide when there are no abnormalities in the roller guide, motor, and sheave is subtracted from the value of the spectral distribution generated by the Fourier transform unit 77 in step S33 for the third time. As explained with reference to Fig. 11, the spectral components indicated by the dotted line and the spectral components indicated by the dashed-dotted line are removed, so that the spectral components in the frequency band lower than frequency fm are emphasized.

[0050] The proportion of the spectral components of noise caused by the items for which the presence or absence of an abnormality has not been determined in the spectral distribution increases by subtracting the value of the spectral distribution caused by the operation of the items for which the presence or absence of an abnormality has not been determined for the items for which the presence or absence of an abnormality has not been determined for the items for which the presence or absence of an abnormality has not been determined for the items from the value of the spectral distribution generated by the Fourier transform unit 77. By making the feature amount of the items for which the presence or absence of an abnormality has not been determined prominent, the abnormality detection unit 79 can accurately detect whether or not there is an abnormality for the items for which the presence or absence of an abnormality has not been determined.

[0051] Next, the abnormality detection device 72 judges whether the analysis has been completed for all the driving patterns (step S37). If the analysis has not been completed for all the driving patterns (step S37: No), the abnormality detection device 72 returns to step S31 and repeats the process from step S31 to step S37. In the second process of step S31, for example, time series data measured in the driving pattern shown in FIG. 6(B) is extracted from the storage unit. The processes from step S32 to step S37 are the same as the first process. In the third process of step S31, for example, time series data measured in the driving pattern shown in FIG. 6(C) is extracted from the storage unit. The processes from step S32 to step S37 are the same as the first process.

[0052] On the other hand, if the analysis has been completed for all the operation patterns (step S37: Yes), the abnormality detection device 72 ends the abnormality detection process.

[0053] Next, the process of adding supervised data will be described with reference to the flowchart shown in Fig. 14. It is assumed that the above-mentioned anomaly detection process has been completed (step S51).

[0054] If there is any product determined to have an abnormality in the abnormality detection process (step S52: Yes), the abnormality detection device 72 displays the product determined to have an abnormality on the input / output device 93 (step S53). The abnormality detection device 72 outputs a warning such as "There is an abnormality in the sheave" or "There is an abnormality in the guide rail" to the input / output device 93.

[0055] When the product is indicated as having an abnormality, the inspector inspects whether the judgment made by the abnormality detection device 72 was correct (step S54). The inspector adds supervised data based on the inspection results (step S55). Specifically, if the judgment made by the abnormality detection device 72 is correct, the inspector uses the input / output device 93 to label the data indicating the spectral distribution on which the judgment of "abnormality" was based as "abnormal" and store the data in the memory unit. On the other hand, if the judgment made by the abnormality detection device 72 is incorrect, the inspector uses the input / output device 93 to label the data indicating the spectral distribution on which the judgment of "abnormality" was based as "normal" and store the data in the memory unit.

[0056] If there is no product that has been determined to have an abnormality in the abnormality detection process (step S52: No), the task of adding supervised data is not performed.

[0057] As described above, the elevator abnormality detection device 72 according to the first embodiment includes the extraction unit 76 that extracts time series data of a specified time length from the time series data acquired by the data acquisition unit 75, the Fourier transform unit 77 that generates a spectral distribution by Fourier transforming the time series data extracted by the extraction unit 76, and the abnormality detection unit 79 that detects an abnormality in the product constituting the elevator device 10 based on the generated spectral distribution by AI using supervised data. By performing the extraction process by the extraction unit 76, the spectral distribution obtained by Fourier transforming the time series data limited to an arbitrary time length Wx has a high proportion of noise feature amounts caused by products whose operation cycle is shorter than fx (Wx is several times 1 / fx). By making the feature amount of the product of interest prominent, the abnormality detection unit 79 can accurately detect whether or not there is an abnormality in the product of interest (product whose operation cycle is shorter than fx). Therefore, the accuracy of abnormality detection of the elevator device 10 can be improved.

[0058] Moreover, the elevator abnormality detection device 72 according to the first embodiment has a subtraction processing unit 78 that subtracts the value of the spectral distribution caused by the operation of the item when no abnormality is present for the item that is determined to be free of abnormality from the value of the spectral distribution generated by the Fourier transform unit 77. By subtracting the value of the spectral distribution caused by the operation of the item when no abnormality is present for the item that is determined to be free of abnormality from the value of the spectral distribution generated by the Fourier transform unit 77, the proportion of the spectral components of noise caused by the item for which the presence or absence of an abnormality is not determined in the spectral distribution increases. By making the feature amount of the item for which the presence or absence of an abnormality is not determined prominent, the abnormality detection unit 79 can accurately detect whether or not there is an abnormality for the item for which the presence or absence of an abnormality is not determined. Therefore, the accuracy of the abnormality detection of the elevator device 10 can be improved.

[0059] Moreover, the elevator abnormality detection device 72 according to the first embodiment performs a process of adding supervised data as described using the flowchart shown in Fig. 14. Since the elevator abnormality detection device 72 according to the first embodiment improves the accuracy of judgment by the AI ​​as the supervised data increases, it is possible to improve the accuracy of abnormality detection of the elevator device 10.

[0060] Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment. For example, in the above description, the data measurement process and the abnormality detection / cutoff process are performed separately, but the abnormality detection process may be performed at any time on the time series data measured while the data measurement process is being performed.

[0061] As described above, the supervised data is stored in advance in the storage unit in correspondence with the combination of the driving pattern and the time length. Although not described in detail above, in the abnormality detection process in step S35, the supervised data corresponding to the combination of the driving pattern and the time length may be selected and applied to the AI's judgment.

[0062] Furthermore, the driving pattern when an abnormality is detected may include not only the example shown in FIG. 6 but also the driving pattern during normal operation.

[0063] In the above description, the case where the supervised data addition process is performed has been described. However, if a sufficient amount of supervised data is stored in advance and the target anomaly detection accuracy is satisfied, the supervised data addition process may be omitted.

[0064] Also, in the above description, a case has been described in which in step S31 the data acquisition unit 75 extracts time series data for each driving pattern from the storage unit, and in step S32 the extraction unit 76 specifies a time length from the time series data acquired by the data acquisition unit 75 and extracts a portion of the time series data. In another embodiment, the processes of steps S31 and S32 may be combined. Specifically, the data acquisition unit 75 may receive a time length specification from the extraction unit 76, and the data acquisition unit 75 may acquire time series data of the specified time length from the storage unit.

[0065] (Embodiment 2) In the description of the first embodiment, the case where the subtraction processing unit 78 performs the subtraction processing of step S34 described with reference to Fig. 13 has been described. As another embodiment, an embodiment in which the subtraction processing unit 78 is omitted and the subtraction processing of step S34 described with reference to Fig. 13 is omitted may be considered. It is expected that the accuracy of anomaly detection will be lower than the detection accuracy in the embodiment described in the first embodiment. However, when it is determined that sufficient detection accuracy can be ensured by adjusting the time length of the time series data by the extraction unit 76, the subtraction processing can be omitted.

[0066] (Embodiment 3) In the description of the first embodiment using FIG. 13, the subtraction processing unit 78 subtracts the spectral distribution value caused by the operation of the equipment when it is determined that there is no abnormality of the equipment from the spectral distribution value generated by the Fourier transform unit 77 in the subtraction process of step S34. As another embodiment, the subtraction processing unit 78 may subtract the spectral distribution value obtained by Fourier transforming the time series data measured when it is determined that there is no abnormality for all the equipment from the spectral distribution value generated by the Fourier transform unit 77. By subtracting the spectral component value when all the equipment is normal, the feature amount of the spectral component caused by the abnormality of the equipment is emphasized. By making the feature amount of the spectral component caused by the abnormality of the equipment prominent, the abnormality detection unit 79 can improve the accuracy of abnormality detection of the elevator device 10.

[0067] (Embodiment 4) In the explanation of the first to third embodiments, the case where an abnormality is detected in the equipment related to the car 31 has been described. However, the presence or absence of an abnormality can also be detected in the opening / closing motor of the door 32 of the car 31 and the equipment related to the opening / closing of the door 32. For example, the presence or absence of an abnormality in the sheave related to the opening / closing of the door 32 can be detected. In this case, the operation pattern shown in FIG. 6 is prepared according to the operation of the lower level of the door 32.

[0068] (Embodiment 5) The elevator anomaly detection device according to the embodiment can inspect whether or not debris or the like is caught in the sill, which is a rail for opening and closing the door 32 of the car 31. As supervised data, a spectrum when debris or the like is caught in the sill and a spectrum when debris or the like is not caught in the sill are prepared. A large amount of supervised data generated based on different conditions, such as a number of materials such as gum or stones, a number of sizes, and the position of the debris caught in the sill, is prepared. Information such as the material, size, and position of the debris is linked as a label for the supervised data.

[0069] After preparing the above supervised data, the abnormality detection process is performed based on the flowchart shown in FIG. 13, making it possible to estimate what type of dirt is stuck in what position on the sill.

[0070] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0071] 10 Elevator equipment 13 Inverter 15. Measurement Equipment 151 Current measuring device 152 Voltage measuring device 21~24 Guide rail 31 Car 31a opening 32 Door 36 Operation Panel 40 Lifting motor 41 Opening and closing motor 42 Pulley 43 Wire 44 Cable 50 Counterweight 70 Control Panel 71 Drive unit control section 72 Anomaly detection device 73 Operation pattern setting section 75 Data Acquisition Section 76 Extraction part 77 Fourier transform section 78 Subtraction processing section 79 Anomaly detection section 80 Control Unit 81 CPU 82 Main memory 83 Auxiliary storage 84 Interface section 85 Bus 91 Drive unit 93 Input / Output Devices 100 Hoistway

Claims

1. An elevator abnormality detection device for detecting an abnormality in an elevator system having a motor for driving the elevation and lowering of a car, a power source for supplying power to the motor, and a control device for controlling the motor, a data acquisition unit that acquires time-series data indicating a waveform of a voltage or a current from a measurement device that measures a waveform of a voltage or a current supplied from the power source to the motor; an extraction unit that extracts time series data of a specified time length from the time series data acquired by the data acquisition unit; a Fourier transform unit that generates a spectral distribution by performing a Fourier transform on the time series data extracted by the extraction unit; An abnormality detection unit that detects an abnormality of the equipment constituting the elevator apparatus based on the generated spectral distribution by AI using supervised data for at least a part of the data; a subtraction processing unit that subtracts a spectral distribution value caused by an operation of the product when the product is determined to be free of abnormality from the spectral distribution value generated by the Fourier transform unit; The abnormality detection unit is an elevator abnormality detection device that detects an abnormality in an item that constitutes the elevator device based on the spectral distribution after subtraction processing.

2. The time length is set based on the operation cycle of each of the components constituting the elevator apparatus. The elevator abnormality detection device according to claim 1.

3. The extraction unit sets the time length by sequentially changing the time length from a time length corresponding to a product having a short operating cycle to a time length corresponding to a product having a long operating cycle, the abnormality detection unit detects the presence or absence of an abnormality in the order of products having shortest operating cycles based on the time series data of the set time length.

3. An elevator abnormality detection device according to claim 1 or 2.

4. a subtraction processing unit that subtracts a value of the spectral distribution when there is no abnormality from the value of the spectral distribution generated by the Fourier transform unit; The abnormality detection unit detects an abnormality in the product constituting the elevator apparatus based on the spectrum distribution after the subtraction process.

3. An elevator abnormality detection device according to claim 1 or 2.

5. An elevator abnormality detection method for detecting an abnormality in an elevator system having a motor that drives the elevation and lowering of a car, a power source that supplies power to the motor, and a control device that controls the motor, comprising: a data acquisition step of acquiring time series data indicating a voltage or current waveform from a measuring device that measures a voltage or current waveform of the power supply; an extraction step of extracting time series data of a specified time length from the time series data; a Fourier transform step of generating a spectral distribution by Fourier transforming the extracted time series data; An abnormality detection process for detecting an abnormality in the equipment constituting the elevator apparatus based on the generated spectral distribution by AI using supervised data for at least a portion of the data; a subtraction process for subtracting a spectral distribution value caused by an operation of the product when the product is determined to be free of an abnormality from the spectral distribution value generated in the Fourier transform process; Including, In the abnormality detection step, an abnormality in an item constituting the elevator apparatus is detected based on the spectral distribution after subtraction processing.

6. In the extraction step, the time length is set by sequentially changing the time length from a time length corresponding to a product having a short operating cycle to a time length corresponding to a product having a long operating cycle, In the anomaly detection step, the presence or absence of an anomaly is detected in order of the product having the shortest operating cycle based on the time series data of the set time length. The elevator abnormality detection method according to claim 5.

Images