Omen detection device and omen detection method

Abstract

The invention relates to a sign detection device and a sign detection method. The embodiment of the invention relates to an omen detection device and an omen detection method. A sign of an elevator device failure is detected. The omen detection device according to an embodiment includes a data acquisition unit and an omen detection unit. The data acquisition unit acquires time series data measured during lifting operation of a passenger car of an elevator device. The omen detection unit detects whether there is an abnormal omen in the articles constituting the elevator device on the basis of the acquired time series data by using the AI of the supervision data.

Description

This application is based on Japanese Patent Application No. 2023-200463 filed on November 28, 2023, and claims the priority of this application. This application incorporates the entire content of this application by reference. Technical Field

[0001] Embodiments of the present invention relate to a sign detection device and a sign detection method. Background Art

[0002] There is an abnormality detection device that determines the supplies that have become abnormal when an abnormality occurs in an elevator device. Existing abnormality detection devices detect the occurrence of an abnormality. Therefore, an abnormality cannot be detected unless after a failure occurs.

[0003] For example, when a failure occurs in the door motor, the doors of the passenger car cannot be opened or closed, and there is a possibility of trapping passengers in the passenger car. In an abnormality detection device that detects the occurrence of an abnormality, the possibility of trapping passengers in the passenger car cannot be reduced.

[0004] In addition, when a failure occurs in the hoisting motor (winch), since it is an operation to replace the hoisting motor, the elevator device cannot be used for a long time. If it is possible to detect signs of an abnormality in the supplies, maintenance can be performed at night or the like before a failure occurs, and the inconvenience to users can be reduced. Summary of the Invention

[0005] The present invention has been completed in view of the above circumstances, and an object thereof is to detect signs of a failure of an elevator device.

[0006] The sign detection device according to an embodiment for solving the above problems includes a data acquisition unit and a sign detection unit. The data acquisition unit acquires time series data measured during the up-and-down operation of the passenger car of the elevator device. The sign detection unit uses AI with supervised data to detect whether there are signs of becoming abnormal in the supplies that make up the elevator device based on the acquired time series data. Thereby, maintenance can be performed before a failure occurs, and for example, the possibility of trapping passengers in the passenger car can be reduced. Brief Description of the Drawings

[0007] Figure 1 It is a perspective view of the elevator device of this embodiment. Figure 2 It is a block diagram showing the control system of the elevator device of this embodiment. Figure 3 It is a block diagram of the drive unit of this embodiment. Figure 4 It is a physical block diagram of the control unit of this embodiment. Figure 5 It is a functional block diagram of the control unit of this embodiment. Figure 6 This is a diagram for explaining the operation mode of the present embodiment. Figure 7 This is a diagram for explaining the time series data of the present embodiment. Figure 8 This is a diagram for explaining the operation cycle of the components constituting the elevator device of the present embodiment. Figure 9 This is a diagram for explaining the time length of the time series data extracted by the extraction unit of the present embodiment. Figure 10 This is a diagram for explaining the frequency spectrum distribution generated by the Fourier transform unit of the present embodiment. Figure 11 This is a diagram for explaining the operation of the subtraction processing unit of the present embodiment. Figure 12 This is a flowchart for explaining the omen detection process of the present embodiment. Figure 13 This is a diagram for explaining the time series data acquired by the data acquisition unit of the present embodiment. Figure 14 This is a diagram for explaining the time series data acquired by the data acquisition unit of the present embodiment. Figure 15 This is a flowchart for explaining the data measurement process of the omen detection device of the present embodiment. Figure 16 This is a flowchart for explaining the determination process of the omen detection device of the present embodiment for components with omens of abnormality. Figure 17 This is a flowchart for explaining the generation process of the supervision data of the omen detection device of the present embodiment. Detailed Embodiment

[0008] Hereinafter, the present embodiment will be described with reference to the drawings. In the description, an XYZ coordinate system composed of mutually orthogonal X-axis, Y-axis, and Z-axis is appropriately used. The diagrams and flowcharts used in the description of the present embodiment show an example.

[0009] (Embodiment 1) Figure 1 This is a perspective view of the elevator device 10 of the present embodiment. The elevator device 10 is disposed inside a hoistway 100 provided in a building such as a commercial facility or a residential facility. As Figure 1 shown, the elevator device 10 includes a passenger car 31, a counterweight 50, a hoist motor 40, guide rails 21 to 24, and a control panel 70 (elevator control device).

[0010] The guide rails 21 to 24 are members with the length direction as the Z-axis direction. The guide rails 21 and 22 are a pair of members for guiding the passenger car 31 to move up and down freely. In addition, the guide rails 23 and 24 are a pair of members for guiding the counterweight 50 to move up and down freely. The guide rails 21 and 22 are arranged separately in the Y-axis direction. In addition, the guide rails 23 and 24 are also arranged separately from each other in the Y-axis direction. In Figure 1 the guide rails 23 and 24 of the counterweight 50 are arranged separately from the guide rails 21 and 22 of the passenger car 31 in the X-axis direction. In addition, the arrangement of the guide rails 21 to 24 is not limited to Figure 1 the arrangement shown.

[0011] The passenger car 31 is a unit that accommodates users and moves up and down in the hoistway 100. The passenger car 31 is arranged between the guide rails 21 and 22 and is installed on the guide rails 21 and 22 so as to be able to move in the vertical direction.

[0012] On the +X side surface of the passenger car 31, an opening 31a for entering and exiting the interior is formed. The opening 31a is closed or opened by a pair of doors 32 that move along the side surface of the passenger car 31. The doors 32 are opened and closed by an opening and closing motor (not shown in Figure 1 ) in the figure.

[0013] The counterweight 50 is installed on the guide rails 23 and 24 so as to be able to move in the vertical direction. The weight of the counterweight 50 is adjusted to a specified ratio with respect to the weight of the passenger car 31.

[0014] The hoist motor 40 is a motor for raising and lowering the passenger car 31. The hoist motor 40 is arranged at the upper part of the hoistway 100 with the rotation axis parallel to the Y-axis. A pulley 42 is fixed on the rotation axis of the hoist motor 40.

[0015] A cable 43 is wound around the pulley 42 of the hoist motor 40. One end of the cable 43 is fixed to the passenger car 31, and the other end is fixed to the counterweight 50.

[0016] The control panel 70 is arranged on the hoistway 100. A control device for controlling the hoist motor 40, equipment provided on the passenger car 31, etc. is accommodated in the control panel 70.

[0017] Figure 2 is a block diagram showing the control system of the elevator device 10. The control system is configured to include a control unit 80 and a drive unit 91 accommodated in the control panel 70, and an operation panel 36 provided in the passenger car 31.

[0018] The operation panel 36 is provided on the inner wall surface of the passenger car 31. The operation panel 36 is an interface for receiving the destination floor and the like from the user of the passenger car 31. By operating the operation panel 36, the user can register the destination floor and the like of the passenger car 31 and open and close the door 32. The operation panel 36 is connected to the control unit 80 accommodated in the control panel 70 via Figure 1 the cable 44 shown.

[0019] Figure 2 The drive unit 91 shown drives the hoist motor 40 and the opening / closing motor 41 for driving the door 32 of the passenger car 31 by supplying electric power to the hoist motor 40 and the opening / closing motor 41 for driving the door 32 of the passenger car 31 (the illustration is omitted in Figure 1 ). Thus, the hoist motor 40 (winch) and the opening / closing motor 41 are driven. The drive unit 91 drives the hoist motor 40 according to an instruction from the control unit 80. In addition, the drive unit 91 drives the opening / closing motor 41 according to an instruction from the control unit 80.

[0020] Figure 3 is a block diagram of the drive unit 91. The drive unit 91 includes an inverter 13 and a measuring device 15. The inverter 13 is a power supply device that supplies electric power to the hoist motor 40 and the opening / closing motor 41. The inverter 13 is constituted by a switching regulator. When the hoist motor 40 and the opening / closing motor 41 are formed of three-phase AC motors, the inverter 13 outputs a three-phase AC voltage.

[0021] The measuring device 15 includes a current measuring device 151 and a voltage measuring device 152. The current measuring device 151 is a device that measures the current supplied from the inverter 13 to the hoist motor 40 and the opening / closing motor 41 for each phase. The voltage measuring device 152 is a device that measures the output voltage of the inverter 13 for each phase.

[0022] In addition, the measuring device 15 includes a vibration sensor 154 and a torque sensor 155. The vibration sensor 154 is installed at the bottom of the passenger car 31 and measures the vibration accompanying the hoisting operation of the passenger car 31 and the opening / closing operation of the door 32. The torque sensor 155 measures the torque of the hoist motor 40 accompanying the hoisting operation of the passenger car 31 and the torque of the opening / closing motor 41 accompanying the opening / closing operation of the door 32.

[0023] Figure 4is 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 storage unit 82, an auxiliary storage unit 83, and an interface unit 84 that are interconnected via a bus 85. The CPU 81 executes the processing described later according to a program stored in the auxiliary storage unit 83. The main storage unit 82 includes a RAM (Random Access Memory) and the like. The main storage unit 82 serves as a working area for the CPU 81. The auxiliary storage unit 83 includes non-volatile memories such as a ROM (Read Only Memory) and a semiconductor memory. The auxiliary storage unit 83 stores programs and various parameters executed by the CPU 81. In addition, the auxiliary storage unit 83 stores supervised data used in AI (Artificial Intelligence)-based omen detection. The supervised data is data with labels such as "no abnormality", "omen", and "abnormality" attached to time series data. In addition, it can also be semi-supervised data with labels attached to a part of it.

[0024] The interface unit 84 includes a serial interface, a parallel interface, a wireless LAN interface, and the like. 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, and the like is connected to the interface unit 84.

[0025] Figure 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 omen 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 according to an input from the operation panel 36 or a call panel on each floor. For example, when the drive unit control unit 71 rotates the lift motor 40 forward via the drive unit 91, the passenger car 31 ascends and the counterweight 50 descends. When the drive unit control unit 71 rotates the lift motor 40 in reverse via the drive unit 91, the passenger car 31 descends and the counterweight 50 ascends. In addition, when the drive unit control unit 71 rotates the opening / closing motor 41 forward via the drive unit 91, the door 32 of the passenger car 31 and the doors in the landing halls on each floor are controlled to open, and when the opening / closing motor 41 is rotated in reverse, the door 32 of the passenger car 31 and the doors in the landing halls on each floor are controlled to close.

[0027] The omen detection device 72 includes an operation mode setting unit 73, a data acquisition unit 74, an omen detection unit 75, an extraction unit 76, a Fourier transform unit 77, a subtraction processing unit 78, and an article determination unit 79.

[0028] The operation mode setting unit 73 sets an operation mode for the moving speed and moving acceleration of the specified passenger car 31 in the drive unit control unit 71. As the operation mode, there are an operation mode during normal operation for transporting users and an operation mode during sign detection for detecting whether there is a sign of abnormality in the supplies constituting the elevator device 10. The operation modes are stored in advance in the storage unit. The operation mode setting unit 73 selects an operation mode from the storage unit and provides it to the drive unit control unit 71.

[0029] Figure 6 This is an example of an operation mode for determining supplies with signs of abnormality. Figure 6 (A) of Figure 6 represents an operation mode in which the speed of the passenger car 31 is changed with an acceleration lower than that during normal operation. By changing the speed with a lower acceleration, it is possible to easily detect noise caused by vibrations or action cycles of supplies occurring at a specific speed. Figure 6 (B) of Figure 6 represents an operation mode in which the passenger car 31 is moved at a constant speed. Figure 6 (C) of Figure 6 represents an operation mode in which the speed of the passenger car 31 is increased with an acceleration higher than that during normal operation and then the lifting speed of the passenger car 31 is decreased with a low acceleration. Figure 6 (D) of Figure 6 represents performing step by step Figure 6 (A) of Figure 6 represents an operation mode for acceleration control. In addition, Figure 6 The operation mode shown in Figure 6 is an example and is not limited thereto. The operation mode for determining supplies with signs of abnormality is set not only for the lifting control of the passenger car 31 but also for the opening and closing control of the door 32.

[0030] Return Figure 5 Then, the data acquisition unit 74 acquires time series data measured during the lifting operation of the passenger car 31. For example, the data acquisition unit 74 acquires time series data of the waveform of the voltage or current supplied from the inverter 13 to the lifting motor 40 for driving the lifting of the passenger car 31 or the opening and closing motor 41 for opening and closing the door 32 of the passenger car 31, which is measured by the measuring device 15, time series data of the vibration waveform during the lifting operation of the passenger car 31 measured by the vibration sensor 154, and time series data of the torque waveform of the opening and closing motor 41 during the opening and closing of the door 32 measured by the torque sensor 155.

[0031] In the first embodiment, the case where the data acquisition unit 74 acquires time series data representing the waveform of voltage or current from the measuring device 15 will be described. The time series data measured by the measuring device 15 is temporarily stored in the storage unit. The data acquisition unit 74 acquires the time series data from the storage unit. The time length of the time series data acquired in one operation mode is, for example, 10 seconds, 1 minute, or 10 minutes. Figure 7This is an example of a part of the time series data for obtaining the voltage waveform. In the time series data, in addition to the frequency components corresponding to the voltage changes accompanying motor control, there is also noise with various frequency components caused by vibrations and operation cycles of the supplies.

[0032] In addition, the supplies constituting the elevator device 10 each have their own operation cycles. This operation cycle is also related to the operation mode. For example, the rotation cycle of the pulley related to the cable connecting the passenger car 31 and the counterweight 50 is determined by the diameter of the pulley and the lifting speed of the passenger car 31. In addition, the cycle of vibrations of the guide rails 21 to 24 is determined by the lifting speed, acceleration, and moving distance of the passenger car 31. In addition, the rotation cycle of the pulley for opening and closing the door 32 is determined by the diameter of the pulley and the opening and closing speed of the door 32. Figure 8 This is an example showing the inherent operation cycles of the respective supplies constituting the elevator device 10. Figure 8 The information shown is stored in the storage unit according to the operation speed of each operation mode.

[0033] Return Figure 5 , the omen detection unit 75 uses AI with supervised data to detect whether there is an omen of abnormality in the supplies constituting the elevator device 10 based on the acquired time series data. The omen detection unit 75 is composed of AI with a learning function. The AI is constituted by, for example, a neural network or a support vector machine.

[0034] During the lifting operation of the passenger car 31 from when there is no abnormality in the supplies constituting the elevator device 10 until an abnormality occurs, supervised data used in the AI possessed by the omen detection unit 75 is generated based on the time series data acquired by the data acquisition unit 74. The time series data is measured, for example, during the regular inspection every month. For the supervised data, the time series data in the case where there is an abnormality in the supplies is attached with the label "abnormal", and the time series data in the case where there is no abnormality in the supplies is attached with the label "no abnormality". Within a specified period (for example, one month) after acquiring the time series data, in the case where an abnormality occurs in the supplies constituting the elevator device 10, the label "no abnormality" of the corresponding time series data is changed and the label "omen" is attached.

[0035] The extraction unit 76 specifies a time length from the time series data acquired by the data acquisition unit 74 and extracts a part of the time series data. Figure 9 This is a diagram for explaining the time length of the time series data. The time length is based on Figure 8It is set according to the inherent operation cycle of the supplies shown. For example, the time length Wm for detecting the omen of abnormality of the pulley is several times (e.g., 2 times or 4 times) the time length of 1 / fm (e.g., about 0.1 second), and the time length Wn for detecting the omen of abnormality of the guide rail is several times the time length of 1 / fn (e.g., about 100 seconds). When checking the supplies related to the AND gate 32, the time length can be set to, for example, about 2 seconds.

[0036] Return Figure 5 , the Fourier transform unit 77 generates a frequency spectrum distribution by performing a Fourier transform on the time series data extracted by the extraction unit 76. Figure 10 is an example of the frequency spectrum distribution generated by the Fourier transform unit 77. The frequency spectrum values corresponding to the operation cycles of the supplies shown in Figure 8 are relatively large values.

[0037] The subtraction processing unit 78 performs a process of subtracting the value of the frequency spectrum distribution caused by the operation of the supplies determined to have no omen from the value of the frequency spectrum distribution generated by the Fourier transform unit 77. For example, when it is determined that there is no omen of abnormality in the roller guide shown in Figure 8 , the value of the frequency spectrum distribution caused by the operation in the case of no omen in the roller guide is subtracted from the value of the frequency spectrum distribution shown in Figure 10 . Regarding the time series data obtained in the operation mode shown in Figure 6 , the frequency spectrum distribution data in the case where there is no omen of each supply corresponding to the time length specified by the extraction unit 76 is pre-stored in the storage unit. Through this subtraction process, the frequency spectrum indicated by the dotted line shown in Figure 11 is deleted, or the value of the frequency spectrum indicated by the dotted line becomes smaller.

[0038] The supply determination unit 79 determines the supplies having an omen of becoming abnormal based on the generated frequency spectrum distribution by using AI of supervised data. The supply determination unit 79 is composed of AI having a learning function. The AI is composed of, for example, a neural network or a support vector machine. The AI constituting the supply determination unit 79 detects whether there is an omen of becoming abnormal in the supplies constituting the elevator device 10 by comparing the feature amounts of the frequency spectrum distribution output by the subtraction processing unit 78 and the feature amounts of the supervised data.

[0039] Generate supervised data for use in the AI incorporated in the in-use item determination unit 79 based on the time-series data representing the waveforms of the voltage or current acquired by the data acquisition unit 74. The supervised data is data with the label "abnormal" attached to the spectral distribution data when there is an abnormality and the label "normal" attached to the spectral distribution data when there is no abnormality. In addition, in the case where an abnormality occurs within a specified period (e.g., one month) after the data with the label "normal" is attached, the label "normal" for the corresponding data is changed and the label "omen" is attached. And it is generated as supervised data associated with data (specific names such as motors or pulleys) of in-use items that are determined to have an omen of becoming abnormal.

[0040] For example, attach the label "abnormal: pulley is deformed" to the spectral distribution in the case of pulley deformation, and attach the label "abnormal: there is damage on the pulley" to the spectral distribution in the case of pulley damage. In addition, in the case where an abnormality occurs in the guide rail within, for example, one month after the acquisition of the time-series data, labels such as "omen: the mounting screws of the guide rail are loose" or "omen: the contact resistance between the pulley and the cable is reduced" can be attached.

[0041] This supervised data generates a large amount of data according to the time length set by the extraction unit 76 and Figure 6 for each operating mode shown, and stores it in the storage unit. Regarding this supervised data, even when there is no omen of becoming abnormal in all in-use items, a large amount of data is generated according to the time length set by the extraction unit 76 and Figure 6 for each operating mode shown, and is stored in the storage unit with the label "normal" or "no omen" attached.

[0042] Next, while referring to the Figure 12 flowchart shown, the omen detection process (omen detection method) for detecting omens of becoming abnormal in the in-use items constituting the elevator device 10 will be described. The following control is performed according to the program stored in the auxiliary storage unit 83, and the main body of the control is the control unit 80 (CPU 81). The measuring device 15 measures the time-series data of the output current and output voltage of the inverter 13 and notifies the measured data to the control unit 80.

[0043] The omen detection device 72 detects whether there is a lifting operation of the passenger car 31 (step S11). In the case where there is no lifting operation of the passenger car 31 (step S11: no), the omen detection device 72 continues to monitor the situation where the passenger car 31 is in a lifting operation. In the case where there is a lifting operation of the passenger car 31 (step S11: yes), the omen detection device 72 acquires the time-series data of the output current and output voltage of the inverter 13 measured by the data acquisition unit 74 (step S12). Step S12 is the data acquisition process.

[0044] Next, the omen detection device 72 determines whether there is an omen of abnormality in the supplies that make up the elevator device 10 based on the acquired time-series data by using the AI that constitutes the omen detection unit 75 and uses supervised data (step S13). Step S13 is the omen detection process.

[0045] Figure 13 is a diagram schematically showing the time-series data attached to the supervised data in the case where there is no omen of abnormality. The time-series data in the case of no omen is generated by attaching the label "no omen" to the time-series data of the current shown in Figure 13 shown. Figure 14 is a diagram schematically showing the time-series data of the current attached to the supervised data in the case where there is an omen of abnormality in the bearing. The supervised data in the case of an omen is generated by attaching the label "with omen" to the time-series data of the current shown in Figure 14 shown. The omen detection device 72 determines whether there is an omen of abnormality in the supplies that make up the elevator device 10 by comparing the feature amounts of the supervised data shown in Figure 13 and Figure 14 shown with the feature amounts of the time-series data of the output current and output voltage acquired by the data acquisition unit 74.

[0046] In the case where there is no omen of abnormality in the supplies (step S14: No), the omen detection device 72 transfers the process to step S16. On the other hand, in the case where there is an omen of abnormality in the supplies (step S14: Yes), the omen detection device 72 sets a flag indicating that there is an omen (step S15) and transfers the process to step S16.

[0047] Next, the omen detection device 72 determines whether the omen detection of all supplies has ended (step S16). In the case where the omen detection of all supplies has not ended (step S16: No), the omen detection device 72 repeats the process from step S11 to step S16. In the case where the omen detection of all supplies has ended (step S16: Yes), the process ends. In the case where the flag indicating that there is an omen is set in step S15, the omen detection device 72 outputs that there is an omen of abnormality to the input / output device 93 and performs the process of determining the supplies that are the cause of the omen of abnormality.

[0048] Next, referring to Figure 15The flowchart shown below explains the measurement process of data for supplies that are the causes of signs of anomalies. The following control is performed according to the program stored in the auxiliary storage unit 83, and the control entity is the control unit 80 (CPU 81). The measurement device 15 measures the time-series data of the output current and output voltage of the inverter 13 and notifies the control unit 80 of the measured time-series data. Here, a case where the passenger car 31 is lifted and lowered in the four operating modes shown below and data is measured will be explained. Figure 6 A case where the passenger car 31 is lifted and lowered in the four operating modes shown below and data is measured will be explained.

[0049] First, the operation mode setting unit 73 sets the operation mode shown in (A) of Figure 6 in the drive unit control unit 71 (step S31). The measurement device 15 measures the time-series data of the voltage and current when the passenger car 31 is moved and controlled or the door 32 is opened and closed according to the corresponding operation mode (step S32) and stores it in the storage unit.

[0050] The operation mode setting unit 73 determines whether all the operation modes have been implemented (step S33). If not all the operation modes have been implemented (step S33: No), the operation mode setting unit 73 returns to step S31 and sets the unimplemented operation mode. For example, in the second step S31, the operation mode setting unit 73 sets the operation mode shown in (B) of Figure 6 in the drive unit control unit 71. In addition, in the third step S31, the operation mode setting unit 73 sets the operation mode shown in (C) of Figure 6 in the drive unit control unit 71. On the other hand, if all the operation modes have been implemented (step S33: Yes), the sign detection device 72 ends the data measurement process.

[0051] Next, referring to the flowchart shown in Figure 16 below, the determination process for determining supplies with signs of anomalies will be explained. After ending the Figure 15 data acquisition process described above, the measured time-series data is stored in the storage unit. The supervision data corresponding to the operation mode and the time length set by the extraction unit 76 is stored in the storage unit in advance.

[0052] First, the data acquisition unit 74 extracts the time-series data from the storage unit for each operation mode (step S51). For example, the data acquisition unit 74 first extracts the time-series data measured in the operation mode shown in (A) of Figure 6 from the storage unit.

[0053] Next, the extraction unit 76 specifies a time length from the time series data acquired by the data acquisition unit 74 and extracts a part of the time series data (step S52). The time length is preset for each item to be inspected according to the operation cycle of the item to be inspected, etc. The extraction unit 76 first sets the shortest time length. For example, in the case where the roller guide shown in Figure 8 is the item to be inspected, the extraction unit 76 sets the time length Wi shown in Figure 9 to a time length that is a multiple (e.g., 2 times or 4 times) of 1 / fi. Then, the extraction unit 76 provides the time series data of the time length Wi to the Fourier transform unit 77.

[0054] Next, the Fourier transform unit 77 generates a frequency spectrum distribution obtained by performing a Fourier transform on the time series data of the time length Wi (step S53). In the frequency spectrum distribution obtained by performing a Fourier transform on the time series data limited to the time length Wi, there is no noise caused by items with an operation cycle longer than fi, such as the pulleys and guide rails shown in Figure 8 , or the proportion of the power of the frequency spectrum caused by items with an operation cycle longer than fi in the transformed frequency spectrum distribution is low.

[0055] Next, the subtraction processing unit 78 performs a process of subtracting the value of the frequency spectrum distribution caused by the operation of the item determined not to be a sign of abnormality from the value of the frequency spectrum distribution generated by the Fourier transform unit 77 (step S54). In the process of the first step S54, since there is no item determined not to be a sign of abnormality, the process of step S54 is skipped.

[0056] Next, the item determination unit 79 detects whether there is a sign of abnormality in the items constituting the elevator device 10 by comparing the feature amount of the frequency spectrum distribution output by the subtraction processing unit 78 and the feature amount of the supervision data (step S55). In the frequency spectrum distribution obtained by performing a Fourier transform on the time series data limited to the time length Wi, there is no noise caused by items with an operation cycle longer than fi, such as the pulleys and guide rails shown in Figure 8 , or the proportion of the power of the frequency spectrum caused by items with an operation cycle longer than fi in the transformed frequency spectrum distribution is low. Therefore, for the frequency spectrum distribution obtained by performing a Fourier transform on the time series data limited to the time length Wi, the proportion of the feature amount of the noise caused by items with an operation cycle shorter than fi becomes higher. By making the feature amount of the item of interest prominent, the item determination unit 79 can correctly detect whether there is a sign of abnormality in the item of interest (the item with an operation cycle shorter than fi).

[0057] The sign detection device 72 determines whether all the products have been analyzed (step S56). Specifically, it determines whether the product to be inspected has been analyzed. Figure 8 The supplies shown correspond to Figure 9 If the analysis of all products has not yet been completed (step S56: No), the sign detection device 72 returns to step S52, changes the time length, and repeats the processing from step S52 to step S56.

[0058] In the subtraction process of the second step S54, for example, if it is determined in the process of the first step S55 that there is no sign of abnormality in the roller guide, the value of the spectrum distribution caused by the operation of the roller guide when there is no sign of abnormality in the roller guide is subtracted from the value of the spectrum distribution generated by the Fourier transform unit 77 in the second step S53. Figure 11 To explain, since the spectral components indicated by the dotted lines are removed, the spectral components in the frequency band lower than the frequency fi are emphasized.

[0059] In the subtraction process of the third step S54, for example, if it is determined in the process of the first step S55 that there is no sign of abnormality in the roller guide and in the process of the second step S55 that there is no sign of abnormality in the motor and the pulley, the value of the spectrum distribution caused by the operation of the roller guide when there is no sign of abnormality in the roller guide, the motor, and the pulley is subtracted from the value of the spectrum distribution generated by the Fourier transform unit 77 in the third step S53. Figure 11 To explain, since the spectral components indicated by the dotted lines and the spectral components indicated by the dashed lines are removed, the spectral components in the frequency band lower than the frequency fm are emphasized.

[0060] By subtracting the value of the spectral distribution caused by the movement of the product determined as not having a sign of abnormality from the value of the spectral distribution generated by the Fourier transform unit 77, the proportion of the spectral component of the noise caused by the product for which the presence or absence of a sign of abnormality has not been determined in the spectral distribution becomes higher. Since the feature amount of the product for which the presence or absence of a sign of abnormality has not been determined becomes significant, the product identification unit 79 can accurately detect whether there is a sign of abnormality for the product for which the presence or absence of a sign of abnormality has not been determined.

[0061] Next, the sign detection device 72 determines whether the analysis of all the operation modes is completed (step S57). If the analysis of all the operation modes is not completed (step S57: No), the sign detection device 72 returns to step S51 and repeats the processing of steps S51 to S57. In the second processing of step S51, for example, the first step of extracting ... Figure 6The time series data measured in the operation mode shown in (B). The processing from step S52 to step S57 is the same as the first time. In the processing of step S51 in the third time, for example, extract from the storage unit the time series data measured in the operation mode shown in Figure 6 (C). The processing from step S52 to step S57 is the same as the first time.

[0062] On the other hand, when the analysis of all operation modes is completed (step S57: Yes), the omen detection device 72 ends the determination process of the supplies with omens. When an omen of abnormality is detected in the processing of step S55, the omen detection device 72 outputs the supplies with an omen of abnormality to the input / output device 93.

[0063] Next, with reference to Figure 17 the flowchart shown, the generation process of the supervised data used in the AI of the omen detection unit 75 will be described.

[0064] For example, during the regular inspection every month, obtain the time series data of the voltage and current of the inverter 13 during the lifting operation of the passenger car 31 (step S71). During the regular inspection, if a failure is found in the elevator device 10 (step S72: Yes), it is stored in the storage unit as supervised data with the label of "abnormal" attached to the obtained time series data (step S73). On the other hand, during the regular inspection, if no failure is found in the elevator device 10 (step S72: No), it is stored in the storage unit as supervised data with the label of "no abnormality" attached to the obtained time series data (step S74).

[0065] The omen detection device 72 determines whether any of the supplies constituting the elevator device 10 has failed before a specified period (for example, 1 month) has passed after obtaining the time series data (step S75). Before a specified period (for example, 1 month) has passed after obtaining the time series data, if any of the supplies constituting the elevator device 10 has failed (step S75: Yes), the omen detection device 72 changes the label of the corresponding time series data from "no abnormality" to "omen", and stores it in the storage unit as supervised data with the label of "omen" attached (step S76).

[0066] On the other hand, before a specified period (for example, 1 month) has passed after obtaining the time series data, if none of the supplies constituting the elevator device 10 has failed (step S75: No), the omen detection device 72 keeps the label of the corresponding time series data as "no abnormality", and stores it in the storage unit as supervised data with the label of "no abnormality" attached (step S77).

[0067] As described above, the omen detection device 72 of Embodiment 1 can detect whether there is an omen of abnormality in the supplies constituting the elevator device 10 based on the acquired time-series data by using the AI using supervised data in the omen detection unit 75. Thus, maintenance can be performed before a failure occurs. Therefore, for example, the possibility of trapping passengers in the passenger car can be reduced.

[0068] In addition, the omen detection device 72 of Embodiment 1 includes: an extraction unit 76 that extracts time-series data of a specified time length from the time-series data acquired by the data acquisition unit 74; a Fourier transform unit 77 that generates a frequency spectrum distribution by performing a Fourier transform on the time-series data extracted by the extraction unit 76; and an article determination unit 79 that determines an article having an omen of abnormality based on the frequency spectrum distribution generated by the AI using supervised data. By the extraction process performed by the extraction unit 76, in the frequency spectrum distribution obtained by performing a Fourier transform on the time-series data restricted to an arbitrary time length Wx, the proportion of the characteristic amount of the noise caused by an article having a cycle shorter than the operation cycle ratio fx (Wx is several times 1 / fx) becomes higher. Since the characteristic amount of the article of interest becomes prominent, the article determination unit 79 can correctly detect whether there is an omen of abnormality in the article of interest (the article having a cycle shorter than the operation cycle ratio fx). Therefore, maintenance such as replacement of the article can be performed before the article fails. Thus, the possibility of trapping passengers in the passenger car can be reduced. In addition, since a spare article can be prepared before the article fails, the maintenance time of the elevator device 10 can be shortened.

[0069] In addition, the omen detection device 72 of Embodiment 1 has a subtraction processing unit 78 that subtracts the value of the frequency spectrum distribution caused by the operation of an article determined not to have an omen of abnormality from the value of the frequency spectrum distribution generated by the Fourier transform unit 77. By subtracting the value of the frequency spectrum distribution caused by the operation of an article determined not to have an omen of abnormality from the value of the frequency spectrum distribution generated by the Fourier transform unit 77, the proportion of the frequency spectrum component of the noise caused by an article whose omen of abnormality has not been determined in the frequency spectrum distribution becomes higher. Since the characteristic amount of the article whose omen of abnormality has not been determined becomes prominent, the article determination unit 79 can correctly determine an article having an omen of abnormality.

[0070] In addition, the omen detection device 72 of Embodiment 1 performs the generation process of the supervised data as described in the flowchart shown Figure 17 As described above. The omen detection device 72 of Embodiment 1 improves the judgment accuracy of the AI by increasing the supervised data, and thus can improve the detection accuracy of the omen of abnormality of the elevator device 10.

[0071] As described above, the embodiments of the present invention have been described, but the present invention is not limited to the above embodiments. For example, in the above description, a case has been described in which it is determined whether there is a sign of abnormality based on the time series data measured during the lifting operation at the time of regular maintenance. As another embodiment, it is also possible to determine whether there is a sign of abnormality based on the time series data measured during the lifting operation during normal operation. In addition, in the case of determining whether there is a sign of abnormality based on the time series data measured during the normal lifting operation, in order to remove non-specific vibrations, measurement data when there are no passengers may also be used. The determination of whether there are passengers in the passenger car 31 can be made using a camera that captures the inside of the passenger car, or a load sensor that measures the weight of the passenger car 31.

[0072] In addition, the operation mode during abnormality detection includes not only Figure 6 the example shown, but may also include the operation mode during normal operation.

[0073] In addition, in the above description, a case has been described in which additional processing of supervision data is performed. However, in the case where a sufficient amount of supervision data is pre-stored and the target abnormality detection accuracy is satisfied, the additional processing of supervision data may be omitted.

[0074] (Embodiment 2) In Embodiment 1, a case has been described in which it is detected whether there is a sign of abnormality in the components constituting the elevator device 10 based on the time series data of the voltage or current of the inverter 13. However, it is also possible to detect whether there is a sign of abnormality in the components by other methods. For example, it is also possible to detect whether there is a sign of abnormality in the components based on the time series data measured by the vibration sensor 154 or the torque sensor 155.

[0075] The sign detection device 72 of Embodiment 2 uses the time series data measured by Figure 3 the vibration sensor 154 or the torque sensor 155 shown. The vibration sensor 154 measures the time series data of the vibration during the lifting operation of the passenger car 31. The torque sensor 155 measures the time series data of the torque of the opening / closing motor 41 during opening / closing. The data acquisition unit 74 acquires the time series data measured by the vibration sensor 154 or the torque sensor 155.

[0076] Supervision data is generated based on the time series data of the vibration waveform and the time series data of the torque waveform. In addition, supervision data is generated corresponding to Figure 6 the operation mode shown and the normal operation mode. In addition, in the case of using the time series data measured by the vibration sensor 154, in order to remove non-specific vibrations caused by the actions of passengers, it is preferable to perform sign detection based on the data when there are no passengers in the passenger car 31.

[0077] Although some embodiments of the present invention have been described, these embodiments are shown by way of example and are not intended to limit the scope of the present invention. These new embodiments can be implemented in various other ways, and various omissions, substitutions, and changes can be made without departing from the gist of the invention. These embodiments or their modifications are included in the scope and gist of the invention and are included in the invention described in the claims and its equivalent scope.

Claims

1. A warning detection device, characterized in that: have: a data acquisition unit that acquires time series data measured during the ascending and descending operation of a passenger car of the elevator device; and A sign detection unit detects whether or not there is a sign of abnormality in components constituting the elevator device based on the acquired time series data by using AI of supervisory data.

2. The omen detection device according to claim 1, characterized in that: In the lifting and lowering operation of the passenger car from when there is no abnormality in the components constituting the elevator device to when an abnormality occurs, supervisory data used in the AI ​​possessed by the sign detection unit is generated based on the time series data acquired by the data acquisition unit, If an abnormality occurs in a component constituting the elevator device within a predetermined period after acquiring the time series data, supervisory data used in the AI ​​provided in the sign detection unit is generated as supervisory data in which a label indicating a sign is added to the corresponding time series data.

3. The omen detection device according to claim 1 or 2, characterized in that: The time series data acquired by the data acquisition unit includes: time series data representing a waveform of a voltage or current measured by a measuring device that measures a waveform of a voltage or current supplied to a motor that drives a car to move up or down or opens or closes a door of the car from a power source that supplies power to the motor; time series data of a torque waveform measured by a torque sensor that measures the torque of the motor when the door is opened or closed; and Time series data of a vibration waveform measured by a vibration sensor that measures vibration during the raising and lowering operation of the passenger car.

4. The omen detection device according to claim 1 or 2, characterized in that: have: an extracting unit that extracts time series data of a specified time length from the time series data acquired by the data acquiring unit; a Fourier transform unit that generates a frequency spectrum distribution by performing Fourier transform on the time series data extracted by the extraction unit; and The product identification unit identifies a product that is likely to become abnormal based on the generated frequency spectrum distribution using AI that uses supervisory data.

5. The omen detection device according to claim 4, characterized in that: The time length is set according to the operation cycle of each component constituting the elevator device.

6. The omen detection device according to claim 4, characterized in that: The sign detection device includes a subtraction processing unit that subtracts the value of the spectrum distribution caused by the movement of the product determined not to be a sign of abnormality from the value of the spectrum distribution generated by the Fourier transform unit. The article identification unit detects a sign of abnormality in articles constituting the elevator device based on the frequency spectrum distribution after the subtraction process.

7. A method for early warning detection, characterized in that: include: A data acquisition step of acquiring time series data measured during the ascending and descending operation of the passenger car of the elevator device; as well as The sign detection step detects whether there is a sign of abnormality in the components constituting the elevator device based on the acquired time series data by using AI of the supervisory data.

8. The omen detection method according to claim 7, characterized in that: In the data acquisition process, acquiring time series data measured during the lifting and lowering operation of the passenger car from when there is no abnormality in the components constituting the elevator device to when an abnormality occurs, If an abnormality occurs in the components constituting the elevator device within a predetermined period after the time series data is acquired, supervisory data is generated in which a label indicating that a sign has occurred is added to the corresponding time series data.

Images