Elevator

Abstract

The invention provides an elevator which can capture the situation that the elevator taking experience becomes poor due to continuous vibration when an elevator car stops. An elevator is provided with a vibration measurement unit (10). The vibration measuring part (10) measures vibration generated in an elevator car (6) when the elevator car (6) stops. The state of the elevator is diagnosed on the basis of the vibration measured by the vibration measurement unit (10), the amplitude of the vibration, and the duration of the vibration.

Description

Technical Field

[0001] The present disclosure relates to elevators. Background Art

[0002] Patent document 1 discloses an example of an elevator. In an elevator, early detection of discomfort in the elevator riding experience caused by vibration is sought. In an elevator, the acceleration calculated based on the measurement value of a load sensor installed between a car room and a support frame is compared with a preset threshold value to determine whether the operating condition is good.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Publication No. 2006-264853 Summary of the invention

[0006] The riding experience of an elevator car is sometimes deteriorated due to vibrations occurring in the car. One of the vibrations that deteriorate the riding experience is the vibration in the up and down directions when the car stops. Such vibrations are caused by the input of passengers' boarding and alighting actions. Here, if an abnormality occurs in the elevator itself, such as a reduction in attenuation characteristics or a reduction in rigidity due to deterioration of the ropes suspending the car, a large degree of vibration will continue for a long time, resulting in a deterioration in the riding experience. In the elevator of Patent Document 1, it is impossible to capture the situation where the riding experience is deteriorated due to the continuous vibration when the car stops.

[0007] The present disclosure is intended to solve such a problem. The present disclosure provides an elevator capable of detecting a situation in which the riding experience is deteriorated due to continuous vibration when the car is stopped.

[0008] The elevator involved in the present disclosure comprises: a vibration measuring unit for measuring the vibration occurring in the elevator car when the elevator car stops; and a vibration analyzing unit for determining whether there is an abnormality in the elevator using the vibration measured by the vibration measuring unit, a first threshold value of a preset amplitude, and a second threshold value of a preset vibration duration.

[0009] The elevator involved in the present disclosure comprises: a vibration measuring unit, which measures the vibration occurring in the elevator car when the elevator car stops; a monitoring unit, which obtains the passenger's behavior data when the elevator car stops; a vibration analyzing unit, which calculates the time during which the vibration measured by the vibration measuring unit continues to vibrate at a magnitude exceeding a first threshold value of a preset vibration amplitude as a vibration duration; a behavior analyzing unit, which analyzes the passenger's behavior data obtained by the monitoring unit; a learning device, which uses the vibration duration calculated by the vibration analyzing unit and the passenger's behavior data analyzed by the behavior analyzing unit to generate a learned model for inferring the state of the elevator; and an inference device, which uses the vibration duration calculated by the vibration analyzing unit, the passenger's behavior data analyzed by the behavior analyzing unit, and the learned model generated by the learning device to diagnose the state of the elevator.

[0010] According to the elevator involved in the present disclosure, it is possible to capture the situation where the elevator riding experience deteriorates due to continuous vibration when the car is stopped. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 This is an overall diagram of the elevator in Embodiment 1.

[0012] Figure 2 : is a graph schematically showing acceleration time changes of vibration data measured by the vibration measuring unit.

[0013] Figure 3 This is a graph showing a profile graph in which the peaks of the vibration waveform are connected with respect to the absolute value of the acceleration change.

[0014] Figure 4 This is a graph schematically showing examples of vibration in a case where the vibration damping characteristic is weakened due to an abnormality of the elevator and vibration in a normal case as time changes in acceleration.

[0015] Figure 5 This is a diagram showing the relationship between the vibration duration time when the elevator is normal, the vibration duration time when an abnormality occurs, and the threshold value as a method of setting the threshold value of the vibration duration time.

[0016] Figure 6 This is a graph showing an example of a vibration waveform when a plurality of passengers are riding in an elevator.

[0017] Figure 7 This is an example of a flowchart summarizing the specific process of diagnosis in the first embodiment.

[0018] Figure 8 yes Figure 7 A variation of the flowchart shown.

[0019] Fig. 9This is a hardware configuration diagram of a main part of the elevator according to the first embodiment.

[0020] Fig.10 This is an overall view of the elevator in the second embodiment.

[0021] Fig.11 This is a graph showing a profile diagram created by processing a plurality of peaks of a vibration waveform related to each riding action in a time series order.

[0022] Fig.12 This is a graph showing a profile diagram created by performing a process of sequentially connecting a plurality of peaks in a time series on the entire vibration waveform.

[0023] Fig.13 This is an example of a flowchart summarizing the specific process of diagnosis in the second embodiment.

[0024] Fig.14 This is an overall diagram of the elevator in Embodiment 3.

[0025] Fig.15 This is an example of a flowchart summarizing the specific process of diagnosis in the third embodiment. DETAILED DESCRIPTION

[0026] The method for implementing the object of the present disclosure is described with reference to the accompanying drawings. In each figure, the same symbol is added to the same or equivalent part, and repeated descriptions are appropriately simplified or omitted. In addition, the object of the present disclosure is not limited to the following embodiments, and any deformation of the constituent elements of the implementation or omission of any constituent elements of the implementation can be implemented within the scope of the gist of the present disclosure.

[0027] Implementation method 1.

[0028] Figure 1 This is an overall diagram of the elevator in Embodiment 1.

[0029] The elevator in this example is a traction elevator. The elevator is used in a building with multiple floors. In the building, an elevator lift passage 1 is provided. The lift passage 1 is a space having a length in the vertical direction across multiple floors. On each floor of the building, an elevator landing is provided. The landing is a place adjacent to the lift passage. In this example, a machine room 2 is provided above the lift passage 1.

[0030] A pulley 3 is provided in the machine room 2. The pulley 3 is connected to a hoist 4. A suspension body 5 is mounted on the pulley 3. As the suspension body 5, for example, a plurality of ropes or a plurality of belts are used. In this example, a car 6 and a counterweight 7 are connected to both ends of the suspension body 5, respectively. That is, Figure 1The case where the suspension body 5 is suspended in a 1:1 roping system is exemplified. The suspension body 5 suspends the car 6 and the counterweight 7 in the elevator passage 1.

[0031] When the pulley 3 is rotated by the drive of the hoist 4, the car 6 is raised and lowered inside the lifting passage 1 by the friction between the suspension body 5 and the pulley 3. The elevator is provided with a control device 8. The elevator operates by controlling the rotation of the hoist 4 by the control device 8. After the car 6 arrives at any floor, passengers get on and off between the elevator boarding area and the car 6. The control device 8 is connected to a diagnostic device 9.

[0032] A vibration measuring unit 10 is provided in the car 6. The vibration measuring unit 10 is composed of, for example, an acceleration sensor, a velocity sensor, a displacement sensor or a scale device installed in the car 6, or a combination of a plurality of them. The diagnostic device 9 has a vibration analyzing unit 11 and a recording unit 12. The vibration analyzing unit 11 of the diagnostic device 9 analyzes vibration data obtained by measuring the vibration occurring in the car 6 when passengers are riding or getting off the elevator by the vibration measuring unit 10. The vibration analyzing unit 11 determines whether there is any abnormality in the elevator based on the analysis result of the vibration data. The recording unit 12 has a storage medium. The recording unit 12 records whether there is any abnormality in the elevator determined by the vibration analyzing unit 11.

[0033] Next, the operations of the vibration measuring unit 10 , the vibration analyzing unit 11 , and the recording unit 12 constituting the diagnostic device 9 will be described in detail.

[0034] The vibration measuring unit 10 measures the vibration occurring in the elevator car 6 when the elevator car 6 stops. When the vibration measuring unit 10 detects that the elevator car 6 stops by a signal from the control device 8, it starts measuring the vibration occurring in the car 6. When the vibration measuring unit 10 detects that the elevator car 6 starts to travel by a signal from the control device 8, it ends the measurement. Hereinafter, as the vibration measured by the vibration measuring unit 10, the vibration in the up-and-down direction of the car 6 is described as an example.

[0035] The vibration analysis unit 11 analyzes the vibration measured by the vibration measuring unit 10. The vibration analysis unit 11 receives the vibration data of the elevator car 6 when it stops, which is measured by the vibration measuring unit 10. The vibration analysis unit 11 performs at least one of the analysis including differentiation of the received vibration data or smoothing processing using a filter, and calculates, for example, the time variation of acceleration. Thereafter, the vibration analysis unit 11 performs a process of calculating the duration of the vibration based on the calculated time variation of acceleration. In the case where the vibration measuring unit 10 is composed of a plurality of sensors and devices such as an acceleration sensor, the vibration analysis unit 11 may also calculate the duration of the vibration using an average value of each analysis result, etc., instead of the time variation of acceleration.

[0036] exist Figure 2 In the graph, a solid line schematically shows a temporal change in acceleration based on vibration data measured by the vibration measuring unit 10. When one passenger gets on the elevator car 6, the waveform of the vibration generated is a waveform in which the acceleration is large at the beginning and gradually decays.

[0037] The vibration analysis unit 11 performs processing to connect a plurality of peak values ​​of the vibration waveform in a time series order, and produces Figure 2 The profile is represented by dotted lines in the figure. Each peak value of the vibration waveform is, for example, the maximum value of the vibration waveform. The values ​​between the peak values ​​in the profile are connected by interpolation processing such as linear interpolation or spline interpolation. Here, in the vibration analysis unit 11, the acceleration threshold value A1 is pre-set. The threshold value A1 is an example of the first threshold value. Figure 2 In FIG. 1 , the threshold value A1 is represented by a single dotted line. In addition, the time length from the time point when the original vibration waveform first exceeds the threshold value A1 to the time point when the profile graph first falls below the threshold value A1 is set as the vibration duration T. In addition, the starting point of the vibration duration can also be the starting point of the profile graph, that is, the time point of the initial peak of the vibration waveform.

[0038] In addition, Figure 3 In FIG. 1 , the absolute value of the time variation of the acceleration based on the vibration data measured by the vibration measuring unit 10 is schematically shown by a solid line on the graph. The vibration analyzing unit 11 may also create a graph connecting the peak values ​​for the absolute value of the time variation of the acceleration. Figure 3 The vibration analysis unit 11 may define the vibration duration T as the time length from the time point when the original vibration waveform first exceeds the threshold value A1 to the time point when the profile first falls below the threshold value A1.

[0039] The vibration analysis unit 11 determines whether there is an abnormality in the elevator based on the magnitude of the vibration duration T determined in this way. Specifically, when the vibration duration T exceeds the time threshold Ta, the vibration analysis unit 11 determines that the elevator riding experience is deteriorated, that is, there is an abnormality in the elevator. The time threshold Ta is pre-set in the vibration analysis unit 11. The threshold Ta is an example of the second threshold.

[0040] exist Figure 4 In the example of the time variation of acceleration, the attenuation characteristics of the acceleration of the vibration weaken due to an abnormality of the elevator and the normal case are schematically shown. Figure 4In the figure, the solid line represents the vibration waveform of the normal situation. On the other hand, the dashed line represents the vibration waveform of the situation where the attenuation characteristics of the vibration weaken due to the abnormality of the elevator. In addition, the dotted line represents the contour diagram in which the peak values ​​of the vibrations of each situation are connected in time series by straight lines. The vibration duration T′ of the contour diagram of the vibration during the abnormality exceeds the threshold value A1, which is greater than the vibration duration T calculated based on the contour diagram of the vibration during the normal situation. In order to judge the vibration during the abnormality as "abnormal", it is necessary to set the threshold value Ta to a value between time T and time T′.

[0041] exist Figure 5 The relationship between the vibration duration when the elevator is normal, the vibration duration when an abnormality occurs, and the threshold value Ta is shown as a method for setting the threshold value Ta. In the way that passengers board the car 6, there are individual differences among passengers and deviations each time, so the vibration duration T is also affected by these and produces deviations. Therefore, by setting the threshold value Ta in a manner that takes these deviations into account, the accuracy of a primary diagnosis can be improved. For example, the vibration data of the elevator when the elevator is normal, such as when the elevator is just installed, is obtained a predetermined number of times, and the maximum value of these vibration durations is set as Ta.

[0042] In addition, it is also possible to assume a situation where multiple passengers are riding the elevator, and the vibration duration is defined as follows. Figure 6 An example of a vibration waveform when multiple passengers are on the elevator car is shown. In this case, if the vibration duration T is calculated as described above, T becomes a larger value compared to the case where only one passenger is on the elevator, resulting in the possibility of misdiagnosis. Therefore, multiple acceleration thresholds are set such as Aa1, Aa2, ..., Aam-2, Aam-1, Aam. The vibration analysis unit 11 defines the time from the time point below the threshold value "smaller than the last maximum value of the contour graph and closest to the maximum value" to the time point below the minimum threshold value as the vibration duration T". Figure 6 In the figure, the threshold value "smaller than the last maximum value of the profile graph and closest to the maximum value" becomes Aam-1. Therefore, the vibration analysis unit 11 calculates the time from the time when the profile graph is lower than the threshold value Aam-1 after the last maximum value to the time point when it is lower than the minimum threshold value Aa1 as the vibration duration T". By using the vibration duration T" calculated in this way to determine the abnormality of the elevator, it is possible to perform a diagnosis with the same accuracy as when one passenger is riding the elevator.

[0043] In addition, for example, when the position of the car 6 changes and the length of the suspension body 5 of the part suspending the car 6 becomes longer, its tensile rigidity becomes smaller, so the elongation of the suspension body 5 when the passenger takes the elevator becomes larger, thereby causing vibration with a larger displacement amplitude. In this case, the acceleration of the vibration and the duration of the vibration also change compared to the original position of the car 6. Considering this situation, the threshold value corresponding to the number of floors of the elevator is set, so that no matter which floor the car 6 is on, the vibration analysis unit 11 can diagnose the abnormality of the elevator. For example, for an elevator in which the car 6 stops at the 1st floor to the nth floor, a threshold group corresponding to each floor (Aam1, Aam2, ..., Aamn; Ta1, Ta2, ..., Tan) is pre-set in the vibration analysis unit 11. Here, the integer i is set to an arbitrary integer from 1 to n, and the acceleration threshold Aami and the time threshold Tai are examples of the first threshold and the second threshold corresponding to the situation where the car 6 stops at the i-th floor. When diagnosing an abnormality of the elevator, the vibration analysis unit 11 receives information on the position of the car 6 from the control device 8 and selects an appropriate threshold value corresponding to the number of floors from the threshold value group, thereby being able to diagnose an abnormality of the elevator at any floor.

[0044] The recording unit 12 records the result of the determination made by the vibration analysis unit 11. The diagnostic device 9 may also transmit the determination result to the outside of the elevator using a network line, etc. Thus, for example, the determination result can be quickly sent to a maintenance person who performs maintenance of the elevator, etc., so that the maintenance person can respond at an early stage.

[0045] By means of the above structure, it is possible to diagnose with high accuracy that the riding experience is deteriorated when the elevator car 6 stops. In addition, the diagnosis is performed based on the duration of vibration affected by the abnormality of the elevator, so it is possible to diagnose the situation in which the riding experience is deteriorated due to the abnormality of the elevator itself.

[0046] Figure 7It is a flowchart summarizing the specific process of diagnosing elevator abnormalities. In STEPPa1, after the control device 8 confirms that the elevator car 6 has stopped, the vibration measuring unit 10 starts measuring the vibration in the up and down directions occurring in the car 6 in STEPPa2. Then, in STEPPa3, if the control device 8 confirms that the elevator car 6 has started to move, the vibration measuring unit 10 sends the measurement data to the vibration analyzing unit 11 after completing the measurement. In STEPa4, the vibration analyzing unit 11 calculates the vibration duration T. In STEPPa5, when the vibration duration T exceeds the predetermined vibration duration threshold Ta, the vibration analyzing unit 11 diagnoses that there is an abnormality in the elevator in STEPPa6. On the other hand, when the measured vibration duration T is the same as the vibration duration threshold Ta or does not exceed the threshold Ta, the vibration analyzing unit 11 diagnoses that there is no abnormality in the elevator in STEPPa7. In STEPa8, the judgment result is recorded in the recording unit 12, and the diagnosis is completed once.

[0047] In addition, the diagnostic device 9 can also detect the recording unit 12 as follows: Figure 8 As shown in the flowchart shown, STEPPa9 and STEPPa10 are added, and not only a diagnosis is performed once, but also a judgment action is added with reference to the past diagnosis history. In STEPPa9, the vibration analysis unit 11 calculates the frequency X of being judged as "abnormal" from the diagnosis history of a preset number of diagnoses. For this frequency X, a frequency threshold Xa is preset. In STEPPa10, when the frequency X becomes greater than the threshold Xa, the diagnostic device 9 reports the situation to the outside. In this case, by appropriately setting the threshold Xa, maintenance personnel can respond quickly. As specific examples of the reporting destination, a display or an alarm device, or a maintenance terminal carried by a maintenance personnel in a remote location, etc. can be cited.

[0048] In addition, the diagnostic device 9 may set a threshold for the total number of times the "abnormality" is determined, or the number of times the "abnormality" is determined continuously, instead of the frequency of the determination, and similarly report the situation to the outside.

[0049] Such a diagnostic device 9 can be applied not only to 1:1 roping but also to all roping traction elevators such as 2:1 roping or a system of a machine room-less elevator without a machine room 2 .

[0050] Furthermore, such an elevator diagnostic device 9 can be applied not only to vibrations caused by passengers riding the elevator when the car 6 stops, but also to vibrations caused by passengers getting off the car 6. Moreover, it can be applied not only to vibrations in the up-and-down direction of the car 6, but also to lateral vibrations caused by the tilt of the car 6, etc. In addition, in the present embodiment, the processing of converting vibration data into an example of time variation of acceleration in the vibration analysis unit 11 is described, but the same processing can also be performed on speed or displacement. That is, the threshold value of the amplitude of vibration is not limited to the threshold value for the amplitude of acceleration, but can also be the threshold value for the amplitude of speed or displacement, etc.

[0051] Next, use Fig. 9 , an example illustrating the hardware structure of an elevator.

[0052] Fig. 9 This is a hardware configuration diagram of a main part of the elevator according to the first embodiment.

[0053] Each function of the elevator can be realized by a processing circuit. The processing circuit includes at least one processor 100a and at least one memory 100b. The processing circuit may include at least one dedicated hardware 200 together with the processor 100a and the memory 100b, or may include at least one dedicated hardware 200 instead of them.

[0054] When the processing circuit includes a processor 100a and a memory 100b, each function of the elevator is realized by software, firmware, or a combination of software and firmware. At least one of the software and firmware is described as a program. The program is stored in the memory 100b. The processor 100a realizes each function of the elevator by reading and executing the program stored in the memory 100b.

[0055] The processor 100a is also called a CPU (Central Processing Unit), a processing device, an arithmetic device, a microprocessor, a microcomputer, or a DSP. The memory 100b is composed of a nonvolatile or volatile semiconductor memory such as a RAM, a ROM, a flash memory, an EPROM, or an EEPROM.

[0056] When the processing circuit has dedicated hardware 200, the processing circuit is implemented by, for example, a single circuit, a complex circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

[0057] Each function of the elevator can be realized separately by the processing circuit. Alternatively, each function of the elevator can be realized collectively by the processing circuit. Regarding each function of the elevator, a part can be realized by dedicated hardware 200, and the other part can be realized by software or firmware. In this way, the processing circuit realizes each function of the elevator through dedicated hardware 200, software, firmware, or a combination thereof.

[0058] Implementation method 2.

[0059] In the second embodiment, the differences from the example disclosed in the first embodiment are described in detail. As for the features not described in the second embodiment, any features of the example disclosed in the first embodiment may be adopted.

[0060] Fig.10 This is an overall view of the traction elevator in the second embodiment.

[0061] The elevator of the second embodiment is provided with a monitoring unit 13 and a behavior analysis unit 14 for monitoring the behavior of passengers in addition to the structure of the elevator of the first embodiment. Specifically, the monitoring unit 13 is composed of a camera 13a for photographing the interior of the car 6, a scale device 13b, a floor reaction force sensor 13c provided on the floor of the car, an acceleration sensor, a speed sensor or a displacement sensor, or a combination of a plurality of these. The monitoring unit 13 obtains the behavior data of the passengers while the car 6 is stopped.

[0062] The behavior analysis unit 14 is connected to the monitoring unit 13. The monitoring unit 13 obtains the behavior data of the passenger when the elevator car 6 stops, and then transmits the obtained behavior data to the behavior analysis unit 14.

[0063] Here, when the monitoring unit 13 is a camera 13a, the monitoring unit 13 transmits the image or video data obtained by photographing the interior of the car 6 as the passenger's behavior data to the behavior analysis unit 14. When the monitoring unit 13 is a scale device 13b, the monitoring unit 13 transmits the time change of the weight of the car 6 as the passenger's behavior data to the behavior analysis unit 14. When the monitoring unit 13 is a floor reaction force sensor 13c, the monitoring unit 13 transmits the time change of the floor reaction force applied to the floor of the car as the passenger's behavior data to the behavior analysis unit 14.

[0064] The behavior analysis unit 14 analyzes the passenger's behavior data, calculates the number of passengers n, the timing of riding τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of the car 6 after each passenger rides as analysis results, and transmits these to the vibration analysis unit 11. In the case where the passenger's behavior data is an image or video taken by the camera 13a, the behavior analysis unit 14 infers the various parameters of the above analysis results through image analysis. In the case where the passenger's behavior data is the weight change of the car 6 measured by the scale device 13b, the behavior analysis unit 14 infers the various parameters of the above analysis results based on its time change. In the case where the passenger's behavior data is the ground reaction force, the behavior analysis unit 14 infers the various parameters of the above analysis results based on the time change of the ground reaction force. In the case where the monitoring unit 13 uses multiple types of sensors and other devices, the behavior analysis unit 14 can also be set to calculate the various parameters of the above analysis results by the average value of the results calculated by each device.

[0065] A plurality of threshold groups are pre-set in the vibration analysis unit 11. The vibration analysis unit 11 selects and sets the acceleration threshold value Aa and the vibration duration threshold value Ta from the threshold group based on the number of passengers n received from the behavior analysis unit 14, the timing of boarding τ1, τ2, ..., τn, the weight M1, M2, ..., Mn of the car 6 after each passenger boarded the car, and the car position information received from the control device 8.

[0066] The vibration analysis unit 11 uses the vibration data measured by the vibration measurement unit 10 to calculate, for example, a temporal change in acceleration. Fig.11 Schematically shows the change in acceleration during vibration when a plurality of passengers board the car 6. Fig.11 , an example is shown when three passengers of the same weight take the elevator continuously. The vibration occurring in the car 6 at this time is different from the situation when only one passenger takes the elevator, and the second and subsequent passengers take the elevator halfway, so that the acceleration increases again from the middle and becomes a decay waveform.

[0067] The vibration analysis unit 11 processes the vibration caused by each passenger riding the elevator divided by the timings τ1, τ2, and τ3 of riding the elevator by connecting multiple peaks of acceleration changes in a time series order to produce a Fig.11 The contour graph is shown by dotted lines in FIG. The values ​​between the peaks in the contour graph are connected by interpolation processing such as linear interpolation or spline interpolation. The vibration analysis unit 11 calculates the total time that the contour graph exceeds the preset acceleration threshold value Aa as the vibration duration T. Fig.11, the vibration duration T is the sum of T1, T2, and T3. In addition, the vibration analysis unit 11 may also set the starting point of the range for calculating the vibration duration of each passenger to the point at which the acceleration change caused by each passenger first exceeds the threshold value Aa, the timing τ1, τ2, τ3 of the elevator ride, or the initial point of the profile. The vibration analysis unit 11 may also set the ending point of the range for calculating the vibration duration of each passenger to the timing at which the profile is lower than the threshold value Aa, or to the final point of each profile when the profile is not lower than the threshold value Aa, etc.

[0068] In addition, the vibration analysis unit 11 may also be Fig.12 As shown, a contour graph is created by connecting a plurality of peak values ​​in a time series order for the entire vibration waveform. The values ​​between the peak values ​​in the contour graph are connected by interpolation processing such as linear interpolation or spline interpolation. The vibration analysis unit 11 may also calculate the total length of time exceeding the threshold value Aa from the point where the acceleration exceeds the preset acceleration threshold value Aa at the first point to the point where the acceleration falls below the threshold value Aa at the last point in the contour graph created in this way as the vibration duration T. Fig.12 , the vibration duration T is the sum of T4 and T5.

[0069] The vibration analysis unit 11 may also create a profile graph by connecting the peaks of the waveform of the absolute value of the acceleration change. The vibration analysis unit 11 may also calculate the vibration duration T by performing the same processing on the profile graph created in this way.

[0070] Here, when a heavy passenger or a heavy cargo takes the elevator car 6, the vibration duration T is also affected by the magnitude of the applied load. In such a case, the vibration analysis unit 11 can perform a more appropriate diagnosis by setting the acceleration threshold value Aa that adopts the weight Mn of the car 6 after n passengers take the elevator calculated by the behavior analysis unit 14. For example, when a heavy passenger takes the elevator, the acceleration increases with the increase of the applied load, so the threshold value Aa is also set to a value that increases according to the degree of load increase.

[0071] In addition, a threshold value Ta of the vibration duration is selected from pre-set thresholds based on the number of passengers n and the timing of the elevator rides τ1, τ2, ..., τn. For example, when the number of passengers increases, the vibration duration T of the vibration generated in the car n becomes longer, so the threshold value Ta is also set to a larger value. On the other hand, when the intervals of the timing of the elevator rides τ1, τ2, ..., τn are extremely short, the vibration duration T obtained by summing the vibration durations of all elevator rides becomes shorter. Therefore, the vibration analysis unit 11 may also set a threshold value τa for the intervals of the elevator rides, set a vibration duration threshold value Ta corresponding to the length of the intervals of τ1, τ2, ..., τn, or remove it from the diagnosis object according to the length of the intervals of τ1, τ2, ..., τn.

[0072] The vibration analysis unit 11 diagnoses whether there is abnormality in the elevator based on the vibration duration T and the vibration duration threshold value Ta calculated in this way. The recording unit 12 records the diagnosis result.

[0073] In addition, by extracting only the vibration data caused by the last passenger, Fig.11 In this example, only vibration data after time τ3 is extracted, and vibration data corresponding to one passenger riding the elevator is obtained. The vibration analysis unit 11 can also diagnose the elevator by the same processing as in the first embodiment based on such vibration data.

[0074] In this way, even when there are multiple people on the elevator, the vibration analysis unit 11 selects a suitable threshold value Ta from the pre-set threshold value group based on the number of passengers n calculated by the behavior analysis unit 14, the timing τ1, τ2, ..., τn of each ride, the weight M1, M2, ..., Mn of the car 6 after each ride, and the car position information transmitted from the control device 8, thereby being able to appropriately diagnose the elevator.

[0075] In addition, an elevator may perform a special operation such as a robot-only operation mode for carrying objects such as robots whose weights are known in advance. In this case, the weight and size of the objects are known, so it is difficult for the magnitude of the vibrations to vary. Therefore, by setting robot-only thresholds Aa and Ta, it is possible to perform diagnosis with higher accuracy than when the passengers are humans. For example, by reducing Figure 5 Set robot-specific thresholds, such as the deviation range in .

[0076] In addition, when multiple passengers or cargoes with greatly different body weights or weights board the car 6, the vibration analysis unit 11 may set an appropriate acceleration threshold value Aa for each passenger. The vibration analysis unit 11 can perform appropriate diagnosis by selecting an appropriate vibration duration threshold value Ta for the combination.

[0077] In this way, even when there are a plurality of passengers, heavy passengers, or heavy cargo on board, the vibration analysis unit 11 can appropriately diagnose the elevator.

[0078] In addition, when the passenger continues to move around inside the elevator car 6 while the elevator car 6 is stopped, the vibration continues at a large acceleration, so the vibration duration also becomes extremely large, which may lead to a misdiagnosis. Therefore, when the vibration analysis unit 11 detects the passenger's movement in the analysis of the passenger's behavior data in the behavior analysis unit 14, it can prevent misdiagnosis by removing the vibration caused by the passenger's movement from the diagnosis object.

[0079] Fig.13This is an example of a flowchart for diagnosing an abnormality of an elevator through such a structure and action. After the control device 8 confirms that the elevator car 6 has stopped in STEPb1, the vibration measuring unit 10 and the monitoring unit 13 start to obtain vibration data and passenger behavior data in STEPb2. After that, in STEPb3, when the control device 8 confirms that the elevator car 6 has started to move, the vibration measuring unit 10 and the monitoring unit 13 end to obtain data. In STEPb4, the behavior analysis unit 14 analyzes the passenger's behavior and confirms problems such as passenger turmoil or the presence or absence of boarding and alighting. If there is no problem, the behavior analysis unit 14 continues to analyze the passenger's behavior data and calculates the number of passengers n, the timing of the elevator τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of the car 6 after each ride. In STEPb5, the vibration analysis unit 11 receives the number of passengers n, the timing of the elevator τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of the car 6 after each ride from the behavior analysis unit 14. The vibration analysis unit 11 selects the threshold value Aa and the threshold value Ta based on the received information. In STEPb6, the vibration analysis unit 11 analyzes the vibration data to calculate the profile diagram for the acceleration change. The vibration analysis unit 11 calculates the vibration duration T based on the calculated profile diagram. In STEPb7 to STEPb9, the vibration analysis unit 11 determines whether the vibration duration T exceeds the threshold value Ta to diagnose whether the elevator has abnormality. In STEPb10, the recording unit 12 records the diagnosis result.

[0080] The diagnostic device 9 may not only perform a single diagnosis up to STEPb10, but also make external reports based on multiple diagnostic histories, as in STEPb11 to STEPb13. That is, the vibration analysis unit 11 calculates the frequency X of being determined as "abnormal" based on the diagnostic history of N times, which is a preset number of diagnostic times. A frequency threshold Xa is preset for this frequency X. At a timing such as when the frequency X exceeds the threshold Xa, the diagnostic device 9 reports the status to the outside. As specific examples of the reporting destination, a display or an alarm device, or a maintenance terminal carried by a maintenance person in a remote location, etc. can be cited.

[0081] In addition, the diagnostic device 9 may set a threshold for the total number of times the "abnormality" is determined or the number of times the "abnormality" is continuously determined instead of the frequency of the determination, and similarly report the situation to the outside.

[0082] With the above-described configuration, the diagnostic device 9 can assign an appropriate threshold value to the vibration by comparing and analyzing the vibration data with the passenger's behavior data, and thus can diagnose an abnormality of the elevator with high accuracy.

[0083] Such a diagnostic device 9 can be applied not only to 1:1 roping but also to all roping traction elevators such as 2:1 roping or a system of a machine room-less elevator without a machine room 2 .

[0084] Furthermore, such an elevator diagnostic device 9 can be applied not only to vibrations caused by passengers riding the elevator when the car 6 stops, but also to vibrations caused by passengers getting off the car 6. Moreover, it can be applied not only to vibrations in the up-and-down direction of the car 6, but also to lateral vibrations caused by the tilt of the car 6, etc. In addition, in the present embodiment, the processing of converting vibration data into an example of time variation of acceleration in the vibration analysis unit 11 is described, but the same processing can also be performed on speed or displacement. That is, the threshold value of the amplitude of vibration is not limited to the threshold value for the amplitude of acceleration, but can also be the threshold value for the amplitude of speed or displacement, etc.

[0085] Implementation method 3.

[0086] In Embodiment 3, the differences from the example disclosed in Embodiment 1 or Embodiment 2 are described in detail. For features not described in Embodiment 3, any features of the example disclosed in Embodiment 1 or Embodiment 2 may be adopted.

[0087] Fig.14 This is an overall view of a traction-type elevator in the third embodiment.

[0088] exist Fig.14 In the traction type elevator shown, a machine room 2 is provided above a lifting passage 1. A pulley 3 is provided in the machine room 2. The pulley 3 is connected to a hoist 4. A suspension body 5 is mounted on the pulley 3. For example, a plurality of ropes or a plurality of belts are used as the suspension body 5. In this example, a car 6 and a counterweight 7 are connected to both ends of the suspension body 5, respectively. That is, Fig.14 The case where the suspension body 5 is suspended in the 1:1 roping system is exemplified. The suspension body 5 suspends the car 6 and the counterweight 7 in the elevator passage 1.

[0089] When the pulley 3 is rotated by the drive of the hoist 4, the car 6 is raised and lowered in the lifting passage 1 by the friction between the suspension body 5 and the pulley 3. The elevator is provided with a control device 8. The elevator operates by controlling the rotation of the hoist 4 by the control device 8. After the car 6 arrives at any floor, passengers get on and off at the elevator entrance and between the cars 6. The control device 8 is connected to a diagnostic device 9.

[0090] The diagnostic device 9 includes a vibration analysis unit 11 , a monitoring unit 13 , a behavior analysis unit 14 , a learning device 15 , an estimation device 16 , and a learned model storage unit 17 .

[0091] The car 6 is provided with a vibration measuring unit 10. The vibration measuring unit 10 is composed of, for example, an acceleration sensor, a velocity sensor, a displacement sensor, or a scale device attached to the car 6, or a combination of a plurality of these.

[0092] The vibration analysis unit 11 analyzes the vibration data sent from the vibration measurement unit 10, calculates acceleration changes, and calculates the vibration duration T based on the preset acceleration threshold Aa. The calculation method of the vibration duration T based on the acceleration threshold Aa is the same as that of the first or second embodiment.

[0093] The monitoring unit 13 obtains the passenger's behavior data when the elevator car 6 is stopped. Specific examples of the monitoring unit 13 are a camera 13a for photographing the interior of the car 6, a scale device 13b, a floor reaction force sensor 13c provided on the floor of the car, an acceleration sensor, a speed sensor, or a displacement sensor, or a combination of any plurality of these. The monitoring unit 13 obtains the passenger's behavior data while the car 6 is stopped.

[0094] Here, when the monitoring unit 13 is a camera 13a, the monitoring unit 13 transmits the image or video data obtained by photographing the interior of the car 6 as the passenger's behavior data to the behavior analysis unit 14. When the monitoring unit 13 is a scale device 13b, the monitoring unit 13 transmits the time change of the weight of the car 6 as the passenger's behavior data to the behavior analysis unit 14. When the monitoring unit 13 is a floor reaction force sensor 13c, the monitoring unit 13 transmits the time change of the floor reaction force applied to the floor of the car as the passenger's behavior data to the behavior analysis unit 14.

[0095] The behavior analysis unit 14 analyzes the passenger's behavior data, calculates the number of passengers n, the timing of riding τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of the car 6 after each passenger rides as analysis results, and transmits these to the vibration analysis unit 11. In the case where the passenger's behavior data is an image or video taken by the camera 13a, the behavior analysis unit 14 infers the various parameters of the above analysis results through image analysis. In the case where the passenger's behavior data is the weight change of the car 6 measured by the scale device 13b, the behavior analysis unit 14 infers the various parameters of the above analysis results based on its time change. In the case where the passenger's behavior data is the ground reaction force, the behavior analysis unit 14 infers the various parameters of the above analysis results based on the time change of the ground reaction force. In the case where the monitoring unit 13 uses multiple types of sensors and other devices, the behavior analysis unit 14 can also be set to the average value of the results calculated by each device to obtain the various parameters of the above analysis results.

[0096] The vibration analysis unit 11 and the behavior analysis unit 14 are connected to a learning device 15. The learning device 15 includes a data acquisition unit 18 and a model generation unit 19. Furthermore, the vibration analysis unit 11 and the behavior analysis unit 14 are connected to an inference device 16. The inference device 16 includes a data acquisition unit 20 and an inference unit 21.

[0097] The data acquisition unit 18 in the learning device 15 and the data acquisition unit 20 of the inference device 16 obtain the car position information L from the control device 8, obtain the vibration duration T from the vibration analysis unit 11, and obtain the number of passengers n, the timing of the ride τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of the car 6 under each passenger from the behavior analysis unit 14 as learning data.

[0098] The model generation unit 19 learns the state of the elevator based on the learning data generated based on the combination of the car position information L, the vibration duration T, the number of passengers n, the timing τ1, τ2, ..., τn of the elevator, and the weights M1, M2, ..., Mn of the car 6 under each elevator output from the data acquisition unit 18. That is, the model generation unit 19 generates a learned model for estimating the state of the elevator based on the car position information L, the vibration duration T, the number of passengers n, the timing τ1, τ2, ..., τn of the elevator, and the weights M1, M2, ..., Mn of the car 6 under each elevator. In this example, the learning data is data that associates the car position information L, the vibration duration T, the number of passengers n, the timing τ1, τ2, ..., τn of the elevator, and the weights M1, M2, ..., Mn of the car 6 under each elevator.

[0099] The learning algorithm used by the model generation unit 19 can use a known algorithm such as supervised learning, unsupervised learning, and reinforcement learning. As an example, the case of applying clustering such as the k-means method as unsupervised learning is described. Unsupervised learning refers to a method of learning the characteristics of learning data by providing learning data that does not contain results or labels to the learning device 15.

[0100] The model generation unit 19 learns the state of the elevator by so-called unsupervised learning according to a grouping method using, for example, the k-means method. The k-means method is a non-hierarchical clustering algorithm that classifies the number of clusters provided into k using the mean of the clusters.

[0101] More specifically, the k-means method is processed according to the following process. First, a cluster is randomly assigned to each data xi. Next, the center Vj of each cluster is calculated based on the assigned data. Next, the distance between each xi and each Vj is calculated, and xi is reallocated to the cluster with the nearest center. Moreover, if the cluster assignment of all xi does not change during the above process, or the amount of change is less than a certain threshold set in advance, it is determined to be converged and the process is terminated.

[0102] In the present application, the model generating unit 19 learns the state of the elevator according to the learning data through so-called unsupervised learning, wherein the learning data is produced based on a combination of the car position information L, the vibration duration T, the number of passengers n, the timing of the ride τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of each car 6 obtained by the data acquiring unit 18.

[0103] The model generation unit 19 generates and outputs a learned model by performing the above learning. The learned model storage unit 17 stores the learned model output from the model generation unit 19. By implementing such learning in a normal elevator, a learned model is obtained that learns the relationship between the car position information L, the vibration duration T, the number of passengers n, the timing of the elevator ride τ1, τ2, ..., τn, the weight M1, M2, ..., Mn of the car 6 under each elevator ride, and the normal state of the elevator.

[0104] The inference unit 21 infers the state of the elevator using the learned model stored in the learned model storage unit 17. That is, the inference unit 21 inputs data produced based on a combination of the car position information L, the vibration duration T, the number of passengers n, the timing of the ride τ1, τ2, ..., τn, and the weights M1, M2, ..., Mn of the cars 6 under each ride, obtained by the data acquisition unit 20, to the learned model, thereby being able to infer which cluster the data belongs to, and outputting the inference result as the state of the elevator. When the data input to the learned model does not belong to any of the clusters representing the normal state of the elevator, the inference unit 21 determines that an abnormality has occurred in the elevator.

[0105] The recording unit 12 records the result determined by the inference unit 21. The diagnostic device 9 may also transmit or report the determination result recorded in the recording unit 12 to the outside of the elevator using a network line, etc. Thus, for example, the determination result can be quickly transmitted or sent to a maintenance person, and the maintenance person can deal with it at an early stage.

[0106] Furthermore, in the present embodiment, the estimation unit 21 is described as outputting the state of the elevator using the learned model learned by the model generation unit 19, but the estimation unit 21 may output the state of the elevator based on the learned model acquired from the outside.

[0107] In addition, the case where the data acquisition unit 18 and the data acquisition unit 20 are respectively provided in the learning device 15 and the inference device 16 is described, but the structure of the diagnostic device 9 is not limited to this case. The diagnostic device 9 may also be configured so that a single data acquisition unit acquires necessary information from the control device 8, the vibration analysis unit 11, and the behavior analysis unit 14, and outputs the acquired information to the model generation unit 19 of the learning device 15 and the inference unit 21 of the inference device 16.

[0108] In this way, the inference unit 21 determines the state of the elevator based on the car position information L, vibration duration T, the number of passengers n, the timing of riding the elevator τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of each car 6 under the elevator.

[0109] Next, use Fig.15 Next, a processing flow for diagnosing the state of an elevator using the learning device 15 and the estimation device 16 will be described. The diagnosis of the elevator in the third embodiment consists of two stages, namely, a learning stage and a diagnosis stage.

[0110] In the learning phase, first, in order to obtain learning data, in STEPc1, after the data acquisition unit 18 receives a signal indicating that the elevator car 6 has stopped from the control device 8, the vibration measurement unit 10 and the monitoring unit 13 measure data.

[0111] In STEPc2, after the data acquisition unit 18 receives a signal indicating that the elevator car 6 starts traveling from the control device 8, the vibration analysis unit 11 and the behavior analysis unit 14 receive the data measured by the vibration measurement unit 10 and the monitoring unit 13. The vibration analysis unit 11 and the behavior analysis unit 14 analyze the received data and calculate the car position information L, the vibration duration T, the number of passengers n, the timing of the passengers τ1, τ2, ..., τn, and the weights M1, M2, ..., Mn of the cars 6 at each passenger.

[0112] In STEPc3, the data acquisition unit 18 acquires the output car position information L, vibration duration T, the number of passengers n, boarding timing τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of each car 6 under the elevator.

[0113] In STEPc4, the model generating unit 19 learns the normal state of the elevator according to the learning data through so-called unsupervised learning to generate a learned model, wherein the learning data is produced based on a combination of the car position information L, the vibration duration T, the number of passengers n, the timing of the elevator τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of each car 6 obtained by the data acquiring unit 18.

[0114] In STEPc5, the learned model storage unit 17 stores the learned model generated by the model generation unit 19. Such learning is performed every time the elevator stops at a stop floor, and the learning is continued until the predetermined number of times is reached.

[0115] In the diagnosis phase, in order to diagnose the state of the elevator using the estimation device 16, the vibration measuring unit 10 and the monitoring unit 13 measure data in STEPc6.

[0116] In STEPc7, the vibration analysis unit 11 and the behavior analysis unit 14 analyze the data measured by the vibration measurement unit 10 and the monitoring unit 13, and calculate the car position information L, vibration duration T, the number of passengers n, the timing of the passengers τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of each car 6.

[0117] In STEPc8, the data acquisition unit 20 acquires the output car position information L, vibration duration T, the number of passengers n, boarding timing τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of each car 6 under the elevator.

[0118] In STEPc9, the inference unit 21 inputs the data of the car position information L, vibration duration T, number of passengers n, boarding timing τ1, τ2, ..., τn, and weight M1, M2, ..., Mn of each car 6 obtained by the data acquisition unit 20 into the learned model stored in the learned model storage unit 17.

[0119] In STEPc10, the estimation unit 21 outputs the state of the elevator obtained from the learned model.

[0120] In STEPc11, the estimation unit 21 diagnoses the state of the elevator based on the output result of the state of the elevator.

[0121] In addition, like the diagnostic device 9 of Embodiment 1 or Embodiment 2, the inference device 16 may report not only based on one diagnosis but based on the diagnostic history of a preset number of diagnosis times. The inference device 16, for example, calculates the frequency X of being determined as "abnormal". A frequency threshold Xa is preset for the frequency X. The inference device 16 may report the abnormality of the elevator to the outside at a timing such as when the frequency X exceeds the threshold Xa. As specific examples of the reporting destination, a display or an alarm device, or a maintenance terminal carried by a maintenance person in a remote location, etc., can be cited.

[0122] In addition, the estimation device 16 may set a threshold for the total number of times the "abnormality" is determined, or the number of times the "abnormality" is determined continuously, instead of the frequency of the determination, and similarly report the situation to the outside.

[0123] With the above-described configuration, the state of the elevator is diagnosed for all vibrations in the vertical direction of the elevator when the car 6 is stopped.

[0124] In addition, in this embodiment, the case where unsupervised learning is applied to the learning algorithm used by the model generation unit 19 and the inference unit 21 is described, but the learning algorithm is not limited to this. Regarding the learning algorithm, in addition to unsupervised learning, reinforcement learning, supervised learning, or semi-supervised learning can also be applied.

[0125] In addition, as a learning algorithm used in the model generation unit 19, deep learning (Deep Learning) for learning the extraction of feature quantities themselves can be used, and other well-known methods can also be used.

[0126] When implementing unsupervised learning in this embodiment, it is not limited to the non-hierarchical clustering using the k-means method as described above, and other known clustering methods may be used. For example, hierarchical clustering such as the shortest distance method may be used.

[0127] In the present embodiment, the case where the learning device 15 and the inference device 16 are provided in the diagnostic device 9 is described, but part or all of the learning device 15 and the inference device 16 may be provided in a device independent of the elevator and connected to the elevator via a network, for example, instead of the diagnostic device 9. In addition, part or all of the learning device 15 and the inference device 16 may be built into the control device 8. Furthermore, part or all of the learning device 15 and the inference device 16 may be installed on a cloud server.

[0128] Furthermore, the model generation unit 19 may be configured to learn the state of an elevator based on learning data created for a plurality of elevators.

[0129] The model generation unit 19 may acquire learning data from a plurality of elevators used in the same area, or may learn the state of the elevator using learning data collected from a plurality of elevators operating independently in different areas. In addition, an elevator may be added to the object for collecting learning data, or removed from the object. Furthermore, the learning device 15 that has learned the state of an elevator for a certain elevator may be applied to a different elevator, and the state of the elevator may be learned again for the different elevator and updated.

[0130] Such an elevator diagnostic device 9 can be applied not only to 1:1 roping but also to all roping traction elevators such as 2:1 roping or a system of a machine room-less elevator without a machine room 2 .

[0131] Furthermore, such an elevator diagnostic device 9 can be applied not only to vibrations caused by passengers riding the elevator when the car 6 stops, but also to vibrations caused by passengers getting off the car 6. Moreover, it can be applied not only to vibrations in the up-and-down direction of the car 6, but also to lateral vibrations caused by the tilt of the car 6, etc. In addition, in the present embodiment, the processing of converting vibration data into an example of time variation of acceleration in the vibration analysis unit 11 is described, but the same processing can also be performed on speed or displacement. That is, the threshold value of the amplitude of vibration is not limited to the threshold value for the amplitude of acceleration, but can also be the threshold value for the amplitude of speed or displacement, etc.

[0132]

Industrial Applicability

[0133] The elevator involved in the present disclosure may be applied to a building having a plurality of floors.

[0134] (Explanation of Reference Numerals)

[0135] 1: lifting passage; 2: machinery room; 3: pulley; 4: hoist; 5: suspension body; 6: car; 7: counterweight; 8: control device; 9: diagnosis device; 10: vibration measurement unit; 11: vibration analysis unit; 12: recording unit; 13: monitoring unit; 13a: camera; 13b: scale device; 13c: ground reaction force sensor; 14: action analysis unit; 15: learning device; 16: inference device; 17: learned model storage unit; 18: data acquisition unit; 19: model generation unit; 20: data acquisition unit; 21: inference unit; 100a: processor; 100b: memory; 200: dedicated hardware.

Claims

1. An elevator having: a vibration measuring section that measures vibration occurring in the elevator car when the car is stopped; and The vibration analysis unit determines whether or not there is an abnormality in the elevator using the vibration measured by the vibration measurement unit, a first threshold value of a preset amplitude, and a second threshold value of a preset vibration duration.

2. The elevator according to claim 1, wherein: The vibration analysis unit determines that there is an abnormality in the elevator when the time for which the vibration amplitude measured by the vibration measurement unit exceeds the first threshold value continues to exceed the second threshold value.

3. The elevator according to claim 1 or 2, wherein: have: a monitoring unit that obtains passenger movement data when the elevator car stops; and a behavior analysis unit, for analyzing the behavior data of the passenger obtained by the monitoring unit, The vibration analysis unit selects the first threshold value and the second threshold value from a preset threshold value group based on the behavior data analyzed by the behavior analysis unit, and determines whether there is an abnormality in the elevator using the selected first threshold value and the second threshold value.

4. The elevator according to any one of claims 1 to 3, wherein: When the elevator is performing robot-only operation, the vibration analysis unit uses the values ​​set for the robot as the first threshold value and the second threshold value to determine whether or not there is an abnormality in the elevator.

5. An elevator having: a vibration measuring unit for measuring vibration occurring in the elevator car when the elevator car is stopped; A monitoring unit, which obtains the movement data of passengers when the elevator car stops; a vibration analysis unit that calculates a time during which the vibration measured by the vibration measurement unit continues to vibrate at a magnitude exceeding a first threshold value of a vibration amplitude that is set in advance as a vibration duration time; a behavior analysis unit for analyzing the behavior data of the passenger obtained by the monitoring unit; a learning device that generates a learned model for inferring the state of the elevator using the vibration duration calculated by the vibration analysis unit and the passenger's behavior data analyzed by the behavior analysis unit; as well as The inference device diagnoses the state of the elevator using the vibration duration calculated by the vibration analysis unit, the passenger's behavior data analyzed by the behavior analysis unit, and the learned model generated by the learning device.

Images