WO2023223490A1

WO2023223490A1 - Elevator

Info

Publication number: WO2023223490A1
Application number: PCT/JP2022/020797
Authority: WO
Inventors: 龍亮谷生
Original assignee: 三菱電機株式会社
Priority date: 2022-05-19
Filing date: 2022-05-19
Publication date: 2023-11-23
Also published as: JPWO2023223490A1; JP7207625B1

Abstract

Provided is an elevator with which it is possible to ascertain worsening ride comfort due to vibrations continuing when the car is stopped. This elevator comprises a vibration measuring unit (10). The vibration measuring unit (10) measures vibrations occurring in the car (6) of the elevator when the car (6) is stopped. The state of the elevator is diagnosed on the basis of the vibrations measured by the vibration measuring unit (10), the vibration amplitude, and the vibration duration.

Description

elevator

The present disclosure relates to an elevator.

Patent Document 1 discloses an example of an elevator. In elevators, efforts are being made to early detect defects in ride comfort caused by vibration. In an elevator, an acceleration calculated from a measured value of a load sensor provided between a car and a support frame is compared with a preset threshold value to determine whether the operating condition is good or bad.

Japanese Patent Publication No. 2006-264853

The ride comfort of an elevator car can be worsened by vibrations generated in the car. One type of vibration that worsens ride comfort is vertical vibration when the car is stopped. Such vibrations are excited by input from passengers getting on and off the train. If an abnormality occurs in the elevator itself, such as a decrease in damping characteristics or a decrease in rigidity due to deterioration of the ropes that suspend the car, severe vibrations will continue for a long time, leading to a worsening of ride comfort. In the elevator disclosed in Patent Document 1, it is not possible to detect the deterioration in ride comfort caused by sustained vibration when the car is stopped.

The present disclosure relates to solving such problems. The present disclosure provides an elevator that can detect deterioration in riding comfort caused by sustained vibration when the car is stopped.

An elevator according to the present disclosure includes a vibration measuring unit that measures vibrations generated in the elevator car when the elevator car is stopped; the vibration measured by the vibration measuring unit; a preset first threshold of amplitude; A vibration analysis unit that determines whether or not there is an abnormality in the elevator using a second threshold value of the vibration duration.
The elevator according to the present disclosure includes a vibration measurement unit that measures vibrations generated in the elevator car when the elevator car stops, a monitoring unit that acquires passenger behavior data when the elevator car stops, and the vibration measurement unit that measures vibrations generated in the elevator car. a vibration analysis unit that calculates a period of time during which the vibration continues to vibrate at a magnitude exceeding a preset first vibration amplitude threshold value as a vibration duration; and a behavior analysis unit that analyzes passenger behavior data acquired by the monitoring unit. a learning device that generates a trained model for inferring the state of the elevator using the vibration duration calculated by the vibration analysis unit and the passenger behavior data analyzed by the behavior analysis unit; and an inference device that diagnoses the state of the elevator using the vibration duration calculated by the vibration duration section, the passenger behavior data analyzed by the behavior analysis section, and the learned model generated by the learning device.

With the elevator according to the present disclosure, it is possible to detect deterioration in ride comfort due to continued vibration when the car is stopped.

1 is an overall view of an elevator in Embodiment 1. FIG. It is a graph schematically showing acceleration time changes of vibration data measured by a vibration measurement unit. It is a graph showing a profile in which the peak of a vibration waveform is connected to the absolute value of acceleration change. 12 is a graph schematically showing an example of vibration when the vibration damping characteristic is weakened due to abnormality of the elevator and vibration when the vibration is normal, as a time change in acceleration. As a method of setting the threshold value of the vibration duration time, the relationship between the vibration duration time when the elevator is normal, the vibration duration time when an abnormality occurs, and the threshold value is shown. It is a graph showing an example of a vibration waveform when a plurality of passengers board the vehicle. 1 is an example of a flowchart summarizing a specific process of diagnosis in the first embodiment. This is a modification of the flowchart shown in FIG. 7. 1 is a hardware configuration diagram of main parts of an elevator according to Embodiment 1. FIG. FIG. 3 is an overall view of an elevator in Embodiment 2. FIG. It is a graph showing a profile created by performing a process of connecting a plurality of peaks in chronological order with respect to vibration waveforms for individual boarding. It is a graph showing a profile created by performing a process of connecting a plurality of peaks in chronological order for the entire vibration waveform. 7 is an example of a flowchart summarizing the specific process of diagnosis in Embodiment 2. FIG. 7 is an overall view of an elevator in Embodiment 3. 7 is an example of a flowchart summarizing the specific process of diagnosis in Embodiment 3.

Embodiments for carrying out the subject matter of the present disclosure will be described with reference to the accompanying drawings. In each figure, the same or corresponding parts are given the same reference numerals, and overlapping explanations are simplified or omitted as appropriate. Note that the subject of the present disclosure is not limited to the following embodiments, and modifications of any constituent elements of the embodiments or modifications of any constituent elements of the embodiments may be made without departing from the spirit of the present disclosure. Can be omitted.

Embodiment 1.
FIG. 1 is an overall view of the elevator in the first embodiment.

The elevator in this example is a traction type elevator. Elevators are applied to buildings with multiple floors. In a building, an elevator hoistway 1 is provided. The hoistway 1 is a vertically long space spanning multiple floors. Elevator landings will be provided on each floor of the building. The landing is located adjacent to the hoistway. In this example, a machine room 2 is provided above the hoistway 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 suspended over the pulley 3. For the suspension body 5, for example, a plurality of ropes or a plurality of belts are used. In this example, a cage 6 and a weight 7 are connected to both ends of the suspension body 5, respectively. That is, FIG. 1 illustrates a case where the suspension body 5 is suspended by a 1:1 roping system. A suspension body 5 suspends a car 6 and a weight 7 in the hoistway 1.

When the pulley 3 is rotated by the drive of the hoist 4, the car 6 moves up and down in the hoistway 1 due to the frictional force between the suspension body 5 and the pulley 3. The elevator is provided with a control device 8. The elevator operates as the control device 8 controls the rotation of the hoist 4. After the car 6 arrives at any floor, passengers board and alight between the landing and the car 6. A diagnostic device 9 is connected to the control device 8 .

The car 6 is provided with a vibration measurement section 10. The vibration measurement unit 10 is configured by, for example, an acceleration sensor, a speed sensor, a displacement sensor, or a weighing device attached to the car 6, or a combination of a plurality of these. The diagnostic device 9 includes a vibration analysis section 11 and a recording section 12. The vibration analysis unit 11 of the diagnostic device 9 analyzes vibration data obtained by the vibration measurement unit 10 measuring vibrations generated in the car 6 when passengers board or exit the car. The vibration analysis unit 11 determines whether there is an 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 or not there is an abnormality in the elevator determined by the vibration analysis unit 11.

Next, details of the operations of the vibration measurement section 10, vibration analysis section 11, and recording section 12 that constitute the diagnostic device 9 will be explained.

The vibration measurement unit 10 measures vibrations generated in the elevator car 6 when the elevator car 6 is stopped. When the vibration measuring unit 10 detects that the elevator car 6 has stopped using a signal from the control device 8, it starts measuring vibrations generated in the elevator car 6. Then, when the vibration measuring unit 10 detects that the elevator car 6 starts running based on a signal from the control device 8, the vibration measuring unit 10 ends the measurement. Hereinafter, vibrations measured by the vibration measurement unit 10 will be explained using vertical vibrations of the car 6 as an example.

The vibration analysis section 11 analyzes the vibrations measured by the vibration measurement section 10. The vibration analysis section 11 receives from the vibration measurement section 10 vibration data measured by the vibration measurement section 10 when the elevator car 6 is stopped. The vibration analysis unit 11 performs an analysis including at least one of differentiation of the received vibration data and smoothing processing using a filter, and calculates, for example, a temporal change in acceleration. Thereafter, the vibration analysis unit 11 performs a process of calculating the duration of the vibration based on the calculated change in acceleration over time. When the vibration measurement unit 10 is composed of multiple sensors and devices such as acceleration sensors, the vibration analysis unit 11 uses the average value of each analysis result instead of the time change of acceleration to determine whether the vibration continues. You may also calculate the time required to do so.

FIG. 2 schematically shows the change in acceleration over time based on vibration data measured by the vibration measurement unit 10 using a solid line on a graph. When a passenger gets into the elevator car 6 alone, the waveform of the generated vibration is such that the initial acceleration is large and gradually attenuates.

The vibration analysis unit 11 performs a process of connecting multiple peaks of the vibration waveform in chronological order, and creates a profile as shown by the dotted line in FIG. 2. Each peak of the vibration waveform is, for example, the maximum value of the vibration waveform. Values between peaks in the profile are connected by an interpolation process, such as linear interpolation or spline interpolation. Here, in the vibration analysis section 11, an acceleration threshold A1 is set in advance. Threshold A1 is an example of a first threshold. The threshold value A1 is indicated by a dashed line in FIG. Further, the length of time from the time when the original vibration waveform first exceeds the threshold value A1 to the time when the profile first falls below the threshold value A1 is defined as the vibration duration time T. Note that the starting point of the vibration duration may be the starting point of the profile, that is, the time of the first peak of the vibration waveform.

Further, FIG. 3 schematically shows the absolute value of the temporal change in acceleration based on the vibration data measured by the vibration measurement unit 10 as a solid line on the graph. The vibration analysis unit 11 may create a profile shown by a dotted line in FIG. 3 in which each peak is connected for the absolute value of such a time change in acceleration. The vibration analysis unit 11 may define the vibration duration T as the time length from the time when the original vibration waveform first exceeds the threshold value A1 to the time when the profile first falls below the threshold value A1.

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 ride comfort of the elevator has deteriorated, that is, that there is an abnormality in the elevator. The time threshold Ta is set in the vibration analysis section 11 in advance. The threshold Ta is an example of a second threshold.

FIG. 4 schematically shows an example of a case where the damping characteristic of vibration acceleration is weakened due to an abnormality in the elevator and a case where it is normal, as a time change in acceleration. In FIG. 4, the vibration waveform in a normal case is shown by a solid line. On the other hand, the vibration waveform when the vibration damping characteristics are weakened due to an abnormality in the elevator is shown by a broken line. Further, a profile in which vibration peaks in each case are connected in chronological order with a straight line is shown by a dotted line. The vibration duration T' during which the abnormal vibration profile exceeds the threshold value A1 is longer than the vibration duration T calculated from the normal vibration profile. In order to determine that the abnormal vibration is "abnormal", it is necessary to set the threshold value Ta to a value between time T and time T'.

FIG. 5 shows the relationship between the vibration duration when the elevator is normal, the vibration duration when an abnormality occurs, and the threshold Ta, as a method of setting the threshold Ta. Since the way passengers board the car 6 varies from person to person and from time to time, the vibration duration T also varies due to these influences. Therefore, by setting the threshold value Ta so as to take these variations into account, it is possible to improve the accuracy of one-time diagnosis. For example, vibration data when the elevator is normal, such as immediately after the elevator is installed, is acquired a predetermined number of times, and the maximum value of the vibration duration is set to Ta.

Furthermore, assuming a case where multiple passengers board the vehicle, the vibration duration may be defined as follows. FIG. 6 shows an example of a vibration waveform when a plurality of passengers board the car. In this case, when the vibration duration time T is calculated as described above, T becomes a larger value than when one passenger gets on the vehicle, and there is a possibility of misdiagnosis. Therefore, a plurality of acceleration thresholds are set such as Aa1, Aa2, . . . , Aam-2, Aam-1, Aam. The vibration analysis unit 11 defines the time from the time when the vibration falls below a threshold value that is smaller than and closest to the last local maximum value of the profile to the time when the vibration falls below the minimum threshold value as a vibration duration T''. . In FIG. 6, the threshold value that is smaller than the last maximum value of the profile and closest to this maximum value is Aam-1. Therefore, the vibration analysis unit 11 calculates the time from the time when the profile falls below the threshold value Aam-1 after the last local maximum value to the time when the profile falls below the minimum threshold value Aa1 as the vibration duration T''. By determining whether there is an abnormality in the elevator using the vibration duration time T'' calculated in this way, it becomes possible to diagnose with the same accuracy as when one passenger gets on the elevator.

Also, for example, if the position of the car 6 changes and the length of the suspension body 5 at the part where the car 6 is hung becomes longer, its tensile rigidity decreases, so the amount of stretch of the suspension body 5 when a passenger gets on board increases. Vibrations with larger displacement amplitudes occur. In this case, the vibration acceleration and vibration duration also change from the original position of the car 6. By taking this into consideration and setting a threshold value according to the number of floors of the elevator, the vibration analysis unit 11 can diagnose abnormalities in the elevator regardless of the number of floors where the car 6 is located. For example, for an elevator where the car 6 stops from the 1st floor to the nth floor, a group of threshold values (Aam1, Aam2, ..., Aamn; Ta1, Ta2, ..., Tan) corresponding to each floor is preset in the vibration analysis section 11. Ru. Here, when the integer i is any integer from 1 to n, the acceleration threshold Aami and the time threshold Tai are examples of the first threshold and the second threshold corresponding to the case where the car 6 stops on the i floor. be. When diagnosing an abnormality in an 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 from the threshold group according to the floor number, thereby detecting an abnormality in the elevator at any floor. be able to diagnose.

The recording unit 12 records the determination result by the vibration analysis unit 11. The diagnostic device 9 may transmit the determination result to the outside of the elevator using a network line or the like. By doing so, the determination result can be quickly sent to the maintenance staff who maintains the elevator, for example, and the maintenance staff can take early action.

With the above configuration, it is possible to accurately diagnose the deterioration of the ride comfort when the elevator car 6 is stopped. Furthermore, since diagnosis is performed based on the vibration duration that is affected by an abnormality in the elevator, it is possible to diagnose a phenomenon in which the ride comfort deteriorates due to an abnormality in the elevator itself.

FIG. 7 is a flowchart summarizing the specific process of diagnosing elevator abnormalities. After the control device 8 confirms that the elevator car 6 has stopped in STEPa1, the vibration measuring unit 10 starts measuring vertical vibrations generated in the car 6 in STEPa2. Thereafter, in STEPa3, when the control device 8 confirms that the elevator car 6 has started running, the vibration measuring section 10 finishes the measurement and then sends the measured data to the vibration analyzing section 11. In STEPa4, the vibration analysis unit 11 calculates the vibration duration T. If the vibration duration T exceeds the predetermined vibration duration threshold Ta in STEPa5, the vibration analysis unit 11 diagnoses that there is an abnormality in the elevator in STEPa6. On the other hand, if the measured vibration duration T is the same as the vibration duration threshold Ta or does not exceed the threshold Ta, the vibration analysis unit 11 diagnoses in STEPa7 that there is no abnormality in the elevator. The determination result is recorded in the recording unit 12 in STEPa8, and one diagnosis is completed.

Furthermore, the diagnostic device 9 adds STEPa9 and STEPa10 to the recording unit 12, as shown in the flowchart shown in FIG. Good too. In STEPa9, the vibration analysis unit 11 calculates the frequency X at which it is determined that there is an abnormality from the diagnosis history for the preset number of diagnoses. For this frequency X, a frequency threshold Xa is set in advance. In STEPa10, the diagnostic device 9 reports the situation to the outside at a timing when the frequency X becomes equal to or higher than the threshold value Xa. In this case, by appropriately setting the threshold value Xa, maintenance personnel can take prompt action. Specific examples of the destination include a display or alarm device, or a maintenance terminal owned by a maintenance worker in a remote location.

In addition, the diagnostic device 9 sets a threshold value for the total number of times it is determined to be "abnormal" or the number of consecutive times it is determined that "abnormality exists" instead of the frequency for determining "abnormality present". The situation may be reported externally.

Furthermore, such a diagnostic device 9 can be applied not only to 1:1 roping, but also to all types of roping traction elevators, such as 2:1 roping or machine room-less elevator systems that do not have a machine room 2.

Further, such an elevator diagnostic device 9 can be applied not only to vibrations caused by passengers getting on the car 6 when the car 6 is stopped, but also to vibrations caused by passengers getting off the car 6. In addition, the present invention can be applied not only to vibrations in the vertical direction of the car 6, but also to vibrations in the lateral direction due to the inclination of the car 6. Note that in this embodiment, a process has been described in which vibration data is converted into a temporal change in acceleration in the vibration analysis unit 11, but the same process may be performed on velocity or displacement. That is, the threshold value for the amplitude of vibration is not limited to the threshold value for the amplitude of acceleration, but may be a threshold value for the amplitude of velocity, displacement, or the like.

Next, an example of the hardware configuration of the elevator will be described using FIG. 9.
FIG. 9 is a hardware configuration diagram of the main parts of the elevator according to the first embodiment.

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 circuitry may include at least one dedicated hardware 200 along with or in place of the processor 100a and memory 100b.

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 written as a program. The program is stored in memory 100b. The processor 100a implements each function of the elevator by reading and executing programs stored in the memory 100b.

The processor 100a is also referred to as a CPU (Central Processing Unit), processing device, arithmetic device, microprocessor, microcomputer, or DSP. The memory 100b is configured of a nonvolatile or volatile semiconductor memory such as RAM, ROM, flash memory, EPROM, and EEPROM.

When the processing circuit comprises dedicated hardware 200, the processing circuit is implemented, for example, as a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

Each function of the elevator can be realized by a respective processing circuit. Alternatively, each function of the elevator can be realized collectively by a processing circuit. Regarding each function of the elevator, some may be realized by dedicated hardware 200, and other parts may be realized by software or firmware. In this manner, the processing circuit implements each function of the elevator using dedicated hardware 200, software, firmware, or a combination thereof.

Embodiment 2.
In Embodiment 2, points that are different from the example disclosed in Embodiment 1 will be explained in particular detail. As for the features not described in the second embodiment, any of the features in the examples disclosed in the first embodiment may be adopted.

FIG. 10 is an overall view of the traction type elevator in the second embodiment.

In addition to the configuration of the elevator of Embodiment 1, the elevator of Embodiment 2 includes a monitoring section 13 and a behavior analysis section 14 that monitor passenger behavior. Specifically, the monitoring unit 13 includes a camera 13a that photographs the inside of the car 6, a weighing device 13b, a floor reaction force sensor 13c provided on the car floor, an acceleration sensor, a speed sensor, a displacement sensor, or a combination of a plurality of these. It consists of The monitoring unit 13 acquires passenger behavior data while the car 6 is stopped.

Additionally, the behavior analysis section 14 is connected to the monitoring section 13. After acquiring behavioral data of the passengers when the elevator car 6 is stopped, the monitoring unit 13 transmits the acquired behavioral data to the behavior analyzing unit 14 .

Here, when the monitoring unit 13 is the camera 13a, the monitoring unit 13 transmits the image or video data captured inside the car 6 to the behavior analysis unit 14 as the passenger behavior data. When the monitoring unit 13 is the weighing device 13b, the monitoring unit 13 transmits the change in weight of the car 6 over time to the behavior analysis unit 14 as passenger behavior data. When the monitoring unit 13 is the floor reaction force sensor 13c, the monitoring unit 13 transmits temporal changes in the floor reaction force applied to the car floor to the behavior analysis unit 14 as passenger behavior data.

The behavior analysis unit 14 analyzes the behavior data of the passengers and calculates the number of passengers n, the timing of boarding τ1, τ2, ..., τn, and the weight M1, M2, ..., Mn of the car 6 after each passenger boarded the car. These are calculated as analysis results and transmitted to the vibration analysis section 11. When the passenger's behavior data is an image or video taken by the camera 13a, the behavior analysis unit 14 estimates various quantities of the above analysis results by image analysis. When the passenger's behavior data is a change in the weight of the car 6 measured by the weighing device 13b, the behavior analysis unit 14 estimates various quantities of the analysis results from the time change. When the passenger's behavior data is a floor reaction force, the behavior analysis unit 14 estimates various quantities of the above-mentioned analysis results from temporal changes in the floor reaction force. When the monitoring unit 13 uses devices such as a plurality of types of sensors, the behavior analysis unit 14 may obtain various amounts of the analysis results as an average value of the values calculated by each device.

A plurality of threshold groups are preset in the vibration analysis section 11. The vibration analysis unit 11 receives from the behavior analysis unit 14 the number of passengers n, boarding timings τ1, τ2, ..., τn, the weights M1, M2, ..., Mn of the car 6 after each passenger has boarded, and the control device. According to the combination of the car position information received from 8, the acceleration threshold Aa and the vibration duration threshold Ta are selected from the threshold group and set.

The vibration analysis unit 11 uses the vibration data measured by the vibration measurement unit 10 to calculate, for example, changes in acceleration over time. Further, FIG. 11 schematically shows changes in acceleration due to vibration when a plurality of passengers get into the car 6. In FIG. 11, an example is shown in which three passengers of similar weight board the vehicle in succession. The vibration generated in the car 6 at this time differs from the case where only one passenger gets on the car, but when the second and subsequent passengers get on the car, the acceleration increases again from the middle and becomes a damped waveform.

The vibration analysis unit 11 performs processing to connect multiple peaks of acceleration changes in chronological order for the vibrations caused by each passenger boarding, which are divided by the boarding timings τ1, τ2, and τ3, as shown by dotted lines in FIG. Create a profile that allows you to Values between peaks in the profile are connected by an interpolation process, such as linear interpolation or spline interpolation. The vibration analysis unit 11 calculates the total time during which this profile exceeds a preset acceleration threshold value Aa as the vibration duration T. In FIG. 11, the vibration duration T is the sum of T1, T2, and T3. The vibration analysis unit 11 sets the starting point of the range for calculating the vibration duration of each passenger to the point at which the acceleration change induced by each passenger first exceeds the threshold Aa, the boarding timings τ1, τ2, τ3, or the profile. It can also be the first point. The vibration analysis unit 11 may set the end point of the range for calculating the vibration duration of each passenger to the timing when the profile falls below the threshold value Aa, or the final point of each profile if the profile does not fall below the threshold value Aa. .

Furthermore, the vibration analysis unit 11 may create a profile by performing a process of connecting a plurality of peaks in chronological order for the entire vibration waveform, as shown in FIG. Values between peaks in the profile are connected by an interpolation process, such as linear interpolation or spline interpolation. The vibration analysis unit 11 vibrates the total length of time during which the profile created in this way exceeds the threshold Aa from the point where the acceleration first exceeds the preset acceleration threshold Aa to the last point where the acceleration falls below the threshold Aa. It may also be calculated as the duration T. In FIG. 12, the vibration duration T is the sum of T4 and T5.

Furthermore, the vibration analysis unit 11 may create a profile by connecting the peak of the waveform to the absolute value of the change in acceleration. The vibration analysis unit 11 may calculate the vibration duration T by performing similar processing on the profile created in this way.

Here, when a heavy passenger or heavy cargo gets into 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 takes into account the weight Mn of the car 6 after n people board, which is calculated by the behavior analysis unit 14. For example, when a heavy passenger gets on board, the acceleration increases as the applied load increases, so the threshold value Aa is also set to a value that increases according to the degree of increase in the load.

Further, the threshold value Ta of the vibration duration is selected from preset threshold values according to the number of passengers n and the boarding timings τ1, τ2, ..., τn. For example, as the number of passengers increases, the vibration duration T of the vibrations generated in the car n becomes longer, and therefore the threshold value Ta is also set to a larger value. On the other hand, if the intervals between the boarding timings τ1, τ2, . For this reason, the vibration analysis unit 11 sets a threshold value τa for the time interval of boarding, and sets a vibration duration threshold value Ta according to the length of the interval of τ1, τ2, ..., τn, or sets a vibration duration threshold value Ta according to the length of the interval of τ1, τ2, ..., τn. , τn may be excluded from the diagnostic target depending on the length of the interval.

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

Furthermore, by extracting only the vibration data from the last passenger who boarded the vehicle, in FIG. 11, only the vibration data after time τ3, vibration data corresponding to the time when one passenger boarded the vehicle can be obtained. The vibration analysis unit 11 may diagnose the elevator using the same process as in the first embodiment based on such vibration data.

In this way, even when multiple people board the car, the number of boarders n calculated by the behavior analysis unit 14, the timings τ1, τ2, ..., τn of each boarding, and the weights M1, M2, ..., Mn of the car 6 after each boarding. , and the car position information transmitted from the control device 8, the vibration analysis unit 11 selects an appropriate threshold value Ta from a preset threshold value group, thereby making it possible to appropriately diagnose the elevator. can.

In addition, the elevator may perform a dedicated operation, such as a robot-only operation mode, for carrying objects such as robots whose weight is known in advance. In this case, since the weight and size of the transported object are clear, variations in the magnitude of the generated vibrations are unlikely to occur. Therefore, by providing a threshold value Aa, a threshold value Ta, etc. exclusively for the robot, diagnosis can be performed with higher accuracy than when the passenger is a human passenger. The robot-specific threshold is set by, for example, reducing the variation range in FIG. 5.

Further, when a plurality of passengers or cargo having significantly different weights or weights get into the car 6, the vibration analysis section 11 may set an appropriate acceleration threshold value Aa for each passenger getting on. . The vibration analysis unit 11 can perform an appropriate diagnosis by selecting an appropriate vibration duration threshold value Ta for the combination.

In this way, even when multiple passengers, heavy passengers, or heavy cargo board the elevator, the vibration analysis unit 11 can appropriately diagnose the elevator.

In addition, if a passenger continues to violently move inside the elevator car 6 while the elevator car 6 is stopped, the vibrations will continue with a large acceleration, and the duration of the vibrations will also become extremely long, which may lead to a misdiagnosis. Therefore, when the behavior analysis unit 14 detects a violent behavior of a passenger by analyzing the passenger behavior data, the vibration analysis unit 11 can prevent misdiagnosis by excluding vibrations caused by the violent behavior of the passenger from the diagnosis target.

FIG. 13 is an example of a flowchart for diagnosing an abnormality in an elevator using such a configuration and operation. After the control device 8 confirms that the elevator car 6 has stopped in STEP b1, the vibration measuring unit 10 and the monitoring unit 13 start acquiring vibration data and passenger behavior data in STEP b2. Thereafter, in STEP b3, when the control device 8 confirms that the elevator car 6 has started running, the vibration measuring section 10 and the monitoring section 13 end data acquisition. In STEP b4, the behavior analysis unit 14 analyzes the behavior of the passengers and checks whether there are problems such as unruly passengers or whether passengers are getting on or off the train. If there is no problem, the behavior analysis unit 14 continues to analyze the behavior data of the passengers, and calculates the number of passengers n, the timings of boarding τ1, τ2, ..., τn, and the weights M1, M2, and M2 of the cars 6 after each boarding. ..., Mn is calculated. In STEPb5, the vibration analysis unit 11 receives from the behavior analysis unit 14 the number of boarders n, the boarding timings τ1, τ2, ..., τn, and the weights M1, M2, ..., Mn of the car 6 after each boarding. 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 and calculates a profile for acceleration changes. The vibration analysis unit 11 calculates the vibration duration T from the calculated profile. In STEPb7 to STEPb9, the vibration analysis unit 11 determines whether the vibration duration T exceeds the threshold value Ta and diagnoses whether there is an abnormality in the elevator. In STEPb10, the recording unit 12 records the diagnosis result.

The diagnostic device 9 may issue an alarm to the outside based not only on the one-time diagnosis up to STEPb10, but also on a plurality of diagnosis histories, such as from STEPb11 to STEPb13. That is, the vibration analysis unit 11 calculates the frequency X at which it is determined that there is an abnormality from the diagnosis history of N times, which is the number of times of diagnosis set in advance. For this frequency X, a frequency threshold Xa is set in advance. The diagnostic device 9 reports the situation to the outside at a timing when the frequency X exceeds the threshold value Xa. Specific examples of the destination include a display or alarm device, or a maintenance terminal owned by a maintenance worker in a remote location.

With the above-described configuration, the diagnostic device 9 can assign an appropriate threshold value to the vibration by comparing the vibration data with the passenger behavior data and analyzing it, so that it can accurately diagnose abnormalities in the elevator. .

Embodiment 3.
In Embodiment 3, points that are different from the examples disclosed in Embodiment 1 or Embodiment 2 will be explained in particular detail. For features not described in Embodiment 3, any of the features disclosed in Embodiment 1 or Embodiment 2 may be adopted.

FIG. 14 is an overall view of the traction type elevator in Embodiment 3.

In the traction type elevator shown in FIG. 14, a machine room 2 is provided above the hoistway 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 suspended over the pulley 3. For the suspension body 5, for example, a plurality of ropes or a plurality of belts are used. In this example, a cage 6 and a weight 7 are connected to both ends of the suspension body 5, respectively. That is, FIG. 14 illustrates a case where the suspension body 5 is suspended using a 1:1 roping system. A suspension body 5 suspends a car 6 and a weight 7 in the hoistway 1.

The diagnostic device 9 includes a vibration analysis section 11, a monitoring section 13, a behavior analysis section 14, a learning device 15, an inference device 16, and a learned model storage section 17.

The car 6 is provided with a vibration measurement section 10. The vibration measurement unit 10 is configured by, for example, an acceleration sensor, a speed sensor, a displacement sensor, or a weighing device attached to the car 6, or a combination of a plurality of these.

The vibration analysis unit 11 analyzes the vibration data sent from the vibration measurement unit 10, calculates, for example, an acceleration change, and calculates the vibration duration T based on a preset acceleration threshold value Aa. The method for calculating the vibration duration T based on the acceleration threshold value Aa is the same as in the first embodiment or the second embodiment.

The monitoring unit 13 acquires behavioral data of passengers while the elevator car 6 is stopped. Specific examples of the monitoring unit 13 include a camera 13a that photographs the inside of the car 6, a weighing device 13b, a floor reaction force sensor 13c provided on the car floor, an acceleration sensor, a speed sensor, a displacement sensor, or any one of these. This includes multiple combinations. The monitoring unit 13 acquires passenger behavior data while the car 6 is stopped.

A learning device 15 is connected to the vibration analysis section 11 and the behavior analysis section 14. The learning device 15 includes a data acquisition section 18 and a model generation section 19. Further, an inference device 16 is connected to the vibration analysis section 11 and the behavior analysis section 14. The inference device 16 includes a data acquisition section 20 and an inference section 21.

The data acquisition unit 18 in the learning device 15 and the data acquisition unit 20 in the reasoning device 16 receive the car position information L from the control device 8, the vibration duration T from the vibration analysis unit 11, the number of boarders n, and the boarding timing τ1. , τ2, . . . , τn and the weights M1, M2, .

The model generation unit 19 generates the car position information L output from the data acquisition unit 18, the vibration duration T, the number of boarders n, the boarding timings τ1, τ2, ..., τn, and the weights M1, M2 of the car 6 at each boarding. , ..., Mn, the state of the elevator is learned based on the learning data created according to the combination of ,...,Mn. That is, the model generation unit 19 generates the following information from the car position information L, the vibration duration T, the number of boarders n, the boarding timings τ1, τ2, ..., τn, and the weights M1, M2, ..., Mn of the car 6 at each boarding time. Generate a trained model that infers the state of the elevator. In this example, the learning data includes car position information L, vibration duration T, number of boarders n, boarding timings τ1, τ2, ..., τn, and weights M1, M2, ..., Mn of the car 6 at each boarding. This is data that is associated with each other.

The learning algorithm used by the model generation unit 19 can be a known algorithm such as supervised learning, unsupervised learning, or reinforcement learning. As an example, a case will be described in which clustering such as k-means method, which is unsupervised learning, is applied. Unsupervised learning refers to a method of learning features in the learning data by giving learning data that does not include results or labels to the learning device 15.

The model generation unit 19 learns the state of the elevator by so-called unsupervised learning, for example, according to a grouping method using the k-means method. The k-means method is a non-hierarchical clustering algorithm, and is a method of classifying a given number of clusters into k groups using the average of the clusters.

More specifically, the k-means method is processed as follows. First, clusters are randomly assigned to each data xi. Next, the center Vj of each cluster is calculated based on the allocated data. The distance between each xi and each Vj is then determined and xi is reassigned to the nearest central cluster. Then, if the cluster assignments of all xi do not change in the above process, or if the amount of change falls below a certain threshold set in advance, it is determined that convergence has been achieved and the process ends.

In the present application, the model generation unit 19 generates the car position information L acquired by the data acquisition unit 18, the vibration duration T, the number of boarders n, the boarding timings τ1, τ2, ..., τn, and the car 6 at each boarding. The state of the elevator is learned by so-called unsupervised learning according to learning data created based on the combinations of weights M1, M2, . . . , Mn.

The model generation unit 19 generates and outputs a trained model by performing the learning described above. The trained model storage unit 17 stores the trained model output from the model generation unit 19. By performing such learning on a normal elevator, the car position information L, the vibration duration T, the number of boarders n, the boarding timings τ1, τ2,..., τn, the weight M1 of the car 6 at each boarding, A learned model that has learned the relationship between M2,...,Mn and the normal state of the elevator is obtained.

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 reasoning unit 21 uses the car position information L acquired by the data acquisition unit 20, the vibration duration T, the number of boarders n, the boarding timings τ1, τ2, ..., τn, and the weights M1, M2 of the car 6 at each boarding. , ..., Mn is input into this learned model, it is possible to infer which cluster the data belongs to, and output the inference result as the state of the elevator. If the data input to the learned model does not belong to any of the clusters indicating the normal state of the elevator, the inference unit 21 determines that an abnormality has occurred in the elevator.

The recording unit 12 records the determination result by the inference unit 21. The diagnostic device 9 may transmit or report the determination result recorded in the recording unit 12 to the outside of the elevator using a network line or the like. By doing so, the determination result can be quickly communicated or sent to, for example, maintenance personnel, allowing the maintenance personnel to take early action.

In the present embodiment, the inference unit 21 has been described as outputting the state of the elevator using the learned model learned by the model generation unit 19, but the inference unit 21 outputs the state of the elevator based on the learned model acquired from the outside. The status of the elevator may also be output.

Furthermore, although a case has been described in which the data acquisition unit 18 and the data acquisition unit 20 are provided in each of the learning device 15 and the inference device 16, the configuration of the diagnostic device 9 is not limited to this case. The diagnostic device 9 uses one data acquisition section to acquire necessary information from the control device 8, vibration analysis section 11, and behavior analysis section 14, and transmits the information to the model generation section 19 of the learning device 15 and the inference section 21 of the inference device 16. The configuration may be such that the acquired information is output.

In this way, the inference unit 21 calculates the car position information L, the vibration duration T, the number of boarders n, the boarding timings τ1, τ2, ..., τn, the weights of the car 6 at each boarding time M1, M2, ..., Mn The state of the elevator is determined based on.

Next, a processing flow for diagnosing the state of the elevator using the learning device 15 and the reasoning device 16 will be described using FIG. 15. Elevator diagnosis in the third embodiment consists of two phases: a learning phase and a diagnosis phase.

In the learning phase, in STEPc1, in order to obtain learning data, the data acquisition section 18 receives a signal indicating the stop of the elevator car 6 from the control device 8, and then the vibration measurement section 10 and the monitoring section 13 acquire the data. measure.

In STEPc2, after the data acquisition unit 18 receives a signal indicating the start of travel of the elevator car 6 from the control device 8, the vibration analysis unit 11 and the behavior analysis unit 14 collect the data measured by the vibration measurement unit 10 and the monitoring unit 13. receive. The vibration analysis unit 11 and the behavior analysis unit 14 analyze the received data and determine the car position information L, the vibration duration T, the number of boarders n, the boarding timings τ1, τ2, ..., τn, and the car 6 for each boarding. The weights M1, M2,..., Mn of are calculated.

In STEPc3, the data acquisition unit 18 acquires the output car position information L, vibration duration T, number of boarders n, boarding timings τ1, τ2, ..., τn, weights of the car 6 at each boarding time M1, M2, ... , Mn.

In STEPc4, the model generation unit 19 generates the car position information L acquired by the data acquisition unit 18, the vibration duration T, the number of boarders n, the timings of boarding τ1, τ2, ..., τn, and the weight of the car 6 at each boarding. According to the learning data created based on the combinations of M1, M2, . . . , Mn, the normal state of the elevator is learned by so-called unsupervised learning, and a learned model is generated.

In STEPc5, the learned model storage section 17 stores the learned model generated by the model generation section 19. This kind of learning is performed every time the elevator stops at a stop floor, and the learning continues until a predetermined number of times is reached.

In the diagnosis phase, the vibration measurement section 10 and the monitoring section 13 measure data in STEPc6 in order to diagnose the state of the elevator using the inference device 16.

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, the vibration duration T, the number of people boarding n, the boarding timing τ1, τ2,...,τn, and the weights M1, M2,..., Mn of the car 6 at each boarding are calculated.

In STEPc8, the data acquisition unit 20 obtains the outputted car position information L, vibration duration T, number of boarders n, boarding timings τ1, τ2, ..., τn, weights of the car 6 at each boarding time M1, M2, ... , Mn.

In STEPc9, the inference unit 21 uses the car position information L acquired by the data acquisition unit 20, the vibration duration T, the number of boarders n, the boarding timings τ1, τ2, ..., τn, the weight M1 of the car 6 at each boarding, The data of M2, . . . , Mn are input to the trained model stored in the trained model storage unit 17.

In STEPc10, the inference unit 21 outputs the state of the elevator obtained from the learned model.

In STEPc11, the inference unit 21 diagnoses the condition of the elevator based on the output result of the condition of the elevator.

Note that, like the diagnostic device 9 of

Embodiment

1 or 2, the inference device 16 may issue an alarm based on the diagnosis history for a preset number of diagnoses, rather than just a single diagnosis. The inference device 16 calculates, for example, the frequency X of determining that there is an "abnormality". For this frequency X, a frequency threshold Xa is set in advance. The inference device 16 may notify the outside of the abnormality of the elevator at a timing when the frequency X exceeds the threshold value Xa. Specific examples of the destination include a display or alarm device, or a maintenance terminal owned by a maintenance worker in a remote location.

In addition, the inference device 16 sets a threshold value for the total number of times it is determined to be "abnormal" or the number of consecutive times it is determined that "abnormality exists" instead of the frequency at which it is determined that "there is an abnormality". The situation may be reported externally.

With the above configuration, the condition of the elevator can be diagnosed with respect to any vibrations in the vertical direction of the elevator when the car 6 is stopped.

Note that in this embodiment, a case has been described in which unsupervised learning is applied to the learning algorithm used by the model generation unit 19 and the inference unit 21, but the learning algorithm is not limited to this. As for learning algorithms, in addition to unsupervised learning, reinforcement learning, supervised learning, semi-supervised learning, etc. can also be applied.

Further, as a learning algorithm used in the model generation unit 19, deep learning that learns to extract the feature amount itself can be used, and other known methods can also be used.

When realizing unsupervised learning in this embodiment, not only non-hierarchical clustering using the k-means method as described above but also other known methods capable of clustering may be used. For example, hierarchical clustering such as the shortest distance method may be used.

In the present embodiment, a case has been described in which the learning device 15 and the reasoning device 16 are provided in the diagnostic device 9. However, some or all of the learning device 15 and the reasoning device 16 may be connected to It may be provided in a separate device from the elevator, which is connected to the elevator via the elevator. Furthermore, 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 implemented on a cloud server.

Furthermore, the model generation unit 19 may be configured to learn the state of the elevator according to learning data created for a plurality of elevators.

The model generation unit 19 may acquire learning data from multiple elevators used in the same area, or may utilize learning data collected from multiple elevators operating independently in different areas. The status of the elevator may also be learned. Furthermore, it is also possible to add the elevator to the objects for which learning data is to be collected, or to remove it from the objects in the middle. Furthermore, the learning device 15 that has learned the elevator status regarding a certain elevator may be applied to another elevator to relearn and update the elevator status regarding the other elevator.

Further, such an elevator diagnostic device 9 can be applied not only to 1:1 roping but also to all types of roping type traction elevators, such as 2:1 roping or machine room-less elevator systems that do not have a machine room 2.

The elevator according to the present disclosure can be applied to buildings having multiple floors.

1 Hoistway, 2 Machine room, 3 Pulley, 4 Hoisting machine, 5 Suspension body, 6 Car, 7 Weight, 8 Control device, 9 Diagnosis device, 10 Vibration measurement section, 11 Vibration analysis section, 12 Recording Department, 13 Monitoring Section, 13a Camera, 13b Weighing device, 13c Floor reaction force sensor, 14 Behavior analysis section, 15 Learning device, 16 Inference device, 17 Learned model storage section, 18 Data acquisition section, 19 Model generation section, 20 data acquisition section, 21 Inference unit, 100a processor, 100b memory, 200 dedicated hardware

Claims

a vibration measurement unit that measures vibrations generated in the elevator car when the elevator car stops;
a vibration analysis unit that determines whether or not there is an abnormality in the elevator using the vibration measured by the vibration measurement unit, a preset first amplitude threshold, and a preset second vibration duration threshold; ,
Equipped with an elevator.
The vibration analysis unit determines that there is an abnormality in the elevator when the time period during which the vibration amplitude measured by the vibration measurement unit continues to vibrate at a magnitude exceeding the first threshold exceeds the second threshold.
The elevator according to claim 1.
a monitoring unit that acquires passenger behavior data when the elevator car is stopped;
a behavior analysis unit that analyzes passenger behavior data acquired by the monitoring unit;
Equipped with
The vibration analysis unit selects the first threshold and the second threshold from a group of thresholds set in advance according to the behavior data analyzed by the behavior analysis unit, and selects the first threshold and the second threshold. Determine whether there is an abnormality in the elevator using a threshold value,
The elevator according to claim 1 or claim 2.
When the elevator is operating exclusively for the robot, the vibration analysis unit determines whether or not there is an abnormality in the elevator using values set exclusively for the robot as the first threshold value and the second threshold value. ,
The elevator according to any one of claims 1 to 3.
a vibration measurement unit that measures vibrations generated in the elevator car when the elevator car stops;
a monitoring unit that acquires passenger behavior data when the elevator car is stopped;
a vibration analysis unit that calculates, as a vibration duration time, a time during which the vibration measured by the vibration measurement unit continues to vibrate at a magnitude exceeding a preset first vibration amplitude threshold;
a behavior analysis unit that analyzes passenger behavior data acquired by the monitoring unit;
a learning device that generates a trained model that infers the state of the elevator using the vibration duration calculated by the vibration analysis unit and passenger behavior data analyzed by the behavior analysis unit;
an inference device that diagnoses the state of the elevator using the vibration duration calculated by the vibration analysis unit, passenger behavior data analyzed by the behavior analysis unit, and the learned model generated by the learning device;
Equipped with an elevator.