JP2022025828A - Elevator abnormal state detection device and elevator abnormal state detection method - Google Patents

Abstract

To identify the location of abnormal sounds, the component that generates the abnormal sounds, and the cause of the occurrence from sound data that includes the abnormal sounds among sounds generated by an elevator.SOLUTION: First, a traveling position of an elevator at the time when sound data including abnormality is generated and a generation position of the sound data are calculated by acquiring the sound data and traveling position data at the time when the elevator is traveling by a sound sensor and a position sensor. Then, information on material of parts of the elevator at the traveling position and the generating position at the time when the sound data including the abnormal sound is generated is extracted so as to calculate the parts with the material that emit this sound data including the abnormal sound through learning the sound data including the abnormal sound. The material information of the parts of the elevator is compared with the material that emits the sound data including the abnormal sound, and the cause of the sound data including the abnormal sound is identified based on the result of the comparison.SELECTED DRAWING: Figure 1

Description

The present invention relates to an elevator abnormality state detection method and an elevator abnormality state detection device.

Generally, in the elevator abnormality diagnosis, the equipment or parts where the elevator abnormality occurs are detected by comparing the learning data and the diagnosis data.
Patent Document 1 describes a diagnostic system and an elevator that can follow aged deterioration by updating learning data so that the influence of outliers can be suppressed by using sensor data diagnosed as normal.

That is, the technique described in Patent Document 1 diagnoses the state of the device to be diagnosed based on the sensor data from the sensor installed in the device to be diagnosed and the learning data for diagnosis. Then, the sensor data at the time of being diagnosed as normal is added to the stored learning data group held in advance, only the center of the stored data group and the data close to it are left, and the distant data is deleted from the stored data group and stored. The data group is being updated.

Japanese Unexamined Patent Publication No. 2015-35118

However, although the technique described in Patent Document 1 has a certain effect in updating learning data due to deterioration over time, there is a problem that a sudden failure, for example, a location where an abnormal sound is generated and a device that generates the abnormal sound cannot be specified. rice field.

In view of the above problems, an object of the present invention is to identify a location where an abnormal sound is generated, a component where the abnormal sound is generated, and a cause of the abnormal sound from sound data including an abnormal sound generated by an elevator due to aging or the like.

In order to solve the above problems, for example, the configuration described in the claims is adopted.
The present application includes a plurality of means for solving the above problems, and to give an example thereof, the present invention is an elevator abnormal state detection method for detecting an abnormal state of an elevator, and the following (a) to (g). ) Includes steps.
(A) Steps to acquire sound data during traveling of an elevator by a sound sensor,
(B) Steps to acquire the traveling position data of the elevator by the position sensor,
(C) A step of calculating the position of the sound source that generates the sound data including the abnormal sound from the sound data including the abnormal sound among the sounds acquired by the sound sensor and the traveling position of the elevator acquired by the position sensor.
(D) A step of extracting material information of elevator parts from the traveling position of the elevator when sound data including abnormal sound is generated and the position of the sound source generating sound data including abnormal sound.
(E) A step of learning sound data including abnormal sounds and calculating a component having a material that emits sound data including abnormal sounds based on the learning data.
(F) A step of comparing the material information of the parts of the elevator with the material that emits the sound data including the calculated abnormal sound.
(G) A step of identifying the cause of generation of sound data including abnormal sound based on the result of comparison.

According to the elevator abnormality state detection method and the elevator abnormality detection device of the present invention, the material of the component at a predetermined position of the elevator is compared with the material information acquired from the sound data, and the cause of the sound data including the abnormal sound is determined. Since it can be specified, it is possible to present the operator with a coping method for the abnormal sound.
Issues, configurations and effects other than those described above will be clarified by the following description of the embodiments.

It is a schematic diagram for demonstrating the whole structure of the elevator abnormality state detection system which concerns on one Embodiment of this invention. It is a figure explaining the hardware configuration of the elevator abnormality state detection system which concerns on one Embodiment of this invention. It is a flowchart which shows the processing procedure of the elevator abnormality state detection system which concerns on one Embodiment of this invention. It is a flowchart explaining the processing method of the position data and sound data at the time of occurrence of an abnormal sound in the elevator abnormal state detection system which concerns on one Embodiment of this invention. It is a figure explaining the method of synchronizing the position information and the sound data of the elevator abnormality state detection system which concerns on one Embodiment of this invention. It is a schematic diagram explaining the analysis method of the sound pattern which concerns on one Embodiment of this invention. It is a figure which shows the display example of the failure report which concerns on one Embodiment of this invention. It is a table for demonstrating the position information and material information of a component which concerns on one Embodiment of this invention. It is a table for demonstrating the feature group of the sound data which concerns on one Embodiment of this invention. It is a figure which shows the whole structure of the elevator abnormality state detection system which concerns on the 2nd Embodiment of this invention.

<Example of the first embodiment>
Hereinafter, specific examples of the elevator abnormality state detection method and the elevator abnormality state detection device according to the embodiment of the present invention will be described with reference to FIGS. 1 to 7.
FIG. 1 shows the overall configuration of the elevator abnormality state detection device of the present embodiment.
The elevator abnormality state detection device of the present embodiment is composed of a measuring device 1 which is a main body, a history database (hereinafter referred to as "history DB") 2, a position sensor 3, and an information processing device 4.

The measuring device 1 includes a control unit 11, a sound sensor 12, a position information acquisition unit 13, a learning database (hereinafter referred to as “learning DB”) 14, a processing unit 15, a comparison unit 16, and a parts database (hereinafter, “parts”). It includes a 17), a display data generation unit 18, a communication unit 19, and a display unit 20.

In the present embodiment, the control unit 11 controls the sound sensor 12. The sound sensor 12 has at least two or more sound sensors. For example, the sound sensor 12 is a microphone capable of measuring at least two or more sounds of 20 Hz to 15000 Hz. The reason why at least two sound sensors 12 are provided is to detect the direction in which abnormal sound is generated due to the difference in sound pressure detected by the two sound sensors 12, as will be described later in FIG.
Here, the abnormal sound is not only a sound different from the normal sound due to the failure of the elevator, but also a sound for which it is necessary to confirm that the sound is not due to the failure and a sound for which maintenance of the elevator needs to be performed. The sound of.

The control unit 11 is connected to the sound sensor 12 by a cable or wirelessly, and controls the start or stop of recording of the sound measured by the microphone. Further, the control unit 11 transmits a control signal for sending the recorded sound data to the processing unit 15 to the sound sensor 12.

The position information acquisition unit 13 acquires data from the position sensor 3. Here, the position sensor 3 is a sensor for acquiring position information including, for example, an acceleration sensor, a pressure pressure sensor, an ultrasonic sensor, a millimeter wave sensor, and the like. For example, the barometric pressure sensor can tell the height of the elevator, and the accelerometer can detect the moving speed of the elevator.
Instead of the position sensor 3, an elevator circuit (control circuit) provided on the switchboard of the elevator can be used to detect the position of the elevator and acquire the running state of the elevator.

The information processing device 4 measures information about the equipment or parts constituting the elevator in advance and supplies it to the processing unit 15. The above-mentioned information from the information processing apparatus 4 is also stored in the learning DB 14.

The learning DB 14 is a database in which data related to elevator equipment or parts are stored in advance by machine learning. Using the data stored in the learning DB 14, the abnormal sound generated in the elevator and the normal sound are discriminated, and the characteristic pattern of the sound is discriminated. That is, in the learning DB 14, teacher data for determining whether an abnormal sound or a normal sound and determining a characteristic pattern of a sound are stored in advance by machine learning.

The processing unit 15 processes the data input from the sound sensor 12, the position information acquisition unit 13, and the learning DB 14. That is, the processing unit 15 performs at least the following processes (1) to (5).
(1) High-speed Fourier conversion processing for sound data from sound sensor 12 (2) Frequency analysis processing for spectralogram of sound data from sound sensor 12 (3) Sound from sound sensor 12 by machine learning using data from learning DB14 Processing to determine whether the data is normal or abnormal (3) Processing to determine the characteristic pattern of the sound of the sound data from the sound sensor 12 by machine learning using the data of the learning DB 14 (4) Acquisition by the position information acquisition unit 13. Processing to calculate position information from the data of the position sensor 3 (5) Processing to synchronize the time axis of the sound data from the sound sensor 12 and the position information data from the position sensor 3.

The comparison unit 16 compares the position information and the material information of each component of the elevator stored in the component DB 17 with the detailed information of the sound feature pattern obtained by the processing unit 15. Then, the processing unit 15 calculates the abnormal state of the parts of the elevator and the cause thereof based on the comparison result.

The display data generation unit 18 generates a failure report (described later in FIG. 7) as a countermeasure for resolving the abnormal state based on the cause of the occurrence of the abnormal state of the elevator parts calculated by the comparison unit 16. The failure report including the countermeasures created by the display data generation unit 18 is displayed on the display unit 20.

The communication unit 19 transmits the failure report created by the display data generation unit 18 and displayed on the display unit 20 to the history DB 2. Further, the communication unit 19 transmits various data from the sound sensor 12, the information processing device 4, and the position sensor 3 acquired by the processing unit 15 to the history DB 2. Therefore, the history DB 2 stores the failure report and various data related to the elevator transmitted from the communication unit 19.

<Hardware configuration of the measuring device used in the elevator abnormality state detection system>
FIG. 2 is a diagram showing a hardware configuration of the measuring device 1 used in the elevator abnormality state detection system of this example.
As shown in FIG. 2, the measuring device 1 is composed of a CPU (Central Processing Unit) 22 connected to the bus 21, a ROM (Read Only Memory) 23, a RAM (Random Access Memory) 24, and a non-volatile storage 25. ..

The CPU 22 reads the program code of the software that realizes the function of the measuring device 1 from the ROM 23 and executes the arithmetic processing. Variables and the like generated in the middle of the arithmetic processing performed by the CPU 22 are temporarily written in the RAM 24. The CPU 22 realizes the function of the measuring device 1 described with reference to FIG. 1 by executing the program code recorded in the ROM 23.

The non-volatile storage 25 is composed of a non-volatile memory such as an SSD (Solid State Drive). The non-volatile storage 25 stores programs, data, and the like necessary for the CPU 22 to operate. The non-volatile storage 25 is provided with storage areas such as the learning DB 14, the component DB 17, and the history DB 2 described with reference to FIG.

Further, the measuring device 1 includes an input unit 27 and an output unit 28. The input unit 27 inputs sound data from the sound sensor 12 and position information data from the position sensor 3. The output unit 28 outputs a failure report created based on the cause of the occurrence of the abnormal state of the elevator and the countermeasure. As described above, this failure report is generated by the display data generation unit 18 shown in FIG. 1 and displayed on the display unit 20. Therefore, the output unit 28 in FIG. 2 may be considered to correspond to the display unit 20 in FIG.

Further, the measuring device 1 includes a communication interface (communication unit IF) 26. As the communication interface 26, for example, a NIC (Network Interface Card) or the like is used. The communication interface 26 is hardware corresponding to the communication unit 19 of FIG.

<Processing procedure of the elevator abnormal sound detection method of the first embodiment>
Hereinafter, a processing flow relating to the elevator abnormal sound detection method according to the embodiment of the present invention will be described with reference to FIGS. 3 to 9. FIG. 3 is a processing flow relating to the elevator abnormal sound detection method according to the embodiment of the present invention. It is a flowchart for demonstrating.
First, the information processing apparatus 4 inputs information such as equipment and parts related to the target elevator to the processing unit 15 as a preparation work for measurement (step S1).

Next, the measuring device 1 installed in the car of the elevator acquires the position information data of the elevator and the sound data generated from each part of the elevator, and measures the running state of the elevator (step S2).
The processing unit 15 of the measuring device 1 collates the sound data obtained during the measurement of the traveling state in step S2 with the data stored in the learning DB 14, and checks whether there is an abnormal sound in the sound data from the elevator. Whether or not it is determined (step S3).

When the processing unit 15 determines in step S3 that there is an abnormal sound (YES in step S3), the processing unit 15 obtains the position information data from the position sensor 3 and the sound from the sound sensor 12 acquired in step S2. Using the data, the traveling position when the abnormal sound is generated and the position where the abnormal sound is generated as seen from the measuring device 1 are calculated (step S4).

Here, the traveling position when the abnormal noise is generated is the position of the elevator car when the abnormal noise is generated. Then, the processing unit 15 outputs the traveling position when the abnormal sound is generated as the position information D1. The method of calculating the position where the abnormal sound is generated will be described later in FIGS. 4 (a) and 5.

Next, the processing unit 15 of the measuring device 1 outputs a spectrogram using the sound data including the abnormal sound, collates this spectrogram with the spectrogram stored in the learning DB 14, and determines the waveform pattern (step S5). ). Then, the processing unit 15 outputs the material information D2 of the source corresponding to the waveform pattern. The method for determining the waveform pattern in step S5 will be described later with reference to FIGS. 4 (b) and 5.

Next, the position information D1'and the material information D2'of the parts of the elevator, which are stored in advance, are extracted from the parts DB 17 (step S6).
Then, the comparison unit 16 of the measuring device 1 compares the position information D1'of the elevator parts extracted in step S6 with the position information D1 calculated in step S4, so that the parts that can be matched with the position information D1. Identify the group (step S7).

Further, the comparison unit 16 compares the material information D2'of the elevator component extracted in step S6 with the material information D2 obtained from the determination of the waveform pattern in step S5, and identifies the causative component of the abnormal sound generation ( Step S8).

That is, the comparison unit 16 identifies the causative component identified in step S8 from the possible component group identified in step S7, and presents the failure estimation result and the failure correction method (step S9). Then, the display data generation unit 18 automatically generates a failure report (described later in FIG. 7) based on the failure estimation result presented by the comparison unit 16 and the failure correction method, and displays the failure report on the display unit 20. , Saved in the history DB 2 via the communication unit 19 (step S10).

If the processing unit 15 determines in step S3 that there is no abnormal sound (NO in step S3), the processing unit 15 determines that it is normal (step S11), and ends the processing.
This completes a series of processing procedures of the elevator abnormality state detection method of the present embodiment.

FIG. 4A is a flowchart showing a procedure for calculating the generation position in step S4 of FIG.
First, the processing unit 15 of the measuring device 1 reads the position information data (data from the position sensor 3) and the sound data (data from the sound sensor 12) of the elevator acquired in step S2 of FIG. Synchronize the axes (step S41).
Next, the processing unit 15 extracts the sound data at the time of occurrence of the abnormal sound and the position data at that time from the time axis (step S42).

Then, the processing unit 15 calculates the position of the elevator car from the position data at the time when the abnormal sound is generated (step S43). Here, when calculating the position of the car, for example, hierarchical information, speed, relative distance from the start position, and the like are used.

Further, the processing unit 15 calculates the positional relationship between the position of the sound sensor 12 and the sound source from the waveform information of the sound data recorded by the plurality of sound sensors 12 (step S44).
Through the above processes of steps S41 to S44, the position information D1 of the abnormal sound generation in step S4 of FIG. 3 is calculated.

FIG. 4B is a flowchart showing a procedure for calculating a waveform pattern of abnormal sound generation in step S5 of FIG.
In step S5 of FIG. 3, the waveform pattern of the extracted abnormal sound is determined. In FIG. 4B, first, the processing unit 15 performs a fast Fourier transform of the sound data including the extracted abnormal sound. , Create a spectrogram of sound data including abnormal sounds (step S51).

Then, based on the spectrogram created in step S51, the processing unit 15 calculates the sound pressure characteristics of each frequency range, the geometric characteristics (waveform characteristics) of the abnormal sound wave shape on the spectrogram, and the generation cycle characteristics of the abnormal sound. , Converted to the coordinate information of the learning DB 14 space (step S52). The position on the coordinate axis of the data A shown in FIG. 6 to be described later corresponds to the coordinate information converted in step S52.

Next, the processing unit 15 compares the coordinate information converted in step S52 with the data group stored in advance in the learning DB 14 shown in FIG. 1 (see the data groups G1, G2, and G3 in FIG. 6), and corresponds to the data group. The material information corresponding to the data group and the contact state are calculated (step S53).
Through the above processes of steps S51 to S53, the material information D2 for generating the abnormal sound in step S5 of FIG. 3 is calculated.
The calculation method of the material information D2 for generating the abnormal sound in step S53 will be described later with reference to FIG.

<Method of specifying the position and direction of abnormal sound generation>
FIG. 5 is a diagram showing the time axis of the position data of the elevator car obtained from the position sensor 3 and the sound data obtained from the sound sensor 12 together.
The upper part of FIG. 5 shows the height at which the elevator car exists and the moving speed of the elevator as position data.

As shown in the upper part of FIG. 5, the height of the elevator car is shown from the first floor to the top floor, and the moving speed is shown from the start of operation on the first floor to the stop of operation on the top floor. There is.
The height of the elevator car can be measured by using a pressure sensor as the position sensor 3, and the moving speed of the elevator can be measured by using an acceleration sensor as the position sensor 3.

Further, as shown in the lower part of FIG. 5, a plurality of sound sensors 12 including a sound sensor A and a sound sensor B extract sound data including an abnormal sound at the same time. The sound sensor A faces the direction C1 and the sound sensor B faces the direction C2. The vertical axis shows the loudness (sound pressure) of the abnormal sound. Due to the difference in sound pressure information of the sound data output from the sound sensor A and the sound sensor B, and the difference in the orientation of the sound sensor A and the sound sensor B, the sound source that generates an abnormal sound and the positions of the respective sound sensors A and B. Relationships can be calculated. This makes it possible to identify the position of the sound source that generates the abnormal sound.

<Sound pattern analysis method>
FIG. 6 is a diagram for explaining a method for analyzing the sound pattern when outputting the material information corresponding to the corresponding data groups G1, G2, and G3 calculated in step S53 and the contact state.
FIG. 6 shows a three-dimensional coordinate axis of the learning DB 14 space, that is, a coordinate axis of the sound pressure feature, a coordinate axis of the waveform feature, and a coordinate axis of the generation cycle feature.

The coordinate axis of the sound pressure feature shows the difference in the sound pressure fluctuation state of each frequency of the sound data including the abnormal sound. Further, the coordinate axis of the waveform feature shows the geometric feature of the abnormal sound wave shape on the spectrogram, that is, the waveform pattern, and the coordinate axis of the generation cycle feature shows the cycle in which the abnormal sound is generated, that is, the number of overlaps per predetermined time.

The data A shown in FIG. 6 is a plot of the positions of the sound data A including the abnormal sound observed by the sound sensor 12 on the three coordinate axes of the sound pressure feature, the waveform feature, and the generation cycle feature. Here, the data group G1 (◯), the data group G2 (Δ), and the data group G3 (□) represent sound data data groups having different waveform pattern characteristics. That is, the figures 〇, Δ, and □ in each data group indicate the difference in the shape of the waveform pattern of each data group.

In FIG. 6, the distances L1, L2, and L3 between the position of the sound data A including the abnormal sound and the center positions of the data groups G1, G2, and G3 are calculated. Then, the smallest distance among the distances L1, L2, and L3, for example, L1 is selected. It is determined that the sound data A including the abnormal sound from the minimum distance L1 is the abnormal sound data belonging to the data group G1.

<Failure report displayed on the display unit 20>
FIG. 7 is an example of a failure report created by the display data generation unit 18 and displayed on the display unit 20.
As shown in FIG. 7, this failure report is displayed on the display unit 20 as an "abnormal sound failure investigation report". In the abnormal sound failure investigation report, the creation time, the elevator unit number, the abnormal sound wave type, the position where the abnormal sound is generated, the place where the abnormal sound is generated, the cause of the occurrence, the remedy, and the corrective state are displayed.

Specifically, as shown in FIG. 7, the failure report "creation time: 2020/07/10 12/00: 00" and "elevator unit: 12345678" are displayed in the upper part. Then, the sound pressure waveform of the abnormal sound, the frequency waveform of the abnormal sound, and the position (height) where the abnormal sound is generated are displayed in the middle stage. Here, the sound pressure waveform is the same as the waveform shown in the lower part of FIG. 5, and the frequency waveform indicates that a high frequency is generated at the time when the abnormal sound is generated. Also, from the height waveform, it can be seen that an abnormal sound occurred on the 3rd floor in this failure report.

Further, in the lower part, it is shown that the generation part is the contact part between the cable (rubber) and the car (metal), and the collision between the cable in the tower and the car is displayed as the cause of the generation of the abnormal sound. Further, as a countermeasure (correction method) for this abnormal sound, it is displayed that the fixed state of the cable has been confirmed, and "correction state: completed" indicating that the correction has been made is displayed.
Then, "Creator: OOOO" that created the failure report and "Customer confirmation: XXX" indicating that the customer confirmed that the correction was made are displayed.

<Data structure stored in the parts database 17>
FIG. 8 shows an example of a data structure stored in the component DB 17 in the elevator of model code 0001. The model code 0001 is a code number of the target elevator input in step S1 of FIG.
As shown in FIG. 8, the component DB 17 has a component field, a position field, a material field, a drawing number field, a dimension field, and a contact / adjacent component field.

Parts used in elevators, such as rails, buffers, limit switches, top pulleys, cables, facer plates, etc., are stored in the parts field column. These parts are, for example, parts that may generate abnormal noise.
In the column of the position field, the position where each part exists, for example, the pit, the top floor, the bottom floor, the top, etc. are stored corresponding to the parts in the parts field. Since the rail parts are related to all layers, they are stored as "running" instead of the position information.
In the material field column, materials such as metal and rubber are stored for each part in the parts field.

In the drawing number field field, the number of the design drawing, for example, when the part is a rail, the number of "A0001" is stored.
In the dimension field column, the dimensions (A1, A2, A3 ...) Corresponding to the parts stored in the part field are stored.

The contact / adjacent parts field contains the parts that each part of the part feel is in contact with or the adjacent parts. That is, since the generation of abnormal sound is often caused by contact between parts, the parts in the contact / adjacent parts field are important information for finding the cause of sound data generation including abnormal sound.

<Data structure stored in the learning database 14>
FIG. 9 shows an example of a data structure stored in the learning DB 14 of the present embodiment.
As shown in FIG. 9, the learning DB 14 has a sound data group field, a representative coordinate field, a generation position field, a generation direction field, a contact state field, a material (1) field, a material (2) field, and an abnormal state field. .. The reason why the two material fields, the material (1) field and the material (2) field, are provided is that there is a high possibility that the generation of abnormal noise is caused by the contact between the two parts.

In the sound data group field, different sound data groups are shown as data group 1, data group 2, data group 3, ....
The representative coordinates, the position where the abnormal sound is generated, the direction (direction of generation) of the sound source where the abnormal sound is generated, the characteristics of the contact state, the material of the sound source, and the abnormal state are stored corresponding to the data group of each data group field above. ing.
That is, the representative coordinates of each data group ([A1, B1, C1] in the data group 1) are stored in the representative coordinate field.

Further, in the generation position field, whether it is an abnormal sound during traveling regardless of the floor, or whether it is an abnormal sound generated on the lowest floor or a specific floor is stored.
In the generation direction field, whether the abnormal sound in each sound data group is generated at the top or at the bottom is stored.

In the contact state field, whether it is an abnormal sound due to rotation of a bearing or the like, an abnormal sound due to rubbing, or an abnormal sound due to a collision is stored for each sound data group.
In the material (1) field, metal is stored in all sound data groups, and in the material (2) field, metal, lining, rubber, etc. are stored for each sound data group.
In the abnormal state field, abnormal noise is generated such as deterioration of the bearing in the data group 1, lining wear in the data group 2, collision of the cable and the car in the data group 3, contact between the rope and the foreign matter prevention rubber in the data group 4. The factor is stored.

<Example of the second embodiment>
FIG. 10 shows the overall configuration of the elevator abnormality state detection system according to the second embodiment of the present invention. The difference from the first embodiment shown in FIG. 1 is that the material sensor 5 is used. Since the configurations other than the material sensor 5 are the same as the configurations shown in FIG. 1, the same reference numerals are given and the description thereof will be omitted.

The material sensor 5 shown in FIG. 10 is installed in an elevator, detects material information of each running component at the time of measurement, and stores it in the component DB 17. Here, the material sensor 5 is required to be a sensor capable of measuring at least the material characteristics (for example, reflectance) of an object and capable of determining the material. Specifically, for example, an infrared sensor, an ultrasonic sensor, or the like is used.

The present invention is not limited to the above-mentioned first and second embodiments, but includes various modifications and applications. Further, the above-described embodiment of the embodiment has been described for the purpose of explaining the present invention in an easy-to-understand manner, and the present invention is not necessarily limited to the one provided with all the described configurations, and other examples are appropriate. It can also be applied to configurations. note that,

For example, in the above-described embodiment, the sound sensor may use a camera capable of recording. Further, in the above-described embodiment, the measuring device 1 itself is placed at a specific position (for example, a pit) in the hoistway instead of in the elevator car, and the abnormal noise generation time and the peculiar sound related to the elevator running (for example). The running state of the elevator may be calculated from the generation time of the brake operation sound).

1 ... Measuring device, 2 ... History database (DB), 3 ... Position sensor, 4 ... Information processing device, 5 ... Material sensor, 11 ... Control unit, 12 ... Sound sensor, 13 ... Position Information acquisition unit, 14 ... learning DB, 15 ... processing unit, 16 ... comparison unit, 17 ... parts DB, 18 ... display data generation unit, 19 ... communication unit, 20 ... display unit

Claims (6)

This is a method for detecting an abnormal state of an elevator, which detects an abnormal state of the elevator.
The step of acquiring the sound data of the elevator while traveling by the sound sensor, and
The step of acquiring the traveling position data of the elevator by the position sensor, and
A step of calculating the position of a sound source that generates sound data including the abnormal sound from the sound data including the abnormal sound among the sounds acquired by the sound sensor and the traveling position of the elevator acquired by the position sensor.
A step of extracting material information of parts of the elevator from the traveling position of the elevator when the sound data including the abnormal sound is generated and the position of the sound source of the sound data including the abnormal sound.
A step of learning sound data including the abnormal sound and calculating a component having a material that emits the sound data including the abnormal sound based on the learning data.
A step of comparing the material information of the parts of the elevator with the material that emits the calculated sound data including the abnormal sound, and
A method for detecting an abnormal state of an elevator including a step of identifying a cause of generation of sound data including the abnormal sound based on the result of the comparison. The sound sensor includes at least two sound sensors facing in different directions, and a sound source that generates sound data including the abnormal sound due to a difference in the magnitude of sound pressure detected by the at least two sound sensors. Detect the direction,
The elevator abnormality state detection method according to claim 1. The elevator abnormality state detection method according to claim 1, wherein the position sensor is a pressure sensor or an acceleration sensor, and the height of the elevator car is detected by the pressure sensor and the moving speed of the elevator car is detected by the acceleration sensor. Further, a step of creating a failure report including the generation site of the sound data including the abnormal sound generated by the elevator, the cause of the occurrence, and the correction method in the display data generation unit, and
The elevator abnormality state detection method according to claim 1, further comprising a step of displaying the failure report created by the display data generation unit on the display unit. The item according to any one of claims 1 to 4, further comprising a step of detecting material information of each moving part of the elevator by a material sensor installed in the elevator and storing the material information in the parts database. Elevator abnormal state detection method. A position information acquisition unit that acquires the position of the elevator by a signal from a position sensor that detects the position of the elevator, and
A sound sensor that detects abnormal sounds emitted by the elevator, and
Data for discriminating between sound data including abnormal sound and normal sound not including abnormal sound among the sounds detected by the sound sensor, and learning and determining the characteristic pattern of the sound detected by the sound sensor. A learning database that stores in advance, and
A parts database that stores position information and material information of parts of the elevator in advance,
The sound in the sound data including the abnormal sound is processed by processing the data input from the sound sensor, the position information acquisition unit, the learning database, and the component database, and using the data stored in the learning database by machine learning. A processing unit that determines the characteristic pattern of
A comparison unit that compares the position information and material information of the parts of the elevator stored in the parts database with the characteristic pattern of the sound obtained by the processing unit.
A display data generation unit that creates a failure report including the location of sound data including the abnormal sound calculated by the comparison unit, the cause of the sound data including the abnormal sound, and the correction method, and displays it on the display unit. And, equipped with an elevator abnormality state detection device.

Images