Elevator operation monitoring method and elevator operation monitoring system

Abstract

Unsafe behavior of workers is automatically identified using surveillance images from cameras to ensure worker safety. [Solution] During normal elevator operation, the movements of workers in the machine room, on the elevator car, and in the elevator pit are monitored by cameras installed in at least one of the machine room, the elevator car, and the elevator pit. If worker movements are detected in the captured camera images, the camera images are compared with images registered in a database as unsafe movements to analyze whether or not the camera images contain unsafe movements. If the camera images contain images containing unsafe movements, an operation stop signal is output to activate a safety switch and power supply to the hoist is cut off to stop operation.

Description

[Technical field]

[0001] An embodiment of the present invention relates to an elevator operation monitoring method and an elevator operation monitoring system. [Background technology]

[0002] When performing maintenance work in the machine room, on the car, under the pit, etc., the safety switch is always turned on to stop the operation of the hoist to ensure the safety of the workers. By actively turning on various safety switches, workers can be protected from accidents such as getting caught, falling, or being pinched.

[0003] Conventionally, there are known techniques for removing a cover that covers a rotating part, or for notifying an inspector of the approach of the cage and counterweight with a vibration or an alarm before the cage and counterweight come close to each other when working on the cage. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2009-190851 A [Patent Document 2] JP 2014-015300 A [Patent Document 3] JP 2017-007830 A [Patent Document 4] Japanese Patent Application Publication No. 6-286952 Summary of the Invention [Problem to be solved by the invention]

[0005] Conventional technology is a method to prevent unsafe behavior by alerting workers with vibrations and alarms to raise their safety awareness, but there is a possibility that workers themselves may be fatigued or not be careful enough to take evasive action late, resulting in the repetition of similar workplace accidents. This causes malfunctions, making the elevator unusable and preventing the provision of service to users.

[0006] In view of the above problems, the present invention aims to provide an elevator work monitoring method and elevator work monitoring system that ensures the safety of workers by automatically determining unsafe behavior of workers based on surveillance images taken by a camera. [Means for solving the problem]

[0007] In an embodiment for achieving the above object, during normal elevator operation, the movements of workers in the machine room, on the elevator car, and in the elevator pit are monitored by cameras installed in at least one of the machine room, the elevator car, and the elevator pit. If a worker's movement is detected in a captured camera image, the camera image is compared with images registered in a database as unsafe movements to analyze whether the camera image contains an image including an unsafe movement. If an image including an unsafe movement is detected in the camera image, an operation stop signal is output to activate a safety switch, and power supply to the hoist is cut off to stop operation. [Brief description of the drawings]

[0008] [Figure 1] 1 is a block diagram showing a configuration of an elevator operation monitoring method and an elevator operation monitoring system according to an embodiment of the present invention; [Diagram 2] 1 is a configuration diagram of an elevator to which an elevator operation monitoring method and an elevator operation monitoring system according to an embodiment are applied; [Figure 3A] 3 is a flowchart showing a processing procedure of an elevator operation monitoring method and an elevator operation monitoring system according to an embodiment. [Figure 3B] 3 is a flowchart showing a processing procedure of an elevator operation monitoring method and an elevator operation monitoring system according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] First Embodiment Configuration of the embodiment FIG. 1 is a block diagram showing the configuration of an elevator operation monitoring system according to an embodiment of the present invention, and FIG. 2 is a configuration diagram of an elevator to which the elevator operation monitoring system according to the embodiment is applied.

[0010] As shown in Fig. 2, in the elevator of this embodiment, a car 2 installed in a hoistway 1 is connected to a counterweight 3 by a main rope 4, and moves up and down by a hoist 6 and a deflection sheave 7 installed in a machine room 5. The operation of the elevator is controlled by an elevator control panel 8 installed in the machine room 5.

[0011] Network cameras 11, 12 are installed in the machine room 5 to monitor the movements of a machine room worker 13 working in the machine room 5. Note that multiple network cameras 11, 12 are installed to eliminate blind spots at the work site, and the number is not limited to two, but three or more may be installed depending on the size of the machine room 5. A safety switch 15 and a hoisting machine power switch 16 are also installed in the machine room 5.

[0012] Moreover, on-car network cameras 21, 22 are installed on top of the car 2 to monitor the movements of an on-car worker 23 working on top of the car 2. A safety switch 25 is also installed on the car 2. Note that multiple network cameras 21, 22 are installed to eliminate blind spots at the work site, and are not limited to two, and three or more may be installed depending on the size of the car 2.

[0013] Furthermore, in-pit network cameras 31, 32 are installed in a pit 30 of the elevator shaft 1 to monitor the movements of an in-pit worker 33 working in the pit 30. A safety switch 35 is also installed in the pit 30. Note that multiple network cameras 31, 32 are installed to eliminate blind spots at the work site, and the number is not limited to two, and three or more may be installed depending on the size of the pit 30.

[0014] These network cameras 11, 12, 21, 22, 31, 32, safety switches 15, 25, 35, and hoist power switch 16 are connected to the elevator control panel 8 and cloud server 60 via a network 40. In addition, the network 40 connects the elevator control panels 8 of each elevator car to the cloud server 60 that manages each elevator control panel 8, and may be wired or wireless.

[0015] The cloud server 60 includes a camera image input unit 61, a camera image analysis unit 62, a database 63, and an operation stop signal output unit 64.

[0016] The camera image input unit 61 inputs camera images from the network cameras 11 , 12 , 21 , 22 , 31 , and 32 via the network 40 .

[0017] The camera image analysis unit 62 analyzes the camera image input by the camera image input unit. Specifically, it compares the input camera image with an image in the database 63 to determine whether or not it matches an image registered as an unsafe operation image in the database 63. If it does not match an image registered in the database 63, it performs an AI analysis of the camera image, and if it is an image that is highly likely to be an unsafe operation, it executes a process of registering the image in the database 63 as an unsafe operation image.

[0018] The database 63 registers and accumulates unsafe operation images. The registered images can be updated manually from outside.

[0019] When the camera image analyzing unit 62 determines as a result of analysis that the operation is unsafe, the operation stop signal output unit 64 outputs an operation stop signal to the corresponding elevator control panel 8.

[0020] The elevator control panel 8 includes a safety switch on / off unit 81 and an alarm output unit 82. When an operation stop signal is output from the operation stop signal output unit 64 of the cloud server 60, the safety switch on / off unit 81 activates the safety switches 15, 25, 35 installed in each unit to cut off the power supply to the elevator equipment. It also cuts off the hoist power switch 16 and activates the hoist brake (not shown). The alarm output unit 82 receives the operation stop signal from the operation stop signal output unit 64 and issues an alarm to the workers 13, 23, 33 who are working.

[0021] <<Function of the embodiment>> Next, the operation (processing procedure) of the elevator operation monitoring method and elevator operation monitoring system according to the embodiment will be described with reference to the flowcharts of FIGS. 3A and 3B.

[0022] While the elevator is operating, network cameras 11 and 12 monitor the inside of machine room 5, network cameras 21 and 22 monitor the top of car 2, and network cameras 31 and 32 monitor the inside of pit 30. When any of network cameras 11, 12, 21, 22, 31, or 32 detects a moving object (YES in step S1), the camera image is transmitted to cloud server 60 via network 40 (step S2).

[0023] When the camera image input unit 61 of the cloud server 60 inputs a camera image from any of the network cameras 11, 12, 21, 22, 31, 32, the camera image is analyzed by the camera image analysis unit 62 (step S3). That is, it is determined whether or not the input camera image matches an image registered in the database 63 as an unsafe operation. Here, an unsafe operation refers to a dangerous operation such as an operator approaching a rotating object such as the hoist 6 or the deflector sheave 7 or a live-line device within a certain distance.

[0024] If the input camera image matches an image registered in the database 63 as an unsafe operation (step S3 YES), an elevator operation stop signal is output from the operation stop signal output unit 64 of the cloud server 60 to the elevator control panel 8 (step S4). In response to the operation stop signal from the operation stop signal output unit 64, the alarm output unit 82 issues an alarm to the workers 13, 23, 33 currently working. The safety switch on / off unit 81 activates the safety switches 15, 25, 35 installed in each unit to cut off the power supply to the elevator equipment. In addition, the power supply to the hoist 6 is cut off, and the hoist brake is activated to stop the elevator operation (step S5).

[0025] If the input camera image does not match an image registered in database 63 as an unsafe operation, the camera image is analyzed by AI (step S6 in FIG. 3B). Then, if the AI ​​analysis shows that there is a high probability of the unsafe operation (YES in step S7), an elevator operation stop signal is output from operation stop signal output unit 64 of cloud server 60 to elevator control panel 8 (step S8). This stops the power supply to hoist 6, and the hoist brake is activated to stop the elevator operation (step S9). Then, the image that is determined to have a high probability of being an unsafe operation is registered in database 63 as an unsafe image (step S10).

[0026] After the elevator stops being stopped, if the unsafe operation at the site is resolved as a result of analyzing the camera images (step S11 YES), normal operation is resumed (step S12). In this way, according to this embodiment, even if a worker is unable to take the safety measures that should be taken during inspection due to lack of attention caused by fatigue, the automatic monitoring system using network cameras 11, 12, 21, 22, 31, and 32 can provide a service that provides safer work and prevents work-related accidents.

[0027] In particular, the AI ​​analysis program of the camera image analysis unit 62 can constantly update the operating conditions during dangerous work by automatically analyzing and learning dangerous actions of workers based on images acquired by the network cameras 11, 12, 21, 22, 31, and 32. The updated dangerous work actions are then compared and analyzed in real time with the images acquired by the network cameras 11, 12, 21, 22, 31, and 32, and an operation stop signal is sent to the elevator control panel 8, thereby constantly providing the latest safety environment. <Other embodiments> In the above embodiment, the database 63 is constructed in the cloud server 60, but this is not limited to the above. For example, the database 63 may be provided in a data center (remote monitoring center) connected to the cloud server 60 via a network.

[0028] In addition, images registered as unsafe operations in the database 63 can be manually registered and updated. In addition to machine learning, active learning is possible by having humans (workers) add and delete images as appropriate, making it possible to detect unsafe operations with high accuracy.

[0029] Furthermore, for elevators that are not group-managed, the elevator control panel 8 may be configured to include a camera image input unit 61, a camera image analysis unit 62, a database 63, and an operation stop signal output unit 64 of a cloud server 60. In this case, by installing the control software of this system in an existing elevator control panel 8, it is possible to provide the same services to elevators that have already been installed as to new elevators.

[0030] Although several embodiments of the present invention have been described above, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0031] 1...hoistway, 2...car, 3...counterweight, 4...main rope, 5...machine room, 6...hoist, 7...deflector sheave, 8...elevator control panel, 11, 12...machine room network camera, 13...machine room worker, 15...safety switch, 16...hoist power switch, 21, 22...car network camera, 23...car worker, 25...safety switch, 30...pit, 31, 32...pit network camera, 33...pit worker, 35...safety switch, 40...network, 60...cloud server, 61...camera image input unit, 62...camera image analysis unit, 63...database, 64...operation stop signal output unit, 81...safety switch on / off unit, 82...alarm output unit

Claims

1. During elevator operation, the movements of workers in the machine room, on the elevator car, and in the elevator pit are monitored by cameras installed in at least one of the machine room, on the elevator car, and in the elevator pit; If the captured camera image shows a worker's movement, the camera image is compared with images registered in the database as unsafe movements to analyze whether the camera image contains any images of unsafe movements; If the camera captures an image containing an unsafe operation, it outputs an operation stop signal to activate the safety switch and cuts off the power supply to the hoist to stop the operation. If the result of analyzing whether or not there is an image including an unsafe operation in the camera image is that there is no image including an unsafe operation in the camera image, the camera image is analyzed to determine whether or not it corresponds to an unsafe operation; An elevator operation monitoring method in which, if it is determined that an operation corresponds to an unsafe operation, an operation stop signal is output to activate a safety switch, the power supply to the hoist is cut off to stop the operation, and the camera image is registered in a database as a new image of an unsafe operation.

2. A camera installed in at least one of an elevator machine room, an elevator car, and an elevator pit; a camera image analysis unit that monitors the machine room, the car, and the elevator pit from the camera images captured by the camera, and when movement is detected in the camera images, compares the input camera image with images registered in a database as unsafe operations, and analyzes whether or not the camera images include images containing unsafe operations; and an operation stop signal output unit that outputs an operation stop signal to activate a safety switch and cuts off the power supply to the hoist to stop the operation when an image including an unsafe operation is found in the camera image by the camera image analysis unit, The camera image analysis unit analyzes the camera image to determine whether or not there is an image including an unsafe operation in the camera image. If there is no image including an unsafe operation in the camera image, the camera image analysis unit analyzes the camera image to determine whether or not it corresponds to an unsafe operation. When the operation stop signal output unit determines that an operation corresponds to an unsafe operation, it outputs an operation stop signal to activate a safety switch and cuts off the power supply to the hoist to stop the operation, and registers the camera image in a database as a new unsafe operation image.

3. The system includes an elevator control panel that controls each elevator individually and a cloud server that manages the elevator control panels, 3. The elevator operation monitoring system according to claim 2, wherein the camera image analysis unit and the operation stop signal output unit are installed in the cloud server, and the database is installed in the cloud server or a data center connected to the cloud server.

4. The elevator operation monitoring system according to claim 2 , wherein the camera image analysis unit and the operation stop signal output unit are set in an elevator control panel.

5. 4. The elevator operation monitoring system according to claim 3, wherein the images registered in the database as unsafe operations can be manually registered and updated.

Images