KR102512792B1 - Black box system for industrial sites - Google Patents

Abstract

The present invention relates to an industrial black box system which comprises: a data collection unit for collecting data; a data storage unit for storing the collected data; a black box including a real-time risk detection unit that compares the data stored in the data storage unit with a risk behavior detection model provided from the outside to stop a facility operation of an industrial site or issues a risk warning according to a comparison result; and a reproduction simulation processing unit for mapping and displaying a result of modeling a facility of an industrial site and the data stored in the data storage unit of the black box in one-to-one correspondence with the black box.

Description

Industrial black box system {Black box system for industrial sites}

The present invention relates to an industrial black box system, and more particularly, to an industrial black box system capable of preventing and analyzing safety accidents at industrial sites.

In general, safety accidents occur with a very high frequency at manufacturing sites due to insufficient protection facilities or carelessness of workers.

Technologies for monitoring and preventing the occurrence of safety accidents at manufacturing sites have been proposed.

Most of the conventional technologies for monitoring and preventing safety accidents at industrial sites are portable devices that detect workers' vital signs, such as Registered Patent No. 10-0846829 (registered on July 10, 2008, Industrial Safety and Hazard Monitoring System for Manufacturing). A method using detection means is used.

More specifically, the biosignal detected by the portable sensing means or the surrounding environment information is provided to the control means, and the information received from the control means is analyzed to warn of the risk of safety accidents.

However, since these conventional methods necessarily require portable sensing means, power management and state management of portable sensing means (for example, personal portable terminals such as smart phones) are required.

That is, when the portable sensing means carried by the individual worker is unusable, a problem arises in that the entire system cannot be operated.

In addition, since personal portable sensing means are used, personal location information or other personal information may be leaked, and thus a new system capable of protecting personal information is required.

An object of the present invention in view of the above problems is to provide an industrial black box system capable of protecting personal information.

In addition, another object of the present invention is to provide an industrial black box system capable of preventing safety accidents to workers, monitoring the state of facilities at industrial sites, and reproducing the situation at the time of an accident as it is.

The industrial black box system of the present invention for solving the above technical problem is a data collection unit for collecting data, a data storage unit for storing the collected data, and the data stored in the data storage unit and the risk provided from the outside. A black box including a real-time risk detection unit that compares behavior detection models and stops the operation of industrial facilities or warns of danger according to the comparison result, and models industrial facilities in one-to-one correspondence with the black box. and a reproduction simulation processing unit for mapping and displaying the result and the data stored in the data storage unit of the black box.

In an embodiment of the present invention, the risk behavior detection model may be generated by a detection model generation unit that performs learning and verification using risk behavior big data.

In an embodiment of the present invention, the detection model generation unit extracts risky behaviors from behavioral image big data, generates learning data and evaluation data, learns with the learning data to create a risky behavior detection model, and evaluates A risk behavior detection model may be evaluated using the data for risk behavior detection, and a risk behavior detection model having an evaluation result of a reference value or higher may be selected.

In an embodiment of the present invention, the risk behavior detection model is an image model, and the similarity between the image data and the risk behavior detection model is obtained by comparing with the image data stored in the data storage unit, and the degree of risk can be determined based on the similarity. .

In an embodiment of the present invention, the data collected by the data collection unit is facility operation information data, video information data, and sound information data, and the facility operation information data is a PLC, a control panel of facilities, a computing device for management, or It can be collected by communicating with the monitoring device.

In an embodiment of the present invention, the data collected by the data collection unit is equipment operation information data, video information data, and sound information data, and the video information data and sound information data include cameras and microphones installed for each element of the industrial site. can be collected through

In an embodiment of the present invention, the black box may further include a data processor for deleting data whose storage period has elapsed among data stored in the data storage unit.

In an embodiment of the present invention, the real-time risk detection unit compares the risk behavior reference value setting unit for setting the risk behavior reference value with the image data stored in the data storage unit and the risk behavior detection model. may include a risk behavior determination unit that determines that a risk behavior has been detected.

In an embodiment of the present invention, the real-time risk detection unit may further include a warning light output unit for controlling a warning light when the dangerous behavior is determined as a risky behavior as a result of the determination of the risk behavior determining unit.

In an embodiment of the present invention, the risk behavior reference value includes a risk reference value and a warning reference value, and the warning light output unit outputs a warning light displayed when there is a risk behavior exceeding the risk reference value and when the risk behavior is less than the risk reference value and is greater than the warning reference value. Colors can be controlled differently.

In an embodiment of the present invention, the real-time risk detection unit may further include a facility stop command output unit for stopping the facility when a result of comparing the image data and the risk behavior detection model exceeds a risk reference value.

In an embodiment of the present invention, the reproduction simulation processing unit may include a facility reproduction model editing unit for generating and storing a facility reproduction model, and a facility situation reproduction unit for reproducing equipment conditions on a screen.

The present invention acquires video and audio data for elements of a manufacturing site using image sensors and acoustic sensors, acquires and stores the operational state of equipment at the time, and reproduces the video and audio of workers using the stored data. is replaced with a simulation image to prevent leakage of personal information.

In addition, there is an effect of preventing the occurrence of safety accidents by learning previous accident occurrence data, comparing currently acquired video, audio data, and facility state data to determine the risk of an accident and giving a warning.

1 is a block diagram of an industrial black box system according to a preferred embodiment of the present invention.
2 is a flowchart of a risk notification method using a black box of the present invention.
3 is a processing flow chart of a data processing unit.
4 is a detailed configuration diagram of a reproduction simulation processing unit.
5 is an exemplary diagram for explaining a mapping process.
6 is an operational flowchart of the facility reproduction model editing unit of the present invention.
7 is a flow chart of a process of reproducing a facility situation.
8 is an exemplary diagram of a reproduction screen.

In order to fully understand the configuration and effects of the present invention, preferred embodiments of the present invention will be described with reference to the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, and may be implemented in various forms and various changes may be made. However, the description of the present embodiment is provided to complete the disclosure of the present invention and to completely inform those skilled in the art of the scope of the invention to which the present invention belongs. In the accompanying drawings, the size of the components is enlarged from the actual size for convenience of description, and the ratio of each component may be exaggerated or reduced.

Terms such as 'first' and 'second' may be used to describe various elements, but the elements should not be limited by the above terms. The above terms may only be used for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a 'first element' may be named a 'second element', and similarly, a 'second element' may also be named a 'first element'. can Also, singular expressions include plural expressions unless the context clearly indicates otherwise. Terms used in the embodiments of the present invention may be interpreted as meanings commonly known to those skilled in the art unless otherwise defined.

1 is a block diagram of an industrial black box system according to a preferred embodiment of the present invention.

Referring to FIG. 1 , the present invention includes a detection model generation unit 200 for learning risk behavior using various image data and analyzing risk behavior patterns under accident conditions to generate a risk behavior detection model; The black box 100 collects and stores information data on sound and facility operation, compares the risk behavior detection model with the stored image data, detects and processes risk behavior in real time, and the black box 100 according to user manipulation. ) may include a reproduction simulation processing unit 300 that simulates and displays a situation at a specific point in time using data stored in .

The black box 100 includes a data collection unit 110 that collects information data, a data storage unit 130 that stores information data collected by the data collection unit 110, and the data storage unit 130. ) of information data stored in the data processing unit 120 for deleting past data according to past data deletion standards, and risk behaviors generated by the information data stored in the data storage unit 130 and the detection model generation unit 200. A real-time danger detection unit 140 detects a real-time danger situation according to the matching degree compared with the detection model and performs processing accordingly.

Hereinafter, the configuration and operation of the industrial black box system according to a preferred embodiment of the present invention configured as described above will be described in more detail.

First, the black box 100 collects information data using collection devices installed for each element of an industrial site. The information data collected at this time includes image information and sound information, and includes operation information of facilities in an industrial site.

Actual accidents include accidents caused by the fault of workers, but also accidents caused by abnormalities in the equipment itself, such as breakdowns or explosions.

To this end, the data collection unit 110 uses a facility operation information collection unit 111 that collects raw material information input to each facility, supplied power information, and operating condition information, and an image information sensor such as a camera installed for each element of the industrial site. It may be configured to include an image data collection unit 112 that collects image data and a sound data collection unit 113 that collects sound data through a sound collection sensor such as a microphone installed for each element of an industrial site.

As described above, the facility operation information collection unit 111 collects current operation information of facilities, and at this time, facility operation information can be collected by communicating with a PLC and a control panel of facilities, and a separate program for production management is installed. It may be collected by communicating with a computing device or monitoring device.

Data collected through the data collection unit 110 having such a configuration is stored in the data storage unit 130 . At this time, the data stored in the data storage unit 130 is not accessible through other external terminals, and the stored data cannot be directly reproduced or transmitted to a separate server and stored.

It is assumed that the data stored in the data storage unit 130 is deleted by the processing of the data processing unit 120 when the time set for security and personal information protection has elapsed.

For example, the data processing unit 120 may selectively delete only data for which the period set by the past data processing unit 122 has elapsed when the period set by the past data processing criterion setting unit 121 has elapsed.

The data processing unit 120 may be a module loaded with software for data deletion processing.

The detection model generation unit 200 learns image data related to various risky behaviors and generates risky behavior detection models for detecting risky behaviors.

The detection model generating unit 200 extracts image data determined to be dangerous behavior through behavioral data purification from behavioral image big data, and generates training data and evaluation data.

In this case, the detection model generation unit 200 assumes that image data obtained from an actual industrial site and stored in the data storage unit 130 is not used. This is to prevent leakage of personal information, such as a worker's image, by preventing data obtained from an industrial site from being stored outside the data storage unit 130 in the black box 100 .

Next, a model for risk behavior detection is created by learning the generated learning data, and after verifying that risk behavior detection is smoothly performed using the evaluation data, the risk behavior detection model to be evaluated beyond the evaluation standard and applied in practice. set

Next, the real-time risk detection unit 140 compares the risk behavior detection model of the detection model generator 200 with the image data stored in the data storage unit 130 to determine whether a risk behavior is detected.

The real-time risk detection unit 140 includes a risk behavior reference value setting unit 141, a risk behavior determination unit 142, a warning light output unit 143, and a facility stop command output unit 144.

The risk behavior determination unit 142 compares the image data of the data storage unit 130 with the risk behavior detection model, and if it is determined that there is a risk behavior, the warning light is operated through the warning light output unit 143 so that there is a risk behavior. inform

At this time, the risky behavior determining unit 142 may check facility operation information data, check whether facilities around the location where the dangerous behavior is operating, and operate the warning light only when operating.

In addition, the risk behavior reference value setting unit 141 sets a risk reference value for the sound data and facility operation information data, compares the sound data and facility operation information with the risk reference value, and determines whether the detected sound and facility operation information are at a risk level. can do.

In this case, when the detected sound data is greater than the sound risk reference value, the warning light may be operated through the warning light output unit 143 .

In addition, when the detected facility operation information is at a dangerous level, the operation of the facility may be forcibly stopped through the facility stop command output unit 144 .

For example, the facility stop command output unit 144 may stop facility operation by communicating with a control panel of the facility.

2 is a flowchart of a risk notification method using a black box of the present invention.

Referring to FIG. 2, the data collection unit 110 collects facility operation information, image information, and sound information data (S21), and stores the collected data in the data storage unit 130 (S22). In step S23, the real-time risk detection unit 140 checks the facility operation information data to determine whether the facility is currently in operation (S23), and if the facility is in operation, the risk behavior determination unit 142 determines the risk behavior detection model and image data. Comparing and calculating a risk behavior probability (S24), determining whether the calculated probability value is greater than the risk reference value (S25), and if greater than the risk reference value, the facility stop command output unit 144 through the facility The step of executing the stop (S26), the step of outputting a red warning light through the warning light output unit 143 together with the equipment stop (S27), and the step of confirming whether the probability value calculated as a result of the determination in the step S25 is equal to or greater than the warning reference value. (S28) and a step (S29) of outputting a yellow warning light through the warning light output unit 143 if it is equal to or greater than the warning reference value.

The specific configuration and operation of the method for notifying the risk of an industrial site using the black box system of the present invention will be described in more detail as follows.

First, as in step S21, the data collection unit 110 collects facility operation information, image information, and sound information data.

Facility operation information can be collected through communication with a PLC, a facility control panel, a separate computing device in which a program for production management is installed, or a monitoring device.

In addition, video information and sound information data can be collected through cameras and microphones installed for each element of the industrial site.

The collected information data is stored in the data storage unit 130 as in step S22.

At this time, the information data stored in the data storage unit 130 is deleted by the data processing unit 120 when the set time elapses.

3 is a processing flow chart of the data processing unit 120 .

Referring to FIG. 3 , the data processing unit 120 reads the deletion reference value (S31) and deletes data whose storage time is before the deletion reference value among the data stored in the data processing unit 120 (S32).

The information data stored in the data storage unit 130 is read and used by the real-time risk detection unit 140 .

In particular, as in step S23, the real-time risk detection unit 140 checks the facility operation information data to determine whether the facility is currently in operation.

Next, in step S24, if the facility is in operation, the real-time risk detection unit 140 compares the risk behavior detection model with the image data in the risk behavior determination unit 142 to calculate a risk behavior probability.

The risk behavior probability at this time is calculated based on the similarity between the detected image data and the risk behavior detection model. If the image data and the worker's behavior in the risky behavior detection model completely match, the risky behavior probability becomes 100%.

Next, it is determined whether the probability value calculated as in step S25 is greater than the risk reference value. At this time, the risk reference value may be set as necessary, such as 70% or 80%.

Then, if the probability value calculated in step S26 is greater than the risk reference value, the facility is forcibly stopped through the facility stop command output unit 144.

In addition, as in step S27, a command for outputting a red warning light may be output through the warning light output unit 143.

Steps S26 and S27 may be performed simultaneously.

Next, in step S28, it is checked whether the probability value calculated as a result of the determination in step S25 is greater than or equal to the warning reference value.

At this time, the warning reference value is set to a value lower than the risk reference value described above, and may be set to, for example, 50% or 60%.

If the determination result of step S28 is less than the warning reference value, the determination is terminated.

Next, in step S29, if the result of the determination in step S28 is equal to or greater than the warning reference value, a signal for controlling the yellow warning light is output through the warning light output unit 143.

As such, the present invention can determine whether there is a risk of an accident through comparison of images, and perform appropriate control according to the degree of risk.

4 is a detailed configuration diagram of the reproduction simulation processing unit 300 .

The reproduction simulation processing unit 300 includes a facility reproduction model editing unit 310 and a facility situation reproduction unit 320 .

The facility reproduction model editing unit 310 includes a facility reproduction item management unit 311, a facility object design unit 312, a data mapping unit 313, and an operation status reproduction model storage unit 314.

In addition, the facility situation reproducing unit 320 includes a selecting unit 321, a video data reproducing unit 322, a sound data reproducing unit 323, and a facility operation data reproducing unit 324.

As such, it is assumed that the reproduction simulation processing unit 300 is a computing device that is only connected to the black box 100 on a 1:1 basis and cannot communicate with the outside.

The configuration and operation of the reproduction simulation processing unit 300 applied to the present invention configured as described above will be described in more detail.

First, the facility reproduction item management unit 311 codes and sets setting items, operation items, quality-related items, video items, and sound items for operating the equipment according to the user's selection.

Input of such code is executed through the data mapping unit 313, and can be directly mapped to the facility reproduction model designed in the facility object design unit 312 using a drag-and-drop method.

This is shown in Figure 5.

The operational status reproduction model storage unit 314 stores a facility reproduction model, which is a result of mapping drawings and equipment data.

In this state, a reproduction module of a specific date is selected through the facility situation reproducing unit 320, and the modeled video data, sound data, and facility operation data are combined and reproduced with the reproducing module as necessary.

The selection unit 321 of the facility situation reproducing unit 320 may select a model stored in the operating status reproducing model storage unit 314 and a specific time. The selected model is displayed on the screen.

In addition, the selector 321 may set start and end points of reproduction on the screen, and set speed of reproduction.

When reproduction start, end time and speed are set by the selector 321 as described above, the video data reproducer 322, the sound data reproducer 323, and the equipment operation data reproducer 324 each perform the black box 100. The video data, sound data, and equipment operation data stored in the data storage unit 130 of the are read and the data are reproduced.

At this time, the facility reproduction model operates based on the facility operation data, and reproduces image data and sound data. At this time, the worker area in the image data can be simulated using an imaged worker object so that the individual worker cannot be identified.

6 is an operation flowchart of the equipment reproduction model editing unit 310 of the present invention.

Referring to this, in step S61, as shown in FIG. 5 by the equipment reproduction item management unit 311, setting items for operating equipment are coded to set equipment reproduction items.

Specifically, the facility reproduction item management unit 311 may code and set setting items, operation items, quality-related items, video items, and sound items for operating the equipment according to the user's selection.

Step S62 sets a facility object. At this time, the facility object setting may be performed through the facility object design unit 312 . At this time, the facility object design is modeled in a drawing format so that the shape or operation of the facility can be recognized on the screen.

Next, as in step S63, the facility object set on the screen and the coded equipment reproduction item are mutually mapped, and stored in the operation status reproduction model storage unit 314 as in step S64.

7 is a flow chart of a process of reproducing a facility situation.

Referring to FIG. 7 , as in step S71 , a reproduction model stored in the operation state reproduction model storage unit 314 is selected through the selection unit 321 of the equipment situation reproduction unit 320 .

In addition, as in step S72, facility objects constituting the selected reproduction model are displayed on the screen. At this time, it is assumed that the facility object was previously designed through the facility object design unit 312 .

Then, as in step S73, the selection unit 321 sets each of the start and end points to be reproduced on the screen as times.

Next, as in step S74, among the information data stored in the data storage unit 130 of the black box 100, data including time information corresponding to the start and end points set in step S73 is received.

The data at this time includes video data, sound data, and facility operation data.

Then, as in step S75, a simulation reproduction speed to be reproduced is determined. For example, you can select 0.5x speed, 1.0x speed, 1.5x speed, etc.

When the reproduction start button is selected, simulation results are reproduced on the screen according to the selected reproduction speed as in step S76.

An example of the reproduction screen is shown in FIG. 8 . It is assumed that the illustrated facility models are operated using the facility operation data, and video data and sound data can be reproduced.

At this time, in the region classified as a person in the image data, it is possible to prevent exposure of worker's personal information through image processing.

When replay ends or select stop replay, the above replay screen replay ends.

In this way, the present invention can protect personal information by preventing leakage of captured image data and preventing specific images of workers from being exposed even during reproduction.

Embodiments according to the present invention have been described above, but these are merely examples, and those skilled in the art will understand that various modifications and embodiments of equivalent range are possible therefrom. Therefore, the true technical protection scope of the present invention should be defined by the following claims.

100: black box 110: data collection unit
120: data processing unit 130: data storage unit
140: real-time risk detection unit 200: detection model generation unit
300: reproduction simulation processing unit

Claims (12)

Risky behaviors are extracted from behavioral image big data, which is image data that captures worker behaviors, and training data and evaluation data are generated, and a risky behavior detection model, which is an image model, is created by learning with the learning data, and then used for evaluation. a detection model generating unit that evaluates risk behavior detection models using the data and selects risk behavior detection models whose evaluation results are greater than or equal to a reference value;
A data collection unit that collects data, a data storage unit that stores the collected data, and image data stored in the data storage unit are compared with the risk behavior detection model to obtain a similarity between the video data and the risk behavior detection model; A black box including a real-time risk detection unit that judges the degree of risk according to the degree of similarity and stops the operation of industrial facilities or warns of danger according to the degree of risk; and
In one-to-one correspondence with the black box, a reproduction simulation processing unit for mapping and displaying a result of modeling industrial facilities and data stored in a data storage unit of the black box,
The real-time risk detection unit,
a risk behavior reference value setting unit configured to set a risk behavior reference value;
a risky behavior determination unit determining that a dangerous behavior has been detected when a result of comparing the image data stored in the data storage unit with the risky behavior detection model is greater than or equal to the risky behavior reference value; and
A warning light output unit for controlling a warning light when the dangerous behavior is determined as a risky behavior as a result of the determination of the risk behavior determining unit;
The reproduction simulation processing unit,
A facility object design unit that designs a facility reproduction model, a facility reproduction item management unit that codes and sets setting items, operation items, quality-related items, video items, and sound items for operating the equipment selected by the user; A facility reproduction model including a data mapping unit for directly mapping the settings of the reproduction item management unit to the equipment reproduction model by drag-and-drop method, and an operation status reproduction model storage unit for storing a facility reproduction model, which is a result of mapping drawings and equipment data. editorial department; and
Selecting a model stored in the operating status reproduction model storage unit and a specific time to be displayed on the screen; An image data reproducing unit that calls and reproduces stored video data, an audio data reproducing unit that calls and reproduces the audio data stored in the data storage unit, and a facility operation data that calls and reproduces facility operation data stored in the data storage unit. Including a reproduction unit, including a facility situation reproduction unit that operates a facility reproduction model, reproduces video data, and reproduces sound data, and simulates using a worker object imaged so that the worker area in the image data cannot be identified. industrial black box system. delete delete delete According to claim 1,
The data collected by the data collection unit is
Facility operation information data, video information data, and sound information data,
Facility operation information data,
An industrial black box system characterized in that it collects by communicating with a PLC, a control panel of a facility, a computing device for management or a monitoring device. According to claim 1,
The data collected by the data collection unit is
Facility operation information data, video information data, and sound information data,
The video information data and the sound information data,
An industrial black box system characterized by collecting through cameras and microphones installed for each element of the industrial site. According to claim 1,
The black box,
Industrial black box system further comprising a data processing unit for deleting data for which the storage period has elapsed among the data stored in the data storage unit. delete delete According to claim 1,
The risk behavior reference value is,
Includes risk threshold and warning threshold,
The warning light output unit controls the color of the warning light displayed differently when there is a risk action exceeding the risk reference value and when the risk value is smaller than the risk reference value and exceeds the warning reference value. According to claim 10,
The real-time risk detection unit,
If the result of comparing the image data and the risk behavior detection model exceeds the risk reference value,
An industrial black box system further comprising a facility stop command output unit for stopping the facility. delete

Images