KR102790384B1 - Hazardous facility management system that monitors real-time information on hazardous facilities and manages safety accident at the hazardous facilities based on big data technology and artificial intelligence technology - Google Patents

Abstract

A risk facility management system is disclosed, which monitors real-time information on risk facilities based on big data technology and artificial intelligence technology and manages safety accidents of the risk facilities. The disclosed system includes a plurality of safety surveillance acquisition devices installed directly on or near a risk facility to acquire first safety surveillance data of the risk facility, a mobile robot that drives in a non-surveillance area where visual safety surveillance of the risk facility is impossible to acquire second safety surveillance data of the risk facility, and a control server in which a big data module and an artificial intelligence module operate in conjunction to derive safety surveillance results of the risk facility based on the first and second safety surveillance data.

Description

{Hazardous facility management system that monitors real-time information on hazardous facilities and manages safety accident at the hazardous facilities based on big data technology and artificial intelligence technology}

Embodiments of the present invention relate to a risk facility management system that monitors real-time information on risk facilities and manages safety accidents of the risk facilities based on big data technology and artificial intelligence technology.

The material described in the background art merely provides background information for the present embodiment and does not constitute prior art.

Hazardous facilities are facilities that manage the processing and storage of hazardous materials, such as liquid hazardous materials (chemicals, hazardous substances, etc.) storage tanks, gas facilities, etc. If a fire, explosion, or other safety accident occurs at a hazardous facility, the loss of life and property will increase rapidly. Therefore, hazardous facilities should be monitored to prevent safety accidents at hazardous facilities. However, in the case of conventional technologies, the effect of disaster prevention is uncertain, and practical applicability is also insufficient due to the duplication of safety management tasks, resulting in great dissatisfaction among safety managers.

Meanwhile, innovation in the engineering industry utilizing technologies based on the 4th Industrial Revolution, such as the Internet of Things (IoT), artificial intelligence, and big data, is recognized as an essential element in enhancing national manufacturing competitiveness. As the development of industrial Internet technology has made it possible to collect various types of data in real time that were previously impossible to obtain, research is actively being conducted to efficiently manage data and derive meaningful results from the data, and in particular, big data technology and artificial intelligence technology are being utilized. Big data technology refers to technology that extracts value by collecting, storing, processing, and analyzing extremely large amounts of data that exceed the capacity of existing databases. Artificial intelligence technology refers to ICT (Information and Communications Technologies) technology that artificially implements part or all of human learning, reasoning, and perception abilities using computer programs.

Korean Patent Registration No. 10-2497919 (2023.02.06) Korean Patent Registration No. 10-1774105 (2017.08.28)

The purpose of the present invention is to provide a big data and artificial intelligence-based risk facility management system that obtains and monitors real-time information on risk facilities based on big data technology and artificial intelligence technology.

The objects of the present invention are not limited to the objects mentioned above, and other objects and advantages of the present invention which are not mentioned can be understood by the following description, and will be more clearly understood by the embodiments of the present invention. In addition, it will be easily understood that the objects and advantages of the present invention can be realized by the means and combinations thereof indicated in the claims.

According to a preferred embodiment of the present invention, a big data and artificial intelligence-based risk facility management system includes a plurality of safety surveillance acquisition devices installed directly on or near a risk facility to acquire first safety surveillance data of the risk facility, a mobile robot driving in a non-surveillance area of the risk facility where visual safety surveillance is impossible to acquire second safety surveillance data of the risk facility, and a control server in which a big data module and an artificial intelligence module operate in conjunction to derive a safety surveillance result of the risk facility based on the first and second safety surveillance data. At this time, the big data module transmits big data on which at least one of collection, storage, processing and analysis of the first and second safety surveillance data has been performed to the artificial intelligence module, and the artificial intelligence module performs learning and inference based on the big data, and the artificial intelligence module transmits an artificial intelligence model related to the collection, storage, processing and analysis of the big data module to the big data module to perform at least one of the collection, storage, processing and analysis.

According to the present invention, based on big data and artificial intelligence technology, monitoring of hazardous facilities is possible without managers having to directly inspect the hazardous facilities, and thus safety accidents at hazardous facilities can be drastically reduced.

In addition, according to the present invention, it is possible to prevent safety accidents by identifying and responding to risk factors of hazardous facilities in advance through digital system analysis alone, and to secure competitiveness by reducing costs such as unnecessary personnel recruitment due to digitalization.

In addition, it should be understood that the effects of the present invention are not limited to the effects described above, but include all effects that can be inferred from the composition of the invention described in the detailed description or claims of the present invention.

FIG. 1 is a diagram schematically illustrating the configuration of a big data and artificial intelligence-based risk facility management system according to one embodiment of the present invention.
Figure 2 is a drawing illustrating an example of a hazardous facility.
FIG. 3 is a drawing illustrating an example of a mobile robot and a robot control terminal in a case where the hazardous facility is an outdoor tank, according to one embodiment of the present invention.
FIG. 4 is a diagram schematically illustrating the configuration of a control server according to one embodiment of the present invention.
FIG. 5 is a drawing for explaining the concept of occurrence of abnormal wall slope when the hazardous facility is an outdoor tank according to one embodiment of the present invention.
FIG. 6 is a drawing for explaining the operation of a risk facility management system that monitors the wall slope of an outdoor tank to prevent accidents according to one embodiment of the present invention.
FIG. 7 is a drawing for explaining the operation of a robot control terminal that controls a mobile robot according to one embodiment of the present invention.

The present invention can be modified in various ways and has various embodiments, and specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the present invention to specific embodiments, but should be understood to include all modifications, equivalents, or substitutes included in the spirit and technical scope of the present invention. In describing each drawing, similar reference numerals are used for similar components.

The terms "first", "second", etc. may be used to describe various components, but the components should not be limited by these terms. These terms are used only to distinguish one component from another. The term "and/or" includes any combination of a plurality of related listed items or any one of a plurality of related listed items.

When it is said that a component is "connected" or "connected" to another component, it should be understood that it may be directly connected or connected to that other component, but that there may be other components in between. On the other hand, when it is said that a component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between.

The terminology used herein is only used to describe specific embodiments and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, it should be understood that the terms "comprises" or "has" and the like are intended to specify the presence of a feature, number, step, operation, component, part or combination thereof described in the specification, but do not exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms defined in commonly used dictionaries, such as those defined in common dictionaries, should be interpreted as having a meaning consistent with the meaning they have in the context of the relevant art, and will not be interpreted in an idealized or overly formal sense unless expressly defined in this application.

Hereinafter, embodiments according to the present invention will be described in detail with reference to the attached drawings.

FIG. 1 is a diagram schematically illustrating the configuration of a big data and artificial intelligence-based risk facility management system (1) according to one embodiment of the present invention, and FIG. 2 is a diagram illustrating an example of a risk facility (2).

The big data and artificial intelligence-based risk facility management system (1) described in Fig. 1 may be a system that manages and monitors the occurrence of safety accidents at risk facilities (2) based on big data and artificial intelligence technology.

Referring to Fig. 2, the hazardous facility (2) may include an outdoor hazardous material tank (Fig. 2 (a)) that stores liquid hazardous materials (chemicals, hazardous materials, etc.), a hazardous material manufacturing plant (Fig. 2 (b)), a hazardous material general handling plant (Fig. 2 (c)), etc.

Referring to FIG. 1, the risk facility management system (1) may include a plurality of safety surveillance acquisition devices (10), a mobile robot (20), a robot control terminal (30), and a control server (40).

A plurality of safety surveillance acquisition devices (10) can be installed directly on or near a hazardous facility (2) to acquire first safety surveillance data of the hazardous facility (2) and transmit it to a control server (40).

A plurality of safety surveillance acquisition devices (10) may include a CCTV device (10a), which is a camera device installed near a dangerous facility (2) to acquire an image of the exterior of the dangerous facility (2), i.e., an exterior image. A first wired/wireless communication module may be further included in the CCTV device (10a), and the exterior image captured through the first wired/wireless communication module may be transmitted to the control server (20). At this time, the exterior image may correspond to the first safety surveillance data.

In addition, the plurality of safety monitoring acquisition devices (10) may further include two or more IoT sensor modules (10b) directly installed on the hazardous facility (2). The IoT sensor module (10b) may include at least one sensor and a second wired/wireless communication module. The IoT sensor module (10b) may obtain (measure) a sensing value through at least one sensor and transmit the sensing value to the control server (40) through the second wired/wireless communication module. At this time, the sensing value measured by at least one sensor may correspond to the first safety monitoring data.

According to an embodiment, at least one sensor may include at least one of a temperature and humidity sensor for measuring temperature and humidity inside and outside the hazardous facility (2), and a gas sensor for measuring hazardous gases. In particular, when the hazardous facility (2) is an outdoor tank (2), at least one sensor may further include at least one of a first inertial measurement unit (IMU) sensor for measuring the inclination (gradient) of a roof surface and/or a wall surface of the outdoor tank (2), and a liquid quantity sensor for measuring the amount of hazardous material stored in the outdoor tank (2).

Meanwhile, among two or more IoT sensor modules (10b), the first IoT sensor module (10b) including the first inertial measurement sensor may be installed in a visual surveillance zone, which is an area where a manager managing a hazardous facility (2) can directly monitor safety with his or her eyes. That is, depending on special restrictions such as the supply of driving power to the sensor, the first IoT sensor module (10b) may be installed in a visual surveillance zone, which is an area where the manager can reach and directly monitor. In addition, an area other than the visual surveillance zone may be referred to as a visual non-surveillance zone. That is, the visual non-surveillance zone may mean an area where a manager cannot directly monitor safety with his or her eyes (i.e., with his or her eyes).

The mobile robot (20) can drive through a visually unmonitored area of a dangerous facility (2) to obtain second safety monitoring data of the dangerous facility (2) and transmit the data to the control server (40). That is, the mobile robot (20) can drive through an area where a manager cannot visually monitor the safety of the dangerous facility (2) to obtain second safety monitoring data.

The mobile robot (20) may include a built-in camera and a third wired/wireless communication module. The built-in camera may capture an area image of an area of a visually unsupervised area, and the third wired/wireless communication module may transmit the area image to the robot control terminal (30). At this time, the area image captured by the built-in camera may correspond to the second safety surveillance data.

In particular, when the hazardous facility (2) is an outdoor tank (2), the mobile robot (20) may be a robot that runs on the roof surface and/or wall surface of the outdoor tank (2). In this case, the mobile robot (20) may further include a second inertial measurement sensor that measures the inclination of the external wall surface of the outdoor tank (2) together with the built-in camera described above. That is, the second inertial measurement sensor may measure the inclination of the wall surface of the outdoor tank (2) with respect to the visually unmonitored area, and the horizontality and verticality of the outdoor tank (2) may be inferred based on the measured inclination. At this time, the area image and the sensing value of the second inertial measurement sensor may correspond to the second safety monitoring data.

The robot control terminal (30) is a processor-based device and may include a display unit (31). The robot control terminal (30) may receive and display second safety monitoring data acquired from the mobile robot (20), and may generate a motion control signal of the mobile robot (20) and transmit it to the mobile robot (20). The motion of the mobile robot (20) may be controlled based on the motion control signal. Meanwhile, the robot control terminal (30) may transmit a received area image to the control server (40).

At this time, the robot control terminal (30) can generate a motion control signal in conjunction with the control server (40). For example, as described below, the robot control terminal (30) can receive a driving path map of the mobile robot (20) from the control server (40) and control the driving of the mobile robot (20) based on the driving path map.

FIG. 3 is a drawing showing an example of a mobile robot (20) and a robot control terminal (30) in a case where the hazardous facility (2) is an outdoor tank (2) according to one embodiment of the present invention.

Referring to FIG. 3, as described above, the mobile robot (20) is a robot that can drive on the roof and/or wall of an outdoor tank (2). The wheels of the mobile robot (20) are equipped with magnets (electromagnets) (21), and through these, the robot can drive on the roof and/or wall of the outdoor tank (2) made of metal.

In addition, the mobile robot (20) may include a built-in camera (22) for acquiring an area image, and the acquired area image may be transmitted to the robot control terminal (30). The robot control terminal (30) may transmit the received area image (31a) to the control server (40) and display the area image (31a) on the display unit (31). Meanwhile, the robot control terminal (30) may further include buttons, a joystick, etc. for the inspector managing the mobile robot (20) to manually control the driving of the mobile robot (20).

Again, referring to FIG. 1, the control server (40) is a processor-based device that can manage and monitor occurrence of safety accidents of the hazardous facility (2) based on the first and second safety monitoring data, thereby deriving safety monitoring results. If the safety monitoring results exceed a preset threshold, the control server (40) can provide a notification message to the manager of the hazardous facility (2). The manager who checks the notification message can take necessary measures for abnormal phenomena that may occur at the hazardous facility (2).

FIG. 4 is a diagram illustrating a schematic configuration of a control server (40) according to one embodiment of the present invention.

Referring to FIG. 4, the control server (40) may include a fourth wired/wireless communication module (41), a big data module (42), an artificial intelligence module (43), and a control module (44).

The fourth wired/wireless communication module (41) can receive the first and second safety monitoring data. The big data module (42) and the artificial intelligence module (43) can operate in conjunction with each other to derive safety monitoring results (risk inspection and inspection results, etc.) of the hazardous facility (2) based on the first and second safety monitoring data. The control module (44) can perform control operations of the CCTV device (10a), two or more IoT sensor modules (10b), and the mobile robot (20).

In particular, the big data module (42) is a module that constructs big data and can generate big data by performing collection, storage, processing, and analysis of the first and second safety monitoring data. In addition, the artificial intelligence module (43) can perform a learning process and an inference process based on the big data.

The big data module (42) and the artificial intelligence module (43) can operate in conjunction with each other. That is, the big data module (42) can transfer big data on which at least one of collection, storage, processing, and analysis has been performed to the artificial intelligence module (43), and the artificial intelligence module (43) can perform learning and inference based on the big data on which at least one of collection, storage, processing, and analysis has been performed. In addition, the artificial intelligence module (43) can transfer an artificial intelligence model related to at least one of collection, storage, processing, and analysis of the big data module (42) to the big data module, and the big data module (42) can perform at least one of collection, storage, processing, and analysis based on the artificial intelligence model.

Referring to FIG. 4, the big data module (42) may include a big data collection means (421), a big data storage means (422), a big data processing means (423), and a big data analysis means (424).

Big data collection is an important key step in determining the analysis quality when analyzing big data, and the big data collection means (421) can collect data by considering whether the target data can be collected and used, and whether it includes detailed items suitable for the purpose of use.

The big data storage means (422) can perform an operation of effectively storing and managing big data in order to extract useful information from big data acquired through a data collection process. That is, the big data storage means (422) safely and permanently stores the collected data in a manner suitable for use in analysis, and can store a large amount of data in various formats with high performance and, if necessary, search for data to modify, delete, or read desired content. The big data storage means (422) can preprocess and postprocess the collected data, and can store preprocessed and postprocessed data and data that has not been preprocessed and postprocessed using a preset storage model. The big data storage means (422) can store big data using a cloud server.

Data preprocessing can include filtering, type conversion, and cleansing. Filtering is the process of removing information that is not suitable for the purpose of data utilization. This shortens the analysis time and allows storage space to be utilized efficiently. Type conversion is the process of converting the type of data into a form that is easy to analyze. Cleansing is the process of correcting inconsistencies in collected data, processing missing values, and removing noise in the data.

Post-processing of data may include data transformation, integration, and reduction. Data transformation is the process of converting data collected in various formats into a consistent format for easy analysis, including smoothing, aggregation, generalization, normalization, and attribute/feature construction. Data integration is the process of combining interrelated data into one, detecting and deleting duplicate data through correlation analysis, and selecting one of the conflicting data. Data reduction is the process of reducing unnecessary data that is expected to not be used in the analysis, while maintaining the unique characteristics.

The big data processing means (423) can batch process and process in real time the data stored in the data storage means (422), and can store the batch-processed and real-time-processed data in the data storage means (422).

Big data processing means (423) must process data in a long-term and strategic manner to perform analysis based on large amounts of data, and has the highest processing complexity for various data sources, complex logic processing, large amounts of data processing, etc., rather than a simple processing model, and usually requires distributed processing technology. Big data processing means (423) can minimize delay by processing continuously occurring stream data in real time, and can show data analysis results to users in real time.

The big data analysis means (424) can analyze data processed in the data processing module (423) using a pre-set analysis model (i.e., an artificial intelligence model). The analysis model for analyzing big data may be a model that improves the algorithm of analysis techniques used in the fields of statistics and computer science, especially machine learning or data mining.

The artificial intelligence module (43) can perform a learning process and an inference process by applying big data to an artificial intelligence model corresponding to a machine learning model and/or a deep learning model. Machine learning is a field of artificial intelligence, a research field that gives computers the ability to learn without an explicit program. Machine learning is a technology that studies and builds systems and algorithms for learning, making predictions, and improving their own performance based on empirical data. Instead of executing strictly defined static program commands, machine learning algorithms build specific models to derive predictions or decisions based on input data.

The artificial neural network that constitutes deep learning may include multiple layers, and each of the layers may include multiple neurons (nodes). In addition, the artificial neural network may include synapses (edges) that connect neurons. In other words, the artificial neural network may refer to the overall model in which artificial neurons that form a network by combining synapses change the binding strength of the synapses through learning and have the ability to solve problems. The artificial neural network may generally be defined by the following three factors: a connection pattern between neurons in different layers, a learning process that updates the weights of the synapses, and an activation function that generates an output value from a weighted sum of inputs received from the previous layer. The artificial neural network may include network models such as DNN (Deep Neural Network), RNN (Recurrent Neural Network), BRDNN (Bidirectional Recurrent Deep Neural Network), MLP (Multilayer Perceptron), and CNN (Convolutional Neural Network), but is not limited thereto.

Meanwhile, if the hazardous facility (2) is an outdoor tank (2) storing a liquid-state hazardous substance, an abnormality in the verticality of the wall slope of the outdoor tank (2) may occur. That is, if the outdoor tank (2) storing the hazardous substance is in a normal state ((a) of FIG. 5), the wall of the outdoor tank (2) is arranged vertically with respect to the ground. However, as the storage of the hazardous substance continues, the pressure in the internal space of the outdoor tank (2) may change, and in particular, if a high negative pressure is generated in the internal space of the outdoor tank ((b) of FIG. 5), the structural change such as the wall slope of the outdoor tank (2) may occur, which may cause a safety accident. Therefore, the control server (40) can monitor the inclination of the wall of the outdoor tank (2) in real time to prevent a safety accident.

Hereinafter, the operation of the safety accident management system (1) that prevents accidents due to changes in the verticality of an outdoor tank (2) will be described in more detail with reference to FIG. 6, etc.

FIG. 6 is a drawing for explaining the operation of a safety accident management system (1) that monitors the slope of a wall surface of an outdoor tank (2) to prevent accidents according to one embodiment of the present invention.

Meanwhile, as described above, the big data module (42) and the artificial intelligence module (43) of the control server (40) can operate in conjunction with each other, and the first IoT sensor module (10b) equipped with a first inertial measurement sensor that measures the inclination of the wall surface of the outdoor tank (2) is installed in the visual surveillance area.

The CCTV device (10a) can capture an external image of the outdoor tank (2) and transmit it to the control server (40).

The control server (40) can analyze the received external image to infer the slope (i.e., vertical slope) of multiple points on the wall surface of the outdoor tank (2) based on the vertical line of the ground. Each of the multiple points can be a point on the wall surface of the outdoor tank (2) located apart from each other.

The control server (40) can infer whether there is a point where a verticality abnormality may occur on the wall surface of the outdoor tank (2) with a slope greater than a preset critical slope among the point-by-point slopes of the wall surface. For example, the critical slope may have a slope value of 2° or more and 10° or less.

If it is inferred that a point where a vertical abnormality may occur has occurred, the control server (40) can determine whether the point where a vertical abnormality may occur exists within a visual surveillance area or within a visual non-surveillance area.

If a point where a verticality abnormality may occur exists within a visual surveillance area, the control server (40) can calculate a first slope of the point where a verticality abnormality may occur based on a slope measured by a first inertial measurement device of at least one first IoT sensor module (10b) installed adjacent to the point where a verticality abnormality may occur. For example, if the point where a verticality abnormality may occur is located between the installation points of two first IoT sensor modules (10b), the first slope of the point where a verticality abnormality may occur may be an average of the slopes measured by each of the two first IoT sensor modules (10b).

Conversely, if a point where a verticality abnormality may occur exists within a visually unmonitored area, the control server (40) can generate a travel route map of the mobile robot (20) on the wall surface of the outdoor tank (2) that includes the point where a verticality abnormality may occur as a transit point, and can receive a second slope of the point where a verticality abnormality may occur measured by the second inertial measurement device of the mobile robot (20) that travels based on the travel route map.

The operation of the mobile robot (20) measuring the second slope of a point where vertical abnormality may occur and transmitting it to the control server (40) is described in more detail as follows.

The control server (40) can create a travel route map of the mobile robot (20) on the wall surface of the outdoor tank (2) that includes points where vertical abnormalities may occur as transit points, and transmit the created travel route map to the robot control terminal (30).

Here, the driving route map is a map that guides the driving of the mobile robot (20) using the main points of abnormality occurrence as transit points where the main types of abnormalities frequently occur in the outdoor tank (2), and can be created so that points where vertical abnormalities are likely to occur are further included in the main points of abnormality occurrence.

The robot control terminal (30) can control the mobile robot (20) based on the received driving path map, and in particular, can control the mobile robot (20) to measure the second slope at a point where verticality abnormality may occur based on the driving path map. The mobile robot (20) can measure the second slope through the second inertial measurement sensor under the control of the robot control terminal (30), and the measured second slope can be transmitted to the control server (40) via the robot control terminal (30).

Hereinafter, referring to Fig. 7, the operation of measuring the second slope at a point where verticality abnormality may occur by the mobile robot (20) under the control of the robot control terminal (30) is described as follows.

FIG. 7 is a drawing for explaining the operation of a robot control terminal (30) that controls a mobile robot (20) according to one embodiment of the present invention.

Meanwhile, the motion control signal generated from the robot control terminal (30) to control the motion of the mobile robot (20) may include first and second motion control signals. The first motion control signal may correspond to a manual driving command of the mobile robot (20) by the inspector. That is, the inspector may perform a driving command event through a button or joystick provided in the robot control terminal (30), and the robot control terminal (30) may generate a first motion control signal corresponding to the manual driving command based on the performance of the driving command event. The second motion control signal may be a signal that controls the motion of the mobile robot (20) without the intervention of the inspector. As an example, the second inertial measurement sensor of the mobile robot (20) may be defaulted to a power-saving state, and the second motion control signal may be a wake-up signal of the second inertial measurement sensor, a slope measurement command signal, etc.

Referring to (a) of Fig. 7, the mobile robot (20) can capture an area image (31a) acquired from a built-in camera in real time and transmit it to the robot control terminal (30), and the robot control terminal (30) can display the received area image (31a) on the display unit (31). Meanwhile, the robot control terminal (30) can transmit the area image (31a) as second safety monitoring data to the control server (40).

At this time, the robot control terminal (30) can further display the driving direction (71) of the mobile robot (20) based on the driving path map on the display unit (31) so that it overlaps with the area image (31a). The inspector can check the driving direction (71) and input a driving command event to the robot control terminal (30), and the robot control terminal (30) can generate a first motion control signal based on the input driving command event and transmit it to the mobile robot (20), and the mobile robot (20) can drive based on the first motion control signal.

Meanwhile, during the process of the mobile robot (20) driving on the wall, the robot control terminal (30) can predict in real time the time required to reach the point where the vertical deviation may occur from the current position of the mobile robot (20) based on the driving path map. Thereafter, the robot control terminal (30) can determine whether the predicted required time is shorter than a preset critical time. Here, the critical time can be set to any one time of 2 seconds or more and 5 seconds or less, for example, 3 seconds.

Referring to (b) and (c) of FIG. 7, when the predicted required time becomes shorter than the critical time, i.e., when the predicted required time reaches the critical time, the robot control terminal (30) can display a guidance message (72) for measuring inclination at a point where verticality abnormality may occur through the second inertial measurement sensor on the display unit (31) for a preset display period. At this time, in order for the inspector to confirm that the second inclination is being measured, the display period can be set to be shorter than the critical time. As an example, when the critical time is 3 seconds, the display period can be set to 2 seconds. That is, when the critical time arrives, the robot control terminal (30) can display the guidance message (72) for measuring inclination for 2 seconds from the critical time.

Here, the slope measurement guidance message (72) may include a first slope measurement guidance message (72a) and a second slope measurement guidance message (72b).

The first slope measurement guidance message (72a) may be a slope measurement guidance message (72) set as a default. Here, in order for the inspector to easily check the message, the robot control terminal (30) may display the first slope measurement guidance message (72a) in a relatively large size (i.e., the first size) in the central portion of the area image (31). For example, referring to (b) of FIG. 7, the first slope measurement guidance message (72a) corresponding to “the wall slope will be measured soon” may be displayed in the first size in the central portion of the area image (31a).

The second slope measurement guidance message (72b) may be a slope measurement guidance message (72) displayed when a specific condition occurs. Here, the specific condition may be a condition in which an abnormality other than a vertical abnormality occurs within the area image (31a) immediately before or during the display period. That is, in order for the inspector to check the occurrence of the other abnormality and the message together, the robot control terminal (30) may display the second slope measurement guidance message (72a) in a relatively small size (i.e., the second size) on the edge area of the area image (31). As an example, referring to (c) of FIG. 7, the second slope measurement guidance message (72b) corresponding to “wall slope measurement scheduled” may be displayed in the second size on the edge area of the area image (31a).

Referring to the above, if it is determined that no abnormality other than verticality abnormality has occurred within the area image (31a) immediately before and during the display period, the robot control terminal (30) may display a first inclination measurement guidance message (72a) having a first size at the center of the area image. Alternatively, if it is determined that the other abnormality has occurred immediately before the display period (i.e., immediately before the critical time period) before the display period, the robot control terminal (30) may display a second inclination measurement guidance message (72b) having a second size smaller than the first size at the edge of the area image (31a). Alternatively, if it is determined that the other abnormality has occurred at a first time within the display period, the robot control terminal (30) may display the first inclination measurement guidance message during the period between the start time of the display period and the first time, and may display the second inclination measurement guidance message during the period between the first time and the end time of the display period. That is, the robot control terminal (30) can reduce the first size inclination measurement guidance message (72) displayed in the central portion of the area image (31a) to a second size and then move it to the edge portion of the area image (31a) to display it.

Meanwhile, whether or not the above-mentioned other abnormality has occurred can be determined by the control server (40). That is, the control server (40) that receives the area image (31a) in real time can determine whether or not the above-mentioned other abnormality has occurred by analyzing the area image (31a) using an artificial intelligence model, etc., and if the above-mentioned other abnormality has occurred, it can notify the robot control terminal (30).

Thereafter, when the mobile robot (20) reaches a point where an abnormal verticality may occur, the robot control terminal (30) may transmit a second motion control signal to the mobile robot (20) to stop the mobile robot (20) for a preset stop time and then change the second inertial measurement sensor included in the mobile robot (20) to a wake-up state to measure the second inclination of the wall surface of the outdoor tank (2). Based on the second motion control signal, the mobile robot (20) may stop at a point where an abnormal verticality may occur, measure the second inclination, and transmit the measured second inclination to the control server (40) via the robot control terminal (30). When the stop time has elapsed, the second inertial measurement sensor of the mobile robot (20) may be changed to a power-saving state.

At this time, during the stop time of the mobile robot (20), the inspector's manual driving command may be input to the robot control terminal (30) to generate the first motion control signal. In this case, the robot control terminal (30) may not transmit the first motion control signal to the mobile robot (20) but may transmit only the second motion control signal to the mobile robot (20). That is, when the inspector inputs a manual driving command during the stop time, the mobile robot (20) may not follow the inspector's manual driving command and may measure the second slope at a point where verticality abnormality may occur. Accordingly, the measurement of the second slope is not interrupted and the measurement of the second slope can be performed efficiently.

Meanwhile, when the first or second slope is acquired, the control server (40) can compare the first or second slope with the critical slope, and further compare the first or second slope with the slope inferred from the point where the verticality abnormality may occur.

Here, the control server (40) can determine that an abnormality in the wall verticality has occurred at a point where an abnormality in verticality may occur if the first or second slope is determined to be greater than a critical slope, and can determine that an abnormality in the wall verticality does not occur at a point where an abnormality in verticality may occur if the first or second slope is determined to be less than a critical slope. In addition, the control server (40) can determine that the inference of the slope at the point where an abnormality in verticality may occur is accurate if the inferred slope at the point where an abnormality in verticality may occur is the same as the first or second slope within an error range, and can determine that the inference of the slope at the point where an abnormality in verticality may occur is inaccurate if the inferred slope at the point where an abnormality in verticality may occur is not the same as the first or second slope within an error range.

Therefore, if the first or second slope is greater than the critical slope and is equal to the inferred slope of the possible point of verticality abnormality within the error range, the control server (40) can provide the administrator with a first notification message notifying that an abnormality in the verticality of the wall surface has occurred at the possible point of verticality abnormality. At this time, the control server (40) does not perform re-learning of the artificial intelligence model. In addition, if the first or second slope is greater than the critical slope but is not equal to the inferred slope of the possible point of verticality abnormality within the error range, the control server (40) can provide the first notification message to the administrator and re-learn the artificial intelligence model to update the artificial intelligence model. In addition, if the first or second slope is less than the critical slope, the control server (40) can re-learn the artificial intelligence model to update the artificial intelligence model. At this time, the first notification message is not provided to the administrator.

Meanwhile, as the artificial intelligence model is updated, the inference of the slope at the point where the verticality abnormality may occur can be made more accurate. Accordingly, the verticality abnormality of the outdoor tank (2) can be accurately determined using only the external image acquired from the CCTV device (10a) in the future.

In short, according to the present invention, based on big data and artificial intelligence technology, monitoring of a hazardous facility (2) is possible without the manager having to directly inspect the hazardous facility (2), and thus, safety accidents at the hazardous facility (2) can be drastically reduced.

In addition, according to the present invention, it is possible to prevent safety accidents by identifying and responding to risk factors of hazardous facilities (2) in advance through digital system analysis alone, and to secure competitiveness by reducing costs such as unnecessary personnel recruitment due to digitalization.

As described above, the present invention has been described by specific matters such as specific components, limited examples, and drawings, but this has only been provided to help the overall understanding of the present invention, and the present invention is not limited to the above examples, and those with common knowledge in the field to which the present invention pertains can make various modifications and variations from this description. Therefore, the spirit of the present invention should not be limited to the described examples, and all things that are equivalent or equivalent to the claims as well as the claims described below are considered to belong to the scope of the spirit of the present invention.

Claims (7)

A plurality of safety surveillance acquisition devices installed directly on or near a hazardous facility to acquire first safety surveillance data of said hazardous facility;
A mobile robot that obtains second safety surveillance data of the above-mentioned hazardous facility by driving in an unmonitored area where visual safety surveillance of the above-mentioned hazardous facility is impossible; and
A control server in which a big data module and an artificial intelligence module operate in conjunction to derive safety surveillance results of the hazardous facility based on the first and second safety surveillance data; including,
The above hazardous facility includes an outdoor tank storing liquid hazardous substances,
The above control server provides a notification message to the manager of the hazardous facility when the safety monitoring result exceeds the preset threshold.
The above-described plurality of safety surveillance acquisition devices include a camera device installed near the outdoor tank and two or more IoT sensor modules directly installed in the outdoor tank, wherein the camera device acquires an external image of the outdoor tank corresponding to the first safety surveillance data, and the two or more IoT sensor modules have at least one sensor among a first inertial measurement sensor, a temperature and humidity sensor, a liquid amount sensor, and a gas sensor to acquire a sensing value of the at least one sensor corresponding to the first safety surveillance data, and among the two or more IoT sensor modules, a first IoT sensor module including the first inertial measurement sensor is installed only in a surveillance zone, which is an area excluding the non-surveillance zone, and
The above mobile robot is a wall-running robot that runs on the wall surface of the outdoor tank, and is equipped with a built-in camera and a second inertial measurement sensor that capture an area image, which is an image of an area of the wall surface of the outdoor tank, and the second safety monitoring data is a sensing value of the area image and the second inertial measurement sensor.
Through the linked operation of the big data module and the artificial intelligence module, the control server analyzes the external image to infer the slope of multiple points on the wall surface of the outdoor tank based on the vertical line of the ground, and infers whether there is a point where the verticality of the wall surface may be abnormal, having a slope greater than a preset threshold slope among the slopes of multiple points.
A risk facility management system based on big data and artificial intelligence.
delete In the first paragraph,
If the above vertical abnormality occurrence point exists within the above surveillance area, the control server,
The first slope of the point where the verticality abnormality may occur is calculated based on the slope measured by the first inertial measurement device of the first IoT sensor module installed adjacent to the point where the verticality abnormality may occur,
If the first slope is greater than the critical slope and is equal to the inferred slope of the point where the verticality abnormality can occur within the error range, a first notification message is provided to the administrator notifying that an abnormality in the verticality of the wall surface has occurred at the point where the verticality abnormality can occur.
If the first slope is greater than the critical slope but is not equal to the inferred slope of the point where the verticality abnormality may occur within the error range, the first notification message is provided to the administrator, and the inference of the slope at the point where the verticality abnormality may occur is judged to be inaccurate, and the artificial intelligence model of the artificial intelligence module is retrained to update the artificial intelligence model.
If the first slope is smaller than the critical slope, the inference of the slope is judged to be inaccurate, and the artificial intelligence model is retrained to update the artificial intelligence model.
A risk facility management system based on big data and artificial intelligence.
In the first paragraph,
If the above vertical abnormality occurrence point exists within the above non-monitored area, the control server,
Generating a driving path map of the mobile robot on the wall surface of the outdoor tank that includes the point where the above verticality abnormality may occur as a transit point, and receiving a second slope of the point where the above verticality abnormality may occur measured by the second inertial measurement device of the mobile robot that drives based on the driving path map,
If the second slope is greater than the critical slope and is within the error range of the inferred slope of the point where the verticality abnormality can occur, a first notification message is provided to the manager to inform that an abnormality in the verticality of the wall surface has occurred at the point where the verticality abnormality can occur.
If the second slope is greater than the critical slope but is not equal to the inferred slope of the point where the verticality abnormality may occur within the error range, the first notification message is provided to the administrator, and the inference of the slope at the point where the verticality abnormality may occur is determined to be inaccurate, and the artificial intelligence model of the artificial intelligence module is retrained to update the artificial intelligence model.
If the second slope is smaller than the critical slope, the inference of the slope is judged to be inaccurate, and the artificial intelligence model is retrained to update the artificial intelligence model.
A risk facility management system based on big data and artificial intelligence.
In paragraph 4,
Further comprising a robot control terminal having a display section, receiving the driving route map from the control server, and transmitting a motion control signal of the mobile robot to the mobile robot;
The above motion control signal includes a first motion control signal corresponding to a manual driving command of the mobile robot by the inspector, and a second motion control signal that controls the motion of the mobile robot without intervention of the inspector.
The second inertial measurement sensor of the above mobile robot is defaulted to a power-saving state,
The above robot control terminal is,
The area image obtained from the built-in camera of the mobile robot is displayed on the display unit, and the driving direction of the mobile robot based on the driving path map is further displayed on the display unit so as to overlap with the area image.
If the predicted time required to reach the point where the verticality abnormality may occur from the position of the mobile robot running on the wall is shorter than the preset threshold time, a guidance message for measuring the slope at the point where the verticality abnormality may occur through the second inertial measurement sensor is displayed on the display unit for a preset display period,
When the mobile robot reaches the point where the verticality abnormality can occur, the mobile robot is stopped for a preset stop time, and the second inertial measurement sensor is changed to a wake-up state to measure the second inclination of the wall, and the second motion control signal is transmitted to the mobile robot, but when a manual driving command of the inspector is input during the stop time and the first motion control signal is generated, the first motion control signal is not transmitted to the mobile robot and only the second motion control signal is transmitted to the mobile robot, and when the stop time has elapsed, the second inertial measurement sensor is changed to the power-saving state.
A risk facility management system based on big data and artificial intelligence.
In paragraph 5,
The above robot control terminal is,
If it is determined that no abnormality other than the vertical abnormality has occurred within the area image immediately before and during the above display period, a first slope measurement guidance message having a first size is displayed in the central portion of the area image among the slope measurement guidance messages.
If it is determined that the above other abnormality occurs just before the above display period arrives, a second slope measurement guidance message having a second size smaller than the first size is displayed at the edge of the area image among the slope measurement guidance messages.
If it is determined that the above other abnormality occurs at the first time within the above display period, the first slope measurement guidance message is displayed during the period between the start time of the above display period and the first time, and the second slope measurement guidance message is displayed during the period between the first time and the end time of the above display period.
A risk facility management system based on big data and artificial intelligence.
In the first paragraph,
The above big data module transmits big data on which at least one of the collection, storage, processing and analysis of the first and second safety monitoring data is performed to the artificial intelligence module, and the artificial intelligence module performs learning and inference based on the big data.
The above artificial intelligence module transmits an artificial intelligence model related to the collection, storage, processing and analysis of the big data module to the big data module to perform at least one of the collection, storage, processing and analysis.
A risk facility management system based on big data and artificial intelligence.