KR102403599B1 - Lng bunkering monitoring system - Google Patents

Abstract

The present invention provides an LNG bunkering monitoring system, which detects a dangerous situation occurring around a user during an LNG bunkering process and forms different alarm signals according to the dangerous situation. An LNG bunkering monitoring system according to an embodiment of the present invention is an LNG bunkering monitoring system which monitors a process of bunkering an LNG to a ship. The system comprises: a user who is located at a work site at which the LNG bunkering process is performed and wears a plurality of safety equipment; a user terminal which receives a work image of the work site captured by the user, user location information, and a plurality of sensing information measured by the plurality of safety equipment or measured within the work site, transmits the received image and information to a server, and receives a first alarm signal from the server; and the server which detects a dangerous situation occurring around the work site in which the user is located by using the work image, the location information, and the plurality of sensing information, forms the first alarm signal which is different according to the dangerous situation, and transmits the first alarm signal to the user terminal.

Description

LNG BUNKERING MONITORING SYSTEM

The present invention relates to an LNG bunkering monitoring system, and more particularly, to an LNG bunkering monitoring system that detects a dangerous situation occurring in the vicinity of a user during a bunkering process and forms different alarm signals according to the dangerous situation.

In general, bunkering mainly refers to the storage and transmission of marine fuel oil known as bunker oil. In ship refueling, large quantities of fuel may be temporarily stored on barges or other containers for fuel transfer from shore to ship, so that refueling may be done directly from docks or other port facilities, or by barges or other refueling vessels. It may be achieved by supplying transported fuel.

On the other hand, LNG (liquefied natural gas) is a colorless and transparent liquid that can be obtained by cooling and liquefying natural gas containing methane to about -162 ° C. It has a volume of about 1/600 compared to natural gas. .

Since the liquefaction temperature of natural gas is cryogenic temperature of -163°C at normal pressure, LNG evaporates easily even when the temperature is slightly higher than normal pressure -163°C. A large amount of boil-off gas is generated according to the change.

Accordingly, it can be easily exposed to mechanical defects due to gas leakage and ultra-low temperature, and even if the exposure to defects is recognized, there is an environment in which it is difficult for the operator to directly solve the problem, and measures to prevent this are required.

Korean Patent Publication No. 10-2130223 (LNG bunkering monitoring system, announced on July 3, 2020)

The present invention provides an LNG bunkering monitoring system that detects a dangerous situation occurring in the vicinity of a user during the bunkering process and forms different alarm signals according to the dangerous situation.

An LNG bunkering monitoring system according to an embodiment of the present invention is an LNG bunkering monitoring system that monitors a process of bunkering LNG on a ship, is located at a work site where the bunkering process of LNG is performed, and a user wearing a plurality of safety equipment ; The work image of the work site photographed by the user, the user's location information, and a plurality of sensing information measured by the plurality of safety equipment or measured within the work site are received and transmitted to a server, and the a user terminal for receiving a first alarm signal from the server; And by using the work image, the location information, and the plurality of sensing information to detect a dangerous situation occurring in the vicinity of the work site in which the user is located, and to form the first alarm signal different according to the dangerous situation, the user and the server transmitting to the terminal.

In addition, the dangerous situation is, when the gas concentration in the vicinity of the work site where the user is located is sensed more than a first threshold value, when the temperature around the work site where the user is located is reduced by more than a second threshold value, It may include any one or more of a case in which the humidity around the work site where the user is located is changed by more than a third threshold value, and a case in which the flow rate or flow rate of the LNG in the bunker is changed by more than a fourth threshold value.

In addition, the plurality of safety equipment may include any one or more of a hard hat, work clothes, and goggles.

In addition, the helmet may include: a first photographing unit configured to form the work image by photographing a work image of the work site; location recognition means for forming location information of the user; a user recognition unit for authenticating the user; second output means for receiving the first alarm signal and outputting a voice signal; and a third output means for displaying whether the user has passed authentication.

In addition, the work clothes, a gas detector for detecting the gas concentration in the vicinity of the work site where the user is located; a temperature sensor for sensing a temperature around the work site where the user is located; And it may include a humidity sensor for detecting the humidity around the work site where the user is located.

In addition, the goggles may include an augmented reality module capable of displaying the augmented reality image generated by the server.

In addition, the plurality of sensing information may include any one or more of wind direction and wind speed information of the work site where the user is located, flow rate information of the LNG in the bunkering, and flow velocity information of the LNG in the bunkering.

In addition, the server may generate a valve control signal for stopping the supply of the LNG in the bunker according to the dangerous situation.

In addition, at a location spaced apart from the user, a second photographing unit for acquiring a work image of the work site in which the user is located; and a first output means capable of outputting the announcement received from the server at a location spaced apart from the user.

In addition, the second photographing unit obtains the face image of the user at a location spaced apart from the user, and the server extracts the pupil object by applying the face image to the pre-learned first neural network, Whether the user is watching the work site is determined using the pupil object, and when it is determined that the user is not looking at the work site, a second alarm signal may be generated.

In addition, the server may analyze the work image of the work site received from the user terminal using a machine learning algorithm to detect the dangerous situation, and form the first alarm signal when the dangerous situation is detected have.

In addition, the server may further include a bunkering operator server or a pre-scanning server for receiving and displaying the work image, the location information, the plurality of sensing information, and the first alarm signal from the server.

According to the embodiments of the present invention, by detecting a dangerous situation in the LNG bunkering operation and notifying the operator and the work site, the operator can easily recognize the dangerous situation and evacuate, thereby preventing human accidents. In addition, it is possible to prevent economic loss by detecting a dangerous situation in the LNG bunkering operation in advance.

1 is an exemplary view showing an LNG bunkering monitoring system according to an embodiment of the present invention.
2 is a view of a user wearing equipment according to an embodiment of the present invention.
3 is an exemplary diagram showing the configuration of a server according to an embodiment of the present invention.
4 is a diagram for explaining learning of a neural network according to an embodiment of the present invention.
5 is a flowchart showing the procedure of the LNG bunkering monitoring method according to an embodiment of the present invention.

Embodiments of the present invention are exemplified for the purpose of explaining the technical spirit of the present invention. The scope of the rights according to the present invention is not limited to the embodiments presented below or detailed descriptions of these embodiments.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same or corresponding components are assigned the same reference numerals. In addition, in the description of the embodiments below, overlapping description of the same or corresponding components may be omitted. However, even if descriptions regarding components are omitted, it is not intended that such components are not included in any embodiment.

1 is an exemplary view showing an LNG bunkering monitoring system according to an embodiment of the present invention, and FIG. 2 is a view of equipment wearing by a user according to an embodiment of the present invention.

As shown in FIG. 1 , the LNG bunkering monitoring system 100 is a system for monitoring the process of bunkering LNG to a ship. Specifically, located at the job site where the LNG bunkering process is performed, a user (operator, 110) wearing a number of safety equipment, user authentication means 120 installed at various locations within the job site, valve 125, flow meter ( 130), first output means 135, flow meter 140, server 145, concentration meter 150, bunkering operator server 155, pipe 160, ship scan server 165, second photographing unit ( 170), the control box 175, the wind direction anemometer 180, the first vessel 185 receiving bunkering, that is, receiving LNG, the fuel tank 190, and the second vessel performing bunkering, that is, supplying LNG ( 195) may be included.

In this case, the second vessel 195 may be an LNG bunkering vessel, but is not limited thereto, and may be an onshore LNG storage tank for supplying LNG or a tank lorry vehicle including an LNG storage tank. Hereinafter, it will be described as the second vessel 195 for convenience.

The user 110 is located at a work site where the LNG bunkering process is performed, and may wear a number of safety equipment, for example, a hard hat 111 , work clothes 112 , and goggles 113 . In this case, the hard hat 111 may include a first photographing unit 114 , a second output unit 115 , a third output unit 116 , a location recognition unit 117 , and a user recognition unit 118 . In addition, the work clothes 112 may include a gas detector 119-1, a temperature sensor 119-2, and a humidity sensor 119-3. In addition, the goggles 113 may include an augmented reality module capable of displaying the augmented reality image generated by the server 145 .

The first photographing unit 114 may generate a work image by photographing a work site attached to the safety helmet 111 to perform LNG bunkering. According to an embodiment, the first photographing unit 114 may include, but is not limited to, a handheld camera, an action cam, and the like. The first photographing unit 114 may transmit the generated work image to the user terminal (UE) using a short-distance communication method.

The location recognition means 117 may be attached to the hard hat 111 to generate location information of the user 110 . According to an embodiment, the location recognition means 117 may include, but is not limited to, a Global Positioning System (GPS) receiver, a Galileo receiver, a Glonass receiver, and the like. The location recognition means 117 may transmit the generated location information to the user terminal (UE) using a short-range communication method.

The user recognition unit 118 may determine whether the person is attached to the hard hat 111 and is allowed to access the work site, whether the man is permitted to perform the work, and the like. According to an embodiment, the user recognition unit 118 generates user authentication information by recognizing the user authentication means 120 installed at various locations within the work site, and transmits the generated user authentication information through the user terminal (UE). It can be transmitted to the server 145 and it can be determined whether the user is an authorized manpower to work at a specific location within the work site where the user authentication means 120 is installed. For example, the user recognition unit 118 may include, but is not limited to, a radio frequency identification (RFID) reader, a bar code reader, a quick response (QR) code reader, and the like.

The second output means 115 may receive the first alarm signal from the server 145 and output a voice signal. According to an embodiment, the second output means 115 may output different types of voice signals according to the type of dangerous situation.

The third output means 116 may display whether the user has passed authentication. According to an embodiment, the third output means 116 receives and displays the determination result as to whether or not a person is allowed to work at a specific location within the job site where the user authentication means 120 is installed from the server 145 . can do. For example, the third output means 116 displays a green color when the server 145 determines that the user authentication means 120 is a person permitted to work at a specific location within the work site installed, and the user authentication means In the case of personnel who are not permitted to work at a specific location within the work site where 120 is installed, red may be displayed so that other users can easily recognize whether or not user authentication has passed.

The gas detector 119-1, the temperature sensor 119-2, and/or the humidity sensor 119-3 may form a plurality of pieces of sensing information. According to an embodiment, the gas detector 119-1 may measure the gas concentration around the work site where the user 110 is located, and the temperature sensor 119-2 may measure the gas concentration around the work site where the user 110 is located. can measure the temperature, and the humidity sensor 119-3 can measure the humidity around the work site where the user 110 is located. For example, the gas detector 119-1, the temperature sensor 119-2, and/or the humidity sensor 119-3 may transmit the formed plurality of sensing information to the server 145 through the user terminal (UE). have.

The goggles 113 may receive the augmented reality image from the server 145 through the user terminal (UE) and display it using the augmented reality module. That is, by receiving information on specific work content or work range from the server 145 and displaying it as an augmented reality image on the goggles 113 , the work efficiency of the user 110 can be increased. According to an embodiment, the goggles 113 may communicate with a user terminal (UE) using a short-range communication method, but is not limited thereto.

The user terminal (UE) is a terminal that the user 110 carries at the LNG bunkering work site during the LNG bunkering operation. 119-1), the temperature sensor 119-2, the humidity sensor 119-3, and the server 145 are connected to be communicatively connected, and the work image, location information, and a plurality of sensing information are transmitted to the server 145 and , the first alarm signal formed by the server 145 may be received. According to an embodiment, the user terminal (UE) is a smart phone, a tablet personal computer (PC), a mobile phone, a video phone, an e-book reader (e-book reader), a desktop PC ( Desktop personal computer), laptop personal computer (PC), netbook computer, personal digital assistant (PDA), portable multimedia player (PMP), or wearable device (HMD) mounted-device), electronic apparel, electronic bracelets, electronic necklaces, electronic accessories, electronic tattoos, or smart watches.

The user terminal UE includes a first capturing unit 114 , a first outputting unit 135 , a second outputting unit 115 , a location recognizing unit 117 , and a user recognizing unit 118 worn by the user 110 . ), the gas detector 119-1, the temperature sensor 119-2, the humidity sensor 119-3, and the like to communicate using a short-distance communication method, and communicate with the server 145 using a wireless communication method. can do. According to an embodiment, the user terminal (UE) may include, but is not limited to, a Bluetooth module, a mobile communication module, an infrared communication module, and the like, and is not limited thereto, and is not limited thereto, and is not limited to eMBB (enhanced mobile broadband) and URLLC (Ultra Reliable Low-Latency Communications). ), MMTC (Massive Machine Type Communications), LTE (long-term evolution), LTE-A (LTE Advance), UMTS (Universal Mobile Telecommunications System), GSM (Global System for Mobile communications), CDMA (code division multiple access) , WCDMA (wideband CDMA), WiBro (Wireless Broadband), WiFi (wireless fidelity), Bluetooth (Bluetooth), NFC (near field communication), GPS (Global Positioning System) or GNSS (global navigation satellite system) Wireless communication can be performed. In addition, the user terminal (UE), USB (universal serial bus), HDMI (high definition multimedia interface), RS-232 (recommended standard232), or POTS (plain old telephone service) according to a method such as to perform wired communication. have.

The server 145 receives the work image, location information, and a plurality of sensing information from the user terminal (UE) to detect a dangerous situation occurring in the vicinity of the work site where the user 110 is located, and a first different according to the dangerous situation An alarm signal can be formed. Here, the plurality of sensing information includes gas concentration around the work site where the user 110 is located, the temperature around the work site where the user 110 is located, the humidity around the work site where the user 110 is located, and the user At least one of wind direction and wind speed information of the work site where 110 is located, flow rate information of LNG in bunkering, and flow velocity information of LNG in bunkering may be included, but is not limited thereto.

According to one embodiment, the server 145, when the gas concentration in the vicinity of the work site in which the user 110 is located is sensed above a first threshold (eg, 100 ppm), the user 110 is located When the ambient temperature of the site decreases by more than a second threshold value (eg, 10° C.), the humidity around the work site where the user 110 is located fluctuates by more than a third threshold value (eg, 15%) , the first alarm signal may be formed in at least one of the cases in which the flow rate or flow rate of LNG during bunkering fluctuates by more than a fourth threshold value (eg, 20%), but is not limited thereto.

According to another embodiment, the server 145 detects a dangerous situation by analyzing the work image received from the user terminal (UE) using a machine learning algorithm such as deep learning, and detects the dangerous situation. In this case, the first alarm signal may be formed. For example, the server 145 extracts a work object such as a connection part of a pipe 160 through which LNG is supplied through image analysis of the work image, and an abnormality occurs in the extracted work object (eg, a pipe A first alarm signal may be formed when the occurrence of cracks in the connection part or occurrence of splaying) is detected.

In addition, the server 145 may receive the user authentication information from the user terminal (UE) and determine whether the user is a person permitted to work at a specific location in the work site where the user authentication means 120 is installed, and , the determination result may be transmitted to the user terminal (UE).

The server 145 may generate a valve control signal for stopping the supply of LNG during bunkering according to a dangerous situation during the LNG bunkering process. According to an embodiment, the opening or closing of the valve 125 installed in the pipe 160 through which LNG is supplied may be automatically controlled by the valve control signal generated by the server 145 .

According to an embodiment, the server 145 includes a database for storing information transmitted from a user terminal (UE), a learning unit that performs machine learning such as deep learning using information stored in the database, and a learning unit Based on the machine learning result, the valve 125, the first output means 135, the second output means 115, the third output means 116, etc. for the LNG bunkering operation are controlled, and the user terminal (UE) is used. It may include a control unit for transmitting the first alarm signal, but the configuration of the server 145 is not limited thereto. For example, the server 145, when a gas leak in the vicinity of the work site where the user 110 is located is detected, when the temperature around the work site where the user 110 is located sharply drops, the user 110 When the humidity in the vicinity of the location of the work site changes rapidly, when the flow rate or flow rate of LNG during bunkering is rapidly changed, etc., the first alarm signal is transmitted to the second output means 115 through the user terminal (UE) to make an alarm sound Control to output and guide voice, and control to output different alarm tones and guide voices according to the type of dangerous situation, so that the user 110 intuitively determines what kind of situation it is even after hearing only the alarm tone and guide voice, and It can help you get out of there quickly. In addition, the database can store information on safety standards (first to fourth thresholds) of the work site, work list, work site situation by work date, work site situation by work content, etc. to determine work content and progress can do. In addition, the learning unit performs machine learning using big data for LNG bunkering, determines whether safety standards are met and whether a dangerous situation occurs, and transmits an alarm sound and a voice guidance to the user 110, , this may encourage the user 110 to manually shut off the valve 125 installed in the pipe 160 through which the LNG is supplied, or control to automatically shut off the valve 125 by forming a valve control signal You may. According to an embodiment, the learning unit may continuously update the safety standard of the work site by machine learning the continuously accumulated information of the database.

In addition, the server 145 may monitor the work environment through the second photographing unit 170 installed at the work site where the user 110 is located at a location spaced apart from the user 110 , and the third output means 116 . ), when the user 110 whose green light is on is working, it can be determined that the work is performed by the designated user 110 . In addition, the server 145, the third output means 116, when the red light or no color is lit when the user 110 is working, or the user 110 who does not wear the hard hat 111 In the case of work, a warning broadcast may be transmitted through the first output means 135 installed in the work site.

In addition, the server 145 may generate an augmented reality image representing descriptions or pictures capable of guiding the LNG bunkering operation, and may transmit the generated augmented reality image to a user terminal (UE) through a network.

The user authentication means 120 is installed at various locations within the work site, and a procedure for authenticating whether the user is a designated worker by the user recognition unit 118 possessed by the user 110 may be performed. According to an embodiment, the user authentication means 120 may include, but is not limited to, an RFID tag, a barcode, a QR code, and the like.

The valve 125 may be installed in the pipe 160 through which the LNG is supplied to stop or resume the LNG supply. According to an embodiment, the valve 125 may receive the valve control signal generated by the server 145 to stop or resume the LNG supply.

The flow meter 130 is installed in the pipe 160 through which the LNG is supplied and can measure the flow rate of the LNG to be bunkered, and the flow meter 140 is installed in the pipe 160 to which the LNG is supplied and the LNG to be bunkered. can measure the flow rate, and the concentration meter 150 is installed in the pipe 160 through which the LNG is supplied and can measure the concentration of the LNG to be bunkered.

The first output means 135 may output the first alarm signal as a voice when a dangerous situation occurs during the LNG bunkering process. According to an embodiment, the first output means 135 may include, but is not limited to, speakers installed at various locations within the work site. For example, the first output means 135 may receive the first alarm signal from the server 145 and output a notice informing that a dangerous situation has occurred.

The second photographing unit 170 may acquire a work image of the work site where the user 110 is located at a location spaced apart from the user 110 . According to an embodiment, the second photographing unit 170 may include a closed circuit television (CCTV) installed at various locations within the work site, but is not limited thereto.

The control box 175 may be located at the job site to allow the user 110 to control the LNG bunkering process. According to an embodiment, the control box 175 may control the valve 125 installed in the pipe 160 through which the LNG is supplied to stop or resume the LNG supply.

The wind direction anemometer 180 may be installed at various locations within the work site to measure the wind direction and wind speed to form sensing information. According to an embodiment, the sensing information formed by the wind direction anemometer 180 may be transmitted to the user terminal (UE) using a short-range communication method.

The first ship 185 is a ship that uses LNG as a fuel for ship propulsion, and can receive LNG from the second ship 195 , and the fuel tank 190 is supplied to the second ship 195 . It may be a tank capable of storing LNG. The fuel tank 190 may be a membrane type or an independent type tank, but is not limited thereto.

The second vessel 195 may include a supply vessel (bunkering vessel) that supplies LNG to the first vessel 185 , but as described above, is not limited thereto, and an onshore LNG storage tank for supplying LNG It may also be a tank lorry vehicle including an LNG storage tank.

In other words, although the LNG bunkering monitoring system according to the present invention is exemplified as a ship to ship method, it can also be applied to a tank lorry (or tank trailer) to ship or terminal to ship. can

delete

The bunkering operator server 155 and/or the pre-inspection server 165 receives the work image, location information, a plurality of sensing information, and the first alarm signal from the server 145 and displays it to display the bunkering operation manager and/or ship scan Managers can monitor the LNG supply process on board and whether dangerous situations occur.

3 is an exemplary diagram showing the configuration of a server according to an embodiment of the present invention.

As shown in FIG. 3 , the server 145 may include one or more processors 145 - 1 , one or more memories 145 - 2 and/or a transceiver 145 - 3 . As an embodiment, at least one of these components of the server 145 may be omitted, or other components may be added to the baby server 145 . Additionally or alternatively, some components may be integrated and implemented, or may be implemented as a singular or a plurality of entities. At least some of the internal and external components of the server 145 are connected to each other through a bus, general purpose input/output (GPIO), serial peripheral interface (SPI), or mobile industry processor interface (MIPI), etc. Data and/or signals may be exchanged.

The one or more processors 145 - 1 may control at least one component of the server 145 connected to the processor 145 - 1 by driving software (eg, instructions, programs, etc.). In addition, the processor 145-1 may perform various operations, processing, data generation, processing, etc. related to the present invention. In addition, the processor 145 - 1 may load data or the like from one or more memories 145 - 2 or store it in one or more memories 145 - 2 .

One or more processors 145-1 receive a work image, location information, and a plurality of sensing information from the user terminal (UE) through the transceiver 145-3 to detect a dangerous situation occurring around the user 110, It is possible to form a different first alarm signal according to the dangerous situation.

According to an embodiment, the one or more processors 145 - 1 extract the pupil object by applying the face image received from the second photographing unit 170 to the first learned neural network in advance, and extract the extracted pupil object. It is used to determine whether the operator is watching the work site, and when it is determined that the operator is not watching the work site, a second alarm signal may be generated. According to an embodiment, the one or more processors 145 - 1 may extract the pupil object from the received face images using a machine learning algorithm such as deep learning. A method of extracting the pupil object will be described later. Due to such a configuration, the bunkering monitoring system 100 generates a second alarm signal when the worker put to the work site does not watch the work site or is dozing, etc. It can provide an environment in which the worker can focus on work. .

The one or more processors 145-1 extract the coordinates of the most protruding pupil part from the extracted pupil object based on the face of the operator, and generate a repair line from the extracted coordinates, and the generated correction works If it does not intersect the work object, it is determined that the operator is not looking at the work object, and if the created water line intersects the work site or the work object in the work site, it can be determined that the operator is looking at the work site or work object. have.

The one or more memories 145 - 2 may store various data. Data stored in the memory 145 - 2 is data obtained, processed, or used by at least one component of the server 145 and may include software (eg, instructions, programs, etc.). Memory 145 - 2 may include volatile and/or non-volatile memory. In the present invention, commands or programs are software stored in the memory 145 - 2 , and various functions so that the operating system, applications and/or applications for controlling the resources of the server 145 can utilize the resources of the server 145 . may include middleware that provides

The one or more memories 145-2 store the work image, location information, a plurality of sensing information, and the first alarm signal formed by the one or more processors 145-1 received from the above-described user terminal (UE) through the network. can be saved Also, the one or more memories 145 - 2 may store instructions that, when executed by the one or more processors 145 - 1 , cause the one or more processors 145 - 1 to perform an operation.

In one embodiment, the server 145 may further include a transceiver 145 - 3 . The transceiver 145 - 3 may perform wireless or wired communication between a user terminal (UE) and/or other devices. For example, the transceiver 145-3 is an enhanced Mobile Broadband (eMBB), Ultra Reliable Low-Latency Communications (URLLC), Massive Machine Type Communications (MMTC), long-term evolution (LTE), LTE Advance (LTE-A) ), UMTS (Universal Mobile Telecommunications System), GSM (Global System for Mobile communications), CDMA (code division multiple access), WCDMA (wideband CDMA), WiBro (Wireless Broadband), WiFi (wireless fidelity), Bluetooth (Bluetooth), Wireless communication according to a method such as near field communication (NFC), global positioning system (GPS), or global navigation satellite system (GNSS) may be performed. For example, the transceiver 145-3 performs wired communication according to a method such as universal serial bus (USB), high definition multimedia interface (HDMI), recommended standard232 (RS-232), or plain old telephone service (POTS). can do.

In one embodiment, the one or more processors 145 - 1 may control the transceiver 145 - 3 to obtain information from the user terminal (UE). Information obtained from the user terminal (UE) may be stored in one or more memories 145 - 2 .

As an embodiment, the server 145 may be various types of devices. For example, the server 145 may be a portable communication device, a computer device, or a device according to a combination of one or more of the devices described above. The server 145 of the present invention is not limited to the devices described above.

Various embodiments of the server 145 according to the present invention may be combined with each other. Each of the embodiments may be combined according to the number of cases, and the embodiment of the server 145 made in combination also falls within the scope of the present invention. In addition, the above-described internal/external components of the server 145 according to the present invention may be added, changed, replaced, or deleted according to embodiments. In addition, the above-described internal/external components of the server 145 may be implemented as hardware components.

4 is a diagram for explaining learning of a neural network according to an embodiment of the present invention.

As shown in FIG. 4 , the learning apparatus may train the neural network 145 - 5 to recognize pupils included in the face image. According to an embodiment, the learning device may be a separate entity different from the server 145 , but is not limited thereto.

The neural network 145-5 includes an input layer 145-4 to which training samples are input and an output layer 145-6 to output training outputs, and learns based on differences between the training outputs and labels. can be Here, the labels may be defined based on body part information corresponding to the feature point object. The neural network 145 - 5 is connected as a group of a plurality of nodes, and is defined by weights between the connected nodes and an activation function that activates the nodes.

The learning apparatus may train the neural network 145 - 5 using a Gradient Decent (GD) technique or a Stochastic Gradient Descent (SGD) technique. The learning apparatus may use a loss function designed by the outputs and labels of the neural network.

The learning apparatus may calculate the training error using a predefined loss function. The loss function may be predefined with labels, outputs, and parameters as input variables, where the parameters may be set by weights in the neural network 145 - 5 . For example, the loss function may be designed in a Mean Square Error (MSE) form, an entropy form, or the like, and various techniques or methods may be employed in an embodiment in which the loss function is designed.

The learning apparatus may find weights affecting the training error by using a backpropagation technique. Here, the weights are relationships between nodes in the neural network 145 - 5 . The learning apparatus may use the SGD technique using labels and outputs to optimize the weights found through the backpropagation technique. For example, the learning apparatus may update the weights of the loss function defined based on the labels, outputs, and weights using the SGD technique.

According to an embodiment, the learning apparatus may obtain training face images and extract training feature point objects from the training face images. The learning apparatus may acquire pre-labeled information (first labels) for each training feature point object, and may acquire first labels indicating body part information predefined in the training feature point objects.

According to an embodiment, the learning apparatus may generate the first training feature vectors based on appearance features, pattern features, and color features of the training feature point objects. Various methods may be employed for extracting the feature.

According to an embodiment, the learning apparatus may obtain training outputs by applying the first training feature vectors to the neural network 145 - 5 . The learning apparatus may train the neural network 145 - 5 based on the training outputs and the first labels. The learning apparatus may train the neural network 145 - 5 by calculating training errors corresponding to the training outputs and optimizing the connection relationships of nodes in the neural network 145 - 5 in order to minimize the training errors. The server 145 may extract the pupil object from the face image using the neural network 145 - 5 on which the learning has been completed.

5 is a flowchart showing a procedure of a bunkering monitoring method according to an embodiment of the present invention. Although process steps, method steps, algorithms, etc. are described in a sequential order in the flowchart of FIG. 5, such processes, methods, and algorithms may be configured to operate in any suitable order. In other words, the steps of the processes, methods, and algorithms described in various embodiments of the present invention need not be performed in the order described in the present invention. Also, although some steps are described as being performed asynchronously, some of these steps may be performed concurrently in other embodiments. Further, the exemplification of a process by description in the drawings does not imply that the exemplified process excludes other changes and modifications thereto, and that the exemplified process or any of its steps is It is not meant to be essential to more than one, nor does it imply that the illustrated process is preferred.

As shown in FIG. 5 , in step S510 , a work image by the user is generated. For example, referring to FIGS. 1 to 4 , the first photographing unit 114 installed on the hard hat 111 of the user 110 may generate a working image of the user 110 during the LNG bunkering process.

In step S520, the user's location information is generated. For example, referring to FIGS. 1 to 4 , the location recognition means 117 installed on the hard hat 111 of the user 110 may generate location information of the user 110 . According to an embodiment, the location recognition means 117 may include a Global Positioning System (GPS) receiver, a Galileo receiver, a Glonass receiver, and the like.

In step S530, a plurality of sensing information is generated. For example, referring to FIGS. 1 to 4 , a plurality of gas detectors 119-1, a temperature sensor 119-2, and a humidity sensor 119-3 installed on the work clothes 112 of the user 110 are used. of sensing information may be generated.

In step S540, the work image, location information, and a plurality of sensing information is transmitted to the server. For example, referring to FIGS. 1 to 4 , the user terminal UE transmits the work image generated in step S510, the location information generated in step S520, and a plurality of sensing information generated in step S530 to the server ( 145) can be sent.

In step S550, a first alarm signal is formed. For example, referring to FIGS. 1 to 4 , the server 145 detects a dangerous situation occurring in the vicinity of the user 110 using the work image, location information, and a plurality of sensing information received in step S540, A different first alarm signal may be formed according to a dangerous situation and transmitted to the user terminal (UE).

Although the technical idea of the present invention has been described by various embodiments above, the technical idea of the present invention includes various substitutions, modifications, and changes that can be made within a range that can be understood by a person of ordinary skill in the art to which the present invention belongs. include It is also intended that such substitutions, modifications and variations be included within the scope of the appended claims.

100: LNG bunkering monitoring system 110: user
120: user authentication means 125: valve
130: flow meter 135: first output means
140: tachometer 145: server
150: Densitometer 155: Bunkering operator server
160: pipe 165: pre-owned server
170: second photographing unit 175: control box
180: wind direction anemometer 185: first vessel
190: fuel tank
195: second vessel 111: hard hat
112: work clothes 113: goggles
114: first photographing unit 115: second output means
116: third output means 117: location recognition means
118: user recognition unit 119-1: gas detector
119-2: temperature sensor 119-3: humidity sensor
145-1: Processor 145-2: Memory
145-3: transceiver 145-4: input layer
145-5: neural network 145-6: output layer

Claims (12)

An LNG bunkering monitoring system that monitors the process of bunkering LNG on a ship, comprising:
Located at the job site where the LNG bunkering process is made, the user 110 wearing a number of safety equipment;
The first work image of the work site photographed by the user 110, the user's location information, and a plurality of sensing information measured by the plurality of safety equipment or measured within the work site are received by the server a user terminal (UE) transmitting to (145) and receiving a first alarm signal from the server (145);
Using the first work image, the location information, and the plurality of sensing information to detect a dangerous situation occurring in the vicinity of the work site where the user 110 is located, the first alarm signal that is different according to the dangerous situation the server (145) forming and transmitting to the user terminal (UE);
a second photographing unit 170 for acquiring a second work image of the work site where the user 110 is located at a location spaced apart from the user 110; and
and a first output means (135) capable of outputting the announcement received from the server (145) at a location spaced apart from the user (110);
The plurality of safety equipment includes a hard hat 111 and work clothes 112,
The safety helmet 111 is,
A first photographing unit 114 which forms the first operation image by photographing a first operation image of the work site;
location recognition means 117 for forming location information of the user 110;
A user recognition unit 118 for performing authentication of the user 110,
second output means 115 for receiving the first alarm signal and outputting a voice signal; and
a third output means 116 for displaying whether the user 110 has passed authentication;
The work clothes 112,
A gas detector 119-1 for detecting the concentration of gas in the vicinity of the work site where the user 110 is located;
A temperature sensor 119-2 that senses the temperature of the work site in which the user 110 is located, and
and a humidity sensor 119-3 for detecting the humidity around the work site where the user 110 is located,
The user recognition unit 118 generates user authentication information by recognizing the user authentication means 120 installed at a plurality of predetermined locations in the work site, and transmits the generated user authentication information through the user terminal (UE). It is transmitted to the server 145 and the user authentication means 120 is installed to determine whether the work is permitted at the predetermined location,
The third output means 116, as a result of the determination of the server 145, displays green when the work at the predetermined location in which the user authentication means 120 is installed is permitted, and displays green, and is not permitted. is displayed in red,
The server 145, in the second work image obtained through the second photographing unit 170, the third output means 116, the green manpower is not detected, or the hard hat 111 Outputs the announcement of a warning through the first output means 135 when the user 110 who does not wear is detected,
The plurality of sensing information,
Further comprising wind direction and wind speed information of the work site where the user 110 is located, flow rate information of the LNG in the bunkering, and flow velocity information of the LNG in the bunkering,
The risk situation is
When the gas concentration in the vicinity of the work site where the user 110 is located is detected by the gas detector 119-1 to be greater than or equal to a first threshold value, the user 110 by the temperature sensor 119-2 When the ambient temperature of the work site in which is located is decreased by more than a second threshold, the humidity in the vicinity of the work site where the user 110 is located by the humidity sensor 119-3 fluctuates by more than a third threshold When the flow rate or flow rate of the LNG during bunkering is changed by more than a fourth threshold, and in the first work image formed by the first photographing unit 114, a crack or splintering phenomenon is detected in the pipe connection part Including any one or more of the cases
The server 145 is
When detecting the dangerous situation, the first alarm signal different according to the dangerous situation is formed, and at the same time, a valve control signal for stopping the supply of the LNG in the bunker is generated or for stopping the supply of the LNG Characterized in that the announcement encouraging valve shutoff is output through the first output means (135),
LNG bunkering monitoring system. delete delete delete delete The method of claim 1,
The plurality of safety equipment further includes goggles 113,
The goggles 113,
It characterized in that it comprises an augmented reality module capable of displaying the augmented reality image generated by the server (145),
LNG bunkering monitoring system. delete delete delete The method of claim 1,
The second photographing unit 170,
Obtaining a face image of the user 110 from a second work image of the work site where the user 110 is located at a location spaced apart from the user 110,
The server 145 is
Extracting the pupil object by applying the face image to the pre-learned first neural network,
It is determined whether the user 110 is watching the work site by using the extracted pupil object,
characterized in that when it is determined that the user 110 is not watching the work site, a second alarm signal is generated,
LNG bunkering monitoring system. delete The method of claim 1,
A bunkering operator server 155 or a pre-scanning server 165 that receives and displays the first operation image, the second operation image, the location information, the plurality of sensing information, and the first alarm signal from the server 145 . ) characterized in that it further comprises,
LNG bunkering monitoring system.

Images