Apparatus and method for sensing and correcting a vessel's maritime environment

Abstract

The present disclosure relates to an apparatus and method for sensing and correcting a marine transportation environment of a ship. The apparatus according to one embodiment of the present disclosure includes an inertial sensor that receives sensing information on the temperature and humidity inside cargo contained in a ship from the outside, measures the roll of the ship, and a processor that verifies the temperature and humidity of the sensing information based on at least one of a GPS that indicates the position of the ship and an ETA that indicates when the ship is expected to arrive at a destination on its navigation route, and corrects transportation environment information including the temperature and humidity based on the verification result.

Description

[Technical Field]

[0001] The present disclosure relates to an electronic device and method, and more particularly to a maritime shipping environment sensing and correction method using a data logger, which uses an OTQ (One Time QR-code) to additionally sense physical quantities specific to the maritime shipping environment and correct conditions that induce abnormal values.

[0002] The present disclosure also relates to electronic devices and methods, and more particularly to systems and methods for detecting and responding to gas leaks due to battery failure on a vessel during transit at sea using artificial intelligence models. [Background technology]

[0003] With the development of technology and industry, distribution networks are gradually expanding, and as this distribution network expands, the storage and transportation conditions of goods during the distribution process have become increasingly important factors. In particular, when transporting expensive products such as large-capacity batteries, if the product is damaged due to high heat, vibration, humidity, etc., it is highly likely that not only one item will be damaged but other products loaded in the same space or environment will also be damaged in a chain reaction. High-precision products are transported in vibration-free vehicles, and low-temperature refrigerated foods require precise control of temperature and humidity during transportation.

[0004] As such, high-value-added products require a different logistics and transportation environment than low-value-added products. Managing transportation quality to prevent damage or loss of transported goods is of utmost importance in logistics management, and technological advances have led to the adoption of a variety of devices and methods for efficient logistics management. However, even in the case of high-value-added products, there is a problem of not monitoring the deterioration or damage of products due to high heat, vibration, humidity, etc. during transportation, which calls for the introduction of a more efficient and systematic transportation quality control system. Summary of the Invention [Problem to be solved by the invention]

[0005] The embodiment disclosed in the present disclosure aims to build a system for verifying and correcting transportation environment information based on GPS and ETA to accurately estimate the transportation environment.

[0006] Furthermore, the disclosed embodiments of the present disclosure aim to build a system that includes a virtual sensor to replace an actual gas sensor.

[0007] In order to achieve the above-mentioned technical objectives, according to one aspect of the present disclosure, a system for detecting an abnormal state of a battery included in a ship in a maritime transportation environment of the ship includes: an artificial neural network processor that generates an artificial intelligence model that learns a change pattern in temperature and humidity within an enclosed space included in sensing data generated from a sensing device; and a virtual sensing processor that detects changes in temperature and humidity within the enclosed space as gas leakage occurring in a battery corresponding to the temperature and humidity within the enclosed space based on the artificial intelligence model and the sensing data received from the ship.

[0008] A method for sensing and correcting a marine transportation environment of a ship according to another aspect of the present disclosure includes an artificial intelligence model generation step of generating an artificial intelligence model that has learned a change pattern in temperature and humidity within an enclosed space contained in sensing data, and a gas detection step of detecting changes in temperature and humidity within the enclosed space as a leakage of gas occurring in a battery corresponding to the temperature and humidity within the enclosed space based on the artificial intelligence model and the sensing data received from the ship.

[0009] In addition, a computer program stored on a computer-readable recording medium for execution to embody the present disclosure may also be provided.

[0010] In addition, a computer program stored on a recording medium that executes the method for embodying the present disclosure in combination with the hardware may also be provided.

[0011] The problems to be solved by the present disclosure are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description. [Means for solving the problem]

[0012] To achieve the above-mentioned technical objectives, one aspect of the present disclosure provides an apparatus for sensing and correcting a marine transportation environment of a ship, including an inertial sensor that measures the roll of the ship and a processor that receives sensing information on the temperature and humidity inside cargo contained in the ship from the outside, verifies the temperature and humidity based on at least one of a Global Positioning System (hereinafter referred to as "GPS") that indicates the position of the ship and an estimated time of arrival (hereinafter referred to as "ETA") when the ship is scheduled to arrive at a destination on its navigation route, and corrects transportation environment information including the temperature and humidity based on the verification results.

[0013] A method for sensing and correcting a marine transportation environment of a ship according to another aspect of the present disclosure includes a sensing information receiving step of receiving sensing information on the temperature and humidity inside cargo contained in the ship from the outside, a measurement step of measuring the roll of the ship, a verification step of verifying the temperature and humidity based on at least one of a GPS indicating the position of the ship and an ETA at which the ship is expected to arrive at a destination on its travel route, and a correction step of correcting transportation environment information including the temperature and humidity based on the verification results.

[0014] In addition, a computer program stored on a computer-readable recording medium for execution to embody the present disclosure may also be provided.

[0015] In addition, a computer program stored on a recording medium that executes the method for embodying the present disclosure in combination with the hardware may also be provided. [Effects of the Invention]

[0016] The above-described means for solving the problems of the present disclosure have the effect of accurately estimating the transportation environment and providing convenience and reliability to users.

[0017] According to the above-described solution to the problems of the present disclosure, by introducing a virtual sensor to replace an actual gas sensor, it is possible to reduce manufacturing costs, improve the accuracy of gas detection, and provide convenience and reliability to users.

[0018] The effects of the present disclosure are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a diagram illustrating an example of a vessel according to the present disclosure. [Figure 2] 1 is a diagram illustrating an example of cargo included in a vessel of the present disclosure. [Figure 3a] 1 is a diagram illustrating an exemplary system according to the present disclosure. [Figure 3b] 1 is a diagram illustrating an exemplary system according to the present disclosure. [Figure 4] 1 is a diagram illustrating an example of a tracker of the present disclosure. [Figure 5] 1 is a diagram illustrating an example of a tracker of the present disclosure. [Figure 6] FIG. 1 is a block diagram illustrating an exemplary tracker of the present disclosure. [Figure 7] 1 is a diagram illustrating an example of an inertial sensor of the present disclosure. [Figure 8] 1 is a diagram for explaining an embodiment of the present disclosure in which ship information is verified and corrected while the ship is moving. [Figure 9] 1 is a flowchart illustrating a method according to the present disclosure. [Figure 10] 1 is a diagram illustrating a system according to an embodiment of the present disclosure. [Figure 11] 1 is a diagram illustrating an example of a tracker of the present disclosure. [Figure 12] 1 is a diagram illustrating an example of a tracker of the present disclosure. [Figure 13a] 10 is a diagram illustrating a system according to another embodiment of the present disclosure. [Figure 13b] 10 is a diagram illustrating a system according to another embodiment of the present disclosure. [Figure 14] 1 is a diagram illustrating a sensing operation of a gas sensor according to the present disclosure. [Figure 15] 1 is a diagram for explaining an artificial intelligence model of the present disclosure. [Figure 16] 1 is a diagram for explaining virtual sensing according to the present disclosure. [Figure 17] 17 is a diagram for explaining the embodiment of FIG. 16. [Figure 18] 17 is a diagram for explaining the embodiment of FIG. 16. [Figure 19] 1 is a flowchart illustrating a method according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0020] The same reference numerals refer to the same elements throughout this disclosure. This disclosure does not describe all elements of the embodiments, and content that is common in the technical field to which the disclosure belongs or that is duplicated between embodiments will be omitted. The terms "unit, module, component, block" used in the specification may be embodied in software or hardware, and depending on the embodiment, multiple "units, modules, components, blocks" may be embodied in one component, or one "unit, module, component, block" may include multiple components.

[0021] Throughout this specification, when a part is said to be "coupled" to another part, this includes not only direct coupling but also indirect coupling, and indirect coupling includes coupling via a wireless communication network.

[0022] Furthermore, when a part "comprises" a certain element, this does not mean that it excludes other elements, but that it may further include other elements, unless otherwise specified.

[0023] Throughout this specification, when an element is said to be "on" another element, this includes not only when the element is in contact with the other element, but also when there is another element between the two elements.

[0024] The terms "first," "second," etc. are used to distinguish one component from another component, and are not intended to limit the components to the terms previously described.

[0025] The singular expression includes the plural expression unless the context clearly dictates otherwise.

[0026] The identification numbers used in each step are for convenience of explanation, and do not describe the order of each step. The steps may be performed in a different order than specified unless the context clearly dictates a specific order.

[0027] Hereinafter, the working principle and embodiments of the present disclosure will be described with reference to the accompanying drawings.

[0028] 1. Example 1 The following embodiments describe an apparatus and method for sensing and compensating for a vessel's maritime environment.

[0029] In this specification, the term "device according to the present disclosure" includes various devices capable of performing computations and providing results to a user. For example, the device according to the present disclosure may include all or any one of a computer, a server device, and a portable terminal.

[0030] Here, the computer may include, for example, a notebook computer, desktop computer, laptop computer, tablet PC, slate PC, etc. equipped with a web browser.

[0031] The server device is a server that communicates with external devices and processes information, and may include an application server, a computing server, a database server, a file server, a game server, a mail server, a proxy server, and a web server.

[0032] The portable terminal is, for example, a wireless communication device that ensures portability and mobility, and may include all kinds of handheld-based wireless communication devices such as PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) terminals, smartphones, etc., as well as wearable devices such as watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or head-mounted devices (HMDs).

[0033] FIG. 1 is a diagram illustrating an example of a vessel according to the present disclosure.

[0034] Referring to FIG. 1, a ship 1 can travel from a departure point to a destination along a specific route on the ocean. For example, the ship 1 can move from a first center coordinate Oc in a first direction Xc and / or a second direction Yc, and can also move in a third direction Zc due to the influence of waves, etc. While the ship 1 is moving, it can be rocked by external influences such as waves, wind, gravity, or geomagnetic field. As a result, the ship 1 can pitch, yaw, and roll. Furthermore, the ship 1 can travel carrying various cargo, and as cargo is loaded onto the ship 1, its center of gravity can shift, and the weight of the cargo can cause the ship 1 and its cargo to rock on the ocean. For example, based on the second center coordinate Ob, surge (u) and roll (p) may occur in a first direction Xb, sway (v) and pitch (q) may occur in a second direction Yb, and heave (w) and yaw (r) may occur in a third direction Zb. The ship 1 may have six or more degrees of freedom. Cargo may be contained in the transportation space of the ship 1, and the transportation space may refer to a space in which cargo such as a container box, reefer container, or liner is transported.

[0035] FIG. 2 is a diagram showing an example of cargo included in a vessel of the present disclosure.

[0036] 2, the ship 1 may include at least one cargo. The cargo may include a device for estimating cargo transportation information and ship information for the ship 1. Such a device may be referred to as a tracker or an environmental information estimation device.

[0037] Figures 3a and 3b are diagrams illustrating an exemplary system according to the present disclosure. Figures 4 and 5 are diagrams illustrating an exemplary tracker according to the present disclosure. Figure 4 is a front view of the tracker. Figure 5 is a rear view of the tracker.

[0038] Referring to FIG. 3a, a cargo space tracker 10A, a user terminal 20A, a temperature / humidity measuring sensor 30A, and a distance measuring sensor 40A may be provided to perform the operations of the present disclosure.

[0039] The tracker 10A includes a variety of devices that can perform computations and provide results to the user.

[0040] The user terminal 20A may be either a computer or a portable user terminal, or may be any one of these. Here, the computer may be, for example, a notebook PC, desktop PC, laptop, tablet PC, slate PC, etc. equipped with a web browser.

[0041] The device (server) is a server that communicates with external devices and processes information, and may include an application server, computing server, database server, file server, game server, mail server, proxy server, web server, etc.

[0042] The portable user terminal is, for example, a wireless communication device that ensures portability and mobility, and may include all kinds of handheld-based wireless communication devices such as a Personal Communication System (PCS), a Global System for Mobile communications (GSM), a Personal Digital Cellular (PDC), a Personal Handyphone System (PHS), a Personal Digital Assistant (PDA), an International Mobile Telecommunication (IMT)-2000, a Code Division Multiple Access (CDMA)-2000, a W-Code Division Multiple Access (W-CDMA), a Wireless Broadband Internet (WiBro) terminal, a smartphone, etc., as well as wearable devices such as a watch, a ring, a bracelet, an anklet, a necklace, glasses, contact lenses, or a head-mounted device (HMD).

[0043] The temperature / humidity measurement sensor 30A may be attached within the cargo transport space (e.g., inside the cargo). The temperature / humidity measurement sensor 30A may be attached at a position predetermined by a user. The temperature / humidity measurement sensor 30A may sense the temperature and humidity inside the cargo and output sensing information regarding the temperature and humidity to the tracker 10A. In one embodiment, the temperature / humidity measurement sensor 30A may include a temperature measurement sensor and a humidity measurement sensor. The unit of temperature may be °C and the unit of humidity may be %, but is not limited thereto.

[0044] The distance measuring sensor 40A may be attached within the cargo transport space and may generate distance data by measuring distances between a plurality of first positions predetermined by a user and a plurality of second positions, which are the remaining vertex positions to which the temperature / humidity measuring sensor 30A is not attached. The distance measuring sensor 40A may include any one of a lidar sensor, an ultrasonic sensor, a short / medium-range radar sensor, a long-range radar sensor, and a camera.

[0045] Referring to Figure 3b, a system 100B may include a tracker 10B, a first user terminal 20B, a second user terminal 50B, and a communication network 60B. Although the number of user terminals is two in Figure 3b, the number is not limited to two and may be one, three, or more.

[0046] The tracker 10B may be a device for sensing and correcting the marine transportation environment of a ship. The tracker 10B can communicate with first and second user terminals 20B and 50B via a communication network 60B. The tracker 10B may include any of a variety of devices that perform computational processing and provide results to users. For example, the tracker 10B may include all or any one of a computer, device (server), and portable terminal. Here, the computer may include, for example, a notebook computer, desktop computer, laptop computer, tablet PC, or slate PC equipped with a web browser. The device (server) is a server that communicates with external devices and processes information, and may include an application server, computing server, database server, file server, game server, mail server, proxy server, and web server.

[0047] Referring to FIG. 4, the front of tracker 10B may include sensing unit 131, switch 132, input units 133 and 134, fingerprint recognition button 121, and a display. A user can input start and end dates on the display through input unit 133. Referring to FIG. 5, the rear of tracker 10B may include various buttons 122 and 123 and a power indicator 135. The input unit is for receiving information from a user, and information may be input through the user input unit. Such user input units may include hardware physical keys (e.g., buttons, dome switches, jog wheels, jog switches, etc., located on at least one of the front, rear, and side of the device) and software touch keys. For example, the touch keys may be virtual keys, soft keys, or visual keys displayed on a touchscreen display through software processing, or they may be touch keys located on a portion other than the touchscreen. Meanwhile, virtual keys or visual keys can be displayed on the touch screen in various forms, for example, graphics, text, icons, videos, or a combination thereof.

[0048] 3b, the first user terminal 20B and the second user terminal 50B may be either one or all of the above-mentioned computers and portable user terminals. The portable user terminal may be, for example, a wireless communication device that ensures portability and mobility, and may include all kinds of handheld-based wireless communication devices such as a Personal Communication System (PCS), Global System for Mobile communications (GSM), Personal Digital Cellular (PDC), Personal Handyphone System (PHS), Personal Digital Assistant (PDA), International Mobile Telecommunication (IMT)-2000, Code Division Multiple Access (CDMA)-2000, W-Code Division Multiple Access (W-CDMA), Wireless Broadband Internet (WiBro) terminal, and a smartphone, as well as wearable devices such as a watch, a ring, a bracelet, an anklet, a necklace, glasses, contact lenses, or a head-mounted device (HMD).

[0049] FIG. 6 is an exemplary block diagram of a tracker of the present disclosure.

[0050] Referring to Figure 6, the tracker 200 can sense physical quantities specific to the maritime transportation environment using an OTQ (One Time QR-code) and correct conditions that cause abnormal values. Due to the distribution of the ship's voyage route and the weight of the cargo loaded on the ship 1, a significant amount of roll, which corresponds to a side-to-side movement (i.e., surge) caused by fluid, can occur due to pitch and yaw. Therefore, the tracker 200 can eliminate the influence of the roll direction by correcting the impact in the roll direction, which is the most frequent noise, and can sense only the actual impact applied to the cargo by eliminating the roll value, which is environmental noise. This has the effect of accurately estimating the transportation environment.

[0051] Tracker 200 may include a processor 210 , a memory 220 , an inertial sensor 230 , a communication module 240 , and a salinity sensor 260 .

[0052] The processor 210 can receive sensing information on the temperature and humidity inside the cargo contained in the ship from the outside. The sensing information can be generated by the temperature / humidity measurement sensor 30A in FIG. 3a. The processor 210 can verify the temperature and humidity based on at least one of a Global Positioning System (GPS) that indicates the position of the ship 1 and an estimated time of arrival (ETA) that indicates when the ship 1 will arrive at the destination on its travel route. The processor 210 can then correct the transportation environment information, including the temperature and humidity, based on the verification results.

[0053] In one embodiment, the processor 210 may include a verification unit 211 , a calculation unit 212 , and a correction unit 213 .

[0054] The verification unit 211 can perform spatial verification and temporal verification. In the case of spatial verification, the verification unit 211 measures the current position of the ship 1 based on GPS, compares the temperature and humidity predicted at the current position of the ship 1 with a preset first reference value, and verifies the temperature and humidity of the current ship 1 based on the comparison result. In the case of temporal verification, the verification unit 211 measures the ETA of the ship 1 on its movement path input from the outside, compares the temperature and humidity predicted from the ETA of the current ship 1 with a preset second reference value, and verifies the temperature and humidity of the current ship 1 based on the comparison result. In one embodiment, the verification unit 211 can primarily perform spatial verification and secondarily perform temporal verification.

[0055] The arithmetic unit 212 can perform necessary calculations within the processor 210. In one embodiment, the arithmetic unit 212 can be implemented as an ALU (Arithmetic and Logical Unit) that receives inputs such as input operands, opcodes, and states, calculates the input values, and outputs the calculation results. In one embodiment, the arithmetic unit 212 can calculate and output correction parameters based on the currently measured temperature and humidity and reference values.

[0056] The correction unit 213 can correct the transportation environment based on the verification result of the verification unit 211 and the correction parameters calculated by the calculation unit 212. At this time, the transportation environment can include temperature and humidity.

[0057] The memory 220 can store data supporting various functions of the tracker 200 and programs for the operation of the processor 210, can store input / output data (e.g., music files, still images, videos, etc.), can store a number of application programs (or applications) run by the tracker 200, and data and commands for the operation of the tracker 200. At least some of these application programs can be downloaded from an external server via wireless communication. The memory 220 may include at least one type of storage medium selected from the group consisting of flash memory, hard disk, solid state disk (SSD), silicon disk drive (SDD), micro multimedia card, card-type memory (e.g., SD or XD memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, and optical disk. The memory 220 may store a lookup table (LUT) including error data measured for each movement route of the ship 1, an LUT including error data measured for each area in which the ship 1 can be located, and an LUT for each cargo loaded on the ship 1.

[0058] The inertial sensor 230 can sense physical quantities caused by wave friction and the engine. The inertial sensor 230 can measure the roll of the vessel 1, the vibration of the vessel 1, and / or the impact applied to the vessel 1. The inertial sensor 230 can sense the movement of the vessel 1 on a 6DoF (Degree of Freedom) or 9DoF basis. This is to ensure the usability of the inertial sensor applied to V2+. In one embodiment, the inertial sensor 230 can be embodied as an IMU (Inertial Measurement Unit).

[0059] The communication module 240 may implement a communication interface. The communication interface may include one or more components that enable communication with an external device. For example, the communication interface may include at least one of a wired communication module, a wireless communication module, and a short-range communication module. The wired communication module may include various wired communication modules such as a local area network (LAN) module, a wide area network (WAN) module, or a value-added network (VAN) module, as well as various cable communication modules such as Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), Digital Visual Interface (DVI), recommended standard 232 (RS-232), power line communication, or plain old telephone service (POTS). The wireless communication module may include a Wi-Fi module, a wireless broadband module, and other wireless communication modules that support various wireless communication methods such as GSM (global system for mobile communication), CDMA (code division multiple access), WCDMA (wideband code division multiple access), UMTS (universal mobile telecommunications system), TDMA (time division multiple access), LTE (long term evolution), 4G, 5G, 6G, etc. The wireless communication module may include a wireless communication interface including an antenna for transmitting signals and a transmitter.The wireless communication module may further include a signal conversion module that modulates a digital control signal output from the processor 210 through the wireless communication interface into an analog wireless signal under the control of the processor 210. The short-range communication module is for short-range communication and is compatible with Bluetooth (registered trademark). TM The device may support short-range communication using at least one of the following technologies: RFID (Radio Frequency Identification), Infrared Data Association (IrDA), UWB (Ultra Wideband), ZigBee (registered trademark), NFC (Near Field Communication), Wi-Fi (Wireless Fidelity), Wi-Fi Direct, and Wireless Universal Serial Bus (Wireless USB).

[0060] The salinity sensor 260 may be a sensor for tracking effects on precision machinery such as the ship 1, such as corrosion occurring on the ship 1. The salinity sensor 260 may measure or sense the salinity occurring on the ship 1 in the sea around the ship 1. In one embodiment, the calculation unit 212 may further receive an input of the salinity sensed by the salinity sensor 260 and calculate and output a correction parameter based thereon, and the correction unit 213 may correct the shipping environment based on the verification result of the verification unit 211 and the correction parameter calculated by the calculation unit 212.

[0061] FIG. 7 is a diagram illustrating an example of an inertial sensor of the present disclosure.

[0062] 6 and 7, the inertial sensor 230 may be implemented as an IMU. The inertial sensor 230 implemented as an IMU may include an angular velocity sensor 231, an acceleration sensor 232, and a geomagnetic sensor 233.

[0063] The angular velocity sensor 231 may be referred to as a gyroscope and may measure the angular velocity of the ship 1 to determine how many degrees (e.g., degrees) the ship 1 rotates per time. The acceleration sensor 232 may be referred to as an accelerometer and may measure the angular velocity to determine how much the ship 1 has tilted by resolving the acceleration of gravity when calculating an initial value. The geomagnetic sensor 233 may be referred to as a magnetometer and may measure the geomagnetic field (magnet) to determine how much the ship 1 has deviated from magnetic north by measuring the strength of magnetic flux based on magnetic north.

[0064] FIG. 8 is a diagram for explaining an embodiment of the present disclosure in which ship information is verified and corrected while the ship is moving.

[0065] Referring to FIG. 8, it is assumed that the departure point (DPT) of vessel 1 is New York, the arrival point (ARV) of vessel 1 is London, and the route MVPTH between New York and London is as shown in FIG.

[0066] The verification unit 211 can primarily estimate the temperature and humidity for each predicted voyage path along the navigation route MVPTH and can primarily calculate the ETA for the navigation route MVPTH. For example, the ship 1 may pass through the Atlantic Ocean while traveling along the navigation route MVPTH. At this time, the latitude may change, and as the ship 1 moves from the shore to the deep sea, the humidity of the cargo loaded on the ship 1 may increase. Then, as the ship 1 moves away from the deep sea, the humidity of the cargo loaded on the ship 1 may decrease. The calculation unit 212 can calculate correction parameters taking these conditions into account. The correction unit 213 can update the time difference and humidity changes. Meanwhile, as the ship 1 does not pass through the equator and its longitude increases, the average temperature of the cargo loaded on the ship 1 may decrease. The calculation unit 212 can calculate correction parameters taking these conditions into account, and the correction unit 213 can update the average temperature change by longitude.

[0067] The verification unit 211 can verify the temperature and humidity values ​​compared to the estimated values ​​in the log for each actual route. If an error occurs in the temperature and humidity log, the corresponding error data can be stored and updated in the memory 220. The updated data can correct the estimated temperature and humidity values ​​and be used to improve the model. This has the effect of enabling precise analysis of error causes such as human error and equipment measurement errors. However, temporal errors can occur when the navigation route is changed, abnormal weather occurs, or the ship 1 takes evasive action. The calculation unit 212 can calculate correction parameters taking these conditions into consideration, and the correction unit 213 can update the average temperature change by longitude. Operational information can be updated based on this analysis of the error causes, and the verification logic can be fine-tuned.

[0068] FIG. 9 is a flowchart illustrating a method according to the present disclosure.

[0069] Referring to FIG. 9, the method for sensing and correcting the marine transportation environment of a ship of the present disclosure may include a sensing information receiving step (S100), a measurement step (S200), a verification step (S300), and a correction step (S400).

[0070] The sensing information receiving step (S100) is a step of receiving sensing information on the temperature and humidity inside the cargo contained in the ship from the outside. The sensing information receiving step (S100) is performed by the processor 210.

[0071] The measuring step (S200) is a step of measuring the roll of the ship. The measuring step (S200) is performed by the inertial sensor 230.

[0072] The verification step (S300) is a step of verifying the temperature and humidity based on at least one of the GPS indicating the position of the ship and the ETA when the ship is expected to arrive at the destination on the travel route. The verification step (S300) is performed by the verification unit 211 of the processor 210.

[0073] The correction step (S400) is a step of correcting transportation environment information including temperature and humidity based on the verification result. The correction step (S400) is performed by the correction unit 213 of the processor 210.

[0074] Meanwhile, the disclosed embodiments may be embodied in the form of a recording medium storing computer-executable instructions. The instructions may be stored in the form of program code, which, when executed by a processor, generates program modules to perform the operations of the disclosed embodiments. The recording medium may be embodied as a computer-readable recording medium.

[0075] Computer-readable recording media include all types of recording media that store computer-readable instructions, such as ROM (Read Only Memory), RAM (Random Access Memory), magnetic tape, magnetic disk, flash memory, and optical data storage devices.

[0076] As mentioned above, the disclosed embodiments have been described with reference to the attached drawings. Those skilled in the art will understand that the present disclosure may be implemented in forms different from the disclosed embodiments without changing the technical idea or essential features of the present disclosure. The disclosed embodiments are illustrative and should not be interpreted as limiting.

[0077] 2. Example 2 Other embodiments are illustrated below. The following embodiments illustrate systems and methods for detecting an abnormal state of a battery in a marine shipping environment. In this disclosure, reference numerals in the second embodiment may be understood to indicate different configurations even if they are given the same numbers or letters as the reference numerals in the first embodiment.

[0078] In this specification, the term "device according to the present disclosure" includes various devices capable of performing computations and providing results to a user. For example, the device according to the present disclosure may include all or any one of a computer, a server device, and a portable terminal.

[0079] Here, the computer may include, for example, a notebook computer, desktop computer, laptop computer, tablet PC, slate PC, etc. equipped with a web browser.

[0080] The server device is a server that communicates with external devices and processes information, and may include an application server, a computing server, a database server, a file server, a game server, a mail server, a proxy server, and a web server.

[0081] The portable terminal is, for example, a wireless communication device that ensures portability and mobility, and may include all kinds of handheld-based wireless communication devices such as PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) terminals, smartphones, etc., as well as wearable devices such as watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or head-mounted devices (HMDs).

[0082] The AI-related functions of the present disclosure operate through a processor and memory. The processor may be composed of one or more processors. In this case, the one or more processors may be general-purpose processors such as a CPU, AP, or DSP (Digital Signal Processor), dedicated graphics processors such as a GPU or VPU (Vision Processing Unit), or dedicated AI processors such as an NPU. The one or more processors control the processing of input data according to predefined operating rules or AI models stored in memory. Alternatively, if the one or more processors are dedicated AI processors, the dedicated AI processors may be designed with a hardware structure specialized for processing a specific AI model. For example, the processor may include an MCU (microcontroller unit), a fan control actuator, an APU (Accelerated Processing Unit), etc.

[0083] The predefined behavioral rules or AI models are characterized by being created through learning. Here, "created through learning" means that a basic AI model is trained by a learning algorithm using a large amount of learning data to create predefined behavioral rules or AI models configured to achieve desired characteristics (or goals). Such learning may be performed within the device itself on which the AI ​​according to the present disclosure is performed, or may be performed through a separate server and / or system. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning, but are not limited to the above examples.

[0084] An AI model may be composed of multiple neural network layers. Each of the multiple neural network layers has multiple weight values, and performs neural network operations by calculating the weight values ​​and the operation results of previous layers. The weight values ​​of the multiple neural network layers may be optimized based on the learning results of the AI ​​model. For example, the weight values ​​may be updated to reduce or minimize the loss or cost values ​​acquired by the AI ​​model during the learning process. The artificial neural network may include a deep neural network (DNN), such as a convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network (RNN), a restricted Boltzmann machine (RBM), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), or a deep Q-network, but is not limited to these examples.

[0085] According to an exemplary embodiment of the present disclosure, a processor may embody artificial intelligence. Artificial intelligence refers to a machine learning method based on an artificial neural network, which allows a machine to learn by imitating human biological neurons. Artificial intelligence methodologies can be classified into supervised learning, in which an answer (output data) to a problem (input data) is determined by providing both input data and output data as training data in a learning method; unsupervised learning, in which only input data is provided without output data, so the answer (output data) to the problem (input data) is not determined; and reinforcement learning, in which a reward is provided from an external environment each time an action is taken in the current state, and learning progresses in a direction to maximize this reward. Furthermore, artificial intelligence methodologies may be classified by their architecture, which is the structure of the learning model. The architectures of widely used deep learning technologies can be classified into convolutional neural networks (CNNs), recurrent neural networks (RNNs), transformers, generative adversarial networks (GANs), etc.

[0086] The present device and system may include an artificial intelligence model. The artificial intelligence model may be a single artificial intelligence model or may be embodied as multiple artificial intelligence models. The artificial intelligence model may be composed of a neural network (or artificial neural network) and may include statistical learning algorithms that mimic biological neurons in machine learning and cognitive science. A neural network may refer to a general model that has problem-solving capabilities, in which artificial neurons (nodes) formed through synaptic connections change the strength of synaptic connections through learning. Neurons in a neural network may include a combination of weights or biases. A neural network may include one or more layers composed of one or more neurons or nodes. For example, a device may include an input layer, a hidden layer, and an output layer. The neural network constituting the device can infer the desired output from any input by changing the neuron weights through learning.

[0087] The processor may generate a neural network, train or learn the neural network, perform operations based on received input data, generate an information signal based on the results of the operations, or retrain the neural network. Neural network models may include various types of models such as CNNs (Convolution Neural Networks) such as GoogleNet, AlexNet, and VGG Network, R-CNNs (Region with Convolution Neural Networks), RPNs (Region Proposal Networks), RNNs (Recurrent Neural Networks), S-DNNs (Stacking-based deep Neural Networks), S-SDNNs (State-Space Dynamic Neural Networks), Deconvolution Networks, DBNs (Deep Belief Networks), RBMs (Restricted Boltzmann Machines), Fully Convolutional Networks, LSTMs (Long Short-Term Memory Networks), and Classification Networks, but are not limited to these. The processor may include one or more processors for performing operations according to the neural network model. For example, the neural network may be a deep neural network (Deep Neural Network). Network).

[0088] Neural networks include CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), perceptron, multilayer perceptron, FF (Feed Forward), RBF (Radial Basis Network), DFF (Deep Feed Forward), LSTM (Long Short Term Memory), GRU (Gated Recurrent Unit), AE (Auto Encoder), VAE (Variational Auto Encoder), DAE (Denoising Auto Encoder), SAE (Sparse Auto Encoder), MC (Markov Chain), HN (Hopfield Network), BM (Boltzmann Machine), RBM (Restricted Boltzmann Machine), DBN (Deep Belief Network), DCN (Deep Convolutional Network), DN (Deconvolutional Network), DCIGN (Deep Convolutional Inverse Graphics Network), GAN (Generative Adversarial Network), and LSM (Liquid State Machine). It will be understood by those of ordinary skill in the art that the neural network may include any neural network, including but not limited to, an ELM (Extreme Learning Machine), an ESN (Echo State Network), a DRN (Deep Residual Network), a DNC (Differentiable Neural Computer), an NTM (Neural Turning Machine), a CN (Capsule Network), a KN (Kohonen Network), and an AN (Attention Network).

[0089] According to an example embodiment of the present disclosure, the processor may support a variety of neural networks, including Convolution Neural Networks (CNNs) such as GoogleNet, AlexNet, and VGG Network, Region with Convolution Neural Network (R-CNN), Region Proposal Network (RPN), Recurrent Neural Network (RNN), Stacking-based deep Neural Network (S-DNN), State-Space Dynamic Neural Network (S-SDNN), Deconvolution Network, Deep Belief Network (DBN), Restricted Boltzmann Machine (RBM), Fully Convolutional Network, Long Short-Term Memory (LSTM) Network, Classification Network, Generative Modeling, eXplainable AI, Continual AI, Representation Learning, AI for Material Design, BERT, SP-BERT, MRC / QA, Text Analysis, Dialog System, GPT-3, GPT-4 for Natural Language Processing, Visual Analytics, Visual Understanding, Video Synthesis for Vision Processing, Anomaly Detection, Prediction, and Time-Series for ResNet Data Intelligence. Various artificial intelligence structures and algorithms may be used, including but not limited to, forecasting, optimization, recommendation, data creation, etc. Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

[0090] Figure 10 is a diagram illustrating a system according to one embodiment of the present disclosure. Figures 11 and 12 are diagrams illustrating exemplary trackers of the present disclosure. Figure 11 is a front view of the tracker. Figure 12 is a rear view of the tracker.

[0091] A ship can travel from its origin to its destination along a fixed route on the ocean. For example, a ship can move in a first and / or second direction, and can also move in a third direction due to the influence of waves and other factors. While a ship moves, it can sway due to external influences such as waves, wind, gravity, or geomagnetic fields. As a result, the ship may pitch, yaw, and roll. A ship can carry at least one cargo. The cargo can include a device for estimating cargo transportation information and ship information for the ship. Such a device may be referred to as a tracker. A ship can travel carrying various cargoes, and its center of gravity may change as cargo is loaded onto the ship, and the ship and cargo may sway on the ocean due to the weight of the cargo itself. Cargo may be included in the ship's transportation space, which may refer to the space in which cargo is transported, such as the transportation space for container boxes, reefer containers, and liners. The tracker, which is included in the ship or cargo loaded on the ship while the ship is operating and tracks and transmits the ship's transportation information, may have a built-in battery that can be damaged. Gas may leak as the ship travels along the ocean.

[0092] Referring to Fig. 10, a system 100-1 may be used to detect an abnormal state of a battery included in a ship in a marine transport environment. The system 100-1 may include a tracker 10-1, a first user terminal 20-1, a second user terminal 30-1, and a communication network 40-1. While Fig. 12 shows two user terminals, the number of user terminals is not limited thereto and may be one, three, or more.

[0093] The tracker 10-1 can communicate with first and second user terminals 20-1 and 30-1 through a communication network 40-1. The tracker 10-1 can include any of a variety of devices that can perform computations and provide results to users. For example, the tracker 10-1 can include all or any one of a computer, a device (server), and a portable terminal. Here, the computer can include, for example, a notebook computer, desktop computer, laptop computer, tablet PC, or slate PC equipped with a web browser. The device (server) is a server that communicates with external devices and processes information, and can include an application server, computing server, database server, file server, game server, mail server, proxy server, and web server.

[0094] Referring to FIG. 11, the front of tracker 10-1 may include sensing unit 131-1, switch 132-1, input units 133-1 and 134-1, fingerprint recognition button 121-1, and a display. A user can input start and end dates on the display through input unit 133-1. Referring to FIG. 12, the rear of tracker 10-1 may include various buttons 122-1 and 123-1 and a power indicator 135-1. The input unit is for receiving information from a user, and information may be input through the user input unit. Such user input units may include hardware physical keys (e.g., buttons, dome switches, jog wheels, jog switches, etc. located on at least one of the front, rear, and side of the device) and software touch keys. For example, the touch keys may be virtual keys, soft keys, or visual keys that are displayed on a touchscreen display through software processing, or may be touch keys located outside the touchscreen. Meanwhile, the virtual keys or visual keys may have various forms and be displayed on the touchscreen, and may be, for example, graphics, text, icons, videos, or combinations thereof.

[0095] 12, the first user terminal 20-1 and the second user terminal 30-1 may be either one or all of the above-mentioned computers and portable user terminals. The portable user terminal is, for example, a wireless communication device that ensures portability and mobility, and may include all kinds of handheld-based wireless communication devices such as a Personal Communication System (PCS), Global System for Mobile communications (GSM), Personal Digital Cellular (PDC), Personal Handyphone System (PHS), Personal Digital Assistant (PDA), International Mobile Telecommunication (IMT)-2000, Code Division Multiple Access (CDMA)-2000, W-Code Division Multiple Access (W-CDMA), Wireless Broadband Internet (WiBro) terminal, and a smartphone, as well as wearable devices such as a watch, a ring, a bracelet, an anklet, a necklace, glasses, contact lenses, or a head-mounted device (HMD).

[0096] As described above, by introducing a virtual sensor to replace an actual gas sensor, there are effects such as reducing manufacturing costs and improving the accuracy of gas detection.

[0097] 13a and 13b are diagrams illustrating a system according to another embodiment of the present disclosure.

[0098] 13a, a system 200a-1 may be for detecting an abnormal condition of a battery included in a vessel within a marine shipping environment of the vessel. The system 200a-1 may include a device 210-1 and a gas sensor 220-1.

[0099] Device 210-1 may include a processor 211-1, a memory 212-1, and a communication module 213-1.

[0100] The processor 210-1 can detect an abnormal state of a battery included in the ship in the ship's maritime transport environment. In one embodiment, the processor 210-1 can include an artificial neural network processing unit 2111-1, a virtual sensing processing unit 2112-1, and an arithmetic unit 2113-1. In the present disclosure, the artificial neural network processing unit 2111-1 can be separately provided on a printed circuit board that physically configures the device 210-1, or can be an operating module that logically operates within a processor chipset. For example, the artificial neural network processing unit 2111-1 can be stored in the memory 212-1 as program code and can refer to a functional unit that implements a machine learning model that is fetched and sequentially analyzed by the processor 211-1 to achieve a specific purpose.

[0101] The artificial neural network processing unit 2111-1 can generate an artificial intelligence model. The artificial neural network processing unit 2111-1 can train the artificial intelligence model. The artificial intelligence model can learn change patterns in temperature and humidity within an enclosed space contained in sensing data generated from a sensing device. The sensing data may be data sensed by a device attached to cargo, and the sensing data may include information on temperature and humidity. For example, the sensing data may be expressed as a two-dimensional electronic code or a three-dimensional electronic code. For example, the sensing data may be expressed in various formats such as a barcode, a QR code, or a hologram code. For example, the sensing data may be expressed in the form of a one-time QR code, in which the form of the electronic code changes at regular intervals.

[0102] The virtual sensing processing unit 2112-1 can receive the sensing data from the outside. Based on the artificial intelligence model and the sensing data received from the ship, it can detect changes in temperature and humidity in the enclosed space as gas leakage occurring in the battery corresponding to the temperature and humidity in the enclosed space. In the present disclosure, the virtual sensing processing unit 2112-1 can perform the function of a virtual sensor that operates to sense gas by implementing a pre-trained machine learning model in the artificial neural network processing unit 2111-1.

[0103] In one embodiment, the virtual sensing processing unit 2112-1 can generate structured data by structuring the chemical structure of the gas on a text basis.

[0104] In one embodiment, the virtual sensing processing unit 2112-1 can convert the structured data into restructured spectroscopic data.

[0105] In one embodiment, the virtual sensing processing unit 2112-1 can calculate the prior probability that a gas behaves above a critical value under specific temperature and humidity conditions based on the spectroscopic data.

[0106] In one embodiment, the virtual sensing processing unit 2112-1 can calculate the posterior probability of gas behavior by applying a Bayesian estimation algorithm to the prior probability based on the actually measured temperature and humidity.

[0107] In one embodiment, the virtual sensing processing unit 2112-1 can estimate the behavior of gas by temperature and humidity based on the posterior probability.

[0108] In one embodiment, the virtual sensing processing unit 2112-1 can acquire the transition of temperature and humidity changes when gases exceeding a critical value are present under standard conditions.

[0109] The arithmetic unit 2113-1 can perform various operations within the processor 211-1. In one embodiment, the arithmetic unit 2113-1 can be implemented as an arithmetic and logic unit (ALU), but is not limited thereto.

[0110] Memory 212-1 can store data supporting various functions of tracker 200-1 and programs for the operation of processor 210-1, can store input / output data (e.g., music files, still images, videos, etc.), can store a number of application programs (or applications) run by tracker 200-1, and data and commands for the operation of tracker 200-1. At least some of these application programs can be downloaded from an external server via wireless communication. The memory 212-1 may include at least one type of storage medium selected from the group consisting of flash memory, hard disk, solid state disk (SSD), silicon disk drive (SDD), micro multimedia card, card-type memory (e.g., SD or XD memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, and optical disk. The memory 212-1 may store a lookup table (LUT) including error data measured for each ship's movement route, an LUT including error data measured for each area in which the ship can be located, and an LUT for each cargo loaded on the ship.

[0111] The communication module 213-1 may implement a communication interface. The communication interface may include one or more components that enable communication with an external device. For example, the communication interface may include at least one of a wired communication module, a wireless communication module, and a short-range communication module. The wired communication module may include various wired communication modules such as a local area network (LAN) module, a wide area network (WAN) module, or a value-added network (VAN) module, as well as various cable communication modules such as Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), Digital Visual Interface (DVI), recommended standard 232 (RS-232), power line communication, or plain old telephone service (POTS). The wireless communication module may include a Wi-Fi module, a wireless broadband module, and other wireless communication modules that support various wireless communication methods such as GSM (global system for mobile communication), CDMA (code division multiple access), WCDMA (wideband code division multiple access), UMTS (universal mobile telecommunications system), TDMA (time division multiple access), LTE (long term evolution), 4G, 5G, 6G, etc. The wireless communication module may include a wireless communication interface including an antenna for transmitting signals and a transmitter.The wireless communication module may further include a signal conversion module that modulates a digital control signal output from the processor 210-1 through the wireless communication interface into an analog wireless signal under the control of the processor 210-1. The short-range communication module is for short-range communication and is compatible with Bluetooth (registered trademark). TM The device may support short-range communication using at least one of the following technologies: RFID (Radio Frequency Identification), Infrared Data Association (IrDA), UWB (Ultra-Wideband), ZigBee (registered trademark), NFC (Near Field Communication), Wi-Fi (Wireless-Fidelity), Wi-Fi Direct, and Wireless Universal Serial Bus (Wireless USB).

[0112] The gas sensor 220-1 can sense gases generated in the battery. In one embodiment, the gases can be, but are not limited to, sulfur oxides.

[0113] In one embodiment, the artificial neural network processing unit 2111-1 can receive gas sensing data from the gas sensor 220-1. The artificial neural network processing unit 2111-1 can then receive the sensing data. The artificial neural network processing unit 2111-1 can train an artificial intelligence model based on the sensing data and the sensed gas sensing data.

[0114] In one embodiment, the artificial neural network processing unit 2111-1 can evaluate the performance of the artificial intelligence model based on the change pattern predicted by the artificial intelligence model and the gas sensing data, and can tune or fit the artificial intelligence model according to the performance of the artificial intelligence model.

[0115] 13b, a system 200b-1 may be used to detect an abnormal state of a battery included in a ship in a marine shipping environment. The system 200a-1 may include a first device 100b-1, a second device 210b-1, and a gas sensor 220b-1. Descriptions of the second device 210b-1 and the gas sensor 220b-1 that overlap with those in FIG. 13a will be omitted.

[0116] The first device 110b-1 can include a sensor subsystem 110b-1, a processor 120b-1, a memory 130b-1, and a communication module 140b-1.

[0117] The sensor subsystem 110b-1 may include, but is not limited to, at least one of a temperature sensor, an illumination sensor, a humidity sensor, a proximity sensor, an acceleration sensor, a gravity sensor (G-sensor), a gyroscope sensor, a motion sensor, an infrared sensor (IR sensor), a finger scan sensor, an optical sensor, an ultrasonic sensor, an infrared ray sensor, a magnetic sensor, an RGB (RGB) sensor, a radar sensor, a current sensor, an environmental sensor (e.g., a barometric pressure sensor, a radiation detection sensor, a heat detection sensor, a gas detection sensor, etc.), a chemical sensor (e.g., a healthcare sensor, a biometric recognition sensor, a gas leak monitoring sensor, etc.), and a virtual sensor that performs a function corresponding to the corresponding hardware sensor. Here, the proximity sensor may be a sensor that detects the presence or absence of an object approaching or present in the vicinity of a predetermined detection surface without mechanical contact by using electromagnetic force or infrared rays, etc. At least one such sensor may be built into the sensor subsystem 110b-1. The function of each sensor can be intuitively inferred by those skilled in the art from its name, so detailed description will be omitted.

[0118] The first device 110b-1 may detect an abnormal state of a battery included in the ship while the ship is in a maritime shipping environment. In one embodiment, the processor 110b-1 may store program code corresponding to an artificial neural network processing unit and a virtual sensing processing unit in the memory 130b-1, or may be installed in the processor 110b-1 as firmware to perform the corresponding functions. In the present disclosure, the artificial neural network processing unit or the virtual sensor processing unit may be an operating module logically operating within the chipset of the processor 120b-1. For example, the artificial neural network processing unit may be stored in the memory 130b-1 as program code and may represent a functional unit that embodies a machine learning model trained to achieve a specific purpose by being fetched and sequentially analyzed by the processor 211-1.

[0119] The second device 210b-1 may be installed outside the marine transport environment of the ship. For example, the second device 210b-1 may be a server that communicates with external devices and processes information, and may include an application server, computing server, database server, file server, game server, mail server, proxy server, cloud and web server, etc. In this case, the artificial neural network processing unit 2111b-1 may train a machine learning model on a large scale in the server environment or calculate hyper-parameter values ​​that minimize a loss function for the trained model.

[0120] The first device 110b-1 can store the machine learning model calculated by the second device 210b-1 and the hyperparameter values ​​that drive the machine learning model in the memory 130b-1, and the processor 120b-1 can analyze the program code for driving the machine learning model to quickly and lightweightly derive computing results that are essentially the same as the inference values ​​of the artificial neural network.

[0121] FIG. 14 is a diagram for explaining the sensing operation of the gas sensor of the present disclosure.

[0122] Referring to Figure 14, the gas sensor may include a sensing electrode, a counter electrode, and a current source. Gas molecules generated by battery damage may be adsorbed onto the sensing electrode. In this case, ions may move from the sensing electrode to the counter electrode, and electrons may move along the conductor connected between the sensing electrode and the counter electrode, generating a current in the current source, thereby detecting the leaked gas.

[0123] FIG. 15 is a diagram for explaining the artificial intelligence model of the present disclosure.

[0124] Referring to FIG. 15, in the training stage of the artificial intelligence model, data 711 on temperature and humidity change patterns in an enclosed space corresponding to a gas leak may be stored in a database. The data 711 on the temperature and humidity change patterns stored in the database may be applied to a machine learning algorithm or an artificial intelligence model. The data 711 may be applied to the artificial intelligence model to train the artificial intelligence model. The fundamentals of the training model 713 may include AI for material design, such as generative modeling, explainable AI, continual AI, and representation learning. The language of the training model 713 may include SP-BERT, MRC / QA, text analysis, dialog system, GPT-3, GPT-4, etc. The vision of the training model 713 may include visual analytics, visual understanding, video synthesis, etc. The data intelligence of the training model 713 may include anomaly detection, prediction, time-series forecasting, optimization, recommendation, data creation, etc. The amount of leaked gas detected may be obtained as the output of the training model 713. The above-described process may be repeated, and the training model 713 may be fitted and tuned based on the output of the training model 713.

[0125] After the training model 713 is developed, in the inference stage of the artificial intelligence model, environmental information 721 about the cargo loaded on the ship can be input into the trained model 723, and the amount of gas leakage can be predicted by the trained model 723.

[0126] FIG. 16 is a diagram for explaining virtual sensing according to the present disclosure.

[0127] 16, in step S100-1, text-based structuring of sulfur oxides may be performed. For example, the virtual sensing processing unit 2112-1 may generate structured data by structuring the chemical structural formula of the gas on a text basis. The chemical structural formula of sulfur oxides may be restructured in a text-based, machine-analyzable manner. Examples of the restructuring may include SMILES, morecular graphs, and message passing neural networks.

[0128] In step S200-1, conversion into restructured spectroscopic data may be performed. For example, the virtual sensing processing unit 2112-1 may convert the structured data into restructured spectroscopic data.

[0129] In step S300-1, a prior probability that sulfur oxides exceeding a critical value will be present under specific temperature and humidity conditions may be calculated based on the spectroscopic data. For example, the virtual sensing processing unit 2112-1 may calculate a prior probability that gases exceeding a critical value will be present under specific temperature and humidity conditions based on the spectroscopic data.

[0130] In step S400-1, a posterior probability of sulfur oxide behavior can be calculated by applying Bayesian estimation to the prior probability based on the actual measurement log (temperature / humidity). For example, the virtual sensing processing unit 2112-1 can calculate the posterior probability of gas behavior by applying a Bayesian estimation algorithm to the prior probability based on the actually measured temperature and humidity.

[0131] In step S500-1, the behavior of sulfur oxides according to temperature and humidity may be estimated based on the posterior probability. For example, the virtual sensing processing unit 2112-1 may estimate the behavior of the gas according to temperature and humidity based on the posterior probability.

[0132] In step S600-1, the transition of temperature and humidity changes when sulfur oxides exceeding a critical value are present under standard conditions may be acquired. For example, the virtual sensing processing unit 2112-1 may acquire the transition of temperature and humidity changes when gases exceeding a critical value are present under standard conditions.

[0133] 17 and 18 are diagrams for explaining the embodiment of FIG.

[0134] Referring to Figures 17 and 18, the virtual sensor learns the temperature / humidity change pattern 710 within the enclosed space that corresponds to gas leakage, and the virtual sensor can execute logic to detect specific abnormal patterns as gas leakage based on the sensing data (temperature / humidity).

[0135] Organic redox-active molecules are suitable as reactants in redox flow batteries (RFB-1) due to their low expected cost and widely tunable properties. Many laboratory-scale flow cells experience rapid material degradation and cycling capacity loss (>0.1% / day) due to chemical and electrochemical decay mechanisms, which can hinder commercial distribution. This work combines UV-visible spectrophotometry and statistical inference techniques to elucidate the Michael attack decay mechanism for 4,5-dihydroxy-1,3-benzenedisulfonic acid (BQDS), a once-promising cathode electrolyte reactant for aqueous organic redox flow batteries. This work uses Bayesian inference and multivariate curve resolution on spectroscopic data to derive the Michael attack reaction sequence and rate with quantified uncertainty, estimate the spectra of intermediate species, and establish a quantitative relationship between molecular decay and capacity loss. This work demonstrates the use of statistical inference to elucidate the chemical and electrochemical mechanisms of capacity loss in organic redox flow batteries, along with quantification of uncertainty in flow cell-based electrochemical systems. (720)

[0136] Given the intermittent characteristics of renewable energy sources, the development of low-cost, grid-scale energy storage devices may be necessary for the widespread adoption of many renewable energy sources. In this context, a redox flow battery (RFB) can consist of a pair of electrolyte reservoirs containing charge-storing redox-active materials separated by an ion-permeable or separator membrane. The electrolyte is pumped through a reactor cell and can be cyclically oxidized and reduced as the RFB is charged and discharged. This architecture gives RFBs the unique ability to independently scale their energy storage capacity (scalable by the volume of the electrolyte reservoir and the concentration of the charge-storing species) and power (scalable by the size of the reactor cell stack). As the energy-to-power ratio (or rated discharge duration) increases, the levelized cost of the stored energy can be very close to the cost of the electrolyte. With very inexpensive electrolytes, this cost can be lower than that of a standard sealed (lithium-ion) battery.

[0137] Water-soluble organic and organometallic redox-active molecules have attracted considerable research interest as potential charge carriers for RFBs due to their expected low cost of scale. Therefore, when combined with low chemical cost and the right combination of solubility, chemical stability, and other electrochemical properties (e.g., redox potentials leading to high cell voltages and fast redox kinetics), organic RFBs can store energy at costs lower than state-of-the-art lithium-ion systems. Unfortunately, many aqueous-organic flow batteries suffer from hourly capacity loss rates exceeding 0.1% per day due to rapid chemical decomposition of the organic active materials. These high degradation rates make most organic RFB chemistries unsuitable for practical deployment in RFB installations expected to last for decades.

[0138] Because redox-active organic molecules encompass a wide variety of species and are susceptible to diverse degradation mechanisms (e.g., nucleophilic attack, swelling, hydrolysis), understanding how reactant transformation or decomposition leads to capacity fade can be an important but often challenging task. Such understanding often requires the deployment of new operando measurement tools and cycling protocols that allow the deconvolution of reactant decomposition from other causes of capacity fade. For various candidate RFB charge carriers, such as quinones, iron-based organometallic complexes, and nitrogen-containing aromatic molecules, numerous hypotheses have been proposed for the relationship between molecular decay and capacity fade, some of which may be mutually exclusive. Other chemistries, such as those recently developed based on fluorenone, may involve complex equilibria between species with different redox and quantum states whose impact on capacity retention is not yet fully understood. These challenges require new techniques to distinguish the probability or relative contribution of various hypothetical mechanisms to the capacity fade observed in flow cells. In particular, understanding and rigorously quantifying the extent to which experimentally collected data confirm or challenge specific hypotheses for the chemical and electrochemical causes of capacity fade is crucial for developing organic RFB chemistries.

[0139] This problem can be approached broadly as an estimation or inference process (730), which involves statistical learning of physical models and their parameters from experimental observations. Estimation can be based on the concept of regression, where the goal is to find optimal parameter values ​​(as described) so that model predictions best fit the observations. However, such optimal values ​​are typically single values ​​and do not quantify uncertainty, which is influenced by factors such as the quantity and quality of the observations. In contrast, inference can employ probabilistic solutions to convey the degree of uncertainty regarding the various possible explanations that may give rise to the observed data. Inference is typically performed using probability axioms and Bayesian theorem, where an initial prior uncertainty distribution can be appropriately updated with a posterior uncertainty distribution to account for newly acquired observations. Bayesian update rules can naturally integrate new data that may be gradually incorporated over time, providing a consistent representation of the evidence aggregate.

[0140] Bayesian inference can also be advantageous for accommodating rare and noisy indirect measurements, integrating datasets from diverse sources and of diverse quality, and injecting domain knowledge and expert opinion into the learning process. Beyond parametric inference, the Bayesian framework can be extended to model selection to compare "packages" of different hypotheses and assumptions represented by different model structures and parametrizations (e.g., different reaction mechanisms).

[0141] Bayesian inference and related probabilistic techniques have been applied to a variety of problems in electrocatalysis and battery science, including failure prediction and development of life-extending charging protocols for lithium-ion batteries, analyte labeling, model / variable selection and parameter estimation for lithium-ion battery electrodes, electrochemical cell design, Tafel slope analysis, and materials discovery.

[0142] In this disclosure, Bayesian inference and multivariate curve resolution-intersection least squares (MCR-ALS) were applied to perform spectroscopic analysis of the decomposition of oxidized 4,5-dihydroxy-1,3-benzenedisulfonic acid (BQDS) or tiron, an ortho-hydroquinone derivative, previously irradiated onto a cathode electrolyte material in an aqueous RFB. While Bayesian inference has broad applications, MCR-ALS is particularly suited to spectrophotometers because it uses iterative optimization under well-defined physical constraints to resolve the mixture signal of a multicomponent system into its pure components. MCR-ALS can be applied to understand the differentiation of multicomponent chemical systems through optical absorbance, where the absorbance at a given wavelength can be linearly proportional to the concentration of each component. Previous disclosures have shown that oxidized BQDS is vulnerable to a self-discharge reaction with water known as Michael addition / attack, resulting in the formation of a series of hydroxyl-substituted parahydroquinone species with lower redox potentials than BQDS. However, the specific rate of Michael addition and whether such rate is modified under operating cycling conditions are unknown. This information is very important for establishing a quantitative relationship between reactant decay / conversion and capacity loss.

[0143] Referring to FIG. 17 , a Bayesian model can be selected to identify the most likely kinetic scheme for the decay of BQDS based on UV-visible spectrophotometry of the sacrificial oxidant. Bayesian parameter inference and MCR-ALS can be applied to the UV-visible data to obtain uncertainty-quantified estimates for the Michael attack rate of BQDS both in situ and within the operational flow cell. MCR-ALS can be applied to spectroscopic data acquired from an operational BQDS-containing flow cell to individually isolate the UV-visible spectra of all oxidation and Michael attack products. This disclosure applies Bayesian model selection and inference and multivariate curve resolution techniques to spectroscopic data acquired in situ and in the operando flow cell to clarify and quantify the Michael attack kinetics of BQDS. In the “Spectroscopic Measurements of Michael Attack of BQDS” section, experimental details and measurements of Michael attack of BQDS via UV versus spectrophotometry versus sacrificial oxidant concentration are reported. In the section "Model Selection and Uncertainty Using Bayesian Inference and Multivariate Curve Resolution Analysis—Quantifying Michael Attack Rates," Bayesian model selection and inference of associated decay rate constants are applied to UV data. In the section "Spectrophotometric Analysis of BQDS Decomposition in an Operating Flow Cell," experimental details regarding BQDS oxidation and Michael attack via electrochemical cycling are reported. In the section "Estimation of Decay Rate Constants and UV Spectra of Oxidation Products," MCR-ALS is applied to operando UV data to extract the spectra for each oxidation / intermediate product as well as the associated decay rate constants. Finally, in the "Discussion" section, nuclear magnetic resonance (NMR) analysis and density functional theory (DFT) calculations are used to calculate the reaction energy for Michael addition to BQDS and confirm the results. In the "Methods" section, all experimental and computational methods are reported. This disclosure can be used to elucidate and differentiate the chemical and electrochemical mechanisms of capacity decline in organic RFBs using statistical inference techniques and to understand molecular transformations over various time scales in other flow cell-based electrochemical applications.

[0144] Referring to Figure 18, assuming first-order kinetics for Michael attack and k and k values ​​set at the highest posterior probability, the expected time evolution of the concentrations of all oxidizing species can be shown when the initial concentrations of KCrO and BQDS are 0.4 and 0.2 mM, respectively. Also, assuming rate constants for Bayesian inference, the calculated evolution of oxidizing species for the cases of initial 0.4 mM [KCrO] and initial 0.2 mM [BQDS] can be illustrated in Figure 18.

[0145] FIG. 19 is a flowchart illustrating a method according to the present disclosure.

[0146] 19, a method according to the present disclosure may be a method for sensing and correcting a maritime shipping environment of a ship. The method may include an artificial intelligence model generation step (S1000-1) and a gas detection step (S2000-1).

[0147] The artificial intelligence model generation step (S1000-1) is a step of generating an artificial intelligence model that learns the change patterns of temperature and humidity in a closed space contained in the sensing data.

[0148] The gas detection step (S2000-1) is a step in which changes in temperature and humidity within the enclosed space are detected as gas leakage occurring in the battery corresponding to the temperature and humidity within the enclosed space based on the artificial intelligence model and sensing data received from the ship.

[0149] Meanwhile, the disclosed embodiments may be embodied in the form of a recording medium storing computer-executable instructions. The instructions may be stored in the form of program code, which, when executed by a processor, generates program modules to perform the operations of the disclosed embodiments. The recording medium may be embodied as a computer-readable recording medium.

[0150] Computer-readable recording media include all types of recording media that store computer-readable instructions, such as ROM (Read Only Memory), RAM (Random Access Memory), magnetic tape, magnetic disk, flash memory, and optical data storage devices.

[0151] As mentioned above, the disclosed embodiments have been described with reference to the attached drawings. Those skilled in the art will understand that the present disclosure may be implemented in forms different from the disclosed embodiments without changing the technical idea or essential features of the present disclosure. The disclosed embodiments are illustrative and should not be interpreted as limiting.

[0152] Components for Example 2: Claim 1: A system for detecting an abnormal state of a battery included in a ship within the maritime transport environment of the ship, comprising: an artificial neural network processor that generates an artificial intelligence model that learns change patterns in temperature and humidity within an enclosed space contained in sensing data generated from a sensing device; and a virtual sensing processor that receives the OTQ sensing data from outside and, based on the artificial intelligence model and the OTQ sensing data received from the ship, detects changes in the temperature and humidity within the enclosed space as a leakage of gas generated in the battery corresponding to the temperature and humidity within the enclosed space.

[0153] Claim 2: The system of claim 1, further comprising a gas sensor for detecting gas generated in the battery, wherein the artificial neural network processor trains the artificial intelligence model based on the sensing data and gas sensing data sensed by the gas sensor.

[0154] Claim 3: The system according to claim 2, wherein the artificial neural network processor evaluates the performance of the artificial intelligence model based on the change pattern predicted by the artificial intelligence model and the gas sensing data, and tunes the artificial intelligence model based on the performance of the artificial intelligence model.

[0155] Claim 4: The system according to claim 3, wherein the virtual sensing processor generates structured data by structuring the chemical structural formula of the gas on a text basis.

[0156] Claim 5: The system according to claim 4, wherein the virtual sensing processor converts the structured data into the restructured spectroscopic data.

[0157] Claim 6: The system according to claim 5, wherein the virtual sensing processor calculates a prior probability that a gas above a critical value will behave under specific temperature and humidity conditions based on the spectroscopic data.

[0158] Claim 7: The system of claim 6, wherein the virtual sensing processor applies a Bayesian estimation algorithm to prior probabilities based on the actually measured temperature and humidity to calculate posterior probabilities of the gas behavior.

[0159] Claim 8: The system according to claim 7, wherein the virtual sensing processor estimates the behavior of the gas by temperature and humidity based on the posterior probability.

[0160] Claim 9: The system according to claim 8, wherein the virtual sensing processor acquires the transition of temperature and humidity changes when gas exists at or above a critical value under standard conditions.

[0161] Claim 10: A method for sensing and correcting the marine transportation environment of a ship, comprising: an artificial intelligence model generation step of generating an artificial intelligence model that learns patterns of changes in temperature and humidity within an enclosed space contained in sensing data generated from a sensing device; and a gas detection step of detecting changes in the temperature and humidity within the enclosed space as leakage of gas generated in the battery corresponding to the temperature and humidity within the enclosed space based on the artificial intelligence model and the sensing data received from the ship.

[0162] Claim 11: A computer program stored on a recording medium that, in combination with hardware, executes the method of claim 10.

Claims

1. 1. An apparatus for sensing and correcting a marine transport environment of a vessel, comprising: an inertial sensor for measuring the roll of the vessel; and A device comprising: a processor that receives sensing information on the temperature and humidity inside cargo contained in a ship from the outside, verifies the temperature and humidity of the sensing information based on at least one of a Global Positioning System (GPS) that indicates the position of the ship and an Estimated Time of Arrival (ETA) that indicates when the ship is scheduled to arrive at a destination on its navigation route, and corrects transportation environment information including the temperature and humidity based on the verification result.

2. The processor: a verification unit that measures the current position of the ship based on GPS, verifies the temperature and humidity inside the cargo based on the temperature and humidity predicted at the current position of the ship and a preset first reference value, and measures an externally input ETA along the ship's navigation route, and verifies the temperature and humidity inside the cargo based on the temperature and humidity predicted by the current ETA of the ship and a preset second reference value; a calculation unit for calculating correction parameters based on the currently measured temperature and humidity and reference values; 2. The apparatus according to claim 1, further comprising a correction unit for correcting the transportation environment based on the verification result of the verification unit and the correction parameter of the calculation unit.

3. The verification unit Preliminarily estimating the temperature, humidity and ETA for each predicted flight path according to the flight route; The arithmetic unit Calculating the correction parameters based on the temperature and humidity and ETA for each predicted flight path along the flight route; The correction unit 3. The device of claim 2, further comprising updating time difference and humidity change, and updating average temperature change by longitude.

4. The verification unit We verified the temperature and humidity by comparing the estimated values ​​with the actual log data. If an error occurs in the temperature and humidity logs, the error data is updated. The arithmetic unit calculating the correction parameters in response to the time error; The correction unit The device according to claim 3, characterized in that it updates flight information by analyzing the cause of the errors based on the log errors and time errors for the temperature and humidity.

5. The method further includes a memory for storing a look-up table (hereinafter, "LUT") including error data measured for each movement route of the ship, an LUT including error data measured for each area where the ship can be located, and an LUT for each cargo loaded on the ship, The verification unit 5. The apparatus of claim 4, wherein the temperature and humidity are verified based on a plurality of LUTs stored in the memory.

6. The inertial sensor an angular velocity sensor (gyroscope) for measuring the angular velocity of the vessel; an acceleration sensor (accelerometer) for measuring the acceleration of the vessel; and 6. The device according to claim 5, wherein the device is an IMU (Inertial Measurement Unit) including a geomagnetic sensor (Magnetometer) for measuring the geomagnetic field of the ship.

7. a salinity sensor for measuring salinity generated on the vessel in the sea around the vessel in order to track the effects on the precision machinery; The verification unit The apparatus of claim 6, wherein the temperature and humidity are verified based on the salinity sensed by the salinity sensor.

8. 8. The apparatus according to claim 7, further comprising a communication module for transmitting the vessel information including the corrected temperature and humidity to an external device via a communication network.

9. 1. A method for sensing and correcting a maritime shipping environment of a vessel, comprising: a sensing information receiving step of receiving sensing information on the temperature and humidity inside the cargo contained in the ship from the outside; a measuring step of measuring the roll of the vessel; a verification step of verifying the temperature and humidity of the continuous sensing information based on at least one of a GPS indicating the position of the vessel and an ETA when the vessel is expected to arrive at the destination on the travel route; and The method further comprises a correction step of correcting the transportation environment information including the temperature and the humidity based on the verification result.

10. A computer program stored on a recording medium which, in combination with hardware, performs the method of claim 9.

Images