LLM-based eco-friendly ship fuel supply and demand path optimal visualization simulator

The LLM-based simulator addresses the challenge of optimizing eco-friendly fuel supply routes by integrating port information and regulatory compliance, enabling real-time verification and visualization of sailing routes for eco-friendly ships.

KR103044632B1Active Publication Date: 2026-11-06NEWGEN I&S CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
KR1020230162981
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2023-11-22
Publication Date
2026-11-06
Estimated Expiration
2043-11-22

AI Technical Summary

Technical Problem

Shipping companies face difficulties in optimizing sailing routes due to the lack of interconnected supply chains for eco-friendly fuels, with limited information on ports capable of supplying fuels like LNG, methanol, and hydrogen, and existing systems fail to provide real-time verification of these ports and their operational conditions.

Method used

An LLM-based optimal visualization simulator that includes a search module for identifying eco-friendly fuel supply ports, a virtual module for simulating entry and bunkering conditions, and a visual module for route visualization, utilizing a Large Language Model (LLM) to integrate specialized site information and construct a self-contained database for real-time port verification.

Benefits of technology

Enables real-time identification of eco-friendly fuel supply ports, verifies entry and bunkering conditions, and provides optimized navigation routes, ensuring compliance with regulatory requirements and reducing operational disruptions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112023130229220-PAT00001_ABST
    Figure 112023130229220-PAT00001_ABST
Patent Text Reader

Abstract

The present invention relates to a simulation of fuel for eco-friendly ships, and more specifically, to an LLM-based optimal visualization simulator for the feasibility of fuel supply routes for eco-friendly ships, in which ports capable of supplying fuel to eco-friendly ships are searched, simulations of various conditions at ports capable of bunkering are provided, and a navigation route to the optimal bunkering port is provided as a visualization. The technical gist of the present invention is that an LLM-based optimal visualization simulator for a fuel supply route for an eco-friendly ship according to the above-mentioned purpose comprises: a search module for searching for ports capable of supplying fuel to an eco-friendly ship; a virtual module for providing simulations of various conditions at ports capable of bunkering; and a visual module for providing a visualization of the ship's operating route including the optimal bunkering port.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a simulation of fuel for eco-friendly ships, and more specifically, to an LLM-based optimal visualization simulator for the feasibility of fuel supply routes for eco-friendly ships, in which ports capable of supplying fuel to eco-friendly ships are searched, simulations of various conditions at ports capable of bunkering are provided, and a navigation route to the optimal bunkering port is provided as a visualization. Background Technology

[0002] As the International Maritime Organization (IMO) implements various regulatory measures to reduce greenhouse gas (GHG) emissions by 50% compared to 2008 levels by 2050, shipping companies are considering carbon neutrality or the use of carbon-free fuels following nitrogen oxide (NOx) and sulfur oxide (SOx) regulations.

[0003] Existing shipping companies have responded to environmental regulations by switching from traditional ship fuel, bunker fuel, to high-sulfur fuel, installing scrubbers, and expanding orders for LNG-powered ships. However, as regulations on reducing greenhouse gas emissions become more stringent, they are now in a situation where they must consider switching to alternative fuels such as hydrogen, electricity, and biomass, which are carbon-free fuels.

[0004] Countries around the world are considering the development of ships capable of using alternative fuels and the operation of vessels using carbon-free fuels, and major shipyards and shipping companies are also rapidly moving to apply and test new alternative fuels to keep pace with this trend.

[0005] In particular, Maersk is actively pursuing the development of new alternative fuels, such as by ordering methanol-powered ships, and is seeking to procure fuel through its own investments.

[0006] Domestic shipping companies are facing difficulties as they need to actively engage in the development of alternative fuels such as hydrogen, ammonia, and ethanol, yet there is still a grace period of about 30 years remaining until 2050 and there is still a lack of clear direction and alternatives for carbon-free alternative fuels.

[0007] Consequently, as the number of vessels using eco-friendly fuel oil increases due to environmental regulations, difficulties are arising in optimizing sailing routes because the supply chain is not sufficiently interconnected.

[0008] Port facilities capable of supplying various eco-friendly ship fuels (ammonia, methanol, hydrogen, etc.) are a prerequisite; however, while the number of such supply ports is continuously increasing, the current situation is that users are not being properly provided with information regarding the latest supply ports.

[0009] Accordingly, there is a growing need for a simulator capable of AI-based optimal ship route visualization, which can acquire information on the supply of eco-friendly ship fuel through various methods while optimizing ship routes using continuously increasing port information. Prior art literature

[0010] (Patent Document 1) KR 10-2023-0026053 A

[0011] (Patent Document 2) KR 10-1346184 B1 The problem to be solved

[0012] The present invention was developed to resolve the aforementioned problems, and aims to provide an LLM-based optimal visualization simulator for the feasibility of fuel supply routes for eco-friendly ships, in which the status of eco-friendly fuel supply ports is searched using a search module based on a Large Language Model (LM), and information from specialized sites is additionally utilized to enable the construction of a self-contained database of the acquired information, simulations in various situations such as entry and exit, bunkering methods, and special conditions are provided by a virtual module to verify the feasibility of docking at the port and the time required, and various information such as the sailing route to the port and service methods is provided through a visual module.

[0013] The purpose of the present invention is not limited to the purposes mentioned above, and other unmentioned purposes will be clearly understood from the description below. means of solving the problem

[0014] The technical features of the LLM-based eco-friendly ship fuel supply route optimization visualization simulator according to the present invention for achieving the above-mentioned purpose include: a search module for searching ports capable of supplying fuel to eco-friendly ships; a virtual module for providing simulations of various conditions at ports capable of bunkering; and a visual module for providing a visualization of the ship's operating route including the optimal bunkering port.

[0015] In addition, the above virtual module is characterized by the technical feature of including: first virtual information for which a simulation of entry and exit acceptance conditions is provided; second virtual information for which a simulation of bunkering possible methods, required time, and optimal port is provided; and third virtual information for which a simulation of special conditions for performing bunkering operations is provided.

[0016] In addition, the technical features of the first virtual information include a fuel matching method in which the type of fuel of the ship is matched with the type of eco-friendly fuel of the port; a depth matching method in which the draft of the ship is matched with the water depth of the port; a height matching method in which the maximum height of the ship is matched with the height of the piers for entry and exit; and a specification matching method in which the specifications of the ship and the piping are matched.

[0017] In addition, the second virtual information includes a performance matching method in which the bunkering performance method of the vessel and the supply side is matched; a time matching method in which the bunkering time required according to the pump capacity of the supply side is matched; and a search matching method in which the optimal port search considering the arrival time of the vessel's final destination is matched, wherein the performance matching method is characterized by the technical feature that the supply side matches one of the methods of a vessel, terminal, truck, or tank truck depending on the situation.

[0018] In addition, the above-mentioned visual module is characterized by the technical feature that the navigation route is distinguished and displayed according to the type of vessel, and the bunkering operation method is distinguished and displayed according to the type of fuel. Effects of the invention

[0019] The present invention based on the above-described configuration can expect the following effects.

[0020] Conventionally, ports capable of supplying eco-friendly fuel oil consisted of a limited supply chain, making it impossible to verify them in real time. However, the present invention enables real-time searching of ports capable of supplying eco-friendly fuel oil by utilizing LLM-based search and the latest information from specialized sites.

[0021] In addition, a database capable of storing retrieved information is built in-house, making it possible to avoid the volatility and disruption of continuity of web information.

[0022] By matching fuel type, water depth, vessel height, and pipeline size to provide a simulation, it becomes possible to verify the entry and exit permission conditions for ports capable of supplying eco-friendly fuel oil.

[0023] In addition, through matching of bunkering methods, time required, and optimal port search, it becomes possible to verify information such as possible bunkering methods, time required for bunkering, and estimated optimal port.

[0024] Regulations and laws regarding the permissibility of vessels exist in various countries around the world, and the present invention enables the verification of information based on special conditions, thereby making it possible to confirm the entry and exit status and procedures in accordance with these regulations and laws.

[0025] The implementation of a global port map including new eco-friendly fuel supply ports makes it possible to provide services regarding the types of eco-friendly fuels and bunkering methods.

[0026] By visualizing the route on the nautical chart considering the draft of the vessel and the depth of the target port of operation, it becomes possible to easily identify the route to the target port.

[0027] In addition, during the visualization process, the vessels are classified according to their type, making it possible to accurately identify the different movement paths depending on the vessel type. Brief explanation of the drawing

[0028] FIG. 1 is a configuration diagram of an LLM-based eco-friendly ship fuel supply route optimal visualization simulator according to a preferred embodiment of the present invention. FIG. 2 is a configuration diagram of a virtual module in an LLM-based eco-friendly ship fuel supply route optimization visualization simulator according to a preferred embodiment of the present invention. FIG. 3 is a configuration diagram of the first virtual information in an LLM-based eco-friendly ship fuel supply route optimal visualization simulator according to a preferred embodiment of the present invention. FIG. 4 is a configuration diagram of the second virtual information in an LLM-based eco-friendly ship fuel supply route optimal visualization simulator according to a preferred embodiment of the present invention. FIG. 5 is a schematic diagram of a screen output from a visual module of an LLM-based eco-friendly ship fuel supply route optimal visualization simulator according to a preferred embodiment of the present invention. Specific details for implementing the invention

[0029] The LLM-based optimal visualization simulator for possible fuel supply routes of eco-friendly ships according to a preferred embodiment of the present invention will be described in detail below with reference to the attached drawings.

[0030] FIG. 1 is a configuration diagram of an LLM-based eco-friendly ship fuel supply route optimal visualization simulator according to a preferred embodiment of the present invention, FIG. 2 is a configuration diagram of a virtual module in an LLM-based eco-friendly ship fuel supply route optimal visualization simulator according to a preferred embodiment of the present invention, FIG. 3 is a configuration diagram of a first virtual information in an LLM-based eco-friendly ship fuel supply route optimal visualization simulator according to a preferred embodiment of the present invention, FIG. 4 is a configuration diagram of a second virtual information in an LLM-based eco-friendly ship fuel supply route optimal visualization simulator according to a preferred embodiment of the present invention, and FIG. 5 is a schematic diagram of a screen output from a visual module (300) of an LLM-based eco-friendly ship fuel supply route optimal visualization simulator according to a preferred embodiment of the present invention.

[0032] An LLM-based optimal visualization simulator for a fuel supply route for an eco-friendly ship according to a preferred embodiment of the present invention may be composed of a search module (100), a virtual module (200), and a visual module (300) as shown in FIG. 1.

[0034] First, we will briefly examine the LLM that forms the basis of the present invention.

[0035] An LM (Language Model) is a type of artificial intelligence model trained to understand and generate human language.

[0036] It has been primarily utilized in narrow AI tasks such as text translation by learning patterns, structures, and relationships within a given language, and the quality of the language model depends on its size, the amount and diversity of the training data, and the complexity of the learning algorithms used during training.

[0037] Then, LLM (Large Language Model) refers to a large-scale language model, and LLM is used to perform Natural Language Processing (NLP) tasks through deep learning algorithms and statistical modeling.

[0038] This model can understand and generate sentence structure, grammar, and semantics by learning large-scale language data in advance.

[0039] For example, in the problem of predicting the next word in a given context, LLM can generate the next word by identifying the similarity between words within a sentence and the context, and this task is utilized in various NLP tasks such as machine translation, text summarization, automatic writing, and question answering.

[0040] LLM includes various models such as GPT (Generative Pre-trained Transformer) and BERT (Bidirectional Encoder Representations from Transformers), and these models have hundreds of billions of parameters.

[0041] Recently, attention has been focused on achieving more sophisticated language understanding and generation by using large volumes of training data and large model architectures.

[0043] Next, we will examine the above search module (100).

[0044] The search module (100) may be equipped with a function to search for ports capable of supplying eco-friendly fuel oil, which is fuel for eco-friendly ships.

[0045] In detail, the search module (100) can obtain port information capable of supplying fuel to eco-friendly ships by utilizing the latest information from a specialized site for LLM-based search and the provision of objective information.

[0046] Ports capable of bunkering eco-friendly fuel oils such as LNG, methanol, ammonia, and hydrogen are composed of a limited supply chain, and although the number of ports capable of supplying fuel is continuously increasing, there is difficulty in verifying this in real time, so it may be possible to obtain port information in real time through the search module (100) of the present invention.

[0047] As described above, the search module (100) may include a search device capable of searching for LLM-based search and utilizing the latest information from specialized sites, and may include a server device to build a database for storing and analyzing the port information and providing information for the simulator of the virtual module (200) to be described later.

[0048] Accordingly, the search module (100) searches for ports capable of supplying fuel to eco-friendly ships through a search device, and the searched information is transmitted to a server for storage and analysis, and the information that has been analyzed and is ready for output is transmitted to the visual module (300) to be described later and output.

[0049] In addition, among the information for which the analysis and output preparation are completed, information required for the virtual module (200) to be described later may be transmitted to the virtual module (200) to be described later.

[0050] Accordingly, the search module (100) may include a transmission device for transmitting information, and generally includes one server, but it may be possible to include multiple servers.

[0051] In detail, the information transmitted from the search module (100) to the visual module (300) described later is information necessary for a port map, such as port information and navigation routes, and the information transmitted from the search module (100) to the virtual module (200) described later is information necessary for a simulation corresponding to various conditions by matching ship information and port information, and it is possible for the information to have different purposes and types to be formed differently.

[0052] Accordingly, to facilitate diverse and complex information processing, it may be possible to apply multiple servers with functions separated by purpose or type.

[0053] Accordingly, the above server may be configured as one or multiple units, and if configured as multiple units, it may be possible to configure them by assembling or separating them in one location or multiple locations.

[0054] For example, the above server may be composed of a first server, a second server, and a third server, wherein the first server processes information transmitted to the visual module (300), the second server processes information transmitted to the virtual module (200), and the third server may be used only for the collection and transmission processing of information obtained from the first server and the second server.

[0056] Next, we will examine the virtual module (200) above.

[0057] The virtual module (200) may be able to provide simulations of various conditions at a port where bunkering is possible.

[0058] Generally, even if a port is capable of bunkering the fuel, entry and exit and bunkering at the port are possible only if the port specifications and the ship conditions match, and a virtual module (200) may be used to provide information and verify this.

[0059] In detail, the virtual module (200) is provided with simulations of acceptance conditions for entry and exit, bunkering methods, bunkering time, optimal bunkering port, and special conditions for performing bunkering operations, each of which may be composed of different information and may include first virtual information (S210), second virtual information (S220), and third virtual information (S230).

[0060] The first virtual information (S210) is simulation information regarding allowable conditions for entry and exit, and may include a fuel matching method (S211), a water depth matching method (S213), a height matching method (S215), and a specification matching method (S217).

[0061] The fuel matching method (S211) provides a simulation by matching the type of fuel oil of the ship with the type of eco-friendly fuel of the port, and confirms whether the fuel of the ship matches the fuel supplied by the port, and the simulation may be provided for various fuels such as LNG, ammonia, methanol, and hydrogen.

[0062] The depth matching method (S213) provides a simulation by matching the draft of the vessel with the depth of the port, and it is possible to check whether the depth of the port allows the vessel to enter or exit.

[0063] The height matching method (S215) provides a simulation in which the maximum height of the vessel according to the draft at the time of entry and the height of the entry / exit bridge are matched, making it possible to verify whether entry and exit to the port is possible according to the height of the vessel.

[0064] The specification matching method (S217) provides a simulation by matching the piping specifications of the vessel and the supply side, and it may be possible to verify whether the specifications of the piping for supplying fuel match.

[0065] Generally, in accordance with the regulations of the International Maritime Organization (IMO), ships are required to have three pipe connection sizes for each LNG liquid line and vapor (gas) line, and this information must be verified in advance to reduce work time.

[0066] The second virtual information (S220) is simulation information regarding bunkering possible methods, required time, and optimal terms, and may include an execution matching method (S221), a time matching method (S223), and a search matching method (S225).

[0067] The execution matching method (S221) is provided by matching the bunkering execution method of the vessel and the supply side at the port and providing a simulation, so that it may be possible to verify the bunkering execution method for various supply methods for the vessel.

[0068] The above supply method can be applied using various methods such as ships, terminals, trucks, or tank trucks, making it possible to provide the user with information on various supply methods that may occur in various situations.

[0069] Accordingly, depending on the supply method, it is possible to predict whether the vessel will perform bunkering at sea or at a terminal (including a dock), and to determine the vessel's arrival time.

[0070] The time matching method (S223) provides a simulation in which the bunkering time required according to the pump capacity of the supply side is matched, making it possible to confirm the time required for bunkering for the vessel.

[0071] Since the capacity of the pump varies at each port, the time required for bunkering also varies, and the user can check the time required for bunkering according to the schedule and select the optimal port for bunkering.

[0072] Accordingly, when the time spent at the port is short, the time required for bunkering is provided by simulation through information such as the use of a large-capacity pump or the possibility of utilizing multiple pumps, making it possible to identify the optimal port.

[0073] The search matching method (S225) provides a simulation by matching the optimal port based on the arrival time of the vessel's final destination, and it is possible to provide and verify simulation information of the optimal bunkering port on the sailing route based on the vessel's schedule.

[0074] The third virtual information (S230) is simulation information regarding the requirements of a vessel required in advance at the relevant port, and generally, the information may be provided based on regulations and laws.

[0075] Some ports have separate requirements for the entry and exit of vessels, and accordingly, the procedures for vessel entry and exit vary, so this information can be essential for vessel bunkering.

[0076] Various conditions used in the third virtual information (S230) are generally used, such as EMSA and USCG regulations, but are applied differently to each port and are provided as simulation information, and the user can easily check the conditions for a vessel to enter or leave a port for bunkering.

[0078] Next, we will examine the above-mentioned visual module (300).

[0079] The visual module (300) may be equipped with a function to implement a map based on information obtained from the search module (100).

[0080] In detail, the visual module (300) can implement a global port map including new eco-friendly fuel supply ports based on information obtained from the search module (100), and can implement a port map that distinguishes and indicates the types of eco-friendly fuel and bunkering operation methods.

[0081] For the service regarding the above bunkering operation method, one or more methods may be used depending on the situation, among a method by ship to ship, a method by terminal to ship, or a method by truck to ship.

[0082] The visual module (300) can visualize a route on a nautical chart that takes into account the draft of the vessel and the depth of the target port of operation, and during the visualization process, it may be possible to visualize it by distinguishing it according to the type of vessel.

[0083] The visual module (300) may apply various visualization technologies, and may be able to visualize information using one or more of the various visualization technologies such as Tableau, Dundas BI, JupyteR, Zoho Reports, Google Charts, Visual.ly, RAW, and IBM Watson.

[0084] Additionally, the visual module (300) may include an output device to visualize information, and various devices such as a monitor, terminal, tablet, and PC may be used as one or multiple devices so that the user can easily check them.

[0085] The visual module (300) may be configured as a whole or separately from the search module (100) and virtual module (200), and may include various types of output devices such as fixed output devices and mobile output devices as described above.

[0086] Accordingly, it may be possible for the user to verify acquired information, such as port information and navigation routes, not only at a fixed location but also while moving or at a location other than the vessel.

[0088] Based on the configuration described above, the operating state of the present invention is examined.

[0089] When operating a vessel, there may be cases where bunkering to supply fuel is necessary during operation, and by using the present invention, it becomes possible to search for ports where bunkering is possible.

[0090] Accordingly, the present invention applies a search module (100) to analyze the type of fuel based on the basic level of the ship, and analyzes and collects bunkering ports, while enabling the use of LLM-based search and the latest information from various objective specialized sites.

[0091] The information described above can be easily verified by the user through the visual module (300), and the visual module (300) can be used in various ways, such as a monitor or a portable terminal, which is easy for the user to verify.

[0092] When various ports are recommended and verified to the user based on the above description, it becomes possible to provide various information regarding bunkering at the port through simulation via the virtual module (200).

[0093] Through the virtual module (200), the bunkering method, time required, port entry / exit permission conditions, and various special conditions are provided to the user as simulation information, and the provided information can be verified through the visual module (300).

[0094] Accordingly, the user can select a port within the port where bunkering is possible, according to the condition of the vessel, the user's convenience, and other circumstances, and detailed information for bunkering can be provided by the present invention, which includes a virtual module (200) and a visual module (300).

[0095] In addition, the visual module (300) implements and provides a global port map including new eco-friendly fuel supply ports, and it is possible to visualize the ship's route using various visualization technologies.

[0096] In the visualization process, it becomes possible to distinguish between types of ships according to the type of ship, and to briefly additionally display the types of eco-friendly fuels and bunkering methods for the port.

[0097] In addition, it becomes possible to visualize the route on the nautical chart, taking into account the draft of the vessel and the depth of the target port of operation.

[0098] Accordingly, users can easily grasp information about the overall port, brief information about each port, and information about the shipping routes to the respective ports.

[0100] The above-described embodiments are merely exemplary, and various other modified embodiments are possible therefrom for those skilled in the art.

[0101] Therefore, the true technical scope of protection of the present invention should include not only the above embodiments but also other embodiments that are variously modified according to the technical concept of the invention described in the following claims. Explanation of the symbols

[0102] 100 : Search module 200 : Virtual module 300 : Visual Module S210 : 1st virtual information S211: Fuel Matching Method S213 : Depth Matching Method S215 : Height matching method S217 : Standard batching method S220 : Second virtual information S221 : Execution Matching Method S223 : Time Matching Method S225 : Search matching method S230 : Third virtual information

Claims

Claim 1 A search module that searches for ports capable of supplying fuel to eco-friendly vessels based on LLM-based search and the latest information obtained from multiple specialized sites, and stores and analyzes the searched port information and navigation route information; and a virtual module that performs a simulation including bunkering methods, bunkering time, entry and exit permission conditions, and special conditions by matching vessel information and port information among the information transmitted from the search module. An LLM-based optimal visualization simulator for possible routes to supply eco-friendly ship fuel, wherein a global port map including new eco-friendly fuel supply ports is provided based on information transmitted from the search module and virtual module, and a visual module in which the navigation route is visualized on a nautical chart by combining the type of ship, the draft of the ship, and the depth of the target port of operation as input data along with port information and navigation route information; wherein the search module includes a plurality of servers separated to process, respectively, information transmitted to the visual module for port map configuration and navigation route visualization and information transmitted to the virtual module for simulation execution among the information obtained from the search module; and wherein the plurality of servers include a first server in which information transmitted to the visual module is processed; a second server in which information transmitted to the virtual module is processed; and a third server in which information obtained from the first server and the second server is collected, transmitted, and processed. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 delete

Citation Information

Patent Citations

  • System and method for providing operating profit of vessel and computer-readable recording medium including the same

    KR1020210128331A

  • Method Of Generating An Optimized Ship Schedule To Deliver Liquefied Natural Gas

    US20150324740A1