Ship repair method, ship repair system, computer device, and storage medium

Abstract

The application relates to a ship maintenance method, a ship maintenance system, a computer device and a storage medium. The method comprises the following steps: acquiring fault information and ship body equipment associated with the fault information; searching for a physical field model corresponding to the ship body equipment and the fault information in a model library; generating a standard solution according to a ternary structure of the physical field model; generating first maintenance instructions and second maintenance instructions which cooperate with each other according to the standard solution and sensing information of the ship body equipment under the condition that available resources of the ship meet resource requirements in the standard solution; and sending the first maintenance instructions to maintenance personnel and sending the second maintenance instructions to maintenance machines. The method can solve the problems of delayed response and high misdiagnosis rate in ship maintenance.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of ship maintenance, and in particular, to a ship maintenance method, a ship maintenance system, a computer device and a storage medium. BACKGROUND

[0002] Traditional ship maintenance relies on manual inspection and regular maintenance, and the selection of tools for ship maintenance relies on manual experience. Engineers need to manually match the task type, which is inefficient and prone to errors. Secondly, engineers need to manually select maintenance tools from the tool list provided by the tool library, which may not be suitable for the current scene. This results in an average fault troubleshooting time of more than 30 minutes for ship maintenance, a response delay problem for ship maintenance, and a fault troubleshooting accuracy of less than 85%, a high misdiagnosis rate problem.

[0003] In view of the problems of ship maintenance response delay and high misdiagnosis rate in the related art, there is currently no effective solution. SUMMARY

[0004] Therefore, it is necessary to provide a ship maintenance method, a ship maintenance system, a computer device and a storage medium capable of solving the problems of ship maintenance response delay and high misdiagnosis rate.

[0005] In a first aspect, a ship maintenance method is provided in the present embodiment, and the method comprises:

[0006] Obtaining fault information and ship body equipment associated with the fault information;

[0007] Searching for a physical field model corresponding to the ship body equipment and the fault information in a model library;

[0008] Generating a standard solution according to the ternary structure of the physical field model;

[0009] When the available resources of the ship meet the resource requirements in the standard solution, generating first maintenance instructions and second maintenance instructions that cooperate with each other according to the standard solution and sensing information of the ship body equipment;

[0010] Sending the first maintenance instructions to maintenance personnel and sending the second maintenance instructions to maintenance machines.

[0011] In some embodiments, searching for a physical field model corresponding to the ship body equipment and the fault information in the model library comprises:

[0012] According to the ship body equipment and the fault information associated with each physical field model in the model library, evaluating the severity, frequency and detection degree of a plurality of physical field models;

[0013] obtaining risk priorities of the object field models based on the severity, the frequency and the detectability;

[0014] searching, in the model library, for an object field model corresponding to the ship equipment and the fault information according to the risk priorities from high to low.

[0015] In some embodiments, the method further comprises:

[0016] In the case where there is no object field model corresponding to the ship equipment and the fault information in the model library, downloading an object field model corresponding to the ship equipment and the fault information based on a first gateway and feeding back update information of the model library based on a second gateway; wherein the first gateway and the second gateway are backup gateways.

[0017] In some embodiments, downloading an object field model corresponding to the ship equipment and the fault information based on a first gateway comprises:

[0018] determining whether the ship equipment and the fault information satisfy a triggering condition of an emergency state;

[0019] if yes, sending a preset emergency instruction to the maintenance personnel and / or the maintenance machine and downloading an object field model corresponding to the ship equipment and the fault information from a satellite by the first gateway;

[0020] if no, sending a preset normal instruction to the maintenance personnel and / or the maintenance machine and downloading an object field model corresponding to the ship equipment and the fault information from a shore base by the first gateway.

[0021] In some embodiments, the obtaining of the fault information comprises:

[0022] establishing an association between a life cycle of the ship equipment and space-time data according to running characteristics of the ship equipment under different time and space conditions;

[0023] mapping the space-time data of the ship running based on the association to obtain a life cycle stage in which the ship equipment is located;

[0024] in the case where the life cycle stage enters a preset stage, generating fault information according to the life cycle stage in which the ship equipment is located.

[0025] In some embodiments, after the first maintenance instruction is sent to the maintenance personnel and the second maintenance instruction is sent to the maintenance machine, the method further comprises:

[0026] integrating the fault information, the ship equipment, field sensing information, the first maintenance instruction and the second maintenance instruction.

[0027] updating the object field model in the model library based on the integration result.

[0028] In some embodiments, updating the model library based on the integration result comprises:

[0029] updating non-standard equipment data of the ship according to the integration result;

[0030] constructing a new object field model according to the non-standard equipment data of the ship, and adding the new object field model in the model library.

[0031] In a second aspect, a ship maintenance system is provided in the embodiments, and the system comprises an operation and alarm information AI generation module, a first gateway module, an artificial intelligence module, and a second gateway module, wherein

[0032] The operation and alarm information AI generation module is configured to determine fault information and ship equipment associated with the fault information.

[0033] The first gateway module is configured to send the fault information and the ship equipment to the artificial intelligence module.

[0034] The artificial intelligence module is configured to search for an object field model corresponding to the ship equipment and the fault information in a model library, generate a standard solution based on a ternary structure of the object field model, and generate a first maintenance instruction and a second maintenance instruction that cooperate with each other based on the standard solution and sensing information of the ship equipment when available resources of the ship meet resource requirements in the standard solution.

[0035] The second gateway module is configured to send the first maintenance instruction to a maintenance personnel and send the second maintenance instruction to a maintenance machine.

[0036] In a third aspect, a computer device is provided in the embodiments, and the computer device comprises a memory and a processor. The memory stores a computer program, and the processor implements the ship maintenance method of the first aspect when executing the computer program.

[0037] In a fourth aspect, a computer readable storage medium is provided in the embodiments, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the ship maintenance method of the first aspect.

[0038] The ship maintenance method, the ship maintenance system, the computer device and the storage medium search for the corresponding object field model of the ship body equipment and the fault information in the model library, obtain the standard solution through the ternary structure reasoning of the object field model, and realize efficient diagnosis of the ship equipment fault; based on the actual resource allocation of the ship, the first maintenance instruction and the second maintenance instruction related to the standard solution and the sensing information of the ship body equipment are automatically generated, the ship maintenance instruction is quickly obtained, the accuracy of the ship maintenance instruction is improved, and the problems of delayed response and high misdiagnosis rate of ship maintenance are solved. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 An application environment diagram of the ship maintenance method in an embodiment is shown;

[0040] Figure 2 A flowchart of the ship maintenance method in an embodiment is shown;

[0041] Figure 3 A structural block diagram of the ship maintenance system in an embodiment is shown;

[0042] Figure 4 A structural block diagram of the ship Internet of Things system in an embodiment is shown;

[0043] Figure 5 A structural block diagram of the ship maintenance system in another embodiment is shown;

[0044] Figure 6 A flowchart of the multi-modal AI ship equipment maintenance and repair method in an embodiment is shown;

[0045] Figure 7 A flowchart of obtaining the standard solution in an embodiment is shown;

[0046] Figure 8 An internal structure diagram of the computer device in an embodiment is shown. DETAILED DESCRIPTION

[0047] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0048] The ship maintenance method provided by the embodiments of the present application can be applied to, for example Figure 1The application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data required by the server 104 to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers. The data storage system can store a model library, search for a physical field model in the model library based on the server, and automatically generate a first maintenance instruction and a second maintenance instruction based on the physical field model, and send the first maintenance instruction and the second maintenance instruction to different terminals. Among them, the terminal 102 can be, but not limited to, various personal computers, notebook computers, tablet computers, Internet of Things maintenance tools and the like. The server 104 can be realized by an independent server or a server cluster composed of multiple servers.

[0049] In one embodiment, as Figure 2 shown, a ship maintenance method is provided, and the method is applied to Figure 1 the server 104 in the above application environment for illustration, including the following steps:

[0050] Step S202, obtaining fault information and ship equipment associated with the fault information.

[0051] Among them, the fault information includes sensor information, fault occurrence time, operation history during fault occurrence, etc. The ship equipment includes mechanical, electrical systems and other devices required during ship operation. Optionally, the fault information can be obtained in one or more of the following ways: monitoring the running state of the system or device through sensors and other means; generating fault information after the technician confirms the device through direct observation; predicting fault information by using frequency, usage time, external environment, etc.

[0052] Step S204, searching for a physical field model corresponding to the ship equipment and the fault information in the model library.

[0053] Among them, the physical field model is composed of matter, field and interaction between the two. The physical field model in the model library is based on the ship equipment to obtain the matter, the matter energy acting on the ship structure as the field, and the fault information that the ship equipment may encounter based on the interaction between the two. Optionally, the model library includes multiple physical field models of each device of the ship body. Based on the preset order, the physical field model is searched in the model library, so that the searched physical field model includes the fault information and the ship equipment obtained in step S202.

[0054] Step S206, generating a standard solution according to the three-element structure of the physical field model.

[0055] The three-element structure of the object field model is the above-mentioned matter, field and interaction between the two. The standard solution is used to convert the incomplete or harmful object field model into an ideal model that effectively solves the problem; the standard solution includes but is not limited to adding new matter or field, removing harmful interaction, etc.

[0056] Optionally, according to the three-element structure of the object field model, the type of the problem between the matter and the field is determined, and the matter, the field or the interaction between the matter and the field is adjusted according to the type of the problem, so as to introduce a suitable standard solution.

[0057] Step S208, generating the first maintenance instruction and the second maintenance instruction which cooperate with each other according to the standard solution and the sensing information of the ship equipment under the condition that the available resources of the ship meet the resource demand in the standard solution.

[0058] The resource demand includes the existing resources such as materials in the ship. The generated first maintenance instruction and the second maintenance instruction can be any one or more multimedia resources such as text, image and video.

[0059] Optionally, under the condition that the available resources of the ship are greater than or equal to the resources required by the standard solution, the multi-source sensing information (such as equipment state parameters, environmental conditions, operator behavior, etc.) of the ship equipment in the maintenance site is collected and transmitted to the artificial intelligence server, and the artificial intelligence server fuses and processes the multi-source sensing information and the standard solution, so as to dynamically generate the first maintenance instruction and the second maintenance instruction based on the standard solution and the real-time sensing information of the ship equipment in the maintenance site, so that the first maintenance instruction and the second maintenance instruction cooperate with each other, and at the same time, the first maintenance instruction and the second maintenance instruction change from static dependence on experience to dynamic adaptation to the site state, thereby improving the safety, accuracy and efficiency of the maintenance work.

[0060] Optionally, under the condition that a plurality of standard solutions are generated according to the three-element structure of the object field model, a standard solution in which the resource demand is met is obtained, and the first maintenance instruction and the second maintenance instruction which cooperate with each other are generated based on the standard solution and the sensing information of the ship equipment.

[0061] Optionally, under the condition that the resource demand of the standard solution generated according to the three-element structure of the object field model is not met at all, it is judged whether the ship equipment and the fault information meet the triggering condition of the emergency state; if yes, a preset emergency instruction is sent to the maintenance personnel and / or the maintenance machine, and then the first gateway downloads the object field model corresponding to the ship equipment and the fault information from the satellite; if not, a preset normal instruction is sent to the maintenance personnel and / or the maintenance machine, and the first gateway downloads the object field model corresponding to the ship equipment and the fault information from the shore base.

[0062] Step S210, sending the first maintenance instruction to the maintenance personnel and sending the second maintenance instruction to the maintenance machine.

[0063] Optionally, the first maintenance instruction is sent in one or more of the following manners: the terminal such as a mobile phone or a tablet of the maintenance personnel sends the first maintenance instruction to the maintenance personnel; the wearable device of the maintenance personnel displays the maintenance steps, operation guidance, remote expert connection, etc.

[0064] In the ship maintenance method, the object field model corresponding to the hull equipment and the fault information searched from the model library is taken as a countermeasure model, a standard solution is obtained based on the ternary structure reasoning of the object field model, efficient diagnosis of the ship equipment fault is realized, the first maintenance instruction and the second maintenance instruction related to the standard solution and the sensing information of the hull equipment are automatically generated based on the actual resource allocation of the ship, so that the first maintenance instruction and the second maintenance instruction adapt to the on-site maintenance state, the possibility of erroneous judgment in the ship equipment fault diagnosis and maintenance decision process is reduced, the accuracy of the ship maintenance instruction is improved, and the problems of delayed response and high misdiagnosis rate of ship maintenance are solved.

[0065] In one embodiment, searching the object field model corresponding to the hull equipment and the fault information in the model library comprises: evaluating the severity, occurrence frequency and detection degree of a plurality of object field models according to the hull equipment and the fault information associated with each object field model in the model library; obtaining the risk priority of the plurality of object field models based on the severity, occurrence frequency and detection degree; and searching the object field model corresponding to the hull equipment and the fault information in the model library in order from high to low according to the risk priority.

[0066] The severity refers to the most serious impact of the fault shown by the object field model on the function and safety of the ship system. The occurrence frequency refers to the possibility of the fault shown by the object field model. The detection degree refers to the ability to discover the fault shown by the object field model based on sensors and ship personnel.

[0067] Optionally, the risk priority of the plurality of object field models is calculated according to the risk priority number, wherein the higher the severity score, the more serious the impact of the fault; and vice versa. The higher the occurrence frequency score, the higher the possibility of the fault; and vice versa. The higher the detection degree score, the lower the ability to discover the fault shown by the object field model; and vice versa.

[0068] In this embodiment, the priority-oriented logic searches the object field model in the model library in order from high to low according to the risk priority, which can preferentially identify and process serious faults that may cause catastrophic consequences, ensure that serious faults of the ship equipment can be preferentially responded and disposed of, thereby improving the ship maintenance response, improving the maintenance decision efficiency, and improving the safety of the ship.

[0069] In an embodiment, the ship maintenance method further comprises: in the case that the object field model corresponding to the ship equipment and the fault information does not exist in the model library, downloading the object field model corresponding to the ship equipment and the fault information based on the first gateway, and feeding back the update information of the model library based on the second gateway; wherein the first gateway and the second gateway are backup gateways for each other.

[0070] The first gateway is configured to receive external request information, and the external request information at least includes a request for the object field model corresponding to the ship equipment and the fault information. The second gateway is configured to feed back information to the outside world, and the feedback information at least includes the update information of the model library. In the case that any gateway fails, the first gateway or the second gateway can replace the failed gateway to undertake all traffic processing tasks, thereby improving the stability of the gateway during the ship voyage.

[0071] Optionally, the first gateway can further be configured to send the fault information and the ship structure to an artificial intelligence server, so that the artificial intelligence server performs the following steps: searching the model library for the object field model corresponding to the ship equipment and the fault information; generating a standard solution according to the three-element structure of the object field model; and generating first maintenance instructions and second maintenance instructions that cooperate with each other according to the standard solution and the sensing information of the ship equipment, in the case that the available resources of the ship meet the resource requirements in the standard solution. The second gateway can further be configured to send the first maintenance instructions and the second maintenance instructions generated by the artificial intelligence server to maintenance personnel and maintenance machines respectively.

[0072] In the embodiment, the receiving and feedback functions of the gateway are separated, and the load balancing configuration can be performed on the first gateway and the second gateway respectively, thereby not only improving the resource utilization rate, but also further enhancing the fault tolerance capability of the ship network system.

[0073] Further, in an embodiment, the downloading of the object field model corresponding to the ship equipment and the fault information based on the first gateway comprises: judging whether the ship equipment and the fault information meet the triggering condition of the emergency state; if yes, sending a preset emergency instruction to the maintenance personnel and / or the maintenance machine, and downloading the object field model corresponding to the ship equipment and the fault information from the satellite by the first gateway; if no, sending a preset normal instruction to the maintenance personnel and / or the maintenance machine, and downloading the object field model corresponding to the ship equipment and the fault information from the shore base by the first gateway.

[0074] The preset emergency instruction is a general maintenance instruction in the emergency state, and the preset normal instruction is a general maintenance instruction for a normal fault. The triggering condition of the emergency state can be judged in the case that the ship equipment is a necessary equipment for the ship operation, and / or the fault information threatens the normal navigation of the ship. Alternatively, the triggering condition of the emergency state can be generated according to the ship equipment and system reliability verification guidelines, and the triggering condition is compared with the ship equipment and the fault information.

[0075] The preset regular instruction and the preset emergency instruction can be preset in the ship knowledge base. Optionally, the ship knowledge base includes maintenance information in an emergency state and a non-emergency state. The preset emergency instruction includes: a ship-wide power failure, the artificial intelligence server generates a third maintenance instruction of the emergency state by using the maintenance information associated with the emergency state in the ship knowledge base to assist in guiding the maintenance operation path; and the third maintenance instruction is pushed to each cabin terminal through a wireless network to synchronize the operation progress.

[0076] In the embodiment, the preset emergency instruction and the preset regular instruction enable the maintenance personnel and / or the maintenance machine to perform preliminary maintenance in the case that there is no object field model corresponding to the ship equipment and the fault information in the model base, so that the ship can obtain a maintenance response in time in various situations and the operation safety of the ship is improved.

[0077] In one embodiment, the fault information is obtained by: establishing an association between a life cycle of the ship equipment and space-time data according to running characteristics of the ship equipment under different time and space conditions; mapping the space-time data of the ship operation based on the association to obtain a life cycle stage in which the ship equipment is located; and generating the fault information according to the life cycle stage in which the ship equipment is located in the case that the life cycle stage enters a preset stage.

[0078] The space-time data of the ship operation includes a navigation log of a speed, a route, a navigation time, weather data such as a wind and wave level, and the like. The space-time data of the ship operation can be used to analyze the influence of the environment on the failure of the ship equipment.

[0079] Optionally, training data is obtained according to the running characteristics of the ship equipment under different time and space conditions, a wear prediction model is constructed based on federated learning and by using the training data, and the wear prediction model is used to indicate the association between the life cycle of the ship equipment and the space-time data.

[0080] The ship equipment is damaged differently in different environments and for different time. The accuracy of directly judging whether the ship equipment is faulty according to the longest service life of the ship equipment is low. In the embodiment, the space-time data of the ship operation is combined to predict whether the life cycle stage of the ship equipment needs to be maintained, and the life cycle stage of the ship equipment is predicted in advance to determine whether the life cycle stage is in a preset stage, so that the maintenance personnel and the maintenance machine can maintain the ship equipment in advance.

[0081] In one embodiment, after the first maintenance instruction is sent to the maintenance personnel and the second maintenance instruction is sent to the maintenance machine, the method further includes: integrating the fault information, the ship equipment, the field sensing information, the first maintenance instruction, and the second maintenance instruction; and updating the object field model in the model base based on the integration result.

[0082] The field sensing information includes equipment operation parameters such as main engine bearing temperature, generator voltage, and the like, which are collected in real time by vibration, temperature, pressure and other sensors deployed in the ship. Optionally, based on the fault information, the first maintenance instruction and the second maintenance instruction, a ship maintenance log and a fault report can be obtained, and structured data such as fault type, occurrence time and maintenance measures can be extracted and used as historical maintenance records. Optionally, multi-source data such as historical maintenance records, field sensing information and fault information are fused, and a physical field model is constructed or updated based on the fusion.

[0083] In this embodiment, new faults are learned and recognized by updating the model library, thereby providing more reliable maintenance instructions and reducing the misdiagnosis rate of ship maintenance.

[0084] In one embodiment, the model library is updated based on the integration result, including: updating the non-standard equipment data of the ship according to the integration result; constructing a new physical field model according to the non-standard equipment data of the ship, and adding the new physical field model to the model library.

[0085] Wherein, after the ship equipment is maintained based on the first maintenance instruction and the second maintenance instruction, the internal equipment of the ship can be replaced from standard equipment to non-standard equipment. Optionally, the non-standard equipment of the ship is scanned to obtain non-standard equipment data, and the non-standard equipment data is analyzed to obtain a triple structure, a physical field model is constructed based on the triple structure, and is added to the model library. In this embodiment, the physical field model is constructed based on the non-standard equipment data, which promotes the continuous improvement of the model library and reduces the misdiagnosis rate of ship maintenance.

[0086] Based on the same inventive concept, the embodiments of the present application also provide a ship maintenance system for implementing the above-mentioned ship maintenance method. The implementation scheme of the system for solving the problem is similar to the implementation scheme described in the above method, so the specific limitations in one or more ship maintenance system embodiments provided below can refer to the limitations of the ship maintenance method described above, and will not be repeated here.

[0087] In one embodiment, Figure 3 A ship maintenance system is provided, and the system comprises: an operation and alarm information AI generation module, a first gateway module, an artificial intelligence module, and a second gateway module; wherein,

[0088] The operation and alarm information AI generation module is configured to determine fault information and ship equipment associated with the fault information;

[0089] The first gateway module is configured to send the fault information and the ship equipment to the artificial intelligence module;

[0090] The artificial intelligence module is used to search the model library for the object-field model corresponding to the ship's equipment and fault information; generate a standard solution based on the ternary structure of the object-field model; and generate a first maintenance command and a second maintenance command that cooperate with each other based on the standard solution and the sensor information of the ship's equipment, provided that the ship's available resources meet the resource requirements in the standard solution.

[0091] The second gateway module is used to send the first maintenance instruction to the maintenance personnel and the second maintenance instruction to the maintenance machine.

[0092] In one embodiment, the artificial intelligence module searches for object-field models corresponding to ship equipment and fault information in a model library, including: assessing the severity, frequency of occurrence, and detectability of multiple object-field models based on the ship equipment and fault information associated with each object-field model in the model library; obtaining the risk priority of multiple object-field models based on the severity, frequency of occurrence, and detectability; and searching for object-field models corresponding to ship equipment and fault information in the model library according to the order of risk priority from high to low.

[0093] In one embodiment, the artificial intelligence module is further configured to, when no object-field model corresponding to the ship's equipment and fault information exists in the model library, download the object-field model corresponding to the ship's equipment and fault information based on a first gateway, and provide feedback on the updated information of the model library based on a second gateway; wherein the first gateway and the second gateway serve as backup gateways for each other. Optionally, downloading the object-field model corresponding to the ship's equipment and fault information based on the first gateway includes: determining whether the ship's equipment and fault information meet the triggering conditions of an emergency state; if so, sending a preset emergency command to maintenance personnel and / or maintenance equipment, and having the first gateway download the object-field model corresponding to the ship's equipment and fault information from the satellite; if not, sending a preset regular command to maintenance personnel and / or maintenance equipment, and having the first gateway download the object-field model corresponding to the ship's equipment and fault information from the shore base.

[0094] In one embodiment, the operation and alarm information AI generation module obtains fault information by: establishing a correlation between the life cycle of the ship equipment and spatiotemporal data based on the operating characteristics of the ship equipment under different time and space conditions; mapping the spatiotemporal data of ship operation based on the correlation to obtain the life cycle stage of the ship equipment; and generating fault information based on the life cycle stage of the ship equipment when the life cycle stage enters a preset stage.

[0095] In an embodiment, after sending the first maintenance instruction to the maintenance personnel and sending the second maintenance instruction to the maintenance machine, the artificial intelligence module is further configured to integrate the fault information, the hull equipment, the on-site sensing information, the first maintenance instruction and the second maintenance instruction; and update the object field model in the model library based on the integration result. Optionally, updating the model library based on the integration result comprises: updating the non-standard equipment data of the ship according to the integration result; and constructing a new object field model according to the non-standard equipment data of the ship, and adding the new object field model to the model library.

[0096] Each module in the ship maintenance system described above can be implemented wholly or partially by software, hardware and combinations thereof. Each module described above can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform the operations corresponding to each module.

[0097] In an embodiment, Figure 4 A ship Internet of Things system constructed based on a distributed computing framework is provided. Under the distributed computing framework, the ship Internet of Things system realizes efficient collection and processing of various data such as PLCs (Programmable Logic Controllers), instruments, power equipment and ship environments on the edge side through the system architecture of AI servers, local area networks and multi-terminal collaboration. The system relies on a high-reliability industrial intelligent gateway with protocol analysis capability and edge computing function as a core edge node to provide strong edge support for ship intelligent applications.

[0098] The local area network is constructed based on the following: a shipboard wireless network constructed based on the IEEE 802.11ax protocol, and the local area network can provide end-to-end encrypted communication with a transmission rate greater than or equal to 1.2 Gbps. The encrypted communication can be implemented based on the SM4 algorithm or other existing algorithms.

[0099] The server principle is an intelligent gateway carrier embedded with digital ships and AI (Artificial Intelligence), which is an edge server with more module integration. The AI tool chain of the server can implement the propagation maintenance method in the above embodiments.

[0100] The terminal devices include AR (Augmented Reality) glasses, maintenance robots, interactive video projectors, sensor terminals, visual monitoring instrument panels, etc. Through the local area network, multi-modal terminal interaction can be achieved, including guiding maintenance personnel to implement steps through the maintenance personnel terminal and controlling maintenance equipment through the local area network to realize human-machine collaborative work. For example, using a device similar to AR glasses, a virtual-real fusion technology is used to project and display maintenance steps and safety warnings (such as high-temperature area identification), and the AR glasses support maintenance personnel to interact using gestures and voice; a mobile inspection robot is equipped with a visible light or infrared camera to take data, which is transmitted back to the local server through the local area network, etc.

[0101] In this embodiment, through the innovation of the system architecture and collaboration of the shipborne AI server, the local area network and the multi-terminal, seamless interaction and deep fusion of multi-modal terminals such as AR, robots and sensor data can be achieved, and the operation and maintenance efficiency can be improved.

[0102] Based on the ship Internet of Things system, another ship maintenance system for ship maintenance can be constructed, as shown in Figure 5 In addition to the operation and alarm information AI generation module, the first gateway module, the artificial intelligence module and the second gateway module, the ship maintenance system also includes a power junction module, a device network junction module and a terminal device.

[0103] The power junction module is based on a redundant power supply design. Optionally, the power junction module supports 220VAC (alternating current) and +48VDC (direct current). The device network junction module is based on an IEEE802.11ax (sixth generation wireless network technology) encrypted local area network and supports more than 20 concurrent devices. The terminal device includes AR glasses, maintenance robots, interactive video projectors, sensor terminals, etc.

[0104] Optionally, based on the server in the ship Internet of Things system, the first gateway module and the second gateway module are configured, and a double-gateway architecture design for sending and receiving can be achieved: the first gateway module downloads models in batches through a 4G / 5G link and uses multicast technology to reduce repeated transmission; the second gateway module synchronously feeds back update status, including feedback information such as "Ship A has received", so that the shore base can dynamically adjust the push strategy based on the feedback information. Through the redundant design of the first gateway module and the second gateway module, the model update time is shortened, the bandwidth occupation is reduced, and network security isolation can also be achieved.

[0105] In an embodiment, the job and alarm information AI generation module further comprises a ship multi-source sensor unit. The ship multi-source sensor unit can realize the deployment of more than 300 nodes, and the sensors cover key areas such as power and cabin. The ship multi-source sensor unit can support hybrid networking transmission of wired and / or wireless. Optionally, the ship multi-source sensor unit can push the collected sensor data to the job and alarm information AI generation module and the artificial intelligence module; so that the job and alarm information AI generation module generates fault information according to the sensor information, and the artificial intelligence module generates the first maintenance instruction and the second maintenance instruction according to the sensor information. Optionally, the collected data of the ship multi-source sensor unit is defined, including data cleaning, time series alignment and other preprocessing of the collected data, and the processed sensor data is transmitted to the ship maintenance knowledge base in the job and alarm information AI generation module and the AI server.

[0106] In an embodiment, the artificial intelligence module comprises an AI server, which is used to carry an edge computing chip (such as NVIDIA Jetson AGX Orin), integrates a ship maintenance knowledge base, historical data of equipment operation and maintenance, and a lightweight AI model library, and realizes real-time fault diagnosis and decision generation in a local area network environment. The AI server is configured to correspond to an acceleration server cluster, which can support GPU (Graphics Processing Unit, graphics processor), FPGA (Field-Programmable Gate Array, field-programmable gate array) or ASIC (Application-Specific Integrated Circuit, application-specific integrated circuit) acceleration card, thereby meeting the requirements of large model training and inference.

[0107] Optionally, the AI server can be used to preprocess the fault information generated by the job and alarm information AI generation module and the ship equipment, and realize multi-source data fusion. The fault information includes sensor data, historical maintenance records obtained from the fault information, the ship equipment, the first maintenance instruction and the second maintenance instruction, expert experience, meteorological data and navigation log, and other external data. Based on multi-source data fusion, a physical field model is constructed to obtain a general fault processing framework. Optionally, after the AI server constructs the physical field model, it identifies the associated fault mode of the key system (main engine, auxiliary engine, propulsion system) of the ship, evaluates the priority order of the physical field model in the model library, so that the fault information and the ship equipment are searched in the model library based on the priority order. In addition, based on the multi-source data fusion result, the AI server can further support intelligent customer service, data analysis visualization board and automatic report generation, and provide intelligent decision and operation guidance for maintenance personnel.

[0108] Optionally, the ship maintenance knowledge base supports structured data (SQL), unstructured data (NoSQL), and vector databases, realizes multi-modal knowledge storage, permission control, and encrypted transmission. In order to realize the management of the ship knowledge base, the AI server has a knowledge management unit, a knowledge retrieval unit, and a knowledge update unit. The knowledge management unit is used to realize knowledge collection, supports web page upload of text data such as PDF or Word, web data crawling, database import, and other functions; the knowledge retrieval unit is used to realize semantic-based retrieval (such as ColBERTv2) and RAG (Retrieval-augmented Generation) technology, and improve the accuracy of answers; the knowledge update unit has an automatic incremental update mechanism, combined with a manual review process, to improve the ship maintenance knowledge base. Optionally, based on the ship maintenance knowledge base, the preset emergency instructions and the preset regular instructions in the above embodiments can be obtained. Further, based on the ship maintenance knowledge base, training data for constructing an AI model can be obtained, wherein the AI model is used to generate the first maintenance instruction and the second maintenance instruction according to the standard solution.

[0109] The AI model is constructed as follows: a federated learning engine is provided in the AI server, based on the federated learning engine and the training data provided by the ship maintenance knowledge base, a ship operation and maintenance special AI model is trained, and uploaded to the cloud through a local area network for global model aggregation. In the core components of the algorithm and model layer, PyTorch, TensorFlow, incremental learning, and other model training frameworks are supported, and an integrated distributed training optimization tool is configured. Optionally, the AI model includes a model training and optimization unit for automatically training the AI model, such as: based on hyperparameter tuning, distributed training acceleration: pruning, quantization, and other ways to realize adaptation and automatic training of edge devices. The AI model can also set an inference service unit for multi-model orchestration, dynamically switch model versions through the orchestration of the inference service unit, support hybrid inference, for example, combine GPT (Generative Pre-trained Transformer) and rule engine to realize AI model inference. The inference service unit can also set low-latency optimization functions such as TensorRT acceleration, KV (Key-Value) cache reuse, and real-time monitoring of inference error rate anomaly detection functions.

[0110] The AI model can be used to predict the current life cycle stage of the ship equipment. For example, after the vibration sensor uploads abnormal data to the AI server, the ship knowledge base is retrieved to determine that in the case of "vibration value > 5 mm / s + abnormal sound frequency 1500 Hz", the corresponding fault is "rotating device wear". Based on the core components of the above algorithm and model layer, the standard model is called to compare and identify the associated configuration between the faulty ship equipment and other equipment, and the AI model analyzes and predicts the remaining service life. In the case of determining that the rotating device needs to be repaired, the AI server can match the "bearing wear" case in the propagation knowledge base to generate a "shutdown inspection + lubricating oil replacement" solution; it can also search the model library for the object field model to generate a maintenance instruction.

[0111] The AI model can also be used to generate maintenance instructions. Optionally, a structured template is provided in the AI server, which includes fault phenomena, possible causes, diagnosis steps, treatment measures, and prevention suggestions. After obtaining the standard solution, the AI model generates a first maintenance instruction and a second maintenance instruction based on the structured template.

[0112] Optionally, the artificial intelligence module further includes a streaming media output unit. The streaming media output unit is used to real-time render the AR projection generated by the first maintenance instruction after the first maintenance instruction and the second maintenance instruction are generated. The information data output by the streaming media output unit can be defined, for example, the information data transmitted in real time by the streaming media output unit guides the generation of video AIGC (Artificial Intelligence Generated Content) based on the Stable Diffusion Video model, and combines the AR superposition technology to realize intelligent augmented content presentation. The data distribution method of the streaming media output unit includes outputting information data to industrial cameras, microphones, AR devices, and other terminals.

[0113] Optionally, the artificial intelligence module further includes a maintenance site interaction unit. The maintenance site interaction unit is used to capture the actions of AR glasses and maintenance personnel to obtain sensing information, thereby assisting the AI model in generating the first maintenance instruction and the second maintenance instruction based on the sensing information and the standard solution.

[0114] Optionally, after the maintenance is completed, the maintenance related records are temporarily stored locally, and after landing, they can be uploaded to the ship management command center using network encryption. Further, the maintenance data can also be encrypted and stored in the AI server to trigger the federal learning to optimize the wear prediction model and update the knowledge.

[0115] In this embodiment, the AI server configured is the key to the transformation of the ship industry from "experience-driven" to "data-driven". Its value not only lies in the improvement of single ship operation and maintenance efficiency, but also lies in the intelligent upgrading of the entire industrial chain.

[0116] In an embodiment, the first gateway module and the second gateway module are each provided with a job service gateway module: instruction routing and protocol conversion. Optionally, end-to-end encrypted communication can also be implemented through the first gateway module and the second gateway module to improve data security. Optionally, an API gateway is configured in the first gateway module and the second gateway module to provide a RESTful (Representational State Transfer) interface or a GraphQL (Graph Query Language) interface, and load balancing and flow limiting and fusing functions are supported through the API (Application Programming Interface) gateway.

[0117] In an embodiment, based on Figure 5 a ship maintenance system as shown in the figure, Figure 6 a flowchart of a multi-modal AI ship equipment maintenance and repair method is provided, as shown in the figure, comprising: Figure 6

[0118] Step S601, receiving equipment failure information or equipment maintenance instructions of ship equipment. The equipment failure information or equipment maintenance instructions include equipment failure location, number, and scene. Taking a main engine fuel pump leakage as an example, the ship equipment that fails, video, liquid signal, and mechanical frequency sensor signals can be uploaded to an AI server as failure information through an industrial camera video and a vibration sensor.

[0119] Step S602, the AI server analyzes sensor data in real time, generates a maintenance process sheet in combination with a model library standard solution, and pushes it to various related terminals. The maintenance process sheet can be pushed to related personnel and terminal equipment through a local area network to build a shipboard equipment collaboration network. The maintenance process sheet at least includes a first maintenance instruction and a second maintenance instruction for realizing maintenance decision command and resource allocation. The related personnel pushed can realize safe operation and quality control according to the maintenance process sheet; at the same time, they can perform high-risk operations such as high-altitude valve adjustment with the assistance of robots.

[0120] Optionally, the AI server searches for a material field model in the model library according to the equipment failure information or equipment maintenance instructions of the ship equipment, quickly judges the main contradiction of the ship equipment failure through the ternary structure (S1, S2, F) of the “material-field” model, and proposes a matching standard solution. For ease of understanding, taking the main engine fuel pump leakage as an example, Figure 7 a flowchart of obtaining a standard solution is provided, as shown in the figure, comprising: Figure 7

[0121] ​​Step S701, obtain the object field model of the leakage of the main engine fuel pump. Among them, S1 is fuel (leakage liquid), S2 is the sealing structure of the pump body, and F is the mechanical pressure field; the harmful effect is that high-pressure fuel breaks through the seal, causing leakage pollution.

[0122] Step S702, obtain standard solution A according to the ternary structure, and execute step S604. The standard solution eliminates the harmful effect by introducing a third substance S3. The standard solution A includes: adding self-repairing materials in the sealing structure, such as microcapsule polymers that break under pressure, as repair materials for plugging leakage gaps.

[0123] Step S703, obtain standard solution B according to the ternary structure, and execute step S705. The standard solution B includes: changing the mechanical seal to a magnetic seal, and optimizing the function by changing the type of the object field to avoid contact wear and leakage.

[0124] Step S704, determine whether the available resources of the ship meet the resource requirements in the standard solution A; if yes, execute solution A; if no, execute step S705. Solution A corresponds to standard solution A.

[0125] Step S705, determine whether the available resources of the ship meet the resource requirements in the standard solution B; if yes, execute solution B; if no, run based on the preset emergency instructions in the knowledge base. Solution B corresponds to standard solution B.

[0126] Figure 7 Through the "problem-solution-verification" three closed-loop structure, the maintenance personnel can improve the standardized actions that can be directly operated, and improve the operation quality and efficiency.

[0127] Step S603, generate a push report to the AI server according to the maintenance process and structure.

[0128] Step S604, the AI server updates the ship knowledge base according to the push report, and optimizes the global AI model in the model library through federated learning.

[0129] In this embodiment, by fusing high-precision sensors, edge computing and augmented reality technology, efficient diagnosis and maintenance of ship equipment failures are realized. Taking the AI server as the technical core and taking fault and task triggering as the entrance, the AI large model capability is referenced in the task to generate a maintenance order to guide the maintenance personnel to automatically handle a large number of complex work, realize interactive intelligent dialogue and broadcast video generation of maintenance guidance specification. Even in a network environment, the core functions of the maintenance system are still available, and the fault diagnosis coverage is very high. At the same time, it has the effect of shortening the maintenance time of complex equipment and reducing the frequency of manual intervention.

[0130] It should be understood that although each step in the flowchart involved in each embodiment as described above is shown in sequence according to the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless explicitly stated herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps. For example, step S605 can be executed first. If step S605 is determined to be no, step S604 is executed, and if step S604 is determined to be no, the preset emergency instruction in the knowledge base is executed.

[0131] In one embodiment, a computer device, which can be a server, is provided, and an internal structure diagram of the computer device can be as shown in Figure 8 The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is used to store ship data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to realize a ship maintenance method.

[0132] Those skilled in the art can understand that Figure 8 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0133] In one embodiment, a computer device is also provided, which includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the steps in each method embodiment described above.

[0134] In an embodiment, a computer readable storage medium is provided, having stored thereon a computer program which, when executed by a processor, implements the steps of any of the above method embodiments.

[0135] In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of any of the above method embodiments.

[0136] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium and can include the processes of the above-mentioned embodiments when executed. Any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of a non-volatile and volatile memory. The non-volatile memory can include a read-only memory (ROM), a magnetic tape, a floppy disk, a flash memory, an optical storage, a high-density embedded non-volatile memory, a resistive memory (ReRAM), a magnetoresistive random access memory (MRAM), a ferroelectric memory (FRAM), a phase change memory (PCM), a graphene memory, etc. The volatile memory can include a random access memory (RAM) or an external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms such as a static random access memory (SRAM) or a dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0137] Any combination of the technical features of the above embodiments can be made. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0138] The above-described embodiments are merely illustrative of several embodiments of the present application, which are described in more detail and in a specific manner, but should not be construed as limiting the scope of the patent of the present application. It should be noted that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method of ship repair, characterized in that, The method comprises: acquiring fault information and ship equipment associated with the fault information; searching for a physical field model corresponding to the ship equipment and the fault information in a model library; generating a standard solution according to a ternary structure of the physical field model; generating first and second maintenance instructions that cooperate with each other according to the standard solution and sensing information of the ship equipment when available resources of the ship meet resource requirements in the standard solution; sending the first maintenance instruction to maintenance personnel and the second maintenance instruction to a maintenance machine; searching for a physical field model corresponding to the ship equipment and the fault information in a model library comprises: evaluating severity, frequency of occurrence and detectability of a plurality of physical field models in the model library according to ship equipment associated with each physical field model in the model library and the fault information; obtaining risk priorities of the plurality of physical field models based on the severity, the frequency of occurrence and the detectability; and searching for a physical field model corresponding to the ship equipment and the fault information in the model library in order from high to low according to the risk priorities.

2. The method of claim 1, wherein, The method further comprises: in the case where there is no physical field model corresponding to the ship equipment and the fault information in the model library, downloading a physical field model corresponding to the ship equipment and the fault information based on a first gateway and feeding back update information of the model library based on a second gateway; wherein the first gateway and the second gateway are backup gateways.

3. The method of claim 2, wherein, Downloading a physical field model corresponding to the ship equipment and the fault information based on a first gateway comprises: determining whether the ship equipment and the fault information meet a triggering condition of an emergency state; if yes, sending a preset emergency instruction to the maintenance personnel and / or the maintenance machine and downloading a physical field model corresponding to the ship equipment and the fault information from a satellite by the first gateway; if no, sending a preset normal instruction to the maintenance personnel and / or the maintenance machine and downloading a physical field model corresponding to the ship equipment and the fault information from a shore base by the first gateway.

4. The method of claim 1, wherein, The acquisition of the fault information comprises: establishing an association between a life cycle of the ship equipment and space-time data according to running characteristics of the ship equipment under different time and space conditions; mapping space-time data of the ship running based on the association to obtain a life cycle phase in which the ship equipment is located; generating fault information according to the life cycle phase in which the ship equipment is located when the life cycle phase enters a preset phase.

5. The method of claim 1, wherein, After sending the first maintenance instruction to the maintenance personnel and the second maintenance instruction to the maintenance machine, the method further comprises: integrating the fault information, the ship equipment, field sensing information, the first maintenance instruction and the second maintenance instruction; updating a physical field model in the model library based on the integration result.

6. The method of claim 5, wherein, Updating the model library based on the integration result comprises: updating non-standard equipment data of the ship according to the integration result; constructing a new physical field model according to the non-standard equipment data of the ship and adding the new physical field model in the model library.

7. A marine vessel repair system characterized by, The system comprises a work and police information AI generation module, a first gateway module, an artificial intelligence module, and a second gateway module. The work and police information AI generation module is configured to determine fault information and ship equipment associated with the fault information. The first gateway module is configured to send the fault information and the ship equipment to the artificial intelligence module. The artificial intelligence module is configured to search for a physical field model corresponding to the ship equipment and the fault information in a model library, generate a standard solution based on a ternary structure of the physical field model, and generate a first maintenance instruction and a second maintenance instruction based on the standard solution and sensing information of the ship equipment when available resources of the ship satisfy resource requirements in the standard solution. The second gateway module is configured to send the first maintenance instruction to maintenance personnel and the second maintenance instruction to a maintenance machine. The artificial intelligence module searches for a physical field model corresponding to the ship equipment and the fault information in the model library by evaluating severity, frequency of occurrence, and detection degree of a plurality of physical field models based on ship equipment and fault information associated with each physical field model in the model library, obtaining risk priorities of the plurality of physical field models based on the severity, the frequency of occurrence, and the detection degree, and searching for a physical field model corresponding to the ship equipment and the fault information in the model library in order from high to low according to the risk priorities.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to implement the steps of the method of any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

Images