A method and device for ship equipment maintenance based on augmented reality

Through augmented reality technology to identify ship equipment failures and provide virtual maintenance guidance, the problems of low maintenance efficiency and difficult quality in the existing technology are solved, and fast and accurate maintenance operations are achieved.

CN115512253BActive Publication Date: 2024-11-08COSCO ZHOUSHAN SHIPYARD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211129266.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-09-16
Publication Date
2024-11-08
Estimated Expiration
2042-09-16

AI Technical Summary

Technical Problem

The maintenance of existing marine equipment depends on the experience of skilled workers, is inefficient and difficult to control, especially in multiple failure situations that require a lot of time to find solutions.

Method used

The marine equipment maintenance method based on augmented reality is adopted to collect equipment images, identify faults using convolutional neural networks, obtain virtual models and maintenance information, provide real-time guidance and predict service life, and improve maintenance efficiency and quality.

Benefits of technology

Quickly identify faults through augmented reality technology, providing accurate maintenance solutions, improving maintenance efficiency and quality, reducing the operating time of maintenance personnel, and ensuring efficient maintenance of equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115512253B_ABST
    Figure CN115512253B_ABST
Patent Text Reader

Abstract

A ship equipment maintenance method based on augmented reality, characterized in that: the method collects the initial image of the equipment to be repaired, determines whether the equipment to be repaired has a fault according to a fault discrimination formula, and generates a maintenance signal if there is a fault; identifies and distinguishes the equipment to be repaired, and obtains the identification information and maintenance information of the equipment; obtains the virtual model of the equipment to be repaired, the items to be repaired, and the location information of the items to be repaired from the database according to the identification information and the maintenance information; superimposes and displays the virtual model on the spatial area within the user's field of view, and marks the items to be repaired on the superimposed virtual model according to the location information of the items to be repaired; repairs the items to be repaired according to the user's field of view prompt information, and maps the contents to be measured. The advantages of the present invention are that through the augmented reality method, the fault points in the ship equipment can be accurately located and solutions can be obtained, which is convenient for maintenance personnel to handle and improves the quality and efficiency of maintenance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of ship equipment maintenance, and in particular to a ship equipment maintenance method and device based on augmented reality. Background Art

[0002] At present, ship equipment repair operations are mostly done manually, such as disassembly, replacement of spare parts, analysis and repair of faulty equipment, etc. The repair of ship equipment benefits from experienced technical workers who use their own technical capabilities to repair related equipment. The entire process is based on the subjective consciousness and experience of technical workers to judge the repair fault points of the object to be repaired, and repair the fault points one by one through teamwork. The current maintenance work relies too much on the experience and level of technical workers, especially when multiple faults occur in a piece of equipment, maintenance personnel need to grope back and forth to find the corresponding solution, which brings many efficiency challenges to the maintenance work. Since maintenance personnel have to spend a lot of energy to find the fault point, the maintenance quality of specific equipment is not effectively controlled. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a ship equipment maintenance method and device based on augmented reality that can improve maintenance quality and efficiency in view of the above-mentioned existing technical status.

[0004] The technical solution adopted by the present invention to solve the above technical problems is: a ship equipment maintenance method based on augmented reality, characterized in that the method comprises the following steps:

[0005] S1, collecting the initial image of the equipment to be repaired;

[0006] S2, obtaining the area of ​​the rusted area and the wear area of ​​the equipment according to the initial image of the equipment to be repaired, judging whether the equipment to be repaired has a fault according to the fault discrimination formula, if there is a fault, generating a repair signal and entering step S3, if there is no fault, generating a no-repair signal, re-locking the next equipment to be repaired, and returning to step S1;

[0007] S3, identifying and distinguishing the equipment to be repaired that generates the maintenance signal by using a convolutional neural network to obtain identification information and maintenance information of the equipment;

[0008] S4, obtaining a virtual model of the equipment to be repaired, the item to be repaired, and the location information of the item to be repaired from a database according to the identification information and the repair information;

[0009] S5, superimposing and displaying the virtual model in the space area within the user's field of vision, and marking the item to be repaired on the superimposed virtual model according to the location information of the item to be repaired;

[0010] S6. Repair the item to be repaired according to the user's field of view prompt information, map the content to be measured, and after the repair is completed, obtain the maintenance parameters of the item to be repaired after processing and the standard parameters in the database that are the same as the identification information of the item to be repaired, and judge whether the maintenance parameters fall within the range of the standard parameters. If so, the item to be repaired is qualified, and the next equipment to be repaired is re-locked, and the process returns to step S1; if not, generate a prompt message instructing the maintenance personnel to process the item to be repaired again, and re-execute step S6.

[0011] In order to determine whether the equipment to be repaired has a fault based on the area of ​​the rusted area and the wear area, in step S2, the fault determination formula is:

[0012]

[0013] Among them, a1 and a2 are fault discrimination proportional coefficients, and a1>a2>0, is the error correction factor, Sx is the area of ​​the rusted area of ​​the equipment to be repaired, Ss is the wear area of ​​the equipment to be repaired, and X1 is the fault discrimination coefficient;

[0014] If the fault discrimination coefficient X1 of the equipment to be repaired is greater than the preset fault discrimination coefficient threshold, it is determined that the current equipment to be repaired needs to be repaired, and a maintenance signal is generated; if the fault discrimination coefficient X1 of the equipment to be repaired is less than or equal to the preset fault discrimination coefficient threshold, it is determined that the current equipment to be repaired does not need to be repaired, and a no-maintenance signal is generated.

[0015] In order to further distinguish the emergency situations of the equipment to be repaired, in step S3, if the fault discrimination coefficient X1 of the equipment to be repaired is greater than the preset fault discrimination coefficient threshold, a1, a2 and Take different values ​​for many times and calculate the corresponding fault discrimination coefficients X2, X3, etc., sum and average these fault discrimination coefficients to get the average fault discrimination coefficient. The fault discrimination coefficient average The fault discrimination coefficient average value threshold is compared with the preset fault discrimination coefficient average value threshold. If the fault discrimination coefficient average value exceeds the threshold, the equipment maintenance signal is marked as a first-level maintenance signal; if the fault discrimination coefficient average value exceeds the threshold If the fault discrimination coefficient does not exceed the average value threshold, the equipment maintenance signal is marked as a secondary maintenance signal.

[0016] In order to obtain maintenance information according to the identification information of the equipment to be repaired, in step S3, the identification information includes the model of the equipment to be repaired, the classification number of the equipment to be repaired, and the production date of the equipment to be repaired. The identification information is de-noised to obtain the model Zx, classification number Zf and production date Zr of the equipment. According to the model Zx, classification number Zf and production date Zr, the similarity value K of the equipment to be repaired is obtained by a similarity calculation formula. The similarity calculation formula is:

[0017]

[0018] Among them, b1, b2 and b3 are similarity coefficient factors;

[0019] The similarity value K is compared with a preset standard equipment similarity value. If the obtained equipment similarity value K is the same as the standard equipment similarity value, the maintenance information of the equipment is retrieved from the database.

[0020] Further, in step S3, when the identification information of the device to be repaired cannot be obtained, the three-view information in the initial image of the device is obtained, the three-view information includes the area S1 of the device in the front view, the area S2 of the device in the right view, and the area S3 of the device in the top view, and the appearance similarity value P of the device is obtained by the appearance similarity value calculation formula, and the appearance similarity calculation formula is:

[0021] P=c1*S1+c2*S2+c3*S3

[0022] Among them, c1, c2, c3 are coefficient factors;

[0023] The appearance similarity value P is compared with a standard appearance similarity value preset in a database. If the obtained device appearance similarity value P is the same as the standard appearance similarity value, the maintenance information of the device is retrieved from the database.

[0024] In order to know the service life of the equipment and understand the approximate usage time of the equipment, so that the staff can perform corresponding maintenance or replacement operations at key nodes, obtain the operating parameters of the equipment to be repaired that generates maintenance signals and the equipment that produces non-maintenance signals, and predict the service life of the equipment based on the operating parameters.

[0025] The present invention also provides a ship equipment maintenance device based on augmented reality, characterized in that: the device can implement the above-mentioned ship equipment maintenance method based on augmented reality, including:

[0026] The measurement module is used to collect the initial image of the equipment to be repaired, the maintenance information of the items to be repaired after processing, and the operating parameters of the equipment to be repaired that generates the maintenance signal and the equipment that generates the non-maintenance signal after maintenance;

[0027] An analysis module is used to make a fault judgment on the equipment to be repaired, generate a repair signal and a no-repair signal, identify and distinguish the equipment that generates the repair signal through a convolutional neural network, obtain the identification information or three-view information of the equipment, obtain a similarity value through the identification information or obtain an appearance similarity value through the three-view information;

[0028] A storage module, used to store the standard equipment similarity value, the standard appearance similarity value, the maintenance information, the virtual model, the items to be maintained and the location information of the items to be maintained of each equipment;

[0029] A maintenance module, used to obtain the virtual model of the equipment to be maintained, the items to be maintained and the location information of the items to be maintained in the storage module according to the maintenance information, and superimpose and display the virtual model in the space area within the user's field of vision, and mark the items to be maintained on the superimposed virtual model according to the location information of the items to be maintained;

[0030] A prediction module, used to predict the service life of the equipment to be repaired based on the operating parameters obtained by the measurement module;

[0031] The central processing unit is used to control the measuring module, the analyzing module, the storing module, the repairing module and the predicting module.

[0032] The present invention also provides a ship equipment maintenance device based on augmented reality, which is characterized by comprising:

[0033] A measuring device for collecting an initial image of the equipment to be repaired, maintenance information after the items to be repaired are processed, and operating parameters of the equipment to be repaired that generates the maintenance signal after the equipment is repaired and the equipment that generates the non-maintenance signal;

[0034] AR display device, which can be worn by users;

[0035] A processor, executing a computer program to implement the ship equipment maintenance method based on augmented reality as described in any one of claims 1 to 7;

[0036] A memory stores a computer program executed by the processor.

[0037] Compared with the prior art, the advantages of the present invention are: determining whether the equipment to be repaired has a fault by identifying the fault of the equipment to be repaired, analyzing and matching the equipment to be repaired, obtaining the maintenance plan for the same equipment, accurately prompting the maintenance information to the ship equipment maintenance personnel, and using the augmented reality method to allow the maintenance personnel to obtain the teaching method for equipment maintenance according to the needs and the actual equipment to be repaired, and quickly find the corresponding solution, so that the maintenance personnel can eliminate the fault in time and quickly, thereby improving the maintenance quality and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a flow chart of a ship equipment maintenance method based on augmented reality in an embodiment of the present invention;

[0039] Figure 2 A schematic diagram of the structure of a ship equipment maintenance device based on augmented reality provided by an embodiment of the present invention;

[0040] Figure 3 A schematic diagram of the structure of a ship equipment maintenance device based on augmented reality provided by an embodiment of the present invention;

[0041] Figure 4 This is a schematic diagram of a virtual model superimposed and displayed on a ship equipment maintenance device based on augmented reality in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The present invention is further described in detail below with reference to the accompanying drawings.

[0043] like Figure 1 As shown, this embodiment provides a ship equipment maintenance method based on augmented reality, the method comprising the following steps:

[0044] S1. Collecting the initial image of the equipment to be repaired.

[0045] S2. Obtain the area of ​​the rusted area and the worn area of ​​the equipment according to the initial image of the equipment to be repaired, and determine whether the equipment to be repaired has a fault according to a fault discrimination formula. The fault discrimination formula is:

[0046]

[0047] Among them, a1 and a2 are both fault discrimination proportional coefficient factors, and a1>a2>0, is the error correction factor, Sx is the area of ​​the rusted area of ​​the equipment to be repaired, Ss is the wear area of ​​the equipment to be repaired, and X1 is the fault discrimination coefficient.

[0048] If the fault discrimination coefficient X1 of the equipment to be repaired is greater than the preset fault discrimination coefficient threshold, it is determined that the current equipment to be repaired needs to be repaired, and a maintenance signal is generated; if the fault discrimination coefficient X1 of the equipment to be repaired is less than or equal to the preset fault discrimination coefficient threshold, it is determined that the current equipment to be repaired does not need to be repaired, and a no-maintenance signal is generated, and the next equipment to be repaired is re-locked, and the process returns to step S1.

[0049] In order to further distinguish the emergency situations of the equipment to be repaired, if the fault discrimination coefficient X1 of the equipment to be repaired is greater than the preset fault discrimination coefficient threshold, a1, a2 and Take different values ​​for many times and calculate the corresponding fault discrimination coefficients X2, X3, etc., sum and average these fault discrimination coefficients to get the average fault discrimination coefficient. The fault discrimination coefficient average The fault discrimination coefficient average value threshold is compared with the preset fault discrimination coefficient average value threshold. If the fault discrimination coefficient average value exceeds the threshold, the equipment maintenance signal is marked as a first-level maintenance signal; if the fault discrimination coefficient average value exceeds the threshold If the fault discrimination coefficient does not exceed the average value threshold, the equipment maintenance signal is marked as a level 2 maintenance signal; if the equipment generates a level 1 maintenance signal or a level 2 maintenance signal, proceed to step S3.

[0050] S3. The equipment to be repaired that generates the maintenance signal is identified and distinguished by using a convolutional neural network to obtain the identification information and maintenance information of the equipment. The identification information includes the model of the equipment to be repaired, the classification number of the equipment to be repaired, and the production date of the equipment to be repaired. The identification information is denoised to obtain the model Zx, classification number Zf and production date Zr of the equipment. For example, the equipment model of a motor is YB2 280-4W, the equipment classification number is A-128, and the equipment production date is March 1, 2012. The corresponding equipment model value is 22804, the equipment classification number value is 128, and the equipment production date value is 20120301. According to the model Zx, classification number Zf and production date Zr, the similarity value K of the equipment to be repaired is obtained through the similarity calculation formula. The similarity calculation formula is:

[0051]

[0052] Among them, b1, b2 and b3 are similarity coefficient factors.

[0053] The similarity value K is compared with a preset standard equipment similarity value. If the obtained equipment similarity value K is the same as the standard equipment similarity value, the maintenance information of the equipment is retrieved from the database.

[0054] When the identification information of the device to be repaired cannot be obtained, the three-view information of the initial image of the device is obtained, and the three-view information includes the area S1 of the device in the front view, the area S2 of the device in the right view, and the area S3 of the device in the top view. The appearance similarity value P of the device is obtained by the appearance similarity value calculation formula. The appearance similarity calculation formula is:

[0055] P=c1*S1+c2*S2+c3*S3

[0056] Among them, c1, c2, c3 are coefficient factors;

[0057] The appearance similarity value P is compared with a standard appearance similarity value preset in a database. If the obtained device appearance similarity value P is the same as the standard appearance similarity value, the maintenance information of the device is retrieved from the database.

[0058] S4. Obtain the virtual model of the equipment to be repaired, the items to be repaired, and the location information of the items to be repaired from the database according to the identification information and the maintenance information. The virtual model applies the idea of ​​3D convolutional neural network, adjusts the traditional 3D convolutional neural network in many aspects to make it effectively applied to the 3D recognition of the object to be repaired, and draws it in proportion according to the shape, structure or outline of the object to be repaired. For example, the object to be repaired is a motor with an aspect ratio of 2:1. The real shape and size of the object to be repaired are output through the software algorithm, and the time maintenance information and fault process point of the object to be repaired are output after matching with the virtual model in the software. The result can be a virtual motor scaled according to a preset ratio, or just a virtual cuboid with an aspect ratio of 2:1. The virtual model can be flat or three-dimensional, static or dynamic; the items to be repaired are items for which the corresponding specifications are measured for the entire or partial objects to be repaired, and the specifications include but are not limited to mass, size, area, volume, density and quantity; the location information is the location coordinates used to mark the items to be repaired on the virtual model.

[0059] S5, superimposing and displaying the virtual model in the spatial area within the user's field of vision, and marking the items to be repaired on the superimposed and displayed virtual model according to the location information of the items to be repaired. The virtual model can be superimposed and displayed in a preset area of ​​the spatial area within the user's field of vision. The preset area can be fixed, such as the upper left corner of the spatial area within the user's field of vision. The preset area can also be unfixed and can be changed according to the user's habits; the actual position of the object to be repaired in the spatial area within the user's field of vision can also be obtained, the transparency value of the virtual model can be adjusted according to the preset threshold, and the virtual model with the adjusted transparency value can be superimposed on the object to be repaired according to the actual position; the object to be repaired is positioned and tracked to obtain the actual position of the object to be repaired in the spatial area within the user's field of vision, and a reference position is provided for the superposition of the virtual model and the object to be repaired; after superimposing the virtual model on the object to be repaired, the transparency value of the virtual model is adjusted on the premise that the object to be repaired is not blocked, so that the user can see the virtual model and the object to be repaired at the same time, so that the repair guidance information is more instructive and clearer.

[0060] S6. Repair the item to be repaired according to the user's field of view prompt information, map the content to be measured, and after the repair is completed, obtain the maintenance parameters of the item to be repaired after processing and the standard parameters in the database that are the same as the identification information of the item to be repaired, and judge whether the maintenance parameters fall within the range of the standard parameters. If so, the item to be repaired is qualified, and the next equipment to be repaired is re-locked, and the process returns to step S1; if not, generate a prompt message instructing the maintenance personnel to process the item to be repaired again, and re-execute step S6.

[0061] The maintenance personnel carry out repairs according to the maintenance items fed back in the virtual model, test the items that need to be measured and tested, record the actual results, complete the actual video record of the repair process, form an equipment maintenance report, realize the record of the maintenance process, and achieve the effect of traceability of the maintenance process.

[0062] In order to know the service life of the equipment and understand the approximate use time of the equipment, so that the staff can perform corresponding maintenance or replacement operations at key nodes, obtain the operating parameters of the equipment to be repaired that generates the maintenance signal and the equipment that produces the non-maintenance signal in the unit time, that is, the real-time temperature and real-time amplitude. Set the initial time of the unit time as T1 and the end time as Tv; use the operating time t as the X-axis, the real-time temperature value A1 and the real-time amplitude value A2 as the Y-axis to construct the time and operating parameter coordinate system, substitute the obtained operating parameters into the coordinate system for constructing the maximum and minimum values ​​of the operating parameters, and obtain the real-time operating parameter curve, so as to obtain the deviation value from the maximum and minimum range of the operating parameters, that is, the real-time temperature deviation value Aai and the real-time amplitude deviation value Bbi, i = {1, 2, 3..., n}; then sum and average the real-time temperature deviation value Aai and the real-time amplitude deviation value Bbi to obtain the real-time temperature deviation average value and real-time amplitude deviation from the average value Then the equipment operation obstruction coefficient Xg is obtained through the equipment operation obstruction formula. The equipment operation obstruction formula is:

[0063]

[0064] Among them, d1 and d2 are coefficient factors.

[0065] The pre-stored relationship table between the equipment operation obstruction coefficient and the equipment remaining service life is retrieved from the database, and the obtained equipment operation obstruction coefficient Xg is substituted into the relationship table, thereby obtaining the corresponding service life of the equipment.

[0066] Among them, the above formulas are all obtained by collecting a large amount of data for software simulation and selecting a formula close to the actual value. The coefficient factors a1, a2, b1, b2, b3, c1, c2, c3, d1 and d2 in the formula are all set according to actual conditions.

[0067] like Figure 2 As shown, this embodiment also provides a ship equipment maintenance device based on augmented reality, which can implement the above-mentioned ship equipment maintenance method based on augmented reality, including:

[0068] The measurement module 11 is used to collect the initial image of the equipment to be repaired, the maintenance information of the items to be repaired after processing, and the operating parameters of the equipment to be repaired that generates the maintenance signal after the maintenance and the equipment that generates the non-maintenance signal;

[0069] The analysis module 12 is used to judge the fault of the equipment to be repaired, generate a repair signal and a non-repair signal, identify and distinguish the equipment that generates the repair signal through a convolutional neural network, obtain the identification information or three-view information of the equipment, obtain the similarity value through the identification information or obtain the appearance similarity value through the three-view information;

[0070] The storage module 13 is used to store the standard equipment similarity value, the standard appearance similarity value, the maintenance information, the virtual model, the items to be maintained and the location information of the items to be maintained of each equipment;

[0071] The maintenance module 14 is used to obtain the virtual model of the equipment to be maintained, the items to be maintained and the location information of the items to be maintained in the storage module according to the maintenance information, and superimpose and display the virtual model in the space area within the user's field of vision, and mark the items to be maintained on the superimposed virtual model according to the location information of the items to be maintained;

[0072] Prediction module 15, used to predict the service life of the equipment to be repaired according to the operating parameters obtained by the measurement module;

[0073] The central processing unit 16 is used to control the measurement module, analysis module, storage module, maintenance module and prediction module.

[0074] like Figure 3 As shown, this embodiment also provides a ship equipment maintenance device based on augmented reality, including:

[0075] The measuring device 21 is used to collect the initial image of the equipment to be repaired, the maintenance information after the items to be repaired are processed, and the operating parameters of the equipment to be repaired that generates the maintenance signal after the equipment is repaired and the equipment that generates the non-maintenance signal;

[0076] AR display device 22, wearable by the user;

[0077] The processor 23 executes the computer program to implement the above-mentioned ship equipment maintenance method based on augmented reality;

[0078] The memory 24 stores computer programs executed by the processor.

[0079] like Figure 4 As shown, the virtual model of the product is a rectangular parallelepiped. One of the maintenance tasks is to measure the size of the product. The text "Please measure" is marked on one side of the rectangular parallelepiped model.

Claims

1. A ship equipment maintenance method based on augmented reality, characterized in that: The method comprises the following steps: S1, collecting the initial image of the equipment to be repaired; S2, obtaining the area of ​​the rusted area and the wear area of ​​the equipment according to the initial image of the equipment to be repaired, judging whether the equipment to be repaired has a fault according to the fault discrimination formula, if there is a fault, generating a repair signal and entering step S3, if there is no fault, generating a no-repair signal, re-locking the next equipment to be repaired, and returning to step S1; S3, identifying and distinguishing the equipment to be repaired that generates the maintenance signal by using a convolutional neural network to obtain identification information and maintenance information of the equipment; S4, obtaining a virtual model of the equipment to be repaired, the item to be repaired, and the location information of the item to be repaired from a database according to the identification information and the repair information; S5, superimposing and displaying the virtual model in the space area within the user's field of vision, and marking the item to be repaired on the superimposed virtual model according to the location information of the item to be repaired; S6, repair the item to be repaired according to the user's visual field prompt information, map the content to be measured, obtain the repair parameters of the item to be repaired after the repair is completed and the standard parameters in the database that are the same as the identification information of the item to be repaired, and judge whether the repair parameters fall within the range of the standard parameters. If so, the item to be repaired is qualified, a no-repair signal is generated, and the next device to be repaired is locked again, and the process returns to step S1; if not, a prompt message is generated to instruct the maintenance personnel to process the item to be repaired again, and step S6 is executed again; In step S3, the identification information includes the model of the equipment to be repaired, the classification number of the equipment to be repaired, and the production date of the equipment to be repaired. The identification information is de-noised to obtain the model Zx, classification number Zf and production date Zr of the equipment. The similarity value K of the equipment to be repaired is obtained according to the model Zx, classification number Zf and production date Zr through a similarity calculation formula. The similarity calculation formula is: Among them, b1, b2 and b3 are similarity coefficient factors; Compare the similarity value K with a preset standard equipment similarity value, and if the obtained equipment similarity value K is the same as the standard equipment similarity value, retrieve the maintenance information of the equipment from the database; In the step S3, when the identification information of the device to be repaired cannot be obtained, the three-view information in the initial image of the device is obtained, the three-view information includes the area S1 of the device in the front view, the area S2 of the device in the right view, and the area S3 of the device in the top view, and the appearance similarity value P of the device is obtained by the appearance similarity value calculation formula, and the appearance similarity calculation formula is: P=c1*S1+c2*S2+c3*S3 Among them, c1, c2, c3 are coefficient factors; The appearance similarity value P is compared with a standard appearance similarity value preset in a database. If the obtained device appearance similarity value P is the same as the standard appearance similarity value, the maintenance information of the device is retrieved from the database.

2. The method for ship equipment maintenance based on augmented reality according to claim 1, characterized in that: In step S2, the fault discrimination formula is: Among them, a1 and a2 are fault discrimination proportional coefficients, and a1>a2>0, is the error correction factor, Sx is the area of ​​the rusted area of ​​the equipment to be repaired, Ss is the wear area of ​​the equipment to be repaired, and X1 is the fault discrimination coefficient; If the fault discrimination coefficient X1 of the equipment to be repaired is greater than the preset fault discrimination coefficient threshold, it is determined that the current equipment to be repaired needs to be repaired, and a maintenance signal is generated; if the fault discrimination coefficient X1 of the equipment to be repaired is less than or equal to the preset fault discrimination coefficient threshold, it is determined that the current equipment to be repaired does not need to be repaired, and a no-maintenance signal is generated.

3. The method for ship equipment maintenance based on augmented reality according to claim 2, characterized in that: In step S3, if the fault discrimination coefficient X1 of the equipment to be repaired is greater than the preset fault discrimination coefficient threshold, a1, a2 and Take different values ​​for many times and calculate the corresponding fault discrimination coefficients X2, X3, etc., sum and average these fault discrimination coefficients to get the average fault discrimination coefficient. The fault discrimination coefficient average The fault discrimination coefficient average value threshold is compared with the preset fault discrimination coefficient average value threshold. If the fault discrimination coefficient average value exceeds the threshold, the equipment maintenance signal is marked as a first-level maintenance signal; if the fault discrimination coefficient average value exceeds the threshold If the fault discrimination coefficient does not exceed the average value threshold, the equipment maintenance signal is marked as a secondary maintenance signal.

4. The method for ship equipment maintenance based on augmented reality according to claim 1, characterized in that: The operating parameters of the equipment to be repaired that generates the maintenance signal and the equipment that generates the non-maintenance signal are obtained, and the service life of the equipment is predicted according to the operating parameters.

5. A ship equipment maintenance device based on augmented reality, characterized in that: The device can implement the ship equipment maintenance method based on augmented reality as described in any one of claims 1 to 4 above, including: The measurement module is used to collect the initial image of the equipment to be repaired, the maintenance information of the items to be repaired after processing, and the operating parameters of the equipment to be repaired that generates the maintenance signal and the equipment that generates the non-maintenance signal after maintenance; An analysis module is used to make a fault judgment on the equipment to be repaired, generate a repair signal and a no-repair signal, identify and distinguish the equipment that generates the repair signal through a convolutional neural network, obtain the identification information or three-view information of the equipment, obtain a similarity value through the identification information or obtain an appearance similarity value through the three-view information; A storage module, used to store the standard equipment similarity value, standard appearance similarity value, maintenance information, virtual model, items to be repaired, location information of the items to be repaired, and standard parameters of each device; A maintenance module, used to obtain the virtual model of the equipment to be maintained, the items to be maintained and the location information of the items to be maintained in the storage module according to the maintenance information, and superimpose and display the virtual model in the space area within the user's field of vision, and mark the items to be maintained on the superimposed virtual model according to the location information of the items to be maintained; A prediction module, used to predict the service life of the equipment to be repaired based on the operating parameters obtained by the measurement module; The central processing unit is used to control the measuring module, the analyzing module, the storing module, the repairing module and the predicting module.

6. A ship equipment maintenance device based on augmented reality, characterized in that: include: A measuring device for collecting an initial image of the equipment to be repaired, maintenance information after the items to be repaired are processed, and operating parameters of the equipment to be repaired that generates a maintenance signal after the equipment is repaired and the equipment that generates a non-maintenance signal; AR display device, which can be worn by users; A processor, executing a computer program to implement the ship equipment maintenance method based on augmented reality as described in any one of claims 1 to 4; A memory stores a computer program executed by the processor.