Design method and system of ship sign system

Abstract

The invention provides a design method of a ship sign system. The design method comprises the following steps: performing type marking on information involved in a ship sign design process; based on a historical ship three-dimensional design model, information of ship sign design is extracted, data processing is carried out, and AI training sample data is formed; training AI models based on the AI training sample data, and selecting the AI model with the highest prediction precision through iterative optimization; features of the three-dimensional model in the new ship design scene are extracted, data processing is carried out, the data are input into the AI model with the highest prediction precision, and prediction information of ship sign design is output according to types; calling the mark basic library based on prediction information, and combining the expert system library to generate a mark three-dimensional model according to types; the generated sign three-dimensional model is subjected to normative inspection and correction based on the expert system library, so that the design efficiency of the ship sign system can be improved, the design period can be shortened, and the design error can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of ship design, and in particular to a design method and system for a ship marking system. Background Art

[0002] As a multi-system and huge combination product, ships have the remarkable characteristics of high complexity and high information density. In the process of ship construction, in order to better and more intuitively display various types of ship information, during the design and construction stage of the ship, the ship information is usually expressed and explained in the form of signs.

[0003] From the perspective of composition, ship markings can be divided into types such as compliance markings, alphanumeric markings, and mixed markings. For example, a certain LNG ship contains a total of 760 markings of varying sizes, and with the continuous expansion of ship systems and the gradual increase in the amount of parts, the number of markings will also show a positive growth trend accordingly.

[0004] The sign design mainly covers the selection of sign information types and the creation and layout of sign objects. The selection of sign information includes the location of sign creation, reference basis, sign form, sign characters, etc. When selecting sign information, it is necessary to fully consider the ship type information (such as length, width, deck height, load line height, etc.), ship cabin space, cabin identification code and other factors. At the same time, when arranging signs, the creation of signs will mainly be based on factors such as the outer plate of the hull structure.

[0005] However, the current sign system design mainly has the following problems: first, various current sign systems still rely mainly on manual operation, which leads to extremely low design efficiency of a single sign; second, due to the huge amount of material required for sign system design, the overall design cycle is relatively long; third, manual operation is prone to errors, which can easily lead to design errors. Summary of the invention

[0006] In view of the defects and shortcomings of the prior art, the present application provides a design method and system for a ship marking system. The design method for a ship marking system can improve the design efficiency of the ship marking system, shorten the design cycle, and reduce the design error.

[0007] An embodiment of the present application provides a method for designing a ship marking system, comprising the following steps:

[0008] Build the logo base library and build the expert system library;

[0009] Mark the type and key information involved in the ship logo design process;

[0010] Based on the historical three-dimensional design model of ships, the information of ship logo design is extracted and data is processed to form AI training sample data;

[0011] Training an AI model based on the AI ​​training sample data, and selecting an AI model with the best prediction result through iterative optimization;

[0012] In the new ship design scenario, the type marking step is performed on the 3D model to perform the information involved in the ship logo design process;

[0013] Extract the three-dimensional model features in the new ship design scenario, perform data processing, input them into the AI ​​model with the best prediction result, and output the prediction result information of the ship logo design by type;

[0014] Based on the prediction results, the model of the sign basic library is called, and at the same time, the standardization check and three-dimensional model design information of the sign are performed in combination with the expert system library;

[0015] Based on the expert system library, the designed three-dimensional model of the logo is modified.

[0016] As an implementation mode, the construction of the logo base library specifically includes the following steps:

[0017] Mark the type and key information of the symbols in the 3D design model of historical ships;

[0018] Output the ship mark type and combination information;

[0019] Analyze the output ship mark type and combination form information, and filter out parts whose combination frequency exceeds a preset threshold by type;

[0020] For parts whose combination frequency exceeds the preset threshold, a three-dimensional model of the parts is constructed to form a basic library of markings.

[0021] As an implementation method, the types and key information of the signs in the three-dimensional design model of the historical ship are marked, including:

[0022] The historical ship three-dimensional design models are distinguished by type and combination of information; and sub-type classification is performed based on the classified models, and the type and key information of each marked model are marked one by one.

[0023] As an implementation method, the types and key information of the signs in the three-dimensional design model of historical ships are marked. The marked information should comprehensively consider the creation method of the three-dimensional model of the ship sign, the combination method of the three-dimensional model, the parts of the three-dimensional model, and the information of the ship sign type.

[0024] As an implementation method, the ship mark type and combination form information is output, specifically including:

[0025] After the current three-dimensional model marking is completed, the marking information of the mark will be extracted;

[0026] When some key information of the mark can be extracted based on the model features, there is no need to mark it and it can be directly output based on the model features.

[0027] As an implementation method, the output ship mark type and combination form information is analyzed, and parts whose combination frequency exceeds a preset threshold are screened out by type, specifically including:

[0028] The output annotation information is analyzed, and the models with the same specifications whose usage frequency exceeds the preset threshold are sorted out according to type and usage frequency.

[0029] As an implementation method, a three-dimensional model of parts is constructed for parts whose combination frequency exceeds a preset threshold to form a basic library of signs, which specifically includes:

[0030] The modeling method of marking parts whose combination frequency (or usage rate) exceeds the preset threshold, the combination form between the parts of each object, etc., determines the construction method of the parts according to the type;

[0031] When the three-dimensional model of the part is completed, the created three-dimensional model is uniformly associated with the specified basic library.

[0032] As an implementation method, the way of constructing the three-dimensional model of the part needs to be combined with the design method of the design platform. Different design platforms have different design methods or design concepts, which may lead to different ways of constructing parts.

[0033] As an implementation mode, the construction of the expert system library specifically includes the following steps:

[0034] Collect information on ship industry specifications and experience of senior design experts;

[0035] The collected information is screened and processed to form structured data information, and the structured data information is stored in the expert system library.

[0036] As an implementation method, collect ship industry specifications and experience information of senior design experts, including:

[0037] Collect regulatory requirements in the shipbuilding industry and the knowledge, experience and design methods of senior design experts.

[0038] As an implementation method, based on the expert system library, the modification of the designed three-dimensional model of the sign specifically includes:

[0039] The designed three-dimensional model of the logo is modified using the standards and rules in the expert system library.

[0040] As an implementation method, based on the historical ship three-dimensional design model, the information of the ship logo design is extracted and data processing is performed to form AI training sample data, and the information includes:

[0041] The ship mark creation requires reference to the structure section name, deck information, ship basic information, ship cabin space information, and mark information for type marking.

[0042] As an implementation method, the three-dimensional model characteristics in the new ship design scenario are extracted, specifically including:

[0043] Extract the structural segment name, deck information, ship basic information, and ship cabin space information required for reference in creating the ship logo.

[0044] Another embodiment of the present application provides a design system for a ship marking system, including:

[0045] A logo basic library building module is used to build a logo basic library;

[0046] Expert system library building module, used to build expert system library;

[0047] An intelligent prediction module is used to mark the type and key information of the information involved in the ship logo design process; and based on the historical ship three-dimensional design model, extract the information of the ship logo design, and perform data processing to form AI training sample data; and train the AI ​​model based on the AI ​​training sample data, and select the AI ​​model with the best prediction result through iterative optimization;

[0048] The intelligent design module is used to perform the steps of type marking of information involved in the ship logo design process on the three-dimensional model in the new ship design scenario; and extract the three-dimensional model features in the new ship design scenario, and perform data processing, input the data into the AI ​​model with the best prediction result, and output the prediction result information of the ship logo design; and based on the prediction result, call the logo basic library model, and at the same time, combine with the expert system library to perform normative inspection and three-dimensional model design on the logo three-dimensional model design information; and based on the expert system library, modify the designed logo three-dimensional model.

[0049] As an implementation method, the mark basic library construction module is used to construct the mark basic library, specifically including: marking the type and key information of the marks in the historical ship three-dimensional design model; outputting the ship mark type and combination form information; analyzing the output ship mark type and combination form information, and filtering out parts whose combination frequency exceeds a preset threshold by type; constructing a three-dimensional model of the parts whose combination frequency exceeds the preset threshold to form a mark basic library;

[0050] The expert system library construction module is used to construct the expert system library, specifically including: collecting ship industry specifications and senior design expert experience information; screening and processing the collected information to form structured data information, and storing the structured data information in the expert system library.

[0051] As described above, the design method and system of the ship marking system of the present application have the following beneficial effects:

[0052] The design method of the ship sign system of the present application applies a machine learning algorithm, which can automatically learn design rules from a large number of historical design results without the need for tedious manual rule formulation, greatly improving the degree of design automation; in the new product design process, it can automatically predict the sign composition type, sign information and related sign symbol specifications, and realize the rapid generation of the sign three-dimensional model. Compared with the traditional manual design method one by one, it greatly improves the design efficiency and shortens the design cycle; based on the expert system library, the abnormal information is corrected, which effectively avoids the errors that may occur in manual design, improves the accuracy and reliability of the design, and ensures the design quality of the ship sign system; reduces the intensity of manual labor, reduces the impact of human factors on the design, and makes the design of the ship sign system more standardized, standardized and intelligent. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 Shown is a flow chart of a method for designing a ship marking system in an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0055] It should be noted that the illustrations provided in this embodiment are only used to illustrate the basic concept of the present invention in a schematic manner, and therefore the illustrations only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0056] When using welding robots to weld small structures of ships, the existing technology often adopts algorithms that solve the traveling salesman problem for path planning. However, such algorithms have disadvantages such as complex structure, too many parameters, and complicated adjustment mechanisms, which to a certain extent limit their application effect in welding path planning of small workpieces of ships.

[0057] As a combination product that integrates multiple complex systems and huge amounts of material, the construction process of a ship involves a high degree of technical complexity and information density. In order to ensure the high quality and efficiency of ship construction, it is crucial to intuitively and accurately reflect and convey various types of ship information. During the design and construction phase of a ship, ship information is usually expressed and explained in the form of signs, which play a key role in ship identification, operation, maintenance, and safety management.

[0058] From the perspective of composition, ship markings are mainly divided into compliance markings, alphanumeric markings and mixed markings. Taking a certain LNG ship as an example, it contains up to 760 large and small markings, which cover various systems and components of the ship. With the increase in the complexity of ship systems and the expansion of the amount of parts, the number of markings will also increase accordingly, which undoubtedly puts higher requirements on the design and management of markings.

[0059] The process of sign design mainly includes two links: the selection of sign information type and the creation / layout of sign objects. When selecting sign information, many factors need to be considered, such as the creation location of the sign and its reference basis, the specific form of the sign, the sign characters, etc. In addition, the ship's ship type information (such as length, width, deck height, load line height), ship cabin space layout, cabin identification code, etc. need to be comprehensively considered to ensure the accuracy and practicality of the sign information. When arranging the sign, it is necessary to make reasonable planning based on factors such as the outer plate of the hull structure to ensure the visibility and readability of the sign.

[0060] However, there are still many problems in the design of ship marking systems. First, the design of various marking systems mainly relies on manual operation, which is not only inefficient, but also difficult to meet the needs of large-scale and efficient ship construction. Secondly, due to the huge amount of materials involved in the design of ship marking systems, the total design cycle is long, which affects the construction progress and delivery time of the ship. In addition, manual operation is prone to human errors, resulting in design errors, which may pose a potential threat to the safety and reliability of the ship.

[0061] In view of the above defects, the present application provides a design method and system for a ship marking system, which are now described in detail through the following embodiments.

[0062] This embodiment provides a design method for a ship marking system. Figure 1 As shown, the following steps are included:

[0063] S1: Build the basic library of signs and build the expert system library;

[0064] S2: Mark the type and key information involved in the ship logo design process;

[0065] S3: Based on the historical ship 3D design model, the information of ship logo design is extracted and data is processed to form AI training sample data;

[0066] S4: Train the AI ​​model based on the AI ​​training sample data, and select the AI ​​model with the best prediction results through iterative optimization;

[0067] S5: In the new ship design scenario, the three-dimensional model is subjected to a type labeling step for the information involved in the ship logo design process;

[0068] S6: Extract the 3D model features in the new ship design scenario, process the data, input it into the AI ​​model with the best prediction result, and output the prediction result information of the ship logo design;

[0069] S7: Based on the prediction results, the sign basic library model is called, and the expert system library is combined to perform a standardization check on the sign three-dimensional model design information and three-dimensional model design;

[0070] S8: Modification of the designed logo 3D model based on the expert system library.

[0071] Wherein, step S1 includes step S11 of constructing a mark base library and step S12 of constructing an expert system library.

[0072] Step S11 builds a basic library of signs, which specifically includes the following steps:

[0073] S111: Mark the type and key information of the signs in the 3D design model of the historical ship;

[0074] S112: Outputting the ship mark type and combination information;

[0075] S113: Analyze the output ship mark type and combination form information, and filter out part information whose combination frequency exceeds a preset threshold value according to the type;

[0076] S114: constructing a three-dimensional model of the parts whose combination frequency exceeds a preset threshold to form a basic library of marks.

[0077] Step S111 specifically includes: distinguishing the three-dimensional models in the previous historical ship three-dimensional design models according to information such as type and combination form. For example, the marks can distinguish steel plate marks, surfacing marks, etc. according to the type; sub-classifying based on the classified models, and marking the type and key information of each marked model one by one, for example, marking the composition and plate thickness of the rib position mark, the component parts of the compartment mark and the corresponding cabin name.

[0078] The information marked on the three-dimensional model in step S111 should comprehensively consider the creation method of the three-dimensional model of the ship mark, the combination method of the three-dimensional model, the parts of the three-dimensional model, the type of ship mark and other information.

[0079] In step S112, the ship mark type and combination form information are output, specifically including: when the marking of the three-dimensional model is completed, the marking information of the mark is extracted; when some key information of the mark can be extracted based on the model feature, it does not need to be marked, and it is directly output based on the model feature.

[0080] Step S113 specifically includes: analyzing the currently output annotation information, and sorting out the models of the same specifications with higher usage rates (or exceeding a preset threshold) according to type and usage frequency.

[0081] Step S114 specifically includes: sorting out the modeling methods of annotations of parts with high usage rates (or exceeding a preset threshold), the combination forms of parts of each object, etc., and determining the construction method of the parts according to the type. For example, some parts can be combined according to the modeling method of standard parts, and some models are basically the same with only small differences, which can be defined as template part modeling; when the three-dimensional model of the part is completed, the constructed three-dimensional model of the part is uniformly associated with the specified basic library.

[0082] The way to construct the three-dimensional model of parts needs to be combined with the design method of the design platform. Different design platforms have different design methods or design concepts, which may lead to different ways of constructing parts.

[0083] Step S12 builds an expert system library, which specifically includes the following steps:

[0084] S121: Collect information on ship industry specifications and experience of senior design experts;

[0085] S122: Screening and processing the collected information to form structured data information, and storing the structured data information in the expert system library.

[0086] Step S121 specifically includes: collating and collecting relevant industry specifications and design experience of senior design experts to form structured data.

[0087] In step S122, the structured data information is structured data that is easy to read.

[0088] Integrate relevant industry specifications and requirements as well as the knowledge, experience, and design methods of design experts, and classify the relevant information.

[0089] The specific classification can be determined based on the actual application scenario and the amount of data collected. The more detailed the type distinction is, the more refined the data classification is, and the more detailed the specification requirements for the signs are.

[0090] The processed structured data can be text, relationship icons, and stored in the corresponding database.

[0091] Step S2 specifically includes: sorting out the key and important information required for sign construction according to the sign type, and marking the sorted information in combination with the actual three-dimensional model. For example, the hold space is marked by specific node types, the structural deck and outer plate are marked manually, and the midship is obtained in the structure tree through rules.

[0092] During the model information marking process, other information can also be marked on the ship mark, for example, the distance of a certain surfacing model from the deck, the load mark and the draft mark need to be marked for the current ship navigation.

[0093] In step S3, the information of ship sign design is extracted, including the structure segment name, deck information, ship basic information, ship cabin space information, and various sign information marked in the above steps required for the creation of the ship sign. The extracted ship sign design information is processed and the processed data is used as the training sample data of the AI ​​model.

[0094] When extracting ship mark design information, it can be classified and extracted according to the design type of the ship mark, type, and region.

[0095] When classifying and extracting according to the type of ship mark, different data objects can be extracted for different types of marks; different design software may require different data objects for the same type of mark due to differences in modeling and design methods.

[0096] Data processing includes handling anomalies, missing values, data type processing, etc. in the data, enhancing data visualization, regularization, standardization, etc., while improving data utilization.

[0097] Step S4 also includes: performing AI model training based on the AI ​​training sample data in step S3; iteratively training relevant parameters of the AI ​​model to obtain the optimal AI model by analyzing the weight ratio of each feature and the performance and accuracy of the AI ​​model; iteratively selecting appropriate AI algorithms and parameters based on the distribution characteristics of sample data and prediction problems to ensure the robustness and correctness of the model.

[0098] In step S4, the AI ​​model with the best prediction result is selected through iterative optimization in order to improve the accuracy of the AI ​​model prediction. It can be understood that the selection of the best prediction result in this step refers to selecting the AI ​​model with the best prediction result under the condition of a preset number of iterations.

[0099] Step S6 specifically includes: based on the three-dimensional model in the new ship setting scene, extracting the structural segment name, deck information, basic ship information, and ship cabin space information required for reference of the ship logo creation, and performing data processing on the extracted information; inputting the processed data into the AI ​​model, and outputting the key prediction information of the logo design according to the type based on the trained AI model.

[0100] Step S7 specifically includes: based on the information predicted by the current AI model, designing a three-dimensional model of the ship sign by type. For example, if the current design type is a rib position sign, the returned sign combination type calls the corresponding parts in the sign basic library one by one based on the name of the combined parts, and then combines and assembles them based on the rules based on the spatial position. For another example, if the current design type is a compartment sign, the cabin point and the normal plane of the cabin point used as a reference for the compartment sign design are returned, and then the solid model contours are constructed and stretched one by one through the design mechanism of the three-dimensional design platform.

[0101] Different types of ship signs have different creation locations and types due to differences in ship types. The design can be combined with the rules and experience in the expert system library constructed in step S12.

[0102] For some similar signs, the ship sign model design can also be realized by calling the modeling method of the template part and returning the template name and related parameters through AI.

[0103] Step S8 specifically includes: after the three-dimensional model design of the ship logo is completed, the correctness of the currently designed ship logo is checked using the standards and rules in the expert system library constructed in step S12.

[0104] If there are differences in the current design model, the actual three-dimensional model can be modified according to the corresponding standards and rules in the expert system library constructed in step S12 to meet the standard requirements.

[0105] If the current model design differs greatly, the ship mark can be corrected, and steps S11, S2 to S4 can be repeated to iteratively train the AI ​​model.

[0106] The design method of the ship sign system provided in this embodiment applies a machine learning algorithm and can automatically learn design rules from a large number of historical design results without the need for tedious manual rule formulation, thereby greatly improving the degree of automation of the design; in the process of new product design, the sign composition type, sign information and related sign symbol specifications can be automatically predicted, thereby realizing the rapid generation of a three-dimensional sign model, which greatly improves the design efficiency and shortens the design cycle compared to the traditional manual design method one by one; abnormal information is corrected based on the expert system library, effectively avoiding possible errors in manual design, improving the accuracy and reliability of the design, and ensuring the design quality of the ship sign system; reducing the intensity of manual labor, reducing the impact of human factors on the design, and making the design of the ship sign system more standardized, standardized and intelligent.

[0107] This embodiment also provides a design system for a ship marking system, including:

[0108] A logo basic library building module is used to build a logo basic library;

[0109] Expert system library building module, used to build expert system library;

[0110] Intelligent prediction module, used to mark the type and key information involved in the ship logo design process; and based on the historical ship 3D design model, extract the information of the ship logo design, and perform data processing to form AI training sample data; and train the AI ​​model based on the AI ​​training sample data, and select the AI ​​model with the best prediction result through iterative optimization;

[0111] The intelligent design module is used to perform type marking steps on the information involved in the ship logo design process on the three-dimensional model in the new ship design scenario; and to extract the three-dimensional model features in the new ship design scenario, and perform data processing, input the data into the AI ​​model with the best prediction result, and output the prediction result information of the ship logo design by type; and based on the prediction result, call the logo basic library model, and at the same time combine with the expert system library to perform normative checks and three-dimensional model design on the logo three-dimensional model design information; and based on the expert system library, modify the designed logo three-dimensional model.

[0112] The mark basic library construction module is used to construct the mark basic library. The process of constructing the mark basic library specifically includes: marking the type and key information of the marks in the historical ship three-dimensional design model; outputting the ship mark type and combination form information; analyzing the output ship mark type and combination form information, and filtering out parts whose combination frequency exceeds the preset threshold by type; and constructing a three-dimensional model of the parts whose combination frequency exceeds the preset threshold to form a mark basic library.

[0113] The expert system library construction module is used to construct the expert system library. The process of constructing the expert system library specifically includes: collecting ship industry specifications and experience information of senior design experts; screening and processing the collected information to form structured data information, and storing the structured data information in the expert system library.

[0114] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A method for designing a ship marking system, characterized in that: The following steps are involved: Build the logo base library and build the expert system library; Mark the type and key information involved in the ship logo design process; Based on the historical ship 3D design model, the information of ship logo design is extracted and data is processed to form AI training sample data; Training an AI model based on the AI ​​training sample data, and selecting an AI model with the best prediction result through iterative optimization; In the new ship design scenario, the type marking step is performed on the 3D model to perform the information involved in the ship logo design process; Extract the three-dimensional model features in the new ship design scenario, perform data processing, input them into the AI ​​model with the best prediction result, and output the prediction result information of the ship logo design; Based on the prediction results, the model of the sign basic library is called, and at the same time, the standardization check and three-dimensional model design of the sign three-dimensional model design information are performed in combination with the expert system library; Based on the expert system library, the designed three-dimensional model of the logo is modified.

2. The design method according to claim 1, characterized in that: The construction of the logo base library specifically includes the following steps: Mark the type and key information of the symbols in the 3D design model of historical ships; Output the ship mark type and combination information; Analyze the output ship mark type and combination form information, and filter out parts whose combination frequency exceeds a preset threshold by type; For parts whose combination frequency exceeds the preset threshold, a three-dimensional model of the parts is constructed to form a basic library of markings.

3. The design method according to claim 2, characterized in that: Mark the types and key information of the signs in the 3D design model of historical ships, including: The historical ship three-dimensional design models are distinguished by type and combination of information; and sub-type classification is performed based on the classified models, and the type and key information of each marked model are marked one by one.

4. The design method according to claim 1, characterized in that: The construction of the expert system library specifically includes the following steps: Collect information on ship industry specifications and experience of senior design experts; The collected information is screened and processed to form structured data information, and the structured data information is stored in the expert system library.

5. The design method according to claim 4, characterized in that: Collect ship industry specifications and experience information from senior design experts, including: Collect regulatory requirements in the shipbuilding industry and the knowledge, experience and design methods of senior design experts.

6. The design method according to claim 4, characterized in that: Based on the expert system library, the modification of the designed three-dimensional model of the logo specifically includes: The designed three-dimensional model of the logo is modified using the standards and rules in the expert system library.

7. The design method according to claim 1, characterized in that: Based on the historical ship 3D design model, the information of ship logo design is extracted and data processing is performed to form AI training sample data. The information includes: The ship mark creation requires reference to the structure section name, deck information, ship basic information, ship cabin space information, and mark information for type marking.

8. The design method according to claim 1, characterized in that: Extract 3D model features for new ship design scenarios, including: Extract the structural segment name, deck information, ship basic information, and ship cabin space information required for reference in creating the ship logo.

9. A design system for a ship marking system, characterized in that: include: A logo basic library building module is used to build a logo basic library; Expert system library building module, used to build expert system library; An intelligent prediction module is used to mark the type and key information of the information involved in the ship logo design process; and based on the historical ship three-dimensional design model, extract the information of the ship logo design, and perform data processing to form AI training sample data; and train the AI ​​model based on the AI ​​training sample data, and select the AI ​​model with the best prediction result through iterative optimization; An intelligent design module is used to perform type marking steps on the three-dimensional model for the information involved in the ship logo design process in a new ship design scenario; and extracting the three-dimensional model features in the new ship design scenario, performing data processing, inputting the features into the AI ​​model with the best prediction result, and outputting the prediction result information of the ship logo design; and based on the prediction result, calling the logo basic library model, and combining with the expert system library, performing the normative inspection and three-dimensional model design of the logo three-dimensional model design information; And based on the expert system library, the designed three-dimensional model of the logo is modified.

10. The design system according to claim 9, characterized in that: The mark basic library construction module is used to construct the mark basic library, specifically including: marking the type and key information of the marks in the historical ship three-dimensional design model; outputting the ship mark type and combination form information; analyzing the output ship mark type and combination form information, and filtering out parts whose combination frequency exceeds the preset threshold by type; constructing the parts three-dimensional model for the parts whose combination frequency exceeds the preset threshold to form the mark basic library; The expert system library construction module is used to construct the expert system library, specifically including: collecting ship industry specifications and senior design expert experience information; screening and processing the collected information to form structured data information, and storing the structured data information in the expert system library.

Images