Intelligent detection method and detection system for ship design maturity

Abstract

The invention provides an intelligent detection method and system for ship design maturity, and the detection system employs a five-layer architecture, covers data collection, analysis detection, application service, edge calculation and cloud platform support, and achieves the coverage of the whole process from data collection to intelligent decision. Multi-source data collection is achieved through three-dimensional model analysis, NLP document processing and time sequence data capture, feature engineering is completed in combination with a PointNet + + algorithm, a graph neural network and a domain ontology library, and indexes such as integrity scores and conflict detection are output through a multi-task deep learning model to form an end-to-end intelligent detection system. The system predicts the progress through deep reinforcement learning, dynamically feeds back and adjusts a design strategy, optimizes resource allocation in combination with hybrid cloud and edge calculation, finally realizes closed-loop management of maturity evaluation, risk early warning and continuous improvement in the ship design process, and comprehensively improves the ship design efficiency and quality.

Description

Technical Field

[0001] This invention relates to the field of ship design technology, and in particular to an intelligent detection method and system for ship design maturity. Background Technology

[0002] In the field of ship design, with the increasing complexity of ship functions and the continuous improvement of design requirements, traditional methods for ship design maturity assessment and control face numerous challenges. From the current industry perspective, in the traditional ship design process, quality control and schedule management mainly rely on human experience and simple document management tools. Regarding design maturity assessment, the review of ship 3D models is often phased and largely based on manual inspection. For example, engineers need to examine each part of the model one by one to ensure it meets design requirements; this method is not only inefficient but also prone to human error. When dealing with multi-source heterogeneous data, such as 3D model data, design documents, and process data, traditional methods lack effective integration tools, making it difficult to fully explore the correlations between data. In terms of design status assessment, there is a lack of adaptive assessment models, making it difficult to flexibly adjust assessment standards according to different design projects and stages. At the same time, feedback control during the design process is not timely or precise enough, failing to form effective closed-loop management and hindering the rapid adjustment of design direction to meet quality and schedule requirements. These limitations seriously affect the efficiency, quality, and overall competitiveness of ship design, thus urgently requiring an intelligent assessment and control system to solve these problems. Summary of the Invention

[0003] In view of the shortcomings of the prior art described above, the present invention provides an intelligent detection method for ship design maturity, comprising the following steps:

[0004] S1: Data acquisition and processing, collecting various types of data in the current ship design process, including 3D model data, design documents, and various process data generated during the design process; then, performing feature extraction on the data, including geometric feature extraction, topological relationship analysis, semantic feature annotation, and temporal feature construction, to obtain a feature set;

[0005] S2: Based on a multi-task deep learning model, the feature set extracted from the current ship design is tested, including integrity testing, conflict detection, and code compliance testing, and a testing index system is established.

[0006] S3: Dynamic monitoring and feedback, presenting test results in multiple forms, providing intelligent early warnings and feedback for rectification of abnormal test results, and predicting the progress of subsequent ship design tasks.

[0007] Optionally, in step S1, for geometric feature extraction, the PointNet++ algorithm is used to process the three-dimensional point cloud data to extract the geometric features of the ship's three-dimensional model; for topological relationship analysis, the assembly relationship data is processed by a graph neural network to obtain the connection methods and dependencies between components; for semantic feature annotation, the three-dimensional model and design documents are semantically annotated by establishing an ontology library in the ship design field; for temporal feature construction, the patterns and rules of design changes are identified by analyzing the design modification history.

[0008] Optionally, in step S2, the multi-task deep learning model includes an input layer, an output layer, and a hidden layer;

[0009] The input layer is used to fuse the extracted feature sets into the input;

[0010] Hidden layers consist of multiple layers of neural networks used for non-linear transformations and feature extraction of input data;

[0011] The output layer is used to output the results of integrity detection, conflict detection, and specification compliance detection in parallel.

[0012] Optionally, in step S2, the hidden layer uses a 5-layer residual network and an attention mechanism.

[0013] Optionally, in step S2, the integrity score is used to reflect the overall completion status of the design, the conflict detection is used to find conflicts between components in the design, and the specification compliance is used to determine whether the design meets the relevant industry specifications and standards.

[0014] Optionally, in step S3, the presentation of the detection results includes a 3D visualization dashboard, a progress Gantt chart, and a quality radar chart.

[0015] Optionally, in step S3, the intelligent early warning includes a schedule deviation early warning based on time series prediction. By analyzing historical schedule data, a time series model is established to predict the future design schedule. If a deviation in the schedule is predicted, an early warning is issued in a timely manner.

[0016] Optionally, in step S3, feedback and rectification include establishing a PDCA closed-loop management process, generating a rectification suggestion report, and distributing task work orders.

[0017] The present invention also provides a detection system for implementing the intelligent detection method, comprising:

[0018] The data acquisition layer includes a multi-source heterogeneous data access module, used to collect various types of data during the ship design process;

[0019] The analysis and detection layer includes a feature extraction engine and a core intelligent detection algorithm group. The feature extraction engine extracts features from the collected multi-source heterogeneous data and transforms the data into feature vectors that can be used for analysis. Based on these feature vectors, the core intelligent detection algorithm group uses artificial intelligence algorithms to detect and analyze the maturity status of ship design.

[0020] The application service layer includes a visual monitoring platform and a collaborative management interface. The visual monitoring platform displays information such as the maturity status, progress, and quality indicators of the ship design in an intuitive graphical interface. The collaborative management interface provides a channel for information exchange and collaborative work between different departments and teams, ensuring smooth communication during the design process.

[0021] Optionally, it also includes:

[0022] The edge computing layer includes a distributed processing node cluster, which is located close to the data source to perform preliminary processing on the collected data.

[0023] The cloud platform support layer includes hybrid cloud infrastructure, which provides computing resources, storage resources, and data management capabilities for the entire system.

[0024] As described above, the intelligent detection method and system for ship design maturity provided by the present invention have the following beneficial effects:

[0025] 1) Feature extraction technology for 3D models based on multimodal data fusion

[0026] Three-dimensional ship models contain a wealth of information, and this technology can integrate multimodal data such as geometric features, topological relationships, and semantic information for feature extraction. For example, when processing three-dimensional models of ship structures, it can not only obtain geometric features such as the shape and size of structural components, but also understand the connection methods between components through topological relationship analysis, and clarify the function of components in the entire ship structure through semantic information.

[0027] 2) Adaptive assessment model for design maturity status

[0028] This model can adaptively adjust the evaluation criteria and weights during testing based on different stages of ship design (such as conceptual design, preliminary design, and detailed design), different types of ships (such as liquefied gas carriers and container ships), and different design requirements (such as different speed and deadweight requirements). For example, in the design of high-speed passenger ships, the preliminary design stage may focus more on the hull styling to meet speed requirements, and in this case, the model will correspondingly increase the evaluation weight of hull styling-related indicators.

[0029] 3) Dynamic feedback closed-loop control system

[0030] The system establishes a dynamic feedback closed-loop control system capable of monitoring the maturity status of ship designs in real time. Once it detects that the design maturity does not meet requirements, the system promptly reports the problem and adjusts the design strategy based on the detection results. For example, if it finds that the design maturity of a certain subsystem is low, the system will suggest increasing design resources for that subsystem or adjusting the design process, and then continue to monitor the effects of the adjustments, forming a continuous improvement cycle.

[0031] 4) Progress prediction algorithm based on deep reinforcement learning

[0032] Deep reinforcement learning algorithms are used to predict the progress of ship design. This algorithm takes into account various uncertainties during the design process, such as design changes, resource allocation, and changes in the external environment. By continuously learning from lessons learned in historical projects, it can make more accurate predictions about the progress of current projects. For example, if a certain type of design change caused a delay in a historical project, the algorithm will consider this factor and adjust its progress predictions when encountering similar changes in the current project.

[0033] 5) Intelligent decision support mechanism based on human-machine collaboration

[0034] This system provides a human-machine collaborative intelligent decision support mechanism, fully leveraging the advantages of artificial intelligence algorithms in data processing and analysis, while combining the experience and creativity of human designers. For example, when selecting design solutions, the system provides multiple options based on data analysis, along with an analysis of the advantages and disadvantages of each. Designers can then further evaluate and optimize these options based on their experience, ultimately determining the best design solution. Attached Figure Description

[0035] Figure 1 The diagram shown is a flowchart of the intelligent detection method in Embodiment 1 of the present invention.

[0036] Figure 2 The diagram shown is an architecture diagram of the detection system in Embodiment 1 of the present invention. Detailed Implementation

[0037] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

[0038] It should be noted that the illustrations provided in this embodiment are only schematic representations of the basic concept of the present invention. Therefore, the illustrations only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0039] Example 1

[0040] like Figure 1 As shown in the figure, this embodiment provides an intelligent detection method for ship design maturity, including the following steps:

[0041] S1: Data Acquisition and Processing. This involves collecting various types of data generated during the current ship design process. Due to the wide range of data sources and diverse formats in ship design, it is necessary to collect data from multiple data sources to ensure the effective acquisition of different data types. These data sources include, but are not limited to, 3D model data in different formats, design documents, and various process data generated during the design process. This process is completed through a multi-source data acquisition module.

[0042] Specifically, the data sources include 3D model data generated by commonly used 3D modeling software in ship design, such as CATIA and SolidWorks, including key information such as the model's geometry, component composition, material properties, and assembly relationships. They also include documents such as technical specifications and calculation reports, which are processed using NLP technology. Through natural language processing, key information in the documents can be identified, such as the ship's performance requirements, design parameters, and calculation results. For example, important indicators such as the ship's maximum speed and cargo capacity can be extracted from the technical specifications, and key data for structural strength calculations can be obtained from the calculation reports.

[0043] Next, feature extraction is performed on the data source, including geometric feature extraction, topological relationship analysis, semantic feature annotation, and temporal feature construction, to obtain a feature set. This process is completed through the feature engineering module.

[0044] Specifically, for geometric feature extraction: the PointNet++ algorithm is used to process 3D point cloud data. This algorithm can effectively extract the geometric features of a ship's 3D model, such as the surface shape of the hull and the shape complexity of structural components. For example, for a ship's propeller 3D model, it can accurately extract geometric features such as the curvature and thickness of its blades.

[0045] For topology analysis: Graph neural networks process assembly relationship data. By constructing a graph structure to represent the assembly relationships between various ship components, graph neural networks can deeply analyze topological information such as connection methods and dependencies between components. For example, when analyzing the assembly relationship between a ship's engine and transmission system, it can determine the topological relationships such as the type of connection interface and the assembly sequence between the two.

[0046] For semantic feature annotation: Establish an ontology library for the ship design domain. This ontology library contains various concepts, terms, and their relationships in ship design. By semantically annotating 3D models and design documents, computers can better understand the design content. For example, a ship's compartment can be labeled "cargo hold" and associated with related concepts such as cargo storage and loading / unloading, so that its function and role in the ship's system can be accurately identified in subsequent analysis.

[0047] For the construction of temporal features: pattern recognition of design change sequences. By analyzing the design modification history, patterns and regularities of design changes are identified. For example, it is found that certain types of design changes are always accompanied by changes to other related components, or that certain types of changes are prone to occur in specific design stages. These temporal features help predict future design change trends.

[0048] Next, proceed to step S2: Based on a multi-task deep learning model, perform detection on the feature set extracted from the current ship design, including integrity detection, conflict detection, and code compliance detection, and establish a detection index system.

[0049] Integrity testing reflects the overall completion status of a design, conflict detection identifies conflicts between components (such as spatial interference), and compliance testing determines whether the design meets relevant industry standards and regulations. For example, when inspecting the cabin layout design of a ship, integrity scoring shows the completion rate of each cabin design, conflict detection identifies conflicts such as spatial overlap between cabins, and compliance testing determines whether the dimensions, ventilation, and other design aspects of the cabins comply with relevant regulations.

[0050] Multi-task deep learning models include an input layer, an output layer, and a hidden layer.

[0051] The input layer is used to fuse the extracted feature sets, enabling the model to comprehensively consider all aspects of the design. For example, when detecting the completion of a ship structural design, geometric feature vectors can provide information on the shape and size of structural components, while semantic feature matrices can provide information on the function and assembly relationships of components. The fusion of the two can more accurately assess the completion of the structural design.

[0052] The hidden layers, including multi-layer neural networks, perform nonlinear transformations and feature extraction on the input data. As a preferred approach, a 5-layer residual network and an attention mechanism are used in the hidden layers. The residual network helps address the vanishing gradient problem in deep neural networks, improving model training performance. The attention mechanism, by introducing weights, allows the model to focus more on important feature information, improving detection accuracy. For example, in detecting the design of ship electrical systems, the attention mechanism can make the model focus on important features such as the connection relationships and parameter settings of key electrical equipment, increasing the weight of these aspects.

[0053] The output layer is used to output the results of integrity checks, conflict detection, and compliance checks in parallel. Integrity scores reflect the overall completion status of the design, conflict detection identifies conflicts between components (such as spatial interference), and compliance checks determine whether the design meets relevant industry specifications and standards. For example, when inspecting the cabin layout design of a ship, integrity scores can show the completion rate of each cabin design, conflict detection can identify whether there are spatial overlaps or other conflicts between cabins, and compliance checks can determine whether the dimensions, ventilation, and other design aspects of the cabins comply with relevant specifications.

[0054] It's important to note that multi-task deep learning models require pre-training with multiple input samples. Simply put, this involves training the neural network model on existing samples to obtain standard values. Then, the feature set of the current ship design is compared with these standard values, and the differences are used to derive various detection results. For example, comparing the geometry of the current ship design with standard values ​​reveals its completeness; comparing the topological relationships allows for conflict detection; and comparing the semantic features of the current ship design with standard values ​​allows for compliance checks, such as whether the thickness of a bulkhead meets industry standards.

[0055] The evaluation index system includes, but is not limited to, design integrity, coordination between subsystems, rationality of design changes, and compliance with regulations. Design integrity can be measured by calculating the proportion of completed design elements to the overall design; coordination between subsystems can be assessed by analyzing interface matching and data exchange smoothness; rationality of design changes can be judged based on whether the changes meet design objectives and whether they have undergone a reasonable approval process; and compliance with regulations is determined by comparing with relevant ship design codes and standards.

[0056] Next, proceed to step S3: dynamic monitoring and feedback, presenting the test results in multiple forms, providing intelligent early warnings and feedback for rectification of abnormal test results, and predicting the progress of subsequent ship design tasks.

[0057] Specifically, the test results are presented in the form of a 3D visualization dashboard, a progress Gantt chart, and a quality radar chart. This process is achieved through a real-time monitoring module.

[0058] For 3D visualization dashboards, WebGL technology is used for browser-side rendering. Through these dashboards, users can intuitively view 3D models of ship designs and annotate them with information related to design maturity, such as which components are not yet complete or which areas pose design risks. For example, when viewing the overall layout model of a ship, users can clearly see areas where equipment locations are not yet determined or components with spatial interference risks.

[0059] For schedule Gantt charts, automatic association with the WBS (Work Breakdown Structure) is achieved. Gantt charts clearly display the schedule of design tasks. When associated with the WBS, they accurately reflect the position and progress of each task within the overall project, facilitating schedule control for project managers. For example, if a ship design task is broken down into multiple sub-tasks such as hull design, power system design, and electrical system design based on the WBS, the schedule Gantt chart can display the start time, end time, and completion progress of each sub-task in real time.

[0060] The quality radar chart dynamically displays six evaluation indicators. It assesses design quality across six dimensions (such as design compliance, reliability, safety, maintainability, economy, and environmental friendliness) and updates data in real time, dynamically showing changes in design quality. For example, if the ship's structure is modified during the design process, the quality radar chart can reflect the impact of this modification on dimensions such as structural reliability and economy in real time.

[0061] For intelligent early warning, this includes schedule deviation warnings based on time series forecasting: by analyzing historical schedule data, a time series model is built to predict future design schedules. If a schedule deviation is predicted, the system will issue a timely warning. For example, if historical data predicts that a certain percentage of the tasks in a design phase should be completed by a specific date, but the actual schedule is behind schedule, the system will issue a warning to project managers.

[0062] For schedule forecasting, an LSTM network is used to build the prediction model: Long Short-Term Memory (LSTM) networks can handle long-term dependencies in time series data, improving the accuracy of schedule forecasting. For example, in ship design, some design tasks may be affected by multiple previous tasks. LSTM networks can effectively capture these long-term dependencies, thus predicting the schedule of these tasks more accurately.

[0063] For feedback and rectification, this includes establishing a PDCA closed-loop management process, generating rectification suggestion reports, and distributing task work orders.

[0064] Establish a PDCA closed-loop management process: a cycle of Plan, Do, Check, and Act. Based on the inspection results, develop improvement plans, implement improvement measures, check the effectiveness of the improvements, and standardize effective improvement measures to continuously improve design maturity. For example, if a ship's subsystem design is found to be non-compliant with specifications, first develop an improvement plan, then implement improvement measures, check whether the improved design meets the requirements, and finally incorporate effective improvement measures into the standard process to prevent similar problems from recurring.

[0065] Automatically Generate Rectification Recommendation Reports: Based on the detected problems, the system automatically generates rectification recommendation reports, which include a problem description, possible cause analysis, and specific rectification suggestions. For example, when an issue of unreasonable space utilization in the ship's cabin layout is detected, the report will describe the cabin where the problem occurs, the possible reasons for this being due to insufficient consideration of equipment dimensions in the initial design phase, and provide rectification suggestions such as adjusting the equipment layout or replanning the cabin space.

[0066] Task Order Distribution: The system automatically distributes rectification tasks as work orders to relevant designers, ensuring timely execution of rectification work. For example, if the strength calculation of a component in a ship's structural design is found to be non-compliant, the system will automatically send a rectification task work order to the engineer responsible for the structural design of that component.

[0067] Based on the above-described intelligent detection method, this embodiment also provides a detection system, such as... Figure 2 As shown, it includes:

[0068] The data acquisition layer, including a multi-source heterogeneous data access module, serves as the data entry point for the entire system, responsible for collecting various types of data during the ship design process. Given the wide range and diverse formats of ship design data, this module can interface with multiple data sources to ensure the effective acquisition of different data types. These data sources include, but are not limited to, 3D model data in various formats, design documents, and various process data generated during the design process. Through a unified interface, this multi-source heterogeneous data is accessed, providing fundamental data support for subsequent analysis and processing.

[0069] The analysis and detection layer, comprising a feature extraction engine and a core intelligent detection algorithm group, extracts features from the collected multi-source heterogeneous data, transforming the data into feature vectors suitable for analysis. Based on these feature vectors, the core intelligent detection algorithm group uses artificial intelligence algorithms to detect and analyze the maturity status of ship design.

[0070] The application service layer includes a visual monitoring platform and a collaborative management interface. The visual monitoring platform displays information such as the maturity status, progress, and quality indicators of the ship design through an intuitive graphical interface, allowing the design team and management personnel to quickly understand the design status. The collaborative management interface provides a channel for information exchange and collaborative work between different departments and teams, ensuring smooth communication throughout the design process.

[0071] The edge computing layer, comprising distributed processing node clusters located close to the data source, performs preliminary processing on the collected data. This helps alleviate the computational burden on the cloud platform while improving data processing efficiency and reducing data transmission latency.

[0072] The cloud platform support layer includes hybrid cloud infrastructure, which provides the entire system with powerful computing resources, storage resources, and data management capabilities. It can flexibly allocate resources according to system needs, meeting the requirements of large volumes of ship design data and complex calculations.

[0073] More specifically, the data acquisition layer includes a multi-source data acquisition module and a feature engineering processing module. The multi-source data acquisition module is used to collect various types of data during the current ship design process.

[0074] In summary, this invention provides an intelligent detection method and system for ship design maturity. The system employs a five-layer architecture, encompassing data acquisition (multi-source heterogeneous data access), analysis and detection (feature extraction and intelligent detection), application services (visual monitoring and collaborative management), edge computing (distributed node processing), and cloud platform support (hybrid cloud resource allocation), achieving full-process coverage from data collection to intelligent decision-making. Multi-source data acquisition is achieved through 3D model analysis, NLP document processing, and time-series data capture. Feature engineering is completed using the PointNet++ algorithm, graph neural networks, and domain ontology libraries. A multi-task deep learning model (MTL-DNN) is used to output indicators such as integrity scores and conflict detection, forming an end-to-end intelligent detection system. The system predicts progress through deep reinforcement learning, dynamically adjusts design strategies based on feedback, and optimizes resource allocation through hybrid cloud and edge computing. Ultimately, it achieves closed-loop management of ship design maturity assessment, risk warning, and continuous improvement, comprehensively enhancing the efficiency and quality of ship design.

[0075] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can 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 those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. An intelligent detection method for ship design maturity, characterized in that, Includes the following steps: S1: Data acquisition and processing, collecting various types of data in the current ship design process, including 3D model data, design documents, and various process data generated during the design process; then, performing feature extraction on the data, including geometric feature extraction, topological relationship analysis, semantic feature annotation, and temporal feature construction, to obtain a feature set; S2: Based on a multi-task deep learning model, the feature set extracted from the current ship design is tested, including integrity testing, conflict detection, and code compliance testing, and a testing index system is established. S3: Dynamic monitoring and feedback, presenting test results in multiple forms, providing intelligent early warnings and feedback for rectification of abnormal test results, and predicting the progress of subsequent ship design tasks.

2. The intelligent detection method for ship design maturity according to claim 1, characterized in that: In step S1, for geometric feature extraction, the PointNet++ algorithm is used to process the 3D point cloud data, thereby extracting the geometric features of the ship's 3D model; for Topological relationship analysis uses graph neural networks to process assembly relationship data and obtain the connection methods and dependencies between components; for semantic feature annotation, a ship design ontology library is established to semantically annotate 3D models and design documents; for temporal feature construction, the design modification history is analyzed to identify the patterns and rules of design changes.

3. The intelligent detection method for ship design maturity according to claim 1, characterized in that: In step S2, the multi-task deep learning model includes an input layer, an output layer, and a hidden layer; The input layer is used to fuse the extracted feature sets into the input; Hidden layers consist of multiple layers of neural networks used for non-linear transformations and feature extraction of input data; The output layer is used to output the results of integrity detection, conflict detection, and specification compliance detection in parallel.

4. The intelligent detection method for ship design maturity according to claim 3, characterized in that: In step S2, the hidden layer uses a 5-layer residual network and an attention mechanism.

5. The intelligent detection method for ship design maturity according to claim 1, characterized in that: In step S2, the integrity score is used to reflect the overall completion status of the design, the conflict detection is used to find conflicts between components in the design, and the specification compliance is used to determine whether the design meets the relevant industry specifications and standards.

6. The intelligent detection method for ship design maturity according to claim 1, characterized in that: In step S3, the presentation of the detection results includes a 3D visualization dashboard, a progress Gantt chart, and a quality radar chart.

7. The intelligent detection method for ship design maturity according to claim 1, characterized in that: In step S3, intelligent early warning includes schedule deviation early warning based on time series prediction. By analyzing historical schedule data, a time series model is established to predict future design schedules. If a schedule deviation is predicted, an early warning is issued in a timely manner.

8. The intelligent detection method for ship design maturity according to claim 1, characterized in that: In step S3, feedback and rectification are carried out, including establishing a PDCA closed-loop management process, generating rectification suggestion reports, and distributing task work orders.

9. A detection system, characterized in that, The detection system is used to implement the intelligent detection method according to any one of claims 1-8, including: The data acquisition layer includes a multi-source heterogeneous data access module, used to collect various types of data during the ship design process; The analysis and detection layer includes a feature extraction engine and a core intelligent detection algorithm group. The feature extraction engine extracts features from the collected multi-source heterogeneous data and transforms the data into feature vectors that can be used for analysis. Based on these feature vectors, the core intelligent detection algorithm group uses artificial intelligence algorithms to detect and analyze the maturity status of ship design. The application service layer includes a visual monitoring platform and a collaborative management interface. The visual monitoring platform displays information such as the maturity status, progress, and quality indicators of the ship design in an intuitive graphical interface. The collaborative management interface provides a channel for information exchange and collaborative work between different departments and teams, ensuring smooth communication during the design process.

10. The detection system according to claim 9, characterized in that, Also includes: The edge computing layer includes a distributed processing node cluster, which is located close to the data source to perform preliminary processing on the collected data. The cloud platform support layer includes hybrid cloud infrastructure, which provides computing resources, storage resources, and data management capabilities for the entire system.

Images