Automatic Defect Detection System, Method, Apparatus and Related Equipment for Gas Turbines

Abstract

This application discloses an automatic defect determination system, method, apparatus, and related equipment for gas turbines. The system includes a 3D model import module for importing and managing a reference 3D model of gas turbine components; a data acquisition module for acquiring inspection data of the gas turbine components; a defect drawing and determination module for: registering the inspection data with the reference 3D model; drawing defect areas based on the registered 3D model; automatically determining the defect level based on the inspection data and / or defect areas, and calling rules from the defect rule standard library management module; a report generation module for generating inspection reports based on the defect level; and a data statistics module for storing defect data and providing statistical analysis functions. This invention can replace traditional manual inspection with automated and standardized intelligent determination, completely solving the problems of subjective judgment results, low efficiency, and difficult data management.

Description

Technical Field

[0001] This application relates to the field of nondestructive testing technology for industrial equipment, and in particular to an automatic defect determination system, method, device and related equipment for gas turbines. Background Technology

[0002] As a core power source in the energy, aviation, and marine sectors, the safety and reliability of gas turbines are of paramount importance. Regular endoscopic inspections of the hot-end components of gas turbines are a crucial step in ensuring their safe operation.

[0003] In current technology, defect detection in gas turbines primarily relies on manual labor. Inspectors use endoscopes to penetrate the unit, observing two-dimensional images on a monitor to identify defects, and then determine the severity of the defects (e.g., "pass" or "fail") based on personal experience and paper or electronic maintenance manuals. This traditional method has the following significant drawbacks: 1. High subjectivity and inconsistent standards: Defect judgment heavily relies on the experience and responsibility of the inspectors. Different personnel may have significantly different judgments on the same defect, lacking objective and unified standards. 2. Low efficiency and high labor intensity: The entire process of manual inspection, recording, drawing, and judgment is time-consuming, and staring at the screen for extended periods can lead to visual fatigue, increasing the risk of missed detections and misjudgments. 3. Difficult data management and insufficient value extraction: Inspection data (images, videos, records) is mostly stored in discrete file formats, making systematic management and retrieval difficult. Vast amounts of historical data are not effectively utilized, making it impossible to analyze defect occurrence patterns, predict component lifespan, or provide data support for maintenance strategy optimization and spare parts management. 4. Low visualization and high communication costs: Traditional two-dimensional images and text descriptions cannot intuitively and accurately express the shape, location, and size of complex defects in three-dimensional space, hindering technical review and maintenance decisions. 5. Difficult knowledge transfer: The experience of senior experts is difficult to systematically and digitally accumulate and transfer. New employee training cycles are long, and personnel turnover can easily lead to technical gaps.

[0004] With the development of artificial intelligence, 3D modeling and big data technologies, there is an urgent need for a system and method that can overcome the above-mentioned shortcomings and realize the automation, standardization and intelligence of defect detection. Summary of the Invention

[0005] Based on the above problems, this application provides an automatic gas turbine defect detection system, method, device and related equipment, which can solve the technical problems of gas turbine defect detection relying on manual labor, low efficiency, inconsistent standards, difficult data management and lack of decision support in the prior art.

[0006] In a first aspect, embodiments of this application provide an automatic defect determination system for gas turbines, the system comprising: The defect rule standard library management module is used to store rules for determining defect levels based on defect characteristics; The 3D model import module is used to import and manage the baseline 3D models of gas turbine components; The data acquisition module is used to acquire test data of gas turbine components; The defect drawing and judgment module communicates with the 3D model import module, data acquisition module, and defect rule standard library management module, and is used for: The detection data is registered with the benchmark 3D model; Based on the registered 3D model, the defect area is drawn; Based on the detection data and / or defect area, the rules in the defect rule standard library management module are called to automatically determine the defect level; The report generation module communicates with the defect drawing and judgment module and is used to generate inspection reports based on defect levels. The data statistics module communicates with the defect drawing and judgment module to store defect data and provide statistical analysis functions.

[0007] In one possible implementation, the defect judgment rules stored in the defect rule standard library management module are associated with the operating condition parameters of the gas turbine components; the detection data acquired by the data acquisition module includes the operating condition parameters; and the defect drawing and judgment module is also used to call the rules associated with the current operating condition parameters when judging the defect level.

[0008] In one possible implementation, based on detection data and / or defect areas, rules from the defect rule standard library management module are invoked to automatically determine the defect level, including: Based on the detection data and / or defect area, extract geometric features; geometric features include area features and depth features; match the geometric features with the rules in the defect rule standard library management module; if a matching rule is found, output the corresponding defect level according to the rule.

[0009] In one possible implementation, the detection report generated by the report generation module is an interactive document with an embedded 3D visualization view that allows users to perform rotation, zoom, sectioning, and measurement operations.

[0010] Secondly, embodiments of this application provide an automatic method for determining defects in a gas turbine, the method comprising: Configure defect judgment rules, which are used to define the mapping relationship between defect characteristics and defect levels; Import the baseline 3D model of the gas turbine component to be inspected and obtain the inspection data of the component; The detection data is registered with the benchmark 3D model; On the registered 3D model, the defect area is identified, and the geometric features of the defect area are extracted; The extracted geometric features are matched with the defect judgment rules to automatically determine the level of the defect.

[0011] In one possible implementation, after matching the extracted geometric features with defect determination rules and automatically determining the level of the defect, the method further includes: Generate a standardized inspection report that includes defect information and a 3D visualization view; store the defect data, geometric features, and judgment results of this inspection for use in historical data statistical analysis.

[0012] In one possible implementation, the extracted geometric features are matched with defect determination rules to automatically determine the level of the defect, including: Feature vectors of geometric features are extracted and input into a hybrid decision system of a deep learning model. The hybrid decision system includes defect determination rules. The hybrid decision system performs matching based on the defect determination rules to determine the defect level.

[0013] Thirdly, embodiments of this application provide an automatic defect determination device for gas turbines, the device comprising: The configuration module is used to configure defect judgment rules, which define the mapping relationship between defect characteristics and defect levels. The import module is used to import the baseline 3D model of the gas turbine component to be inspected and to obtain the inspection data of the component; The registration module is used to register the detection data with the reference 3D model; The determination module is used to identify the defect area on the registered 3D model and extract the geometric features of the defect area; The judgment module is used to match the extracted geometric features with the defect judgment rules and automatically determine the level of the defect.

[0014] Fourthly, embodiments of this application provide a computer storage medium storing multiple instructions adapted for loading by a processor and executing the steps of the above-described method.

[0015] Fifthly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being adapted to be loaded by the processor and to execute the steps of the method described above.

[0016] The beneficial effects of the technical solutions provided in some embodiments of this application include at least the following: by constructing a structured defect rule standard library and integrating 3D models and multimodal detection data, the automation, standardization, and intelligence of defect judgment are realized, overcoming the problems of strong subjectivity, low efficiency, and inconsistent standards in traditional manual inspection; based on high-precision registration, 3D defect annotation and visualization make the location, shape, and size of defects more intuitive and accurate, greatly improving the efficiency of technical communication and decision-making; at the same time, the digital and automated management of the entire process not only realizes the effective accumulation and inheritance of detection knowledge, but also provides strong data-driven decision support for equipment condition assessment, predictive maintenance, and full life cycle management through multi-dimensional statistical analysis and trend prediction of historical data, ultimately significantly improving the safe operation level and operation and maintenance economy of gas turbines. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A system architecture diagram of an automatic gas turbine defect determination system provided in this application embodiment; Figure 2 This is a schematic diagram of the structure of an automatic defect determination system for gas turbines provided in an embodiment of this application; Figure 3 A flowchart illustrating an automatic defect determination method for gas turbines provided in this application embodiment; Figure 4 A logic block diagram of an automatic defect determination method for gas turbines provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] To make the features and advantages of this application more apparent and understandable, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0021] In the description of this application, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances. Furthermore, in the description of this application, unless otherwise stated, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship.

[0022] As mentioned earlier, in the energy, aviation, and shipbuilding industries, gas turbines, as core power equipment, are of paramount importance for their operational safety and reliability. To ensure safety, regular endoscopic inspections of critical hot-end components such as turbine blades and combustion chambers are essential to detect and assess potential defects such as cracks, ablation, and material reduction. Currently, the industry's commonly used inspection method is highly manual: inspectors operate endoscopic equipment to obtain internal two-dimensional images, observe the screen, and, based on personal experience and referring to paper or electronic maintenance manuals, judge and classify the severity of defects. This traditional, manual-dominated model presents a series of inherent technical challenges that urgently need to be addressed.

[0023] In view of this, this application provides an automatic defect judgment system, method, apparatus, and related equipment for gas turbines. The aim of this application is to fundamentally eliminate subjectivity in the judgment process by introducing a structured defect rule standard library, transforming expert experience and industry standards into calculable objective rules; by integrating high-precision registration of 3D models and inspection data, automated feature extraction, and intelligent judgment, the inspection process is automated and intelligent, significantly improving efficiency and accuracy; by generating standardized reports containing interactive 3D views, the intuitiveness and communication efficiency of defect information are greatly improved; finally, by constructing a centralized defect data center and performing multi-dimensional statistical analysis, data assets are transformed into powerful tools supporting predictive maintenance and scientific decision-making, thereby achieving intelligent and automatic judgment of gas turbine defect inspection data.

[0024] Please see Figure 1 , Figure 1 This is an exemplary system architecture diagram of an automatic defect determination system for gas turbines provided in an embodiment of this application.

[0025] like Figure 1 As shown, the system architecture may include terminals, a network, and servers. The network serves as the medium for providing communication links between terminals and servers. The network may include various types of wired or wireless communication links, such as wired communication links including fiber optic cables, twisted-pair cables, or coaxial cables, and wireless communication links including Bluetooth communication links, Wireless-Fidelity (Wi-Fi) communication links, or microwave communication links, etc.

[0026] A terminal can interact with a server via a network to receive messages from or send messages to the server, or it can interact with the server via a network to receive messages or data sent to the server by other users. A terminal can be hardware or software. When the terminal is hardware, it can be various electronic devices, including but not limited to smartwatches, smartphones, tablets, laptops, and desktop computers. When the terminal is software, it can be installed in the aforementioned electronic devices and can be implemented as multiple software programs or software modules (e.g., to provide distributed services) or as a single software program or software module; no specific limitations are made here.

[0027] In this embodiment, the terminal can configure defect judgment rules, which are used to define the mapping relationship between defect features and defect levels; import the reference three-dimensional model of the gas turbine component to be inspected and obtain the inspection data of the component; register the inspection data with the reference three-dimensional model; determine the defect area on the registered three-dimensional model and extract the geometric features of the defect area; match the extracted geometric features with the defect judgment rules and automatically determine the level of the defect.

[0028] A server can be a business server that provides various services. It should be noted that a server can be either hardware or software. When a server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When a server is software, it can be implemented as multiple software programs or software modules (e.g., used to provide distributed services), or as a single software program or software module; no specific limitations are made here.

[0029] Alternatively, the system architecture may not include a server. In other words, the server may be an optional device in the embodiments of this specification. That is, the methods provided in the embodiments of this specification can be applied to a system architecture that only includes terminals. The embodiments of this application do not limit this.

[0030] It should be understood that Figure 1 The number of terminals, networks, and servers shown is only illustrative; the number can be any number of terminals, networks, and servers depending on the implementation requirements.

[0031] Please see Figure 2 , Figure 2 This is a schematic diagram of an automatic defect detection system for gas turbines provided in an embodiment of this application. Figure 2 As shown, the automatic defect judgment system 200 for gas turbines includes: a defect rule standard library management module 210, a 3D model import module 220, a data acquisition module 230, a defect drawing and judgment module 240, a report generation module 250, and a data statistics module 260. Among them: The defect rule standard library management module 210 stores rules for determining defect levels based on defect features; the 3D model import module 220 imports and manages the reference 3D model of the gas turbine components; the data acquisition module 230 acquires the inspection data of the gas turbine components; the defect drawing and judgment module 240, which is communicatively connected to the 3D model import module 220, the data acquisition module 230, and the defect rule standard library management module 210, is used to: register the inspection data with the reference 3D model; draw the defect area based on the registered 3D model; and automatically determine the defect level by calling the rules in the defect rule standard library management module 210 based on the inspection data and / or the defect area; the report generation module 250, which is communicatively connected to the defect drawing and judgment module 240, is used to generate an inspection report based on the defect level; and the data statistics module 260, which is communicatively connected to the defect drawing and judgment module 240, is used to store defect data and provide statistical analysis functions.

[0032] The defect rule standard library management module 210 supports the categorization and management of defect rules in tabular form. Categorization criteria include, but are not limited to, component type, location, defect type, and corresponding defect judgment thresholds for each defect type. It supports rule version control, records the creation, modification, and activation history of rules, and supports batch importing rules from external files or exporting existing rules. The data acquisition module 230 collects detection data including the operating condition parameters of the gas turbine, which at least include operating hours, number of starts, load rate, and fuel type. The rules in the defect rule standard library are associated with these operating condition parameters, ensuring that defect judgment results reflect the actual service status of the component. The data statistics module 260 provides multi-dimensional statistical analysis functions, including at least: statistical analysis of the number and frequency of defects by time, component model, defect type, and defect level (qualified / unqualified); analysis of the correlation between defect incidence and operating condition parameters; and prediction of future defect trends.

[0033] In one possible implementation, the defect judgment rules stored in the defect rule standard library management module 210 are associated with the operating condition parameters of the gas turbine components; the detection data acquired by the data acquisition module 220 includes the operating condition parameters; and the defect drawing and judgment module 240 is also used to call the rules associated with the current operating condition parameters when judging the defect level.

[0034] In one possible implementation, based on detection data and / or defect areas, rules in the defect rule standard library management module 210 are invoked to automatically determine the defect level, including: Based on the detection data and / or defect area, geometric features are extracted; geometric features include area features and depth features; the geometric features are matched with the rules in the defect rule standard library management module 210; if a matching rule is found, the corresponding defect level is output according to the rule.

[0035] In one possible embodiment, the detection report generated by the report generation module 250 is an interactive document, whose embedded 3D visualization view supports users in performing rotation, zoom, sectioning, and measurement operations.

[0036] This application provides an automatic defect judgment system for gas turbines. This system, through an integrated modular architecture, organically integrates functions such as defect rule standard library management, 3D model import, data acquisition, defect drawing and judgment, report generation, and data statistics, achieving full-process automation from data acquisition to intelligent judgment. Specifically, by dynamically linking defect judgment rules with operating condition parameters, the system ensures that the judgment results accurately reflect the actual service status of the components, improving the adaptability and accuracy of the judgment. Furthermore, by automatically extracting defect geometric features and intelligently matching them with the rule library, the system achieves objective and rapid judgment of defect levels, effectively eliminating subjective human bias. Finally, the interactive 3D inspection report generated by the system greatly enhances the visualization and operability of the results, thereby significantly improving overall inspection efficiency, standardization, and decision support capabilities.

[0037] Please see Figure 3 , Figure 3 This is a flowchart illustrating an automatic gas turbine defect determination method provided in an embodiment of this application. The execution entity in this embodiment can be an electronic device that performs automatic gas turbine defect determination, a processor within the electronic device performing the automatic gas turbine defect determination method, or an automatic gas turbine defect determination service within the electronic device performing the automatic gas turbine defect determination method. For ease of description, the following uses a processor within an electronic device as an example to describe the specific execution process of the automatic gas turbine defect determination method.

[0038] like Figure 3As shown, the automatic defect determination method for gas turbines may include at least: S301. Configure defect judgment rules. Defect judgment rules are used to define the mapping relationship between defect characteristics and defect levels.

[0039] Specifically, before the inspection task begins, the system first executes the defect judgment rule configuration step. The core purpose of this step is to transform the vague qualitative judgment standards that rely on human experience into a structured and quantitative rule system that can be recognized and executed by a computer, thereby providing an objective and unified basis for subsequent automatic judgment. Specifically, the system provides users with a graphical interface through its defect rule standard library management module, enabling users to create and edit judgment rules in a structured manner based on industry specifications, manufacturer standards, or their own expert experience. Each rule clearly defines the pass / fail judgment threshold corresponding to the geometric characteristics (such as length, area, and depth) of a certain type of defect (such as cracks or material reduction) at a specific location on a specific component, thus clearly establishing a precise mapping relationship from "defect characteristics" to "defect levels." To cope with the complexity of actual service conditions, the rule configuration process also supports incorporating operating condition parameters, such as operating hours, number of starts, or load rate, as correlation conditions into the rule definition. This allows the system to dynamically select applicable judgment standards based on the actual usage of the component, achieving a more accurate condition assessment. For example, rules are stored in a structured "IF-THEN" format. For instance, a rule for "first-stage moving blades" could be defined as: If component = 'first stage moving blade' AND defect morphology = 'reduced material' AND location = 'blade tip' AND defect tolerance <= 1.0mm THEN Defect Grade (Pass / Fail) = 'Pass' The defect rule standard library management module provides a graphical interface, allowing administrators to add, delete, modify, and query rules. Rules can be managed in a hierarchical structure, such as "component-defect type-maintenance standard." Meanwhile, the 3D model import module supports version control; any modifications to rules are recorded, ensuring the traceability of the standards. The rule library can be exported to a universal format for easy sharing and backup across different systems.

[0040] S302. Import the reference 3D model of the gas turbine component to be tested and obtain the test data of the component.

[0041] Specifically, the system first imports a 3D model file from a pre-built model library, retrieving a baseline 3D model file that precisely corresponds to the workpiece identification information in the current inspection order and loading it into the visualization interface. This model serves as a non-destructive digital prototype, providing a precise coordinate system and geometric reference for the spatial location of defects. Simultaneously, the system starts synchronously through a data acquisition module, acquiring multimodal inspection data reflecting the true surface condition of the component from connected 3D scanners and other inspection hardware. This data includes not only high-precision point cloud data for generating the 3D topography but also operating condition parameters retrieved from the unit control system or database, such as cumulative operating hours, number of start-up cycles, and average load rate. Please refer to [link to relevant documentation]. Figure 4 , Figure 4 This is a logic block diagram of an automatic defect determination method for gas turbines provided in an embodiment of this application. Figure 4 As shown: After configuring the defect judgment rules, the baseline 3D model and inspection order are imported, and then multi-source data acquisition is performed. It should be noted that the 3D model import module supports loading mainstream 3D model file formats (such as STL). Before the inspection task begins, the operator first imports the baseline 3D model of the component to be inspected (e.g., the first-stage moving blade). After receiving the baseline 3D model of the component to be inspected, the system provides a user-friendly 3D interactive interface, supporting model translation, rotation, scaling, dynamic sectioning, and defect drawing. The operator can pre-annotate key areas on the model to provide a reference for subsequent defect localization.

[0042] S303. Register the detection data with the reference 3D model.

[0043] For details, please continue reading Figure 4 ,like Figure 4As shown, after importing the baseline 3D model and acquiring the inspection data, the system immediately performs a registration step between the inspection data and the baseline 3D model. The core purpose of this step is to establish a precise spatial correspondence between the actual geometric shape of the component reflected in the inspection data and the ideal design shape represented by the baseline 3D model, thus laying the foundation for subsequent precise defect location and quantitative analysis under a unified coordinate system. In specific implementation, the system first preprocesses the acquired 3D point cloud and other inspection data to eliminate noise using a built-in registration algorithm. Then, based on algorithms such as the iterative nearest point algorithm or its improved versions, it automatically extracts and matches common geometric features (such as edges, holes, surface curvature, etc.) on the inspection data and the baseline 3D model to calculate an optimal spatial coordinate transformation parameter. The system continuously optimizes this transformation parameter through iterative calculation, ultimately accurately superimposing the inspection data onto the corresponding position of the baseline 3D model, achieving alignment between the two in 3D space. During this process, if the inspection data contains associated operating condition parameters, the system will store them as contextual metadata of the registration result, ensuring that complete registration background information can be obtained in the subsequent judgment stage.

[0044] S304. On the registered 3D model, determine the defect area and extract the geometric features of the defect area.

[0045] For details, please continue reading Figure 4 ,like Figure 4 As shown, after data and model registration is completed, the region can be determined, and the defect outline can be drawn and its geometric features extracted. First, the operator can manually outline the two-dimensional or three-dimensional contour of the defect in the registered model view that integrates point cloud data using interactive drawing tools, or the system can automatically identify abnormal areas using edge detection algorithms. Then, based on a pre-established coordinate system and scale, the system calculates and displays the actual dimensions of the contour in three-dimensional space in real time, such as length, area, depth, and volume. After the defect contour is determined, the system automatically calculates the geometric features of the marked region, including but not limited to key quantitative indicators such as the defect's projected area, maximum depth, volume, and location coordinates. During this process, the system simultaneously records the operating condition parameters related to the defect as a judgment context, ensuring that the extracted feature data has a complete operating condition background.

[0046] S305. Match the extracted geometric features with the defect judgment rules to automatically determine the level of the defect.

[0047] Specifically, the system first retrieves judgment rules applicable to the current component type and defect morphology from the defect rule standard library. These rules, in a structured form, clearly define the judgment thresholds for different defect types at different locations and under different operating conditions. Then, the system automatically matches the extracted geometric features, such as defect area and depth, with the threshold conditions in the rule library. When the feature value exceeds the preset threshold, it is automatically judged as unqualified; otherwise, it is qualified. During this matching process, the system simultaneously considers the current component's operating condition parameters and dynamically selects the judgment rules applicable to that condition, ensuring that the evaluation results match the actual service status of the component.

[0048] Optionally, for complex or ambiguous defect morphologies, the system can also enable a hybrid decision-making mechanism that integrates deep learning models to assist in the determination of the defect level by analyzing deep patterns in feature vectors.

[0049] In one possible implementation, after automatically determining all defects, the system can automatically generate a standardized inspection report and store the inspection data. The system first calls the report generation module, which automatically integrates key information from the inspection based on a preset template. This includes, but is not limited to, order details, component information, a list of all defects and their corresponding geometric features and determination levels, and embeds a 3D model view with defect annotations into the report. The generated inspection report uses an interactive document format, allowing users to rotate, scale, and section the defect model in the report using a built-in lightweight 3D engine, thus intuitively examining defect details. Subsequently, the system transmits all data from this inspection task, including but not limited to raw inspection data, extracted geometric features, automatically determined level results, and associated operating parameters, to the data statistics module for structured storage and archiving. This systematic storage process lays the data foundation for subsequent multi-dimensional statistical analysis, enabling managers to query historical data and analyze defect trends by time, component model, defect type, and other dimensions. It also provides data-driven decision support for the formulation of predictive maintenance strategies, ultimately achieving full lifecycle management and value mining of inspection data.

[0050] This application provides an automatic defect determination method for gas turbines. By constructing a complete automated process from rule configuration and data registration to intelligent determination, it achieves standardization and intelligence across the entire detection process. Specifically, by accurately matching extracted geometric features with a structured rule base, the objectivity and consistency of defect level determination are ensured, effectively overcoming the problem of reliance on human experience. Furthermore, the method automatically generates a standardized report containing an interactive 3D view after determination, greatly improving the intuitiveness and communication efficiency of the detection results. Simultaneously, by systematically storing and managing detection data, geometric features, and determination results, a solid foundation is laid for multi-dimensional statistical analysis and trend prediction based on historical data, thereby improving detection efficiency while providing reliable data support for equipment condition assessment and predictive maintenance. In addition, the method further enhances the intelligence and adaptability of determination in complex defect scenarios by introducing a hybrid decision-making mechanism that integrates a rule engine and a deep learning model.

[0051] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device 500 may include: at least one processor 501, at least one network interface 504, user interface 503, memory 505, and at least one communication bus 502.

[0052] The communication bus 502 is used to enable communication between these components.

[0053] The user interface 503 may include a display screen and a camera. Optional user interfaces 503 may include standard wired interfaces and wireless interfaces.

[0054] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0055] The processor 501 may include one or more processing cores. The processor 501 connects to various parts within the electronic device 500 using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 505, and by calling data stored in the memory 505. Optionally, the processor 501 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 501 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 501 and may be implemented as a separate chip.

[0056] The memory 505 may include random access memory (RAM) or read-only memory. Optionally, the memory 505 may include a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 505 may also be at least one storage device located remotely from the aforementioned processor 501. Figure 5 As shown, the memory 505, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an automatic gas turbine defect detection application.

[0057] exist Figure 5In the illustrated electronic device 500, the user interface 503 is mainly used to provide an input interface for the user and to acquire user input data; while the processor 501 can be used to call the gas turbine defect automatic judgment application stored in the memory 505 and specifically perform the following operations: Configure defect judgment rules, which define the mapping relationship between defect features and defect levels; import the reference 3D model of the gas turbine component to be inspected and obtain the component's inspection data; register the inspection data with the reference 3D model; on the registered 3D model, determine the defect area and extract the geometric features of the defect area; match the extracted geometric features with the defect judgment rules to automatically determine the defect level.

[0058] In some possible embodiments, after the processor 501 performs the task of matching the extracted geometric features with defect determination rules and automatically determining the level of the defect, it is also used to perform: Generate a standardized inspection report that includes defect information and a 3D visualization view; store the defect data, geometric features, and judgment results of this inspection for use in historical data statistical analysis.

[0059] In some possible embodiments, when processor 501 performs the action of matching the extracted geometric features with defect determination rules to automatically determine the level of the defect, it specifically performs the following: Feature vectors of geometric features are extracted and input into a hybrid decision system of a deep learning model. The hybrid decision system includes defect determination rules. The hybrid decision system performs matching based on the defect determination rules to determine the defect level.

[0060] This application also provides a computer-readable storage medium storing instructions that, when executed on a computer or processor, cause the computer or processor to perform the above-described instructions. Figure 2 One or more steps in the illustrated embodiment. If the constituent modules of the above-described automatic gas turbine defect detection device are implemented as software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0061] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital versatile discs (DVDs)), or semiconductor media (e.g., solid state disks (SSDs)).

[0062] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The aforementioned storage medium includes various media capable of storing program code, such as Read Only Memory (ROM), Random Access Memory (RAM), magnetic disks, or optical disks. Unless otherwise specified, the technical features of this embodiment and its implementation schemes can be combined arbitrarily.

[0063] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. It will be apparent to those skilled in the art that various modifications can be made to these embodiments, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An automatic defect detection system for gas turbines, characterized in that, The system includes: The defect rule standard library management module is used to store rules for determining defect levels based on defect characteristics; The 3D model import module is used to import and manage the baseline 3D models of gas turbine components; The data acquisition module is used to acquire the test data of the gas turbine components; The defect drawing and judgment module is communicatively connected to the 3D model import module, data acquisition module, and defect rule standard library management module, and is used for: The detection data is registered with the reference 3D model; Based on the registered 3D model, the defect area is drawn; Based on the detection data and / or the defect area, rules in the defect rule standard library management module are invoked to automatically determine the defect level; the step of automatically determining the defect level based on the detection data and / or the defect area includes: Based on the detection data and / or the defect region, geometric features are extracted; the geometric features include area features and depth features. The geometric features are matched with the rules in the defect rule standard library management module; If a matching rule is found, the corresponding defect level is output according to the rule. The report generation module is communicatively connected to the defect drawing and judgment module and is used to generate a detection report based on the defect level. The data statistics module, which is communicatively connected to the defect drawing and judgment module, is used to store defect data and provide statistical analysis functions; the defect judgment rules stored in the defect rule standard library management module are associated with the operating condition parameters of the gas turbine components; the detection data acquired by the data acquisition module includes the operating condition parameters; when judging the defect level, the defect drawing and judgment module is also used to call the rules associated with the current operating condition parameters.

2. The system as described in claim 1, characterized in that, The report generation module generates an interactive document, and its embedded 3D visualization view supports users in performing rotation, zoom, sectioning, and measurement operations.

3. The system as described in claim 1, characterized in that, The defect rule standard library management module supports the classification and management of defect rules in tabular form, supports rule version control, records the creation, modification and activation history of rules, and supports batch import of rules from external files or export of existing rules.

4. An automatic defect determination method for gas turbines, characterized in that, Applied to the system according to any one of claims 1-3, the method comprises: Configure defect judgment rules, which are used to define the mapping relationship between defect characteristics and defect levels; Import the reference 3D model of the gas turbine component to be inspected, and obtain the inspection data of the component; The detection data is registered with the reference 3D model; On the registered 3D model, the defect area is identified, and the geometric features of the defect area are extracted; The extracted geometric features are matched with the defect determination rules to automatically determine the level of the defect.

5. The method as described in claim 4, characterized in that, After automatically determining the level of the defect by matching the extracted geometric features with the defect determination rules, the method further includes: Generate standardized inspection reports that include defect information and 3D visualization views; The defect data, geometric features, and judgment results of this detection are stored for use in historical data statistical analysis.

6. The method as described in claim 4, characterized in that, The step of matching the extracted geometric features with the defect determination rules to automatically determine the level of the defect includes: Extract the feature vector of the geometric features, and input the feature vector into a hybrid decision system of a deep learning model, wherein the hybrid decision system includes the defect determination rule; The hybrid decision-making system matches defects based on the defect determination rules to determine the defect level.

7. A computer storage medium, characterized in that, The computer storage medium stores a plurality of instructions adapted for loading by a processor and executing the steps of the method as described in any one of claims 4 to 6.

8. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method as described in any one of claims 4 to 6.

Images