A method and system for intelligent evaluation of residual life of metal components in a thermal power plant

Abstract

This invention discloses an intelligent assessment method and system for the residual life of metal components in thermal power plants, belonging to the field of thermal power plant equipment inspection technology. The method includes: acquiring the current creep, hardness, and metallographic aging level of the metal component to be assessed; calculating three corresponding predicted residual life values ​​based on pre-established independent functional relationships between each parameter and service duration; and selecting the minimum value among the three predicted values ​​as the final residual life based on the "weakest link" principle. This invention employs a parameter decoupling modeling approach, avoiding the parameter coupling problem caused by traditional joint fitting. By independently analyzing the evolution of different damage mechanisms and combining the weakest link control principle, the accuracy and safety of life assessment are significantly improved. Simultaneously, multi-parameter synchronous acquisition is achieved through integrated testing equipment, improving testing efficiency. This method provides a scientific basis for the preventive maintenance and replacement of critical metal components in thermal power plants.

Description

Technical Field

[0001] This invention C relates to the field of thermal power plant equipment testing and life assessment technology, specifically to an intelligent assessment method and system for the residual life of metal components in thermal power plants. Background Technology

[0002] Metal components of thermal power plants, such as boiler superheater tubes, reheater tubes, main steam pipes, turbine rotors, and blades, are subjected to harsh operating environments of high temperature, high pressure, and high stress for extended periods. During long-term operation, they undergo creep deformation, hardness degradation, and metallographic aging, resulting in damage that gradually reduces their mechanical properties and reliability. When this damage accumulates to a certain extent, it can lead to component failure, causing serious safety accidents such as leaks and explosions. This not only affects the safe and stable operation of the thermal power plant but can also cause significant economic losses and casualties. Therefore, accurately assessing the remaining lifespan of metal components in thermal power plants, promptly identifying potential safety hazards, and developing reasonable maintenance and replacement strategies are crucial for ensuring the safe, economical, and stable operation of thermal power plants.

[0003] Currently, methods for assessing the residual life of metal components in thermal power plants mainly fall into two categories: single-parameter assessment methods and multi-parameter combined assessment methods. Single-parameter assessment methods primarily rely on one parameter—creep strain, hardness value, or metallographic aging level—to predict the component's residual life through empirical formulas or simple fitting. For example, some methods predict residual life solely by measuring the creep strain of the metal component and combining it with creep curve extrapolation; others assess life by measuring hardness values ​​and utilizing the empirical relationship between hardness and service time. The advantages of these methods are their simplicity and low testing cost, but they have significant drawbacks: damage to metal components in thermal power plants is the result of multiple factors, including creep, hardness degradation, and metallographic aging. A single parameter cannot comprehensively reflect the actual damage state of the component, leading to low assessment accuracy and an inability to accurately predict the component's failure time.

[0004] Multi-parameter joint assessment methods combine multiple damage parameters for service life evaluation. This typically involves jointly fitting multiple parameters, such as creep, hardness, and metallographic aging level, to construct a comprehensive fitting function, which is then used to predict residual life. However, this approach has two key drawbacks: First, the influence of creep, hardness, and metallographic aging level on service life differs, and their rates of change and damage mechanisms vary significantly. Jointly fitting these parameters masks their independent effects, leading to decreased accuracy of the fitting function and consequently affecting the accuracy of residual life assessment. Second, existing multi-parameter assessment methods do not adequately consider the weakest points of metal components. The actual lifespan of a component is determined by the parameter with the most severe damage, much like the capacity of a barrel is determined by its shortest stave. Lifespan values ​​obtained solely through a comprehensive fitting function cannot reflect the true weaknesses of the component, potentially leading to overly optimistic assessments and posing safety risks. Summary of the Invention

[0005] To address the problems of high parameter coupling and insufficient fitting accuracy in the existing technology for assessing the residual life of metal components in thermal power plants, this invention proposes an intelligent assessment method and system for the residual life of metal components in thermal power plants. This method constructs a parameter decoupling model to achieve independent analysis and accurate fitting of each key influencing factor, significantly improving the assessment accuracy and providing a scientific basis for the life management of key metal components in thermal power plants.

[0006] This invention is achieved through the following technical solution: A method for intelligent assessment of the residual life of metal components in thermal power plants includes the following steps: Obtain the current creep, hardness value, and metallographic aging level of the metal component to be evaluated; Based on the pre-established functional relationships between creep, hardness, and metallographic aging level and service time, the predicted residual life values ​​corresponding to creep, hardness, and metallographic aging level are calculated. Based on the barrel effect, the minimum value among the creep, hardness, and metallographic aging levels corresponding to the predicted residual life is determined as the residual life of the metal component.

[0007] Preferably, the method for establishing the functional relationship is as follows: Acquire historical data of metal components under service conditions, including service time, as well as creep, hardness values ​​and metallographic aging levels that change with service time; A fitting algorithm was used to fit the creep variable, hardness value, and metallographic aging level to their respective service time, so as to obtain the functional relationship between creep variable and service time, the functional relationship between hardness value and service time, and the functional relationship between metallographic aging level and service time.

[0008] Preferably, the fitting algorithm includes a nonlinear fitting algorithm or a machine learning fitting algorithm; The fitting algorithm includes one or more combinations of least squares method, support vector regression algorithm, and random forest algorithm.

[0009] Preferably, obtaining the current creep, hardness value, and metallographic aging level of the metal component to be evaluated includes: In-situ non-destructive testing was performed on the metal parts to be evaluated using an integrated creep-hardness-metallography testing equipment, which simultaneously acquired creep, hardness values ​​and metallographic images. The metallographic structure image is processed by image recognition based on pre-stored standard metallographic structure feature data to determine the aging level of the metallographic structure.

[0010] Preferably, the method for obtaining the current metallographic aging level of the metal component to be evaluated is as follows: Image recognition algorithms are used to extract characteristic parameters of metallographic structures, such as grain size, grain boundary features, and distribution of second-phase particles. These parameters are then compared with the characteristic parameters of standard metallographic images in the database to determine the aging level of the metallographic structure.

[0011] Preferably, the method for calculating the creep deformation, hardness value, and metallographic aging level corresponding to the predicted residual life is as follows: Substitute the current creep value, current hardness value, and current metallographic aging level into the corresponding function relationship, and solve inversely to obtain the current service time corresponding to the current creep value, current hardness value, and current metallographic aging level, respectively. Substitute the predefined creep failure threshold, hardness failure threshold, and metallographic aging level failure threshold into the corresponding function relationships, and solve them inversely to obtain the limit service time corresponding to the creep, hardness, and metallographic aging levels. The predicted residual life values ​​corresponding to the current creep, current hardness, and current metallographic aging level are determined based on the maximum service life and the current service life.

[0012] Preferably, after acquiring the current mechanical data and before calculating the predicted residual life, the method further includes a step of verifying the current creep, hardness value, and metallographic aging level, wherein the verification includes: The current creep and hardness values ​​are compared with the pre-stored historical data ranges of creep and hardness values ​​for the same material and service temperature. If the current value exceeds the corresponding historical data range, a re-detection is triggered. The acquired metallographic image is compared with the pre-stored standard metallographic image for image feature similarity. If the similarity is lower than the preset threshold, the metallographic image acquisition is judged to be abnormal and re-acquisition is triggered.

[0013] A smart assessment system for the residual life of metal components in thermal power plants includes: The data acquisition module is used to obtain the current creep, hardness value, and metallographic aging level of the metal part to be evaluated. The prediction module is used to calculate the predicted residual life values ​​corresponding to creep, hardness, and metallographic aging levels based on the pre-established functional relationships between creep, hardness, and metallographic aging levels and service time. The residual life module is used to determine the minimum value among the predicted residual life values ​​corresponding to the creep, hardness, and metallographic aging levels as the residual life of the metal component, based on the barrel effect.

[0014] An electronic device, comprising: Memory, used to store computer programs; A processor is used to implement the steps of the intelligent assessment method for the residual life of metal components in thermal power plants when executing the computer program.

[0015] A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the intelligent assessment method for the residual life of metal components in thermal power plants.

[0016] Compared with the prior art, the present invention has the following beneficial technical effects: This application provides an intelligent assessment method for the residual life of metal components in thermal power plants. First, it independently collects three key damage parameters: creep deformation, hardness value, and metallographic aging level. These parameters characterize different failure mechanisms of the material: creep deformation, hardness degradation, and microstructure evolution. Based on three pre-established independent functional relationships, the residual life corresponding to each parameter is calculated separately. This multi-channel modeling approach avoids the parameter coupling problem caused by traditional multi-parameter joint fitting and can more realistically reflect the development law of each damage mechanism. Finally, the minimum value is selected as the final life using the "weakest link" principle, derived from the weakest link control theory in engineering safety design, i.e., the system reliability is determined by the weakest link. This method improves model accuracy and physical interpretability through parameter decoupling; achieves spatiotemporal consistency of data acquisition through integrated detection; and ensures the conservatism and safety of the assessment results through minimum value selection, providing a reliable basis for preventive maintenance. It effectively solves the problems of insufficient assessment accuracy and low safety margin in existing methods, achieving accurate assessment of the life of metal components through decoupling analysis.

[0017] This application also proposes an intelligent assessment system for the residual life of metal components in thermal power plants, an electronic device, and a computer storage medium, which possess all the advantages of the aforementioned intelligent assessment method for the residual life of metal components in thermal power plants. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of the intelligent assessment method for the residual life of metal components in thermal power plants according to the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of 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. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0021] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] A method for intelligent assessment of the residual life of metal components in thermal power plants, comprising: Step 1: Obtain the current creep, hardness value, and metallographic aging level of the metal part to be evaluated.

[0023] This method acquires creep, hardness, and metallographic images of the metal component to be evaluated (and then converts them into aging levels), enabling in-situ, non-destructive acquisition of multi-source damage data. This approach eliminates the temporal and spatial inconsistencies inherent in traditional step-by-step testing, ensuring data comparability in both time and space. The acquisition method is highly efficient, easy to operate, and avoids component damage or data shifts caused by multiple tests.

[0024] Step 2: Based on the pre-established functional relationships between creep, hardness, and metallographic aging levels and service duration, calculate the predicted residual life values ​​corresponding to creep, hardness, and metallographic aging levels.

[0025] Based on three pre-established independent functional relationships—creep-duration, hardness-duration, and metallographic aging level-duration—measured data are substituted into the corresponding models to calculate three independent residual lifetime predictions. This prediction method acknowledges that different damage mechanisms (creep, hardening / softening, and tissue evolution) have independent evolutionary patterns and failure thresholds. It employs a strategy of independent modeling for each parameter and independent prediction for each channel, avoiding the obscuring of the true degradation trajectory due to variable coupling during multi-parameter joint fitting. This prediction method significantly improves the physical interpretability and model accuracy of lifetime prediction. Each prediction value strictly corresponds to the failure process of a specific damage mechanism, overcoming the systematic defect of existing comprehensive assessment methods that lead to overly optimistic prediction results due to parameter averaging.

[0026] Step 3: Based on the barrel effect, the minimum value among the creep, hardness, and metallographic aging level-corresponding residual life prediction values ​​is determined as the residual life of the metal component.

[0027] Based on the barrel effect principle, the minimum value among the three independent life predictions is selected as the final residual life. The overall failure of a metal component is dominated by the parameter with the most severe damage; that is, the weakest link in life determines the actual remaining safe service time of the component. This method fundamentally ensures the conservatism and safety of the assessment results, avoiding the risk that a good state of one parameter may mask the fact that other parameters have reached a critical state. It makes life decisions more aligned with actual engineering safety requirements, provides clear and reliable time points for preventive maintenance and replacement, and significantly reduces the probability of sudden failure accidents caused by overly optimistic assessments.

[0028] Example 1 See Figure 1 A method for intelligent assessment of the residual life of metal components in thermal power plants, comprising the following steps: Step 1: Based on the historical data of the metal components under service conditions, including service time and mechanical data that change with service time, construct an independent functional relationship between each mechanical data and service time. This mechanical data includes the creep, hardness, and metallographic aging level of the metal component.

[0029] Specifically, it includes the following processes: S1.1 Obtain historical data of metal components under service conditions.

[0030] For example, for metal components in thermal power plants (such as boiler superheater tubes and main steam pipes), under the same service materials and service temperatures, three independent historical data sample sets are extracted from a pre-established metallographic structure and hardness database, namely: Creep variables and their corresponding service duration datasets; Dataset of hardness values ​​and corresponding service durations; Data set of metallographic aging levels and corresponding service durations.

[0031] S1.2. Use a fitting algorithm to fit each dataset to obtain the corresponding functional relationship.

[0032] Using either nonlinear fitting algorithms or machine learning fitting algorithms, we independently model the three datasets mentioned above, establishing three independent functional relationships: 1) Using historical data on creep variables of metal components at the same service temperature and corresponding service duration as samples, a fitting algorithm is used to fit the data to obtain the functional relationship between creep variables and service duration. Creep variables increase with the increase of service duration, and the two have a non-linear relationship. The fitting function adopts a non-linear fitting form (such as an exponential function or a power function), which can accurately reflect the change law of creep variables with service duration.

[0033] The functional relationship between creep variables and service duration is y1 = f1(t); Where y1 is a creep variable and t is the service duration.

[0034] 2) Using hardness values ​​and historical hardness data of metal parts of the same material and service temperature in the database, along with their corresponding service durations, as samples, a fitting algorithm is used to fit the data and obtain the functional relationship between hardness values ​​and service duration. Hardness values ​​decrease with increasing service duration, and the two exhibit a non-linear relationship. The fitting function adopts a non-linear fitting form, which can accurately reflect the change law of hardness degradation with service duration.

[0035] The functional relationship between hardness value and service life is y2= f2(t); Where y2 is the hardness value and t is the service life.

[0036] 3) Image recognition is performed on metallographic images to determine their metallographic aging level. Using historical data on the metallographic aging levels of metal parts of the same material and service temperature and their corresponding service durations as samples from the database, a fitting algorithm is used to obtain the functional relationship between the metallographic aging level and service duration. The metallographic aging level increases with the increase of service duration, and the two are in a non-linear relationship. The fitting function adopts a non-linear fitting form, which can accurately reflect the change law of metallographic aging with service duration.

[0037] The functional relationship between metallographic aging level and service life is y3= f3(t); Where y3 represents the metallographic aging level and t represents the service life.

[0038] Furthermore, the method for determining the aging level of the metallographic structure is as follows: the metallographic structure photograph is processed by an image recognition algorithm to extract characteristic parameters such as grain size, grain boundary features, and distribution of second phase particles of the metallographic structure, and these parameters are compared with the characteristic parameters of the standard metallographic structure images in the database to determine the corresponding aging level of the metallographic structure. The aging level is divided into 1 to 5 levels. Grade 1 is unaged (no obvious changes in metallographic structure, uniform grains, and clear grain boundaries). Level 2 indicates slight aging (grains have grown slightly, and grain boundaries are slightly blurred). Level 3 is moderate aging (grains grow significantly, grain boundaries become blurred, and a small amount of second-phase particles aggregate). Level 4 indicates severe aging (significant grain growth, severely blurred grain boundaries, and a large amount of second-phase particles aggregated). Level 5 indicates severe aging (reaching the failure threshold, with coarse and uneven grains, grain boundary cracking, severe aggregation of second-phase particles, and a significant decrease in the mechanical properties of the component, approaching failure).

[0039] Furthermore, the fitting algorithm is a nonlinear fitting algorithm or a machine learning fitting algorithm, including one or more combinations of least squares method, support vector regression algorithm, and random forest algorithm; when constructing the fitting function, mean squared error (MSE) and coefficient of determination (R²) are used. 2 The fitting effect is verified to ensure that the fitting accuracy of the fitted function meets the evaluation requirements, where the coefficient of determination (R²) is used. 2 ≥0.95, mean squared error (MSE) ≤10 -4 This ensures that the fitting function can accurately reflect the relationship between the changes in each parameter and the service life.

[0040] Furthermore, the metallographic structure and hardness database is a pre-established standardized database specifically adapted to the evaluation needs of metal components in thermal power plants, and its construction process includes: The system collects material information (such as austenitic stainless steel, martensitic stainless steel, carbon steel, alloy steel, etc.), service temperature information, service duration information, standard metallographic images, standard hardness values ​​and corresponding characteristic parameters, and failure data (such as ultimate creep, ultimate hardness value, and ultimate metallographic aging level) of commonly used metal components in thermal power plants (including boiler superheater tubes, reheater tubes, main steam pipes, turbine rotors, and turbine blades). After filtering, removing outliers, and standardizing the collected data (such as unifying the hardness value unit, unifying the metallographic image resolution, and unifying the service duration calculation standard), the data is classified and stored according to material and service temperature to form a standardized database.

[0041] This database supports data import, query, matching, and updating. It can continuously optimize the integrity and accuracy of the database based on new detection and failure data, providing sufficient and reliable sample support for the construction of fitting functions and the prediction of residual lifetime.

[0042] It should be noted that no joint fitting is performed between the various functional relationships to ensure that the variation law of each parameter is expressed independently and to avoid mutual interference between parameters.

[0043] Step 2: Obtain the current mechanical data of the metal component to be evaluated, and verify the current mechanical data against the collected data based on the historical database.

[0044] Specifically, it includes the following processes: S2.1. Using an integrated creep-hardness-metallography device, in-situ synchronous testing is performed on the metal components of the thermal power plant to be evaluated to obtain the mechanical data of the metal components (creep, hardness values, and metallographic images).

[0045] S2.2 Verify the mechanical data of the metal component to be evaluated; The collected hardness values ​​and metallographic images are imported into a metallographic structure and hardness database. The database's built-in comparison algorithm is used to match the current data with historical standard data of the same material and service temperature. The data is then verified based on the matching results.

[0046] For hardness values, the database has a built-in comparison algorithm that compares them with the standard hardness values ​​of the same material and service temperature, calculates the error value, and verifies the hardness value based on the error value. If the error value is greater than the set range, it means that the measurement value is inaccurate and the measurement is repeated. The verification eliminates the influence of equipment detection errors and environmental interference.

[0047] For creep variables, the database's built-in comparison algorithm compares them with standard creep variable values ​​of the same material and service temperature, calculates the error value, and if the error value is greater than the set range, it indicates that the measurement value is inaccurate and the measurement is repeated. The verification eliminates the influence of equipment detection errors and environmental interference.

[0048] For metallographic images, the database's built-in image recognition algorithm preprocesses them (e.g., noise reduction, enhancement, segmentation), extracting characteristic parameters such as grain size, grain boundary features, and second-phase particle distribution. These parameters are then compared with the characteristic parameters of standard metallographic images in the database to determine the distortion and errors generated during image acquisition. If the distortion and errors exceed a threshold, the metallographic image is re-acquired to ensure the accuracy of the metallographic characteristic parameters. Based on the calibrated metallographic image, the aging level is determined.

[0049] Step 3: Based on the functional relationships of each mechanical data and the verified mechanical data, determine the predicted residual life value corresponding to each mechanical data.

[0050] The remaining life prediction value is the remaining service time of a metal component from the current moment until it reaches the preset failure threshold. The calculation formula is: Residual life = Limit service time - Current service time.

[0051] 1) The method for calculating the residual lifetime L1 corresponding to the creep variable is as follows: Substituting the verified creep variable into the function y1=f1(t) of creep variable-service duration, we can solve for the current service duration t. 10 ; Based on industry standards and historical failure data from the database, set the creep variable limit value y. 1max (i.e., the creep variable when a component reaches a failure state), y 1max Substituting the fitted function y1=f1(t), the inverse solution yields the limiting service time t corresponding to the creep variable. 1max ; Based on the maximum service life t 1max and length of service t 10 Determine the residual lifetime L1 of the creep variable: L1=t 1max - t 10 2) The calculation method for the residual life L2 corresponding to the hardness value is as follows: The hardness value y after passing the test 20 Substituting the hardness value into the function y2=f2(t) of service time, we can solve for the current service time t. 20 ; Based on industry standards and historical failure data from the database, a hardness limit value y is set. 2min (That is, the hardness value at which the component reaches a failure state; if it is lower than this value, the mechanical properties of the component cannot meet the service requirements.) y 2min Substituting into the fitting function y2=f2(t), the inverse solution yields the limit service time t corresponding to the hardness value. 2max ; Based on the hardness limit value y 2min and current service time t 20 Determine the residual life L2 of the hardness value: L2=t 2max - t 20 3) The calculation method for the residual life L3 corresponding to the metallographic aging level is as follows: The aging level of the metallographic structure is y 30Substituting the function y3=f3(t) of the metallographic aging level versus service time, we can solve for the current service time t. 30 ; Set the limit value y for metallographic aging level 3max =Level 5 (i.e., severe aging state, reaching the failure threshold), y 3max Substituting the fitting function y3=f3(t), the inverse solution yields the limit service life t corresponding to the metallographic aging level. 3max ; According to the metallographic aging level limit value y 3max and maximum service life t 3max Determine the residual life L3 corresponding to the metallographic aging level: L3=t 3max - t 30 The failure threshold (creep variable limit y) 1max Hardness limit value y 2min Metallographic aging level limit value y 3max The method for determining the failure threshold is as follows: combining industry safety standards for metal components of thermal power plants, equipment design requirements, and historical failure data in the database, corresponding failure thresholds are set for metal components of different materials and service temperatures to ensure the rationality and accuracy of the failure thresholds; at the same time, the failure thresholds can be dynamically adjusted and optimized based on database updates and actual engineering experience.

[0052] Step 4: Based on the residual life corresponding to each mechanical data, determine the final residual life of the metal component to be evaluated.

[0053] In this embodiment, the minimum value among the three predicted residual life values ​​L1, L2, and L3 is selected as the final residual life of the metal components of the thermal power plant to be evaluated.

[0054] This assessment method is based on the "barrel effect," where the capacity of a barrel is determined by its shortest stave. Similarly, the actual residual life of metal components in a thermal power plant is determined by the parameter with the most severe damage, namely, creep, hardness, and metallographic aging level. The parameter with the shortest residual life reflects the weakest link in the metal component, and the residual life corresponding to this parameter is the component's actual maximum possible remaining service time. If service continues beyond this time, the component will fail because this parameter reaches the failure threshold. Therefore, selecting the minimum value among the three predicted residual lifespans as the final residual lifespan accurately reflects the component's true lifespan and avoids safety hazards caused by overly optimistic assessment results.

[0055] Compared with existing technologies, the intelligent assessment method for the residual life of metal components in thermal power plants has the following advantages: 1. This method first collects a large amount of sample data to construct independent fitting functions. It uses three core damage parameters—creep variable, hardness value, and metallographic aging level—to construct independent fitting functions with service duration, avoiding the defect of multi-parameter joint fitting that masks the independent influence of each parameter. This method can accurately reflect the change law of each damage parameter with service duration. At the same time, based on the barrel effect, the minimum value among the three residual life prediction values ​​is selected as the final life, which can accurately capture the weak link of the component and reflect the true damage state of the component. This method effectively solves the problems of low evaluation accuracy and optimistic results of existing methods. The evaluation error is ≤5%, which can accurately reflect the true life state of the component.

[0056] 2. The integrated equipment for creep, hardness, and metallographic measurements enables in-situ synchronous acquisition of the three core parameters. This eliminates the need to disassemble components or use multiple separate devices for testing, significantly simplifying the testing process and improving data acquisition efficiency. Simultaneously, a large amount of sample data is acquired through a pre-established database, enabling the automatic construction of fitting functions. The database facilitates automatic data matching, verification, and residual life calculation. The entire process requires no manual intervention, reducing labor costs, avoiding human error, and achieving intelligent assessment of residual life.

[0057] 3. This method establishes a pre-built database of metallographic structures and hardness suitable for thermal power plant metal components, containing historical and failure data for different materials, service temperatures, and service durations, which can provide sufficient sample support for the construction of fitting functions. At the same time, the database supports updates and dynamic optimization, adapting to different types of thermal power plant metal components (such as boiler superheater tubes, main steam pipes, turbine blades, etc.), with a wide range of applications and strong versatility.

[0058] 4. This assessment method can quickly and accurately determine the final residual life of metal components, identify the weak points of the components, and provide a scientific and reliable basis for the maintenance, repair and replacement of thermal power plant equipment. It can help thermal power plants formulate reasonable maintenance strategies, avoid economic waste caused by over-maintenance, avoid safety hazards caused by insufficient maintenance, reduce the risk of equipment failure, and ensure the safe, economical and stable operation of thermal power plants. It has important engineering application value.

[0059] Example 2 A smart assessment system for the residual life of metal components in thermal power plants includes: The data acquisition module is used to obtain the current creep, hardness value, and metallographic aging level of the metal part to be evaluated. The prediction module is used to calculate the predicted residual life values ​​corresponding to creep, hardness, and metallographic aging levels based on the pre-established functional relationships between creep, hardness, and metallographic aging levels and service time. The residual life module is used to determine the minimum value among the predicted residual life values ​​corresponding to the creep, hardness, and metallographic aging levels as the residual life of the metal component, based on the barrel effect.

[0060] It should be noted that, in the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another device, or some features may be ignored or not executed. The modules described as separate components may or may not be physically separated. The components shown as modules may be one or more physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs.

[0061] Furthermore, in the various embodiments of the present invention, the modules can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0062] An electronic device provided in this application includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the intelligent assessment method for the residual life of metal components in thermal power plants as described in any of the above embodiments.

[0063] Another electronic device provided in this application embodiment may further include: an input port connected to a processor for transmitting multimodal data collected by an external acquisition device to the processor; a display unit connected to the processor for displaying the processor's processing results to the outside world; and a communication module connected to the processor for enabling communication between the electronic device and the outside world. The display unit may be a display panel, a laser scanning display, etc.; the communication method adopted by the communication module includes, but is not limited to, Mobile High Definition Link (HML), Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), and wireless connection (including Wi-Fi, Bluetooth, Bluetooth Low Energy, and IEEE 802.11s-based communication technology).

[0064] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the intelligent assessment method for the residual life of metal components in thermal power plants as described in any of the above embodiments.

[0065] For descriptions of relevant parts in the intelligent assessment system for residual life of metal components in thermal power plants, electronic devices, and computer-readable storage media provided in this application's embodiments, please refer to the detailed descriptions of the corresponding parts in the intelligent assessment method for residual life of metal components in thermal power plants provided in this application's embodiments; they will not be repeated here. Furthermore, parts of the technical solutions provided in this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0066] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method for intelligent assessment of the residual life of metal components in thermal power plants, characterized in that, Includes the following steps: Obtain the current creep, hardness value, and metallographic aging level of the metal component to be evaluated; Based on the pre-established functional relationships between creep, hardness, and metallographic aging level and service time, the predicted residual life values ​​corresponding to creep, hardness, and metallographic aging level are calculated. Based on the barrel effect, the minimum value among the creep, hardness, and metallographic aging levels corresponding to the predicted residual life is determined as the residual life of the metal component.

2. The intelligent assessment method for the residual life of metal components in a thermal power plant according to claim 1, characterized in that, The method for establishing the functional relationship is as follows: Acquire historical data of metal components under service conditions, including service time, as well as creep, hardness values ​​and metallographic aging levels that change with service time; A fitting algorithm was used to fit the creep variable, hardness value, and metallographic aging level to their respective service time, so as to obtain the functional relationship between creep variable and service time, the functional relationship between hardness value and service time, and the functional relationship between metallographic aging level and service time.

3. The intelligent assessment method for the residual life of metal components in thermal power plants according to claim 1, characterized in that, The fitting algorithm includes a nonlinear fitting algorithm or a machine learning fitting algorithm; The fitting algorithm includes one or more combinations of least squares method, support vector regression algorithm, and random forest algorithm.

4. The intelligent assessment method for the residual life of metal components in thermal power plants according to claim 1, characterized in that, The process of obtaining the current creep, hardness value, and metallographic aging level of the metal component to be evaluated includes: In-situ non-destructive testing was performed on the metal parts to be evaluated using an integrated creep-hardness-metallography testing equipment, which simultaneously acquired creep, hardness values ​​and metallographic images. The metallographic structure image is processed by image recognition based on pre-stored standard metallographic structure feature data to determine the aging level of the metallographic structure.

5. The intelligent assessment method for the residual life of metal components in a thermal power plant according to claim 4, characterized in that, The method for obtaining the current metallographic aging level of the metal component to be evaluated is as follows: Image recognition algorithms are used to extract characteristic parameters of metallographic structures, such as grain size, grain boundary features, and distribution of second-phase particles. These parameters are then compared with the characteristic parameters of standard metallographic images in the database to determine the aging level of the metallographic structure.

6. The intelligent assessment method for the residual life of metal components in a thermal power plant according to claim 1, characterized in that, The method for calculating the creep variable, hardness value, and metallographic aging level corresponding to the predicted residual life is as follows: Substitute the current creep value, current hardness value, and current metallographic aging level into the corresponding function relationship, and solve inversely to obtain the current service time corresponding to the current creep value, current hardness value, and current metallographic aging level, respectively. Substitute the predefined creep failure threshold, hardness failure threshold, and metallographic aging level failure threshold into the corresponding function relationships, and solve them inversely to obtain the limit service time corresponding to the creep, hardness, and metallographic aging levels. The predicted residual life values ​​corresponding to the current creep, current hardness, and current metallographic aging level are determined based on the maximum service life and the current service life.

7. The intelligent assessment method for the residual life of metal components in a thermal power plant according to claim 1, characterized in that, After acquiring the current mechanical data and before calculating the predicted residual life, the method also includes a step of verifying the current creep, hardness value, and metallographic aging level. This verification includes: The current creep and hardness values ​​are compared with the pre-stored historical data ranges of creep and hardness values ​​for the same material and service temperature. If the current value exceeds the corresponding historical data range, a re-detection is triggered. The acquired metallographic image is compared with the pre-stored standard metallographic image for image feature similarity. If the similarity is lower than the preset threshold, the metallographic image acquisition is judged to be abnormal and re-acquisition is triggered.

8. An intelligent assessment system for the residual life of metal components in thermal power plants, characterized in that, include: The data acquisition module is used to obtain the current creep, hardness value, and metallographic aging level of the metal part to be evaluated. The prediction module is used to calculate the predicted residual life values ​​corresponding to creep, hardness, and metallographic aging levels based on the pre-established functional relationships between creep, hardness, and metallographic aging levels and service time. The residual life module is used to determine the minimum value among the predicted residual life values ​​corresponding to the creep, hardness, and metallographic aging levels as the residual life of the metal component, based on the barrel effect.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the intelligent assessment method for residual life of metal components in thermal power plants as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the intelligent assessment method for residual life of metal components in thermal power plants as described in any one of claims 1-8.

Images