Microstructure aging analysis method and system for steam turbine steam guide pipe, and analysis equipment

Abstract

The embodiment of the application provides a steam turbine guide pipe microstructure aging analysis method and system, which first synchronously collects microstructure images and residual stress data of the guide pipe in operation by using non-destructive testing and an ultrasonic stress sensor. Secondly, the stress data is used for temperature load timing matching, the difference in grain boundary carbide precipitation is quantified, and then the creep cracking risk is estimated. Subsequently, the aging grade is matched and output by combining the stress fluctuation and the organization precipitation data, and the maintenance replacement control strategy is formulated according to the grade and the risk data. Finally, the aging state analysis system firmware is designed and uploaded to the operation and maintenance cloud platform to execute the method. The method realizes non-destructive real-time monitoring, accurately evaluates the dynamic aging degree and failure risk by constructing a stress-structure coupling model, supports predictive maintenance decision-making, and realizes standardized management and remote collaboration through cloud deployment, which significantly improves the operation safety and operation and maintenance efficiency of the unit.

Description

Technical Field

[0001] This invention relates to the field of thermal power equipment analysis and testing technology, specifically to a method and system for aging analysis of the microstructure of steam turbine guide pipes, and analytical equipment. Background Technology

[0002] As the core power equipment in thermal power, nuclear power, and combined cycle power generation systems, the operational safety and reliability of steam turbines directly affect the stable operation and economic benefits of the entire unit. Steam pipes, as crucial pressure-bearing components connecting the boiler (or steam generator) and the various cylinders of the steam turbine, operate under harsh conditions such as high temperature, high pressure, and alternating loads for extended periods, making them highly susceptible to microstructural evolution and structural aging. With prolonged service, typical aging characteristics appear within the steam pipe material, including grain coarsening, grain boundary carbide precipitation, creep porosity formation, and residual stress redistribution. These microstructural evolutions not only reduce the material's mechanical properties but may also induce creep crack initiation and propagation, potentially leading to major safety accidents such as pipe rupture.

[0003] Currently, aging assessments of turbine ducts largely rely on periodic shutdown inspections, metallographic sampling and analysis, and empirical judgment, which suffer from significant drawbacks such as long inspection cycles, irreversible damage, and poor real-time performance. While traditional non-destructive testing techniques can identify macroscopic defects, they struggle to quantitatively assess the degree of microstructural evolution and its impact on structural lifespan. Furthermore, existing assessment methods generally lack modeling of the dynamic relationship between thermo-mechanical coupling loads and material response during service, failing to achieve time-series tracking and risk prediction of aging conditions. Especially under complex and variable operating conditions, stress concentration and microstructural degradation in localized areas of the ducts often exhibit non-uniformity and abruptness, further increasing the difficulty of aging assessments. Summary of the Invention

[0004] The present invention aims to at least solve one of the technical problems existing in the prior art, and provides a method, system and analysis equipment for microstructure aging analysis of steam turbine guide pipes.

[0005] In a first aspect, embodiments of the present invention provide a method for aging analysis of the microstructure of a steam turbine duct, the method comprising: Step S1: The microstructure of the steam turbine's steam pipe under operating conditions is collected using non-destructive testing equipment to obtain a microstructure scanning image of the steam pipe; based on the microstructure scanning image of the steam pipe, residual stress is monitored in the steam turbine's steam pipe under service conditions using an ultrasonic stress sensor to obtain residual stress monitoring data of the steam pipe in service. Step S2: Match the time-series stress fluctuations of temperature loads based on the residual stress monitoring data of the steam turbine duct during service to obtain the time-series stress fluctuation data under operating conditions; quantify the difference in the degree of grain boundary carbide precipitation based on the time-series stress fluctuation data under operating conditions to obtain the difference data in grain boundary structure precipitation; estimate the creep cracking risk of the steam turbine duct based on the difference data in grain boundary structure precipitation to obtain the creep cracking risk data of the steam turbine duct. Step S3: Match the aging level output of the steam pipe according to the operating condition time-series stress fluctuation data and grain boundary precipitation difference data to obtain the steam pipe aging level output data; Based on the aging level output normalized data and the steam turbine steam pipe creep cracking risk data, adjust the steam pipe maintenance and replacement to obtain the steam pipe maintenance and replacement control data. Step S4: Based on the aging level output data of the steam duct and the maintenance and replacement control data of the steam duct, design the steam turbine steam duct aging status analysis system, obtain the steam turbine steam duct microstructure aging analysis firmware, and send the steam turbine steam duct microstructure aging analysis firmware to the operation and maintenance cloud platform to execute the steam turbine steam duct microstructure aging analysis method.

[0006] Optionally, step S1 includes the following steps: Step S11: Deploy electronic non-destructive scanning equipment and ultrasonic stress sensors on the outside of the steam turbine's steam pipe to be inspected; Step S12: The microstructure of the steam turbine's steam pipe under operating conditions is collected using an electronic non-destructive scanning device to obtain a microstructure scanning image of the steam pipe; Step S13: Perform regional segmentation processing on the scanned image of the steam pipe microstructure to obtain a continuous segmented image of the steam pipe microstructure; Step S14: Based on the continuous block image of the microstructure of the steam pipe, and by using an ultrasonic stress sensor to monitor the residual stress of the steam pipe in service, the residual stress monitoring data of the steam pipe in service is obtained.

[0007] Optionally, step S2 includes the following steps: Step S21: Extract the morphological features of aged tissue from the microscopic tissue scanning image of the steam pipe to obtain the morphological features of aged tissue; Step S22: Based on the residual stress monitoring data of the steam pipe during service, the morphological characteristics of the aged microstructure are matched with the time-series stress fluctuations under operating conditions to obtain the time-series stress fluctuation data under operating conditions. Step S23: Based on the operating condition time-series stress fluctuation data and aging microstructure morphology characteristics, perform dynamic growth analysis of steam pipe grains to obtain dynamic growth data of steam pipe grains; Step S24: Quantify the difference in the degree of grain boundary carbide precipitation in the dynamic growth data of the steam pipe grains to obtain the difference data of grain boundary structure precipitation. Step S25: Estimate the risk of creep cracking in the turbine steam pipe based on the grain boundary precipitation difference data and the operating condition time-series stress fluctuation data, and obtain the creep cracking risk data of the turbine steam pipe.

[0008] Optionally, step S23 includes the following steps: Step S231: Estimate the wall thickness of the steam turbine through the microstructure scanning image of the steam turbine, and obtain the estimated wall thickness data of the steam turbine, wherein the steam turbine steam turbine includes a high-pressure steam turbine, a medium-pressure steam turbine, and a low-pressure steam turbine. Step S232: Based on the operating condition time-series stress fluctuation data and the estimated wall thickness of the steam pipe, the grain growth rate range of different regions of the steam pipe is deduced to obtain the grain growth rate range of different regions of the steam pipe. Step S233: Based on the estimated wall thickness of the steam pipe, perform stress gradient correlation coefficient analysis on the grain growth rate range to obtain the wall thickness-related stress gradient correlation coefficient; Step S234: Spatial orientation identification of the morphological characteristics of the aged tissue is performed to obtain grain orientation identification data; Step S235: Based on the grain orientation identification data, extrapolate the grain boundary migration rate distribution during the aging process of the steam pipe at different temperature ranges within the grain growth rate range, and obtain the grain boundary migration rate distribution data. Step S236: Perform dynamic growth analysis of steam pipe grains based on the correlation coefficient of wall thickness-related stress gradient and grain boundary migration rate distribution data to obtain dynamic growth data of steam pipe grains.

[0009] Optionally, step S3 includes the following steps: Step S31: Match the aging level output of the steam pipe according to the stress fluctuation data and grain boundary precipitation difference data under the working condition to obtain the aging level output data of the steam pipe; Step S32: Normalize the aging level output data of the steam pipe to obtain normalized aging level output data; Step S33: Based on the aging level output normalized data, grain boundary precipitation difference data, and turbine steam pipe creep cracking risk data, steam pipe maintenance and replacement control is carried out to obtain steam pipe maintenance and replacement control data.

[0010] Optionally, step S31 includes the following steps: Step S311: Plot the stress time-series fluctuation curves from the working condition time-series stress fluctuation data to obtain the residual stress time-series fluctuation curves; Step S312: Calculate the instantaneous stress rise gradient of the residual stress time-series fluctuation curve to obtain the instantaneous stress rise gradient of the residual stress. Step S313: Based on the instantaneous stress rise gradient of residual stress, perform stress multi-peak region pattern identification on the residual stress time-series fluctuation curve to obtain the residual stress intensity multi-peak region pattern. Step S314: Dynamically match the aging level of the steam pipe based on the multi-peak region pattern of residual stress intensity to obtain dynamic classification data of the steam pipe region; Step S315: Match the aging range of the steam pipe based on the difference data of grain boundary precipitation, and perform multi-region level correction optimization on the dynamic classification data of the steam pipe region to obtain multi-region range classification correction data. Step S316: Match the aging level output of the steam pipe based on the dynamic grading data of the steam pipe area and the grading correction data of the multi-area range, and obtain the aging level output data of the steam pipe.

[0011] Optionally, the following steps are included after step S316: Step S3161: Obtain the design service life and service duration data of the steam turbine duct to be analyzed, and calculate the percentage of the remaining design service life of the duct. Step S3162: Based on the remaining design life percentage data of the steam pipe, perform life loss weight correction on the initially obtained steam pipe aging level output data to obtain the life correction coefficient. Step S3163: Combine the over-temperature alarm record data in the operation history of the steam pipe, perform over-temperature aging gain correction on the aging level output data after life correction, and obtain the over-temperature corrected aging level output. Step S3164: Update the over-temperature correction aging level output to the final steam pipe aging level output data.

[0012] Secondly, embodiments of the present invention provide a microstructure aging analysis system for steam turbine ductwork, the system comprising: The dual-source sensing module is used to collect the microstructure of the steam turbine's steam pipe under operating conditions using non-destructive testing equipment, and obtain a microstructure scanning image of the steam pipe; based on the microstructure scanning image of the steam pipe, residual stress is monitored in the steam turbine's steam pipe under service conditions using an ultrasonic stress sensor, and residual stress monitoring data of the steam pipe in service is obtained. The risk quantification module is used to match the time-series stress fluctuations of temperature load based on the residual stress monitoring data of the steam turbine duct during service, and obtain the time-series stress fluctuation data under operating conditions; to quantify the difference in the degree of precipitation of carbides at grain boundaries based on the time-series stress fluctuation data under operating conditions, and obtain the difference data of precipitation of grain boundaries; and to estimate the risk of creep cracking of the steam turbine duct based on the difference data of precipitation of grain boundaries, and obtain the risk data of creep cracking of the steam turbine duct. The decision control module is used to match the aging level output of the steam pipe according to the operating condition time-series stress fluctuation data and grain boundary precipitation difference data to obtain the aging level output data of the steam pipe; and to control the maintenance and replacement of the steam pipe based on the aging level output normalized data and the steam turbine steam pipe creep cracking risk data to obtain the steam pipe maintenance and replacement control data. The cloud deployment module is used to design a system for analyzing the aging status of turbine steam pipes based on the output data of steam pipe aging level and the control data of steam pipe maintenance and replacement. It obtains the microstructure aging analysis firmware of the turbine steam pipe and sends the microstructure aging analysis firmware of the turbine steam pipe to the operation and maintenance cloud platform to execute the microstructure aging analysis method of the turbine steam pipe.

[0013] Thirdly, embodiments of the present invention also provide a steam turbine duct microstructure aging analysis device, the device comprising: a memory, a processor, and a steam turbine duct microstructure aging analysis program stored in the memory and executable on the processor, the steam turbine duct microstructure aging analysis program being configured to implement the steps of the steam turbine duct microstructure aging analysis method as described above.

[0014] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a microstructure aging analysis program for a steam turbine duct, wherein when the steam turbine duct microstructure aging analysis program is executed by a processor, the steps of the steam turbine duct microstructure aging analysis method as described above are implemented.

[0015] The present invention discloses a method and system for aging analysis of the microstructure of steam turbine ducts. The analysis equipment, by combining non-destructive testing equipment and an ultrasonic stress sensor, can simultaneously acquire microstructure scanning images and residual stress data under service conditions without damaging the duct structure or requiring shutdown for sampling. This overcomes the shortcomings of traditional metallographic sampling methods, which are destructive, have long testing cycles, and cannot reflect real-time operating conditions, significantly improving the safety and efficiency of the testing.

[0016] Furthermore, this invention innovatively establishes a quantitative correlation between temporal stress fluctuations under temperature load and the degree of grain boundary carbide precipitation. By analyzing the impact of temporal stress fluctuations on microstructures (such as grain boundary carbide precipitation), the dynamic mechanism of material aging under thermo-mechanical coupling conditions can be revealed more accurately. This solves the problem that static assessments in existing technologies are difficult to adapt to complex and variable operating conditions, and improves the physical accuracy of aging analysis. Based on the difference data in grain boundary precipitation, creep cracking risk is estimated, directly mapping the degree of microstructure deterioration to macroscopic failure risk. This method can not only identify the current damage state, but also predict future crack initiation trends through temporal fluctuation matching, thereby providing early warning before catastrophic accidents occur and effectively preventing unplanned unit shutdowns caused by steam pipe ruptures.

[0017] Furthermore, by combining aging level output data with creep cracking risk data, this invention can generate specific maintenance and replacement control data. This transforms maintenance decisions from traditional periodic or reactive maintenance to predictive maintenance based on actual health status, avoiding resource waste caused by over-maintenance and eliminating safety hazards caused by under-maintenance, thus optimizing the overall lifecycle maintenance costs. By embedding the analysis logic into firmware and uploading it to the maintenance cloud platform, this invention enables remote algorithm updates, centralized data management, and experience sharing among multiple units. This not only reduces the computational burden on field terminals but also promotes the standardization of aging analysis, providing a scalable digital solution for equipment health management in large power plant clusters. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a schematic flowchart of the microstructure aging analysis method for steam turbine guide pipes according to an embodiment of the present invention; Figure 2 This is a structural block diagram of the microstructure aging analysis system for steam turbine guide pipes according to an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0021] Unless otherwise specifically stated, the technical or scientific terms used in the embodiments of this invention should be understood in their ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains. The terms "comprising" or "including," as used in the embodiments of this invention, do not limit the shapes, numbers, steps, actions, operations, components, elements, and / or groups thereof mentioned, nor do they exclude the appearance or addition of one or more other different shapes, numbers, steps, actions, operations, components, elements, and / or groups thereof, or the inclusion of these.

[0022] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale, and techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail; however, where appropriate, the illustrated techniques, methods, and apparatus should be considered part of the specification. In all the examples shown and discussed herein, any other specific example may have different values. It should be noted that similar symbols and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0023] In the description of the embodiments of the present invention, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In the embodiments of the present invention, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in the embodiments of the present invention, as well as the features of different embodiments or examples.

[0024] Hereinafter, exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0025] Reference Figure 1 , Figure 1 This is a flowchart illustrating the aging analysis method for the microstructure of steam turbine guide pipes according to an embodiment of the present invention, and an embodiment of the aging analysis method for the microstructure of steam turbine guide pipes according to the present invention is presented.

[0026] In one embodiment, the method for aging analysis of the microstructure of the steam turbine guide pipe includes: Step S1: The microstructure of the turbine's steam pipe under operating conditions is collected using non-destructive testing equipment to obtain a microstructure scanning image of the steam pipe. Based on the microstructure scanning image of the steam pipe, residual stress is monitored in the turbine's steam pipe under service conditions using an ultrasonic stress sensor to obtain residual stress monitoring data of the steam pipe.

[0027] The non-destructive testing (NDT) equipment can be a detection device used to obtain the internal microstructure of the inspected object without damaging its structural integrity. It can be used to non-invasively acquire microstructure images while the steam pipe is in operation, avoiding downtime and sampling. In this embodiment, the NDT equipment can utilize signals generated by the interaction of physical fields such as ultrasound, electromagnetic fields, or radiation with the material's microstructure to invert the microstructure. Furthermore, the NDT equipment can include, but is not limited to, one or more of ultrasonic microscopic imaging equipment, electromagnetic ultrasonic probe arrays, and phased array ultrasonic detection systems. The steam turbine steam pipe under operating conditions can be a high-temperature, high-pressure steam transmission pipeline in actual thermodynamic cycle operation, and can be used as the test object, its microstructure and stress state reflecting the actual aging process. The steam pipe microstructure scanning image can be two-dimensional or three-dimensional image data characterizing the spatial distribution of grains, grain boundaries, and precipitated phases in the steam pipe material, and can be used to provide visualization of aging characteristics such as grain boundary carbide precipitation. For example, the steam pipe microstructure scanning image can be generated by scanning the operating steam pipe using the NDT equipment.

[0028] An ultrasonic stress sensor can be a sensing device that measures the residual stress inside a material based on the acoustoelastic effect. It can be used to monitor the residual stress distribution of a steam turbine duct in real time under service conditions. In an exemplary embodiment, the ultrasonic stress sensor can infer the stress magnitude by measuring the change in the propagation speed of ultrasonic waves in a stress field. Furthermore, the ultrasonic stress sensor can be used in conjunction with non-destructive testing equipment to simultaneously acquire microstructure and stress data at the same spatiotemporal location. The steam turbine duct in service conditions can be the duct body itself subjected to actual temperature, pressure, and alternating loads, which can be used as a physical carrier for stress monitoring to ensure that the data reflects the actual service conditions. The residual stress monitoring data of the steam turbine duct can be a time-series set of internal residual stress values ​​during the operation of the duct, which can be used to match temperature load fluctuations and construct a thermo-mechanical coupling response relationship. In a specific embodiment, the residual stress monitoring data of the steam turbine duct can be continuously acquired and recorded by an ultrasonic stress sensor.

[0029] The microstructure of the turbine's steam pipe under operating conditions can be acquired using non-destructive testing (NDT) equipment. This can be done without shutting down or disassembling the steam pipe, by emitting a detection signal and receiving the echo to reconstruct a microstructure image. Further, this operation can be achieved by using high-frequency ultrasonic microscopy to scan the outer wall region of the steam pipe and invert the near-surface grain boundary structure, or by using an electromagnetic ultrasonic transducer array to scan along the pipe's axial direction to obtain circumferential and axial microstructure distribution maps. This provides an image reflecting the aging characteristics of the material's microstructure under actual operating conditions. Based on the microstructure scan image of the steam pipe, residual stress monitoring of the turbine's steam pipe under service conditions can be performed using an ultrasonic stress sensor. This can be done simultaneously with acquiring the microstructure image, using the ultrasonic stress sensor to measure the change in sound velocity in the same area to calculate the residual stress. For example, this operation can be achieved by using a dual-modal probe to integrate ultrasonic imaging and stress measurement functions for synchronous spatial acquisition, or by using a time-synchronous triggering mechanism to coordinate the working cycle of independent NDT equipment and ultrasonic stress sensors. This allows for the simultaneous acquisition of microstructure and mechanical response data, establishing a micro-macro correlation foundation.

[0030] Step S2: Match the time-series stress fluctuations under temperature load based on the residual stress monitoring data of the steam turbine duct during service to obtain the time-series stress fluctuation data under operating conditions. Quantify the difference in the degree of grain boundary carbide precipitation based on the time-series stress fluctuation data under operating conditions to obtain the grain boundary microstructure precipitation difference data. Estimate the creep cracking risk of the steam turbine duct based on the grain boundary microstructure precipitation difference data to obtain the creep cracking risk data of the steam turbine duct.

[0031] The temperature load time-series stress fluctuation can be a dynamic stress response process caused by the time-varying changes in the steam pipe wall temperature and internal pressure due to changes in operating conditions. It can be used as an external excitation variable driving the evolution of the microstructure. The operating condition time-series stress fluctuation data can be a stress time-series sequence synchronized with the temperature load after matching processing. It can be used as an input variable to quantify the difference in the degree of grain boundary carbide precipitation. In an exemplary embodiment, the operating condition time-series stress fluctuation data can be obtained by time-aligning and associating operating condition tags with the residual stress monitoring data of the steam pipe during service. The degree of grain boundary carbide precipitation can be an indicator of the density, size, and continuity of carbide phase precipitation at the material grain boundaries. It can be used to characterize key microscopic features of material creep damage and microstructure aging.

[0032] Grain boundary precipitation difference data can be a quantitative comparison of the precipitation state of grain boundary carbides under different operating conditions and stress fluctuations, and can be used as a direct input parameter for creep cracking risk estimation. In a specific embodiment, grain boundary precipitation difference data can be obtained by joint analysis of operating condition time-series stress fluctuation data and microstructure images. Creep cracking risk estimation of turbine ducts can be a process of predicting the probability or trend of creep cracking in the ducts based on the degree of microstructure deterioration, and can be used to map from microscopic damage to macroscopic failure risk. Turbine duct creep cracking risk data can be numerical or level information that quantitatively expresses the likelihood of future creep cracking in the ducts, and can be used to support early warning and maintenance decisions. For example, turbine duct creep cracking risk data can be generated by calculating risk models from grain boundary precipitation difference data.

[0033] Matching temperature load time-series stress fluctuations based on residual stress monitoring data from the steam pipe can be achieved by aligning the residual stress time-series data with boiler outlet temperature, pressure, and other operating parameters for the corresponding time period and performing correlation modeling. Further, this operation can be implemented by using a dynamic time warping algorithm to align non-uniformly sampled stress and temperature sequences, or by constructing a stress fluctuation clustering model based on operating condition labels (such as start-up / shutdown, variable load) to extract typical fluctuation patterns. This allows for the separation of stress fluctuation components dominated by temperature load changes, eliminating other interfering factors. Quantifying the difference in grain boundary carbide precipitation degree based on operating condition time-series stress fluctuation data can be achieved by extracting features from microstructure images corresponding to different stress fluctuation patterns and calculating the statistical differences in grain boundary carbide precipitation. In a specific embodiment, this operation can be achieved by using image segmentation algorithms to identify grain boundary regions and statistically analyzing the number and area ratio of carbide particles per unit length of grain boundary, or by using a deep learning model to perform semantic parsing of the microstructure image and outputting a carbide precipitation continuity score. This establishes a quantitative mapping relationship between thermo-mechanical coupling loads and microstructure evolution.

[0034] Estimating the risk of creep cracking in turbine ducts based on grain boundary precipitation differences can be achieved by inputting the data into a pre-trained risk prediction model, which then outputs the cracking probability or risk level. For example, this operation can be implemented by using a physical mechanism-based model, combined with precipitate size and grain boundary weakening theory, to calculate the local creep strain rate, or by employing a data-driven approach, using historical duct burst cases to train a classifier to predict high-risk areas. This allows for a direct conversion from microscopic damage state to macroscopic failure risk.

[0035] Step S3: Match the aging level output of the steam duct based on the operating condition time-series stress fluctuation data and grain boundary precipitation difference data to obtain the steam duct aging level output data. Based on the aging level output normalized data and the turbine steam duct creep cracking risk data, adjust the steam duct maintenance and replacement to obtain the steam duct maintenance and replacement control data.

[0036] The process of matching the aging level output of the steam pipe can be a process of mapping the differences in stress fluctuations and grain boundary precipitation under operating conditions to a preset aging level standard, which can be used to generate a standardized description of the aging state. The aging level output data of the steam pipe can be a classification or grading result characterizing the current aging degree of the steam pipe, which can be used to provide a health status benchmark for maintenance and control. In an exemplary embodiment, the aging level output data of the steam pipe can be determined by fusing multi-source data through matching rules or machine learning models. The aging level output normalized data can be standardized steam pipe aging level data with unified dimensions and range, which can be used to facilitate joint calculations with creep cracking risk data. The steam pipe maintenance and replacement control can be a decision-making behavior for maintaining, replacing, or continuing to operate the steam pipe based on aging and risk data, which can be used to generate predictive maintenance strategies. The steam pipe maintenance and replacement control data can be a specific instruction set including maintenance priorities, replacement recommendation time windows, and operation instructions, which can be used to directly guide on-site operation and maintenance. In a specific embodiment, the steam pipe maintenance and replacement control data can be generated by joint inference from the aging level output normalized data and creep cracking risk data.

[0037] Matching the aging level output of the steam turbine duct based on operating condition time-series stress fluctuation data and grain boundary precipitation difference data can be achieved by fusing the two types of data and comparing them with a preset aging level judgment rule library to determine the current aging level. Further, this operation can be implemented by using a fuzzy comprehensive evaluation method to weighted synthesize an aging index from multi-dimensional indicators and divide it into level intervals, or by constructing a decision tree model to determine the aging level based on a combination of stress fluctuation amplitude and precipitation difference threshold, thereby generating a standardized and interpretable description of the aging state. Controlling the maintenance and replacement of the steam turbine duct based on the normalized aging level output data and the creep cracking risk data of the steam turbine duct can be achieved by inputting the normalized aging level and risk data into a maintenance strategy generation engine to output specific maintenance recommendations. For example, this operation can be achieved by setting a risk-aging joint matrix, with different quadrants corresponding to different maintenance actions (such as monitoring, planned replacement, immediate shutdown), or by combining a remaining life prediction model to calculate the optimal replacement time window and generate a work order, thereby achieving a closed loop from state assessment to operation and maintenance decision-making.

[0038] Step S4: Based on the aging level output data of the steam duct and the maintenance and replacement control data of the steam duct, design the steam turbine steam duct aging status analysis system, obtain the steam turbine steam duct microstructure aging analysis firmware, and send the steam turbine steam duct microstructure aging analysis firmware to the operation and maintenance cloud platform to execute the steam turbine steam duct microstructure aging analysis method.

[0039] The design of the turbine steam duct aging status analysis system can be an engineering process of encapsulating aging analysis logic into an executable program module, which can be used to form deployable and reusable analysis capabilities. The turbine steam duct microstructure aging analysis firmware can be an embedded software module that solidifies the entire aging analysis algorithm logic, which can be used to achieve localized real-time analysis and reduce cloud dependence. In a specific embodiment, the turbine steam duct microstructure aging analysis firmware can be compiled into code that can run on edge devices through system design. Furthermore, the turbine steam duct microstructure aging analysis firmware can collaborate with the operation and maintenance cloud platform to support remote updates and data backhaul. The operation and maintenance cloud platform can be a cloud-based information system that centrally manages the health data and analysis models of multiple units, and can be used to achieve remote algorithm updates, centralized data storage, and cross-unit experience sharing. For example, the operation and maintenance cloud platform can include, but is not limited to, equipment health management platforms, power plant group digital twin centers, and predictive maintenance scheduling systems.

[0040] The design of a turbine steam pipe aging status analysis system based on steam pipe aging level output data and steam pipe maintenance and replacement control data can encapsulate the aforementioned analysis logic into a modular software architecture, forming a deployable analysis system. In an exemplary embodiment, this operation can be implemented by using a microservice architecture to decompose each step into an independent service, supporting flexible combination and expansion, or by designing a state machine model to execute analysis tasks in the S100-S400 process sequence, thereby forming a reusable and portable aging analysis capability unit. Sending the turbine steam pipe microstructure aging analysis firmware to the operation and maintenance cloud platform can be achieved by uploading the compiled firmware package to the cloud platform for version management and distribution via a secure communication protocol. Furthermore, this operation can push the firmware to edge computing nodes via OTA (Over-The-Air) technology, or establish a firmware repository on the cloud platform to support automatic matching and deployment of the latest version according to the unit model, thereby achieving centralized management and continuous iteration of the analysis algorithm.

[0041] Taking the health monitoring of the steam duct during variable load operation of a thermal power unit as an example, the microstructure aging analysis method of the steam turbine steam duct in this embodiment can be used when a 600MW coal-fired unit frequently experiences load increases and decreases during grid peak shaving, leading to severe temperature fluctuations in the main steam duct. Non-destructive testing equipment deployed on-site and ultrasonic stress sensors simultaneously acquire images of the pipe wall microstructure and residual stress time-series data. The system matches the stress data with the load curve recorded by the DCS, identifying areas where grain boundary carbide precipitation is significantly aggravated during periods of high stress fluctuation. Based on this difference data, the risk estimation module determines that the creep cracking risk in this area has risen to the warning threshold. Simultaneously, the aging level matching module outputs a level-three aging conclusion. The maintenance control engine, considering both factors, recommends scheduling a shutdown inspection within 30 days. This conclusion, along with the original data, is packaged and uploaded to the group's operation and maintenance cloud platform for reference by similar units, triggering a firmware update to optimize subsequent risk threshold parameters.

[0042] In one embodiment, step S1 includes the following steps: Step S11: Deploy electronic non-destructive scanning equipment and ultrasonic stress sensors on the outside of the steam turbine's steam pipe to be inspected; The turbine's steam duct under inspection is the main body of the duct that is planned for aging condition assessment but for which inspection equipment has not yet been deployed. It can be used as the target object for inspection deployment, clarifying the sensor installation location and scanning range. In this embodiment, the turbine's steam duct under inspection, the electronic non-destructive scanning equipment, and the ultrasonic stress sensor constitute a spatially coordinated sensing system. The electronic non-destructive scanning equipment can be a non-destructive testing device that uses non-mechanical contact methods such as electron beams, electromagnetic fields, or high-frequency electrical signals to image the internal microstructure of materials. It can be used to achieve high-resolution, non-invasive acquisition of the microstructure state outside the steam duct. Furthermore, the operating principle of the electronic non-destructive scanning equipment can be explained in context: by emitting electronic or electromagnetic signals and receiving the response signals after their interaction with the material's grain boundaries and precipitates, a microstructure image can be reconstructed. For example, the electronic non-destructive scanning equipment can include, but is not limited to, one or more of the following: electromagnetic ultrasonic microscopic imager, eddy current array scanning system, and microwave near-field imaging probe.

[0043] Deploying electronic non-destructive scanning equipment and ultrasonic stress sensors on the exterior of the steam turbine's duct to be inspected can be achieved by fixing both types of sensors to a designated inspection area on the outer wall of the duct, ensuring that the detection direction covers the volume of the target material. Furthermore, the deployment of electronic non-destructive scanning equipment and ultrasonic stress sensors on the exterior of the steam turbine's duct can be accomplished by using magnetic or clamp-type mounting brackets, allowing the electronic non-destructive scanning equipment and ultrasonic stress sensors to be coplanarly attached to the duct wall, or by arranging multiple sensor arrays along the axial direction of the duct to cover high-risk areas such as elbows, tees, and welds. This establishes a physical sensing foundation for the simultaneous acquisition of microstructure and stress data during operation.

[0044] Step S12: The microstructure of the steam turbine's steam pipe under operating conditions is collected using an electronic non-destructive scanning device to obtain a microstructure scanning image of the steam pipe; The microstructure of a steam turbine's steam pipe under operating conditions can be acquired by using electronic non-destructive scanning equipment. This can be achieved by activating the equipment during normal unit operation, emitting a detection signal, and receiving the echo to generate a microstructure image. In one exemplary embodiment, this acquisition can be achieved using phased array electronic scanning to dynamically focus on different depth layers to obtain three-dimensional microstructure information, or by combining a temperature compensation algorithm to correct the impact of high temperatures on electronic signal propagation, thereby improving image fidelity. This allows for the acquisition of aging images that reflect the material's microstructure under actual thermal conditions.

[0045] Step S13: Perform regional segmentation processing on the scanned image of the steam pipe microstructure to obtain a continuous segmented image of the steam pipe microstructure; The continuous block image of the microstructure of the steam pipe can be an ordered set of image blocks formed by dividing the original scanned image of the microstructure of the steam pipe into spatial regions. This can be used to improve the accuracy of microstructure characterization in local areas (such as elbows and welds) and support spatial alignment with stress data at corresponding locations. Furthermore, the acquisition method of the continuous block image of the microstructure of the steam pipe can be described in context: the entire scanned image is segmented according to geometric coordinates or feature boundaries using rules or adaptive methods. In a specific embodiment, the continuous block image of the microstructure of the steam pipe can include, but is not limited to, axial segmented image blocks, circumferential sector image blocks, and image blocks based on curvature adaptive segmentation. The regional segmentation processing of the scanned image of the microstructure of the steam pipe can be performed by dividing the entire scanned image into multiple spatially continuous sub-image blocks according to preset rules. For example, the microstructure scanning image of the steam pipe can be divided into regions by uniformly dividing the image according to a fixed-size grid (applicable to straight pipe sections), or by automatically identifying high-variable regions (such as dense grain boundary regions) based on image gradients or edge features to perform non-uniform segmentation. This can enhance the resolution of local non-uniform aging regions and provide structured input for subsequent stress-structure spatial matching.

[0046] Step S14: Based on the continuous block image of the microstructure of the steam pipe, and by using an ultrasonic stress sensor to monitor the residual stress of the steam pipe in service, the residual stress monitoring data of the steam pipe in service is obtained.

[0047] Based on continuous block images of the microstructure of the steam turbine's steam pipe, residual stress monitoring in service conditions using an ultrasonic stress sensor can be achieved by mapping the spatial location of each image block to the measurement points of the ultrasonic stress sensor, simultaneously acquiring stress data in the same area. Furthermore, residual stress monitoring in service conditions using continuous block images of the steam turbine's steam pipe microstructure can be achieved by using the probe unit closest to the center of the image block in the sensor array to read the corresponding stress value, or by reconstructing the stress distribution field of the image block coverage area using an interpolation algorithm based on multiple neighboring stress measurement points. This allows for precise spatial alignment of the microstructure state and residual stress, improving the local accuracy of thermo-mechanical coupling analysis.

[0048] Taking the aging monitoring of the steam generator outlet duct bend area in a nuclear power plant as an example, the microstructure aging analysis method of the steam turbine duct in this embodiment can be as follows: at the 90-degree bend of the steam generator outlet duct in a pressurized water reactor nuclear power unit, maintenance personnel deploy a ring-shaped electronic non-destructive scanning device and a distributed ultrasonic stress sensor array; the device collects microstructure scanning images of the inner and outer arc sides of the bend during full-power operation; the system automatically divides the images into 24 continuous blocks at 15-degree intervals in the circumference; each image block corresponds to a stress measurement point, and the inner arc side image block shows that the grain boundary carbide precipitation density is significantly higher than that of the outer arc side, and the corresponding stress value is 23% higher; this spatially aligned data is used to quantify the degree of thermo-mechanical coupling aging in the bend area and trigger a local key monitoring strategy.

[0049] In one embodiment, step S2 includes the following steps: Step S21: Extract the morphological features of aged tissue from the microscopic tissue scanning image of the steam pipe to obtain the morphological features of aged tissue; The aging microstructure features can be a set of structured image features representing the aging state of the material extracted from the microstructure scanning image of the steam pipe. These features can provide calculable aging characterization inputs for subsequent stress matching and grain growth analysis. In this embodiment, the aging microstructure features can be quantitatively described using image processing or machine learning algorithms to characterize grain size, porosity, grain boundary morphology, etc. Furthermore, the aging microstructure features can include, but are not limited to, one or more of the following: grain coarsening index, creep porosity distribution density, and grain boundary tortuosity.

[0050] Extracting aging morphology features from the microstructure scan images of the steam pipe can be achieved by applying image analysis algorithms to the scan images to identify and quantify aging-related microstructural features. In one exemplary embodiment, this operation can be implemented by automatically labeling grain boundaries, pores, and matrix regions using a semantic segmentation network, outputting geometric and topological feature vectors. Alternatively, conventional image processing methods (such as edge detection and morphological operations) can be used to extract the average grain size and pore area ratio, thereby transforming the original image into structured aging characterization data that can be used for modeling.

[0051] Step S22: Based on the residual stress monitoring data of the steam pipe during service, the morphological characteristics of the aged microstructure are matched with the time-series stress fluctuations under operating conditions to obtain the time-series stress fluctuation data under operating conditions. The operating condition time-series stress fluctuation matching can be a process of associating and aligning the morphological features of aged tissue with the residual stress data within the corresponding time window. This can be used to establish a dynamic mapping relationship between micromorphological evolution and thermo-mechanical load history. Matching the morphological features of aged tissue with operating condition time-series stress fluctuations based on residual stress monitoring data from the steam pipe service can involve aligning the extracted morphological features of aged tissue with the corresponding residual stress time-series data according to the acquisition timestamp, establishing a feature-stress mapping relationship. In a specific embodiment, this operation can be achieved by using a sliding time window to statistically summarize the residual stress sequence (such as mean, peak-to-valley difference) and binding it with the tissue features within the same window. Furthermore, a graph neural network can be constructed to jointly embed the tissue feature nodes with the time-series stress edges, thereby achieving a dynamic association between micromorphological evolution and thermo-mechanical load history, enhancing the realism of aging mechanism modeling.

[0052] Step S23: Based on the operating condition time-series stress fluctuation data and aging microstructure morphology characteristics, perform dynamic growth analysis of steam pipe grains to obtain dynamic growth data of steam pipe grains; The dynamic growth analysis of steam pipe grains can be used to deduce the growth behavior of grains over time during service based on stress fluctuations and microstructure characteristics. This analysis can reveal the physical mechanism between grain boundary migration and stress-driven processes, supporting carbide precipitation modeling. The dynamic growth data of steam pipe grains can be time-series quantitative results describing the changes in grain size, orientation, or boundary position with stress fluctuations under operating conditions. This data can serve as a preliminary basis for quantifying differences in grain boundary carbide precipitation, reflecting the dynamic process of microstructure degradation. In this embodiment, the dynamic growth data of steam pipe grains can be combined with aging microstructure morphology characteristics and time-series stress fluctuation data under operating conditions to invert the grain evolution trajectory using physical models or data-driven methods.

[0053] Dynamic growth analysis of steam pipe grains based on time-series stress fluctuation data and aging microstructure characteristics can integrate stress fluctuation patterns with current microstructure to predict grain growth or coarsening trends under thermo-mechanical coupling. For example, this operation can be achieved by numerically simulating future grain evolution paths using a physical model based on grain boundary migration rate and stress gradient. Furthermore, a time-series prediction model (such as LSTM) can be trained, using historical stress-morphology pairs as input to output a sequence of grain size changes, thereby revealing the dynamic laws of grain evolution driven by load and providing a spatiotemporal context for carbide precipitation.

[0054] Step S24: Quantify the difference in the degree of grain boundary carbide precipitation in the dynamic growth data of the steam pipe grains to obtain the difference data of grain boundary structure precipitation. The quantification of differences in grain boundary carbide precipitation can be a numerical process based on grain dynamic growth data to calculate the differences in grain boundary carbide precipitation states in different regions or time periods. This can be used to generate refined micro-deterioration indicators for risk estimation. Quantifying the differences in grain boundary carbide precipitation in steam pipe grain dynamic growth data can be achieved by assessing the differences in carbide precipitation tendencies in different grain boundary regions based on parameters such as grain boundary migration rate and residence time in the grain dynamic growth data. In one specific embodiment, this operation can be achieved by combining grain boundary curvature and local stress residence time to calculate the carbide nucleation probability distribution; for example, grain growth trajectory clustering can also be used to assign different precipitation weights to high-mobility grain boundaries and low-mobility grain boundaries, thereby accurately characterizing the spatial non-uniformity and temporal cumulative effect of carbide precipitation.

[0055] Step S25: Estimate the risk of creep cracking in the turbine steam pipe based on the grain boundary precipitation difference data and the operating condition time-series stress fluctuation data, and obtain the creep cracking risk data of the turbine steam pipe.

[0056] Estimating the risk of creep cracking in turbine ducts based on grain boundary precipitation difference data and time-series stress fluctuation data can be achieved by jointly inputting grain boundary precipitation difference data with current and historical stress fluctuation data into a risk model and outputting a cracking risk level. In one exemplary embodiment, this operation can be achieved by constructing a multivariate Bayesian network and fusing precipitation continuity, stress peak frequency, and temperature cycle number for risk inference. Furthermore, a physical-informed neural network can be used to embed the creep damage equation into a loss function to achieve risk prediction constrained by mechanism, thereby enabling risk estimation to simultaneously consider microscopic degradation state and load history dynamics, improving the confidence level of early warning.

[0057] For example, in the scenario of long-term service evaluation of the steam generator outlet duct of a nuclear power plant, the microstructure aging analysis method of the steam turbine duct in this embodiment can be as follows: The steam duct of a pressurized water reactor nuclear power unit has been in operation for 15 years. The system obtains its microstructure scanning image through non-destructive testing equipment and extracts the morphological characteristics of the aged structure, including an average increase of 30% in grain size and the appearance of chain-like pores in local areas; the residual stress data obtained simultaneously shows that it has experienced stress cycles caused by 5 start-ups and shutdowns in the past year; through the matching of stress fluctuations in the operating conditions, it is found that the dense pore area corresponds to the high frequency of stress fluctuations; the grain dynamic growth analysis shows that the grain boundary migration in this area is active; based on this, the difference in grain boundary carbide precipitation is quantified, showing that some grain boundaries precipitate in a continuous network; finally, combined with the precipitation difference and the recent stress fluctuation intensity, the risk estimation module determines that the creep cracking risk in this area has risen to a high-risk level, triggering maintenance recommendations.

[0058] In one embodiment, step S23 includes the following steps: Step S231: Estimate the wall thickness of the steam pipe by scanning the microstructure of the steam pipe to obtain the estimated wall thickness data of the steam pipe, wherein the steam turbine steam pipe includes high pressure steam pipe, medium pressure steam pipe and low pressure steam pipe; The estimation of the steam pipe wall thickness can be achieved through a non-contact geometric parameter inversion process based on microscopic tissue scanning images. This process can be used to obtain the actual geometric structure information of the steam pipe and to distinguish the structural differences between high-pressure, medium-pressure, and low-pressure regions. Furthermore, the estimation of the steam pipe wall thickness can be achieved by identifying the inner and outer wall boundaries of the steam pipe in the microscopic tissue scanning images and calculating their normal distances to estimate the local wall thickness. In a specific embodiment, the estimation of the steam pipe wall thickness can be achieved by automatically annotating the inner and outer contours of the pipe wall using semantic segmentation networks such as U-Net, and then converting the physical thickness using pixel spacing; or by using image gradient extremum detection combined with a calibrated scale to directly measure the pixel values ​​of the wall thickness and convert them to millimeters. This incorporates the geometric structure parameters into the aging modeling system, supporting the differentiation of heterogeneous aging behavior of high-pressure / medium-pressure / low-pressure steam pipes.

[0059] The estimated wall thickness data for steam pipes can be a set of numerical values ​​characterizing the pipe wall thickness at different axial or circumferential positions, which can be used as geometric input parameters for stress gradient modeling and regional aging rate extrapolation. Furthermore, the estimated wall thickness data can be obtained by edge detection and distance calculation of the inner and outer wall boundaries in the microstructure scan image of the steam pipe. For example, the estimated wall thickness data for steam pipes can include, but is not limited to, high-pressure steam pipe wall thickness data, medium-pressure steam pipe wall thickness data, and low-pressure steam pipe wall thickness data. High-pressure steam pipes can be high-temperature, high-pressure steam transmission pipelines connecting the boiler outlet to the high-pressure cylinder of the turbine, and can be used as one of the specific object types for steam pipe wall thickness estimation and aging analysis. Medium-pressure steam pipes can be medium-temperature, medium-pressure steam transmission pipelines connecting the reheater outlet to the medium-pressure cylinder of the turbine, and can be used as one of the specific object types for steam pipe wall thickness estimation and aging analysis. Low-pressure steam pipes can be low-temperature, low-pressure steam transmission pipelines that connect the exhaust steam from the intermediate-pressure cylinder of a steam turbine to the low-pressure cylinder. They can be used as one of the specific object types for steam pipe wall thickness estimation and aging analysis.

[0060] Step S232: Based on the operating condition time-series stress fluctuation data and the estimated wall thickness of the steam pipe, the grain growth rate range of different regions of the steam pipe is deduced to obtain the grain growth rate range of different regions of the steam pipe. The estimation of grain growth rate intervals in different regions of the steam pipe can be achieved by combining wall thickness and stress fluctuation data to estimate the possible range of grain coarsening rates in each region of the steam pipe, which can be used to quantitatively describe the spatially non-uniform aging dynamics. Furthermore, the estimation of grain growth rate intervals in different regions of the steam pipe can be achieved by inputting wall thickness data along with the stress fluctuation amplitude and frequency characteristics of the corresponding region into the grain growth model, and outputting the upper and lower bounds of the rate. In an exemplary embodiment, the estimation of grain growth rate intervals in different regions of the steam pipe can employ an Arrhenius-type grain growth equation, equating local stress to activation energy correction terms, and calculating the rate interval by combining the stress concentration factor caused by wall thickness; or by constructing a random forest regression model, using wall thickness, stress peak-to-valley difference, and temperature mean as inputs to predict the quantile interval of the grain growth rate, thereby achieving quantitative resolution of the spatially non-uniform aging dynamics of the steam pipe. The grain growth rate intervals in different regions of the steam pipe can be estimated as upper and lower bounds of the average grain size growth rate divided according to the spatial location of the steam pipe, which can be used to reflect the differences in local aging sensitivity under the combined action of structural geometry and load. Furthermore, the grain growth rate range in different regions of the steam pipe can be derived through physical models or statistical regression based on operating condition time-series stress fluctuation data and steam pipe wall thickness estimation data.

[0061] Step S233: Based on the estimated wall thickness of the steam pipe, perform stress gradient correlation coefficient analysis on the grain growth rate range to obtain the wall thickness-related stress gradient correlation coefficient; Among these methods, stress gradient correlation coefficient analysis can be used to assess the statistical correlation strength between the local stress gradient caused by wall thickness variation and the grain growth rate, and can be used to quantify the influence of geometric discontinuities on microstructure evolution. Furthermore, stress gradient correlation coefficient analysis can be achieved by calculating the statistical correlation between the wall thickness variation rate and the grain growth rate interval, generating a normalized correlation coefficient. For example, stress gradient correlation coefficient analysis can use Pearson or Spearman correlation coefficients to measure the linear or monotonic relationship between the wall thickness gradient and the median growth rate; or it can extract the latent variable correlation strength among wall thickness, stress, and growth rate through partial least squares regression, thereby quantifying the modulating effect of geometric discontinuities on microstructure evolution through stress gradients. The wall thickness-related stress gradient correlation coefficient can be a normalized correlation index describing the joint influence of steam pipe wall thickness variation and local stress gradient on grain growth rate, and can be used to provide structural-mechanical coupling constraint parameters for grain dynamic growth models. Furthermore, the wall thickness-related stress gradient correlation coefficient can be obtained by coupling wall thickness estimation data with grain growth rate intervals through multiple regression or mutual information calculation.

[0062] Step S234: Spatial orientation identification of the morphological characteristics of the aged tissue is performed to obtain grain orientation identification data; Spatial orientation identification can be a process of analyzing the crystallographic orientation distribution of grains from the morphological features of aged microstructures, which can be used to reveal the modulating effect of material texture on grain boundary migration behavior. Furthermore, spatial orientation identification can be achieved by extracting the geometric and textural information of grain boundaries from the morphological features of aged microstructures to infer their crystallographic orientation. In a specific embodiment, spatial orientation identification can be based on the distribution of grain boundary angles and the topological features of triangular grain boundaries, matching a known orientation library for classification; or a pre-trained convolutional neural network can be used to classify the orientation of microstructure image blocks, outputting Euler angles or orientation indices, thereby obtaining grain orientation information and providing a crystallographic basis for temperature-dependent grain boundary migration modeling. Grain orientation identification data can be a set of spatial crystallographic orientation identifiers of each grain within the microscopic region of the steam pipe, which can be used as a key input for temperature-dependent grain boundary migration modeling. Furthermore, grain orientation identification data can be generated by applying orientation imaging algorithms (such as EBSD simulation or deep learning orientation estimation) to the morphological features of aged microstructures.

[0063] Step S235: Based on the grain orientation identification data, extrapolate the grain boundary migration rate distribution during the aging process of the steam pipe at different temperature ranges within the grain growth rate range, and obtain the grain boundary migration rate distribution data. The simulation of grain boundary migration rate distribution during the aging process of the steam pipe at different temperature ranges can be achieved by combining grain orientation with historical operating temperatures to calculate the spatial distribution of grain boundary migration rates at different temperature ranges. This can be used to establish a temperature-orientation coupled aging kinetic model. Furthermore, the simulation can be achieved by combining grain orientation identification data with historical temperature curves of the unit, calculating the migration rate of various grain boundaries for each temperature range. For example, the simulation can be based on a grain boundary energy-orientation relationship model to calculate the migration rate of grain boundaries with specific orientations at different temperatures, and then using a weighted average to obtain the distribution; or by using time series clustering to divide the operating conditions into several temperature ranges, estimating the kernel density of the orientation-growth rate pairs within each range, thereby establishing a dual-dependence model of grain boundary migration on temperature and crystallographic orientation, improving the fidelity of aging kinetics. Grain boundary migration rate distribution data can characterize the probabilistic or deterministic distribution of migration velocities of various grain boundaries (classified by orientation) within different operating temperature ranges. It can be used to depict the dual dependence of grain boundary evolution on thermal history and crystallographic properties. Furthermore, grain boundary migration rate distribution data can be obtained by deducing from grain orientation identification data and temperature load time series using grain boundary migration theoretical models or data-driven methods.

[0064] Step S236: Perform dynamic growth analysis of steam pipe grains based on the correlation coefficient of wall thickness-related stress gradient and grain boundary migration rate distribution data to obtain dynamic growth data of steam pipe grains.

[0065] Among these methods, analyzing the dynamic growth of steam pipe grains based on the correlation coefficient of wall thickness-related stress gradient and grain boundary migration rate distribution data can integrate the stress gradient effect caused by the fusion structure with the temperature-orientation dependent grain boundary migration law, comprehensively deducing the grain evolution trajectory over time. Furthermore, this operation can be achieved by constructing a multi-physics coupled phase-field model, using the stress gradient as the driving force and the grain boundary migration rate distribution as the constitutive parameter; or by designing a graph neural network where nodes represent grains (including orientation), and edge weights are modulated by the stress gradient correlation coefficient to dynamically update grain size, thereby generating high-fidelity dynamic grain growth data with embedded geometric, mechanical, thermal, and crystallographic constraints to support subsequent accurate aging assessments.

[0066] Taking the aging assessment of the high-pressure steam pipe bend area of ​​a supercritical thermal power unit as an example, the microstructure aging analysis method of the steam turbine steam pipe in this embodiment can be as follows: Due to the manufacturing process, the wall thickness of the high-pressure steam pipe bend of a 1000MW supercritical unit is locally reduced to 85% of the design value. The system estimates the wall thickness of this area from the microstructure scanning image and identifies that the grain orientation presents a <110> texture; combined with the variable load conditions recorded by DCS in the past year, it is deduced that the grain growth rate range of this area is significantly higher than that of the straight pipe section; stress gradient correlation coefficient analysis shows that the wall thickness gradient is strongly positively correlated with the growth rate (r = 0.78); further, according to the temperature range (520–560°C), the distribution of grain boundary migration rate is deduced, and it is found that the low ΣCSL grain boundary migration is slow while the random grain boundary coarsens rapidly; finally, the above parameters are integrated to perform grain dynamic growth analysis, and the output shows that the grain size of this area will exceed the critical threshold within the next 6 months, triggering an early warning of accelerated carbide precipitation.

[0067] In one embodiment, step S3 includes the following steps: Step S31: Match the aging level output of the steam pipe according to the operating condition time-series stress fluctuation data and grain boundary precipitation difference data to obtain the aging level output data of the steam pipe.

[0068] This step involves fusing operating condition time-series stress fluctuation data that reflects the history of macroscopic operating loads with grain boundary precipitation difference data that characterizes the evolution of microstructure, establishing a mapping relationship between the two and the aging state of the steam pipe material, thereby outputting aging level output data that quantitatively characterizes the degree of equipment deterioration.

[0069] Step S32: Normalize the aging level output data of the steam pipe to obtain normalized aging level output data.

[0070] Normalization processing can be a process of converting aging level data with different dimensions, magnitudes, or distribution ranges into numerical values ​​within a unified standard range. This can be used to eliminate evaluation biases caused by unit differences, material batches, or sensor calibration, making aging levels comparable across equipment. In this embodiment, normalization processing can map the original aging level output data to a preset unified numerical range through mathematical transformation. For example, normalization processing can map the aging levels of different units to the [0, 1] range using linear scaling, or calculate quantiles based on historical data distribution to convert the current level into a percentile ranking, thereby generating an aging state input with consistent dimensions and interpretation benchmarks, supporting multi-source data fusion decision-making. Furthermore, normalization processing can include, but is not limited to, one or more of Min-Max normalization, Z-score normalization, and quantile normalization.

[0071] Step S33: Based on the aging level output normalized data, grain boundary precipitation difference data, and turbine steam pipe creep cracking risk data, steam pipe maintenance and replacement control is carried out to obtain steam pipe maintenance and replacement control data.

[0072] The maintenance and replacement control of the steam turbine duct based on aging level output normalized data, grain boundary precipitation difference data, and steam turbine duct creep cracking risk data can be achieved by integrating three types of heterogeneous data and generating specific maintenance or replacement instructions according to preset rules or models. In an exemplary embodiment, this operation can be implemented by constructing a three-dimensional decision cube, with the three axes corresponding to the normalized aging level, precipitation difference intensity, and risk probability, respectively, triggering different maintenance actions in different regions; or by training a multi-input classifier, using the three types of data as features, to output discrete maintenance strategy labels (such as "continue operation", "enhance monitoring", "planned replacement"), thereby achieving collaborative decision-making on microstructure degradation, mesoscopic state classification, and macroscopic failure risk, improving the physical consistency and predictive accuracy of maintenance strategies.

[0073] Taking a horizontal comparison of the health status of steam ducts of multiple identical units as an example, the microstructure aging analysis method for steam turbine steam ducts in this embodiment can be as follows: An aging analysis system has been deployed on all three 660MW supercritical units in a power plant. Due to different manufacturing batches, the initial grain size of the steam duct in Unit A is finer, while that in Units B and C is coarser, resulting in a systematic shift in the original aging level output. Through normalization, the aging levels of the three units are uniformly mapped to the same benchmark. Simultaneously, combining the differences in grain boundary carbide precipitation data and creep cracking risk data, the maintenance and control module determines that although Unit A's initial level is lower, its precipitation difference growth rate is fast and its risk data is approaching the threshold, suggesting priority for inspection; while Unit C's three indicators are all at low levels, allowing for an extended monitoring cycle. This decision avoids misjudgments caused by initial material differences, achieving precise differentiated operation and maintenance.

[0074] In one embodiment, step S31 includes the following steps: Step S311: Plot the stress time-series fluctuation curves from the working condition time-series stress fluctuation data to obtain the residual stress time-series fluctuation curves; The residual stress time-series fluctuation curve can be a continuous curve reflecting the dynamic changes of stress during the service of the steam pipe, plotted with time as the horizontal axis and residual stress value as the vertical axis. It can be used to provide the basic data format for transient stress feature extraction and multi-peak pattern recognition. In this embodiment, the residual stress time-series fluctuation curve can be generated from operating condition time-series stress fluctuation data through visualization or function fitting. For example, plotting the stress time-series fluctuation curve from the operating condition time-series stress fluctuation data can be achieved by connecting or fitting discrete operating condition time-series stress fluctuation data in chronological order to a continuous function curve. Furthermore, plotting the stress time-series fluctuation curve from the operating condition time-series stress fluctuation data can be achieved by using spline interpolation to smoothly reconstruct sparse sampling points to generate a high-resolution stress curve, or by directly plotting a line graph from the original sampling point sequence to retain the original data features. This transforms abstract time-series data into a continuous signal format that can be differentiated and recognized as a pattern.

[0075] Step S312: Calculate the instantaneous stress rise gradient of the residual stress time-series fluctuation curve to obtain the instantaneous stress rise gradient of the residual stress. The instantaneous stress rise gradient of residual stress can be the first derivative of the stress value with respect to time at a certain moment on the residual stress time-series fluctuation curve. It characterizes the rate of stress abrupt change and can be used to quantify the intensity of transient stress response caused by thermal shock or sudden load changes, serving as a key driving indicator for accelerated microstructural degradation. In an exemplary embodiment, the instantaneous stress rise gradient of residual stress can be obtained by numerical differentiation or sliding window slope calculation of the residual stress time-series fluctuation curve. Calculating the instantaneous stress rise gradient of the residual stress time-series fluctuation curve can be achieved by calculating the local slope at each time point on the curve and extracting the instantaneous rate of change during the stress rise phase. Furthermore, the instantaneous stress rise gradient of the residual stress time-series fluctuation curve can be calculated by using the central difference method to calculate the first derivative for equally spaced data points, or by using a Savitzky-Golay filter to estimate the derivative while reducing noise, thereby improving gradient stability. This allows for the identification of key transient characteristics of stress abrupt changes and provides a criterion for identifying multi-peak regions.

[0076] Step S313: Based on the instantaneous stress rise gradient of residual stress, perform stress multi-peak region pattern identification on the residual stress time-series fluctuation curve to obtain the residual stress intensity multi-peak region pattern. The residual stress intensity multi-peak region pattern can be a set of stress cycle regions with multiple local maxima and high upward gradients identified in the residual stress time-series fluctuation curve. This pattern can be used to locate high-risk spatiotemporal regions in steam pipes prone to local aging due to repeated alternating loads. In one specific embodiment, the residual stress intensity multi-peak region pattern can be based on a threshold set according to the instantaneous stress rise gradient of the residual stress, combined with a peak detection algorithm to identify repeated high-stress intervals. Furthermore, the residual stress intensity multi-peak region pattern can include, but is not limited to, one or more of the following: periodic high-gradient peak patterns, asymmetric stress cycle patterns, and transient impact superimposed on steady-state fluctuation patterns. Identifying the stress multi-peak region pattern of the residual stress time-series fluctuation curve based on the instantaneous stress rise gradient can be achieved by setting dual thresholds for the rise gradient and peak amplitude, filtering stress cycle intervals that meet the conditions. For example, this operation can be achieved by combining a sliding window to statistically analyze local peak density to identify high-frequency stress fluctuation segments, or by using a clustering algorithm to automatically discover typical multi-peak types by grouping gradient-peak combination features into patterns, thereby accurately locating repeated high-stress areas and avoiding misjudgment of stable stress segments.

[0077] Step S314: Dynamically match the aging level of the steam pipe based on the multi-peak region pattern of residual stress intensity to obtain dynamic classification data of the steam pipe region; The dynamic grading data for the steam pipe region can be a set of local regional states divided and assigned aging levels in the spatial-temporal dimension based on the multi-peak regional pattern of residual stress intensity. This data can provide a spatiotemporally resolved distribution of aging states, reflecting local non-uniform degradation characteristics. In this embodiment, the dynamic grading data for the steam pipe region can map the multi-peak regional pattern to the physical location of the steam pipe and generate grading by dynamically matching preset aging rules according to stress history. Dynamic matching of steam pipe aging levels based on the multi-peak regional pattern of residual stress intensity can be achieved by mapping the identified multi-peak regions to a dynamic aging level rule library according to their stress amplitude, frequency, and duration. Furthermore, this operation can be achieved by establishing a stress cycle counting model (such as rainflow counting), combining it with the Miner's rule to estimate and grade cumulative damage, or by directly outputting the regional aging level using a pre-trained temporal classification network, thereby generating a spatiotemporal distribution of aging states reflecting the local thermo-mechanical history.

[0078] Step S315: Match the aging range of the steam pipe based on the difference data of grain boundary precipitation, and perform multi-region level correction optimization on the dynamic classification data of the steam pipe region to obtain multi-region range classification correction data. The multi-region range grading correction data can be an optimized grading result after spatial consistency correction of the dynamic grading data of the steam pipe region by integrating grain boundary precipitation difference data. This can be used to achieve bidirectional verification between mechanically driven predictions and actual microstructure measurements, improving the spatial accuracy of aging grading. In an exemplary embodiment, the multi-region range grading correction data can spatially align the measured microstructure degradation range with the dynamic grading region, adjusting the grade assignment of inconsistent regions. Matching the aging range of the steam pipe based on grain boundary precipitation difference data can be achieved by aligning the spatial coordinates of significant grain boundary carbide precipitation regions in the microstructure scan image with the geometric model of the steam pipe. For example, this operation can be achieved by mapping the microstructure image to the three-dimensional coordinate system of the pipe using image registration technology, or by inferring the axial and circumferential ranges of the pipe corresponding to the scanned area based on the sensor installation position, thereby determining the actual spatial distribution range of microstructure degradation on the steam pipe. Multi-region grade correction optimization of the dynamic grading data of the steam pipe region can be achieved by comparing the spatial overlap between the dynamic grading region and the grain boundary precipitation degradation range, adjusting the grade of inconsistent regions. Furthermore, this operation can be achieved by downgrading a region if the dynamic grade is high but there is no significant precipitation, or upgrading a region if there is severe precipitation but the grade is low, or by introducing confidence weights to weight and fuse mechanical predictions and tissue measurement results to generate a corrected grade. This can integrate mechanism predictions and measured evidence to improve the spatial authenticity and reliability of aging grading.

[0079] Step S316: Match the aging level output of the steam pipe based on the dynamic grading data of the steam pipe area and the grading correction data of the multi-area range, and obtain the aging level output data of the steam pipe.

[0080] Matching the aging level output of the steam pipe based on the dynamic grading data of the steam pipe area and the grading correction data of multiple areas can be achieved by aggregating or integrating the corrected multi-area grading data to generate a final unified aging level output. Furthermore, this operation can be achieved by taking the most severe area level as the overall pipe output level and marking high-risk locations, or by outputting a data package in the form of a grading map containing the levels and confidence scores of each area, thus forming a final aging status judgment result that is both representative of the overall situation and retains local details.

[0081] Taking the assessment of localized aging of the steam pipe caused by frequent start-ups and shutdowns of a combined cycle unit as an example, the microstructural aging analysis method for the turbine steam pipe in this embodiment can be as follows: A gas-steam combined cycle unit starts and shuts down twice a day, and the main steam pipe experiences severe thermal shock. The system collects residual stress data over a month, plots time-series fluctuation curves, calculates the instantaneous upward gradient, and identifies three high-gradient multi-peak areas (located 50cm downstream of the elbow, near the weld, and at the support constraint point, respectively). Based on this model, dynamic grading is performed, and the above areas are initially determined to be "Level IV aging". At the same time, non-destructive testing images show that only the elbow and weld areas have obvious continuous precipitation of grain boundary carbides, while the precipitation at the support is slight. After multi-regional level correction and optimization, the support area is downgraded to "Level II aging", while the other two remain at Level IV. The final output is aging level data containing spatial location labels, guiding maintenance personnel to focus on inspecting or partially replacing only high-risk sections, avoiding the scrapping of the entire pipe.

[0082] In one embodiment, step S316 is followed by the following steps: Step S3161: Obtain the design service life and service duration data of the steam turbine duct to be analyzed, and calculate the percentage of the remaining design service life of the duct. The design service life of the turbine's steam duct to be analyzed can be the expected safe service life determined during the design phase based on material properties, operating parameters, and safety margins. This can serve as a benchmark timescale for assessing remaining service life and aging risks. Furthermore, the service duration data can be the cumulative operating time of the steam duct since commissioning, which can be extracted from the power plant's equipment ledger or operation log system and used to calculate the equipment's current position in its life cycle. For example, the percentage of remaining design service life of the steam duct can be the proportion of remaining serviceable time to the original design service life, calculated by dividing (design service life - service duration) by the design service life. This can be used to quantify the aging stage of the equipment throughout its life cycle, providing a basis for weight adjustment.

[0083] To obtain the design service life and actual service time data of the steam turbine's ductwork to be analyzed, the remaining design life percentage of the ductwork can be calculated. This can be achieved by reading the design life and cumulative operating time from equipment files and operation logs, and then performing the percentage calculation. In an exemplary embodiment, this operation can be achieved by automatically linking the equipment's unique code to the design parameters and operating log in the ERP system; or by manually entering the design life and automatically converting it to years using the DCS's cumulative operating hours. This allows for the establishment of a benchmark linking aging assessment with the entire equipment lifecycle.

[0084] Step S3162: Based on the remaining design life percentage data of the steam pipe, perform life loss weight correction on the initially obtained steam pipe aging level output data to obtain the life correction coefficient. The life loss weight correction can be a mapping rule that applies a nonlinear adjustment to the initial aging level based on the remaining lifespan percentage, which can be used to reflect the risk differences of the same micro-deterioration at different service stages. In a specific embodiment, the life correction coefficient can be a numerical factor used to modulate the aging level, generated by mapping the remaining design lifespan percentage. The remaining lifespan percentage can be converted into a correction weight through a preset function (such as exponential decay or piecewise linearity), which can be used to dynamically adjust the aging level to reflect the cumulative damage effect.

[0085] The life loss weight correction is performed on the initially obtained aging level output data of the steam duct based on the remaining design life percentage data of the steam duct. This can be achieved by inputting the remaining life percentage into a preset weight function to generate a correction coefficient, which is then applied to the initial aging level. Furthermore, this operation can be performed using an exponential function, where the correction coefficient equals one divided by one plus the product of the material-related parameters α and e raised to the power of β (1 minus the remaining percentage); or a piecewise rule can be used, for example, a coefficient of 1.0 when the remaining life is greater than 70%, 1.2 when it is between 30% and 70%, and 1.5 when it is below 30%. This allows the aging level to reflect the non-linear amplification effect of risk sensitivity during the service phase.

[0086] Step S3163: Combine the over-temperature alarm record data in the operation history of the steam pipe, perform over-temperature aging gain correction on the aging level output data after life correction, and obtain the over-temperature corrected aging level output. The operating history of the steam pipe can be a collection of all operating events, parameter records, and abnormal alarms since its commissioning, providing contextual information for accelerated aging events such as overheating. Furthermore, overheating alarm records can be historical event data recorded in the DCS or monitoring system showing that the steam pipe wall temperature or steam temperature exceeded a set threshold. A time-stamped list of overheating events can be extracted from the power plant's historical database, quantifying the accelerated contribution of temperature anomalies to material aging. For example, the aging level output data after lifespan correction can be an intermediate aging level result corrected for lifespan loss weights. The initial aging level can be multiplied by the lifespan correction coefficient or adjusted according to rules, serving as the input basis for overheating correction. Overheating aging gain correction can be a correction mechanism that nonlinearly enhances the aging level based on the frequency, duration, and overshoot amplitude of overheating events, quantifying the nonlinear impact of high-temperature anomalies on accelerated microstructure degradation. The overheating corrected aging level output can be the final intermediate aging level result after dual correction of lifespan stage and overheating history. An overheating gain factor can be superimposed on the lifespan correction, comprehensively reflecting the true aging state under the combined effects of time accumulation and event impact.

[0087] By combining over-temperature alarm records from the operating history of the steam pipe, the over-temperature aging gain correction is applied to the aging level output data after life correction. This can be done by statistically analyzing the number, duration, and overshoot of over-temperature events, calculating the equivalent aging acceleration factor, and adding it to the aging level. In one specific embodiment, this operation can construct an over-temperature equivalent life loss model, where each over-temperature event is converted into an equivalent number of days of normal temperature operation according to the Arrhenius equation, and the gain coefficient is calculated by summing them up; or an over-temperature event level can be set, such as +0.1 for minor over-temperature, +0.3 for moderate over-temperature, and +0.5 for severe over-temperature, weighted by the highest single or cumulative number of events. This allows for the quantification of the nonlinear accelerating effect of operational anomalies on material aging, improving the accuracy of risk assessment.

[0088] Step S3164: Update the over-temperature correction aging level output to the final steam pipe aging level output data.

[0089] The final aging level output data of the steam pipe can be an authoritative aging level for maintenance decisions after being corrected for both lifespan and over-temperature, and can serve as the final basis for generating predictive maintenance strategies.

[0090] Updating the over-temperature corrected aging level output to the final steam pipe aging level output data can replace the original initial aging level with the over-temperature corrected result as input for subsequent maintenance and control. Furthermore, this operation can directly overwrite the original data and mark the correction source; or retain multiple versions of level data but designate the over-temperature corrected version as the primary decision version, thereby forming a comprehensive aging assessment result that integrates physical conditions, time dimensions, and event impacts.

[0091] Taking the risk reassessment of the steam pipe of an aging unit after multiple over-temperature events as an example, the microstructure aging analysis method of the steam pipe in this embodiment can be as follows: The steam pipe of a subcritical unit that has been in service for 18 years has a designed life of 20 years, with only 10% of its life remaining. The initial aging level assessment is level three, mainly due to the intermittent precipitation of grain boundary carbides. The system reads the operating history and finds that there have been 5 over-temperature alarms in the past two years, 2 of which lasted for more than 4 hours. The life correction module improves the level three to level 3.6 (coefficient 1.2), and the over-temperature correction module further adds a gain of 0.4, finally outputting a level four aging level. This result triggers an immediate shutdown and inspection order, avoiding potential pipe rupture accidents caused by underestimating the dual risks of aging and over-temperature due to relying solely on the current microstructure.

[0092] In addition, refer to Figure 2 To achieve the above objectives, embodiments of the present invention also provide a microstructure aging analysis system for steam turbine guide pipes, the system comprising: The dual-source sensing module 10 is used to collect the microstructure of the steam turbine duct under operating conditions using non-destructive testing equipment, and obtain a microstructure scanning image of the duct; based on the microstructure scanning image of the duct, the residual stress of the steam turbine duct under service conditions is monitored using an ultrasonic stress sensor, and the residual stress monitoring data of the duct is obtained. The risk quantification module 20 is used to match the time-series stress fluctuation of temperature load based on the residual stress monitoring data of the steam duct during service, and obtain the time-series stress fluctuation data under operating conditions; to quantify the difference in the degree of precipitation of grain boundary carbides based on the time-series stress fluctuation data under operating conditions, and obtain the difference data of precipitation of grain boundary structure; and to estimate the creep cracking risk of the steam duct under the turbine based on the difference data of precipitation of grain boundary structure, and obtain the creep cracking risk data of the steam duct under the turbine. The decision control module 30 is used to match the aging level output of the steam pipe according to the operating condition time-series stress fluctuation data and grain boundary structure precipitation difference data to obtain the aging level output data of the steam pipe; and to control the maintenance and replacement of the steam pipe based on the aging level output normalized data and the steam turbine steam pipe creep cracking risk data to obtain the steam pipe maintenance and replacement control data. The cloud deployment module 40 is used to design a system for analyzing the aging status of the turbine steam pipe based on the output data of the aging level of the steam pipe and the maintenance and replacement control data of the steam pipe. It obtains the microstructure aging analysis firmware of the turbine steam pipe and sends the microstructure aging analysis firmware of the turbine steam pipe to the operation and maintenance cloud platform to execute the microstructure aging analysis method of the turbine steam pipe.

[0093] Other embodiments or specific implementations of the turbine steam pipe microstructure aging analysis system described in this invention can be referred to the above-mentioned method embodiments, and will not be repeated here.

[0094] Furthermore, to achieve the above objectives, embodiments of the present invention also provide a steam turbine duct microstructure aging analysis device, the device comprising: a memory, a processor, and a steam turbine duct microstructure aging analysis program stored in the memory and executable on the processor, the steam turbine duct microstructure aging analysis program being configured to implement the steps of the steam turbine duct microstructure aging analysis method as described above.

[0095] In addition, to achieve the above objectives, embodiments of the present invention also provide a computer-readable storage medium storing a steam turbine duct microstructure aging analysis program, wherein when the steam turbine duct microstructure aging analysis program is executed by a processor, the steam turbine duct microstructure aging analysis method as described above is implemented.

[0096] It is understood that the above embodiments are merely exemplary implementations used to illustrate the principles of the present invention, and the present invention is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A method for aging analysis of the microstructure of steam turbine guide pipes, characterized in that, The method includes: Step S1: The microstructure of the steam turbine's steam pipe under operating conditions is collected using non-destructive testing equipment to obtain a microstructure scanning image of the steam pipe; based on the microstructure scanning image of the steam pipe, residual stress is monitored in the steam turbine's steam pipe under service conditions using an ultrasonic stress sensor to obtain residual stress monitoring data of the steam pipe in service. Step S2: Match the time-series stress fluctuations of temperature loads based on the residual stress monitoring data of the steam turbine duct during service to obtain the time-series stress fluctuation data under operating conditions; quantify the difference in the degree of grain boundary carbide precipitation based on the time-series stress fluctuation data under operating conditions to obtain the difference data in grain boundary structure precipitation; estimate the creep cracking risk of the steam turbine duct based on the difference data in grain boundary structure precipitation to obtain the creep cracking risk data of the steam turbine duct. Step S3: Match the aging level output of the steam pipe according to the operating condition time-series stress fluctuation data and grain boundary precipitation difference data to obtain the steam pipe aging level output data; Based on the aging level output normalized data and the steam turbine steam pipe creep cracking risk data, adjust the steam pipe maintenance and replacement to obtain the steam pipe maintenance and replacement control data. Step S4: Based on the aging level output data of the steam duct and the maintenance and replacement control data of the steam duct, design the steam turbine steam duct aging status analysis system, obtain the steam turbine steam duct microstructure aging analysis firmware, and send the steam turbine steam duct microstructure aging analysis firmware to the operation and maintenance cloud platform to execute the steam turbine steam duct microstructure aging analysis method.

2. The method for aging analysis of the microstructure of steam turbine guide pipes according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Deploy electronic non-destructive scanning equipment and ultrasonic stress sensors on the outside of the steam turbine's steam pipe to be inspected; Step S12: The microstructure of the steam turbine's steam pipe under operating conditions is collected using an electronic non-destructive scanning device to obtain a microstructure scanning image of the steam pipe; Step S13: Perform regional segmentation processing on the scanned image of the steam pipe microstructure to obtain a continuous segmented image of the steam pipe microstructure; Step S14: Based on the continuous block image of the microstructure of the steam pipe, and by using an ultrasonic stress sensor to monitor the residual stress of the steam pipe in service, the residual stress monitoring data of the steam pipe in service is obtained.

3. The method for aging analysis of the microstructure of steam turbine guide pipes according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Extract the morphological features of aged tissue from the microscopic tissue scanning image of the steam pipe to obtain the morphological features of aged tissue; Step S22: Based on the residual stress monitoring data of the steam pipe during service, the morphological characteristics of the aged microstructure are matched with the time-series stress fluctuations under operating conditions to obtain the time-series stress fluctuation data under operating conditions. Step S23: Based on the operating condition time-series stress fluctuation data and aging microstructure morphology characteristics, perform dynamic growth analysis of steam pipe grains to obtain dynamic growth data of steam pipe grains; Step S24: Quantify the difference in the degree of grain boundary carbide precipitation in the dynamic growth data of the steam pipe grains to obtain the difference data of grain boundary structure precipitation. Step S25: Estimate the risk of creep cracking in the turbine steam pipe based on the grain boundary precipitation difference data and the operating condition time-series stress fluctuation data, and obtain the creep cracking risk data of the turbine steam pipe.

4. The method for aging analysis of the microstructure of steam turbine guide pipes according to claim 3, characterized in that, Step S23 includes the following steps: Step S231: Estimate the wall thickness of the steam turbine through the microstructure scanning image of the steam turbine, and obtain the estimated wall thickness data of the steam turbine, wherein the steam turbine steam turbine includes a high-pressure steam turbine, a medium-pressure steam turbine, and a low-pressure steam turbine. Step S232: Based on the operating condition time-series stress fluctuation data and the estimated wall thickness of the steam pipe, the grain growth rate range of different regions of the steam pipe is deduced to obtain the grain growth rate range of different regions of the steam pipe. Step S233: Based on the estimated wall thickness of the steam pipe, perform stress gradient correlation coefficient analysis on the grain growth rate range to obtain the wall thickness-related stress gradient correlation coefficient; Step S234: Spatial orientation identification of the morphological characteristics of the aged tissue is performed to obtain grain orientation identification data; Step S235: Based on the grain orientation identification data, extrapolate the grain boundary migration rate distribution during the aging process of the steam pipe at different temperature ranges within the grain growth rate range, and obtain the grain boundary migration rate distribution data. Step S236: Perform dynamic growth analysis of steam pipe grains based on the correlation coefficient of wall thickness-related stress gradient and grain boundary migration rate distribution data to obtain dynamic growth data of steam pipe grains.

5. The method for aging analysis of the microstructure of steam turbine guide pipes according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Match the aging level output of the steam pipe according to the stress fluctuation data and grain boundary precipitation difference data under the working condition to obtain the aging level output data of the steam pipe; Step S32: Normalize the aging level output data of the steam pipe to obtain normalized aging level output data; Step S33: Based on the aging level output normalized data, grain boundary precipitation difference data, and turbine steam pipe creep cracking risk data, steam pipe maintenance and replacement control is carried out to obtain steam pipe maintenance and replacement control data.

6. The method for aging analysis of the microstructure of steam turbine guide pipes according to claim 5, characterized in that, Step S31 includes the following steps: Step S311: Plot the stress time-series fluctuation curves from the working condition time-series stress fluctuation data to obtain the residual stress time-series fluctuation curves; Step S312: Calculate the instantaneous stress rise gradient of the residual stress time-series fluctuation curve to obtain the instantaneous stress rise gradient of the residual stress. Step S313: Based on the instantaneous stress rise gradient of residual stress, perform stress multi-peak region pattern identification on the residual stress time-series fluctuation curve to obtain the residual stress intensity multi-peak region pattern. Step S314: Dynamically match the aging level of the steam pipe based on the multi-peak region pattern of residual stress intensity to obtain dynamic classification data of the steam pipe region; Step S315: Match the aging range of the steam pipe based on the difference data of grain boundary precipitation, and perform multi-region level correction optimization on the dynamic classification data of the steam pipe region to obtain multi-region range classification correction data. Step S316: Match the aging level output of the steam pipe based on the dynamic grading data of the steam pipe area and the grading correction data of the multi-area range to obtain the aging level output data of the steam pipe.

7. The method for aging analysis of the microstructure of steam turbine guide pipes according to claim 6, characterized in that, Following step S316, the following steps are also included: Step S3161: Obtain the design service life and service duration data of the steam turbine duct to be analyzed, and calculate the percentage of the remaining design service life of the duct. Step S3162: Based on the remaining design life percentage data of the steam pipe, perform life loss weight correction on the initially obtained steam pipe aging level output data to obtain the life correction coefficient. Step S3163: Combine the over-temperature alarm record data in the operation history of the steam pipe, perform over-temperature aging gain correction on the aging level output data after life correction, and obtain the over-temperature corrected aging level output. Step S3164: Update the over-temperature correction aging level output to the final steam pipe aging level output data.

8. A microstructure aging analysis system for steam turbine guide pipes, characterized in that, The system includes: The dual-source sensing module is used to collect the microstructure of the steam turbine's steam pipe under operating conditions using non-destructive testing equipment, and obtain a microstructure scanning image of the steam pipe; based on the microstructure scanning image of the steam pipe, residual stress is monitored in the steam turbine's steam pipe under service conditions using an ultrasonic stress sensor, and residual stress monitoring data of the steam pipe in service is obtained. The risk quantification module is used to match the time-series stress fluctuations of temperature load based on the residual stress monitoring data of the steam turbine duct during service, and obtain the time-series stress fluctuation data under operating conditions; to quantify the difference in the degree of precipitation of carbides at grain boundaries based on the time-series stress fluctuation data under operating conditions, and obtain the difference data of precipitation of grain boundaries; and to estimate the risk of creep cracking of the steam turbine duct based on the difference data of precipitation of grain boundaries, and obtain the risk data of creep cracking of the steam turbine duct. The decision control module is used to match the aging level output of the steam pipe according to the operating condition time-series stress fluctuation data and grain boundary precipitation difference data to obtain the aging level output data of the steam pipe; and to control the maintenance and replacement of the steam pipe based on the aging level output normalized data and the steam turbine steam pipe creep cracking risk data to obtain the steam pipe maintenance and replacement control data. The cloud deployment module is used to design a system for analyzing the aging status of turbine steam pipes based on the output data of steam pipe aging level and the control data of steam pipe maintenance and replacement. It obtains the microstructure aging analysis firmware of the turbine steam pipe and sends the microstructure aging analysis firmware of the turbine steam pipe to the operation and maintenance cloud platform to execute the microstructure aging analysis method of the turbine steam pipe.

9. A device for analyzing the microstructure and aging of steam turbine guide pipes, characterized in that, The device includes: a memory, a processor, and a turbine duct microstructure aging analysis program stored in the memory and executable on the processor, the turbine duct microstructure aging analysis program being configured to implement the steps of the turbine duct microstructure aging analysis method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a microstructure aging analysis program for a steam turbine duct, which, when executed by a processor, implements the steps of the microstructure aging analysis method for a steam turbine duct as described in any one of claims 1 to 7.

Images