Stop valve control method for power station boiler

Abstract

The invention provides a stop valve control method for a power station boiler, which belongs to the technical field of intelligent control, and comprises the following steps: embedding an FBG (Fiber Bragg Grating) optical fiber sensor array in a key stress area of a valve body to collect a temperature gradient delta T in real time, and arranging an AE sensor on a valve shell to capture a characteristic frequency band signal of microcrack propagation; the thermal stress prediction model dynamically calculates the real-time thermal stress of the valve body and dynamically analyzes the real-time abrupt change energy of the valve according to the signal analysis model, the multi-mode control strategy is automatically switched according to the operation working condition of the power station boiler in combination with the real-time thermal stress and the abrupt change energy, the valve action speed is dynamically adjusted, and the valve action speed can be dynamically adjusted; and the safety, the reliability and the intelligent level of valve operation are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a stop valve control method for a power station boiler. Background Art

[0002] Traditional globe valves operate at a fixed speed. Sudden temperature fluctuations (>300°C / min) during boiler startup and shutdown trigger localized thermal stress concentration on the valve body, leading to microcracks on the sealing surface. (A typical example: a 660MW supercritical unit experienced a 47% valve replacement rate over three years.) The existing DCS system only monitors valve position and opening without real-time coupling to steam temperature field data, making it impossible to predict thermal shock risks. Excessively rapid valve movement during cold startup causes steam flow fluctuations (±15% of the design value), triggering thermal fatigue in the turbine.

[0003] Therefore, the present invention proposes a stop valve control method for a power station boiler. Summary of the Invention

[0004] The present invention provides a stop valve control method for a power station boiler, which is used to solve the above-mentioned technical problems.

[0005] The present invention provides a stop valve control method for a power station boiler, comprising: An FBG fiber optic sensor array is embedded in the key stress area of ​​the valve body to collect the temperature gradient ΔT in real time. At the same time, an AE sensor is arranged on the valve housing to capture the characteristic frequency band signal of microcrack extension. Dynamically calculate the real-time thermal stress of the valve body based on the thermal stress prediction model and dynamically analyze the real-time sudden energy of the valve based on the signal analysis model; Automatically switch multi-modal control strategies and dynamically adjust valve actuation speed based on the operating conditions of power plant boilers and in combination with real-time thermal stress and sudden energy changes; The arrangement of the FBG optical fiber sensor array satisfies: ≥8 measuring points are evenly distributed on the circumference of the valve seat cone surface, and the sensor diameter is ≤0.5mm; Three layers of annular measuring points are arranged axially on the valve stem neck, and the layer spacing is 0.8-1.2 times the valve stem diameter.

[0006] Preferably, the switching conditions of the multimodal control strategy include: When the energy mutation of the acoustic emission signal in the 20-150kHz frequency band is detected to be ≥5dB, the system is forced to enter the creep mode; When the steam temperature change rate dT / dt>300℃ / min, the safety priority mode is activated; The specific step-by-step deceleration control in creep mode is as follows: Opening 0-30% stage: speed limit ≤ 2% / min, valve body temperature rise rate ≤ 5℃ / min; Opening 30-70% stage: speed limit ≤ 5% / min, valve body temperature rise rate ≤ 10℃ / min; The speed limit is lifted when the opening is >70%.

[0007] Preferably, the thermal stress prediction model is as follows:

[0008] Among them, V1 represents the maximum allowable speed of the valve; 1.8 represents the basic speed coefficient; represents the stress change damping coefficient; 0.6 represents the temperature compensation coefficient; 120 represents the temperature sensitive characteristic value; represents the temperature gradient; Indicates the real-time thermal stress of the valve body Differentiable function with respect to time t.

[0009] Preferably, an FBG optical fiber sensor array is embedded in the key stress area of ​​the valve body, comprising: Use sequential coupling to simulate the thermal-structural transient response of the valve seat cone, define transient thermal boundaries and structural constraints, and cover the actual operating conditions of the valve seat; The stress linearization tool is used to define the surface path at the geometric mutation of the cone surface, decompose it into membrane stress and bending stress, and then calculate the stress according to the stress linearization tool. filter The grid area is marked, where represents the maximum stress on the path, represents membrane stress; represents the stress concentration factor; The mesh of the cone surface is encrypted and verified through mesh convergence analysis; Perform a local search on the marked area and combine it with transient time history analysis to determine whether the stress extreme point distribution is 8 local peaks; If yes, the corresponding marked area will be retained; otherwise, the judgment analysis will continue for the next marked area; The FBG fiber optic sensor array is arranged based on all the reserved areas.

[0010] Preferably, the conical surface is meshed and verified by mesh convergence analysis, including: Defining a grid density gradient along the normal and circumferential directions of the cone surface, and determining the geometric features of each grid in the cone surface, wherein the geometric features are gradient geometric features, which are divided into core geometric features, transition geometric features, and peripheral geometric features; Match the verification conditions based on each geometric feature from the geometry-mesh verification comparison table, and control the corresponding mesh to converge towards the matching verification condition direction; Among them, when it is a core geometric feature, the corresponding grid size is set to L0 / 5~L0 / 3, where L0 represents the basic grid size of the peripheral geometric feature; When it is a transitional geometric feature, the corresponding grid is gradually transitioned with a preset geometric scaling factor; When it is a peripheral geometric feature, the corresponding mesh will be deleted.

[0011] Preferably, the mesh type under the core geometric feature is a hexahedron-dominated hybrid mesh, and the mesh type under the transition geometric feature is a tetrahedron mesh.

[0012] Preferably, dynamically analyzing the real-time sudden change energy of the valve according to the signal analysis model includes: Establish a finite element model of the valve with an initial crack, simulate the crack propagation process, and extract the theoretical spectrum of the acoustic emission signal; Prefabricate microcracks in valve specimens, apply cyclic loads to simulate actual working conditions, and collect the experimental spectrum of acoustic emission signals; The cross-correlation-denoising algorithm is used to select characteristic frequency bands (f1, f2) whose overlap between the theoretical spectrum and the experimental spectrum is greater than or equal to the preset threshold; Wavelet packet decomposition is used to extract the energy time series within the characteristic frequency band; Use long short-term memory network to learn the dynamic law of energy sequence to predict the energy at the next moment and calculate the deviation ,in, For the next moment The predicted energy and actual energy; Set the sliding time window to calculate the energy mutation rate within the window; When the energy mutation rate is greater than or equal to a first threshold and the deviation is greater than or equal to a second threshold, it is determined to be a microcrack propagation energy mutation.

[0013] Preferably, setting a sliding time window to calculate the energy mutation rate within the window includes:

[0014] in, represents the energy mutation rate; Represents the signal mean under the corresponding sliding time window; Represents the noise mean under the corresponding sliding time window; Indicates the maximum signal amplitude among all signals in the corresponding sliding time window; Indicates the minimum signal amplitude among all signals in the corresponding sliding time window; Represents the average amplitude of all signals in the corresponding sliding time window; It represents the maximum signal amplitude among all signals in the corresponding sliding time window after removing outliers. It represents the minimum signal amplitude among all signals in the corresponding sliding time window after removing outliers. It represents the average amplitude of the remaining signal after removing outliers from all signals in the corresponding sliding time window.

[0015] Preferably, it also includes: Construct a damage evolution model for valve body materials; Acquire the current surface image of the valve body and quantitatively analyze it to obtain the surface vector; Inputting the surface vector into the valve body material damage evolution model to predict the damage factor; When the predicted damage factor is greater than a preset factor, the valve action of the corresponding valve body is triggered to be suspended.

[0016] Preferably, obtaining a current surface image of the valve body and quantitatively analyzing to obtain a surface vector includes: A structured light 3D scanner and a short-wave infrared camera are used to synchronously collect 3D point cloud data and thermal imaging images of material defects on the valve body surface; Extract the local surface curvature pi, normal vector deviation Δni, and corrosion depth hj from the 3D point cloud. At the same time, perform multi-scale wavelet transform on the material defect thermal image to obtain the 3D feature vector W3, and obtain the surface vector Vk=[pi,Δni,hj,W3]; The calculation formula of corrosion depth hj is as follows:

[0017] in, Indicates the number of points within the localized corrosion area; Indicates a localized corrosion area; represents the pth sub-region in the local corrosion region; They represent the measured depth after and before corrosion of the p-th sub-region respectively; represents the corrosion volume of the jth local corrosion area; The surface area of ​​the jth localized corrosion region.

[0018] Compared with the prior art, the present invention has the following advantages: By placing FBG fiber optic sensor arrays and AE sensors at key locations on the valve body, they collect temperature gradients and acoustic emission signals from microcracks, respectively. A thermal stress prediction model is then used to calculate real-time thermal stress, and a signal analysis model is used to analyze the sudden change energy. Based on the power plant boiler's operating conditions, combined with thermal stress and sudden change energy, a multimodal control strategy is automatically switched to dynamically adjust the valve's actuation speed. This ensures valve operation efficiency under complex operating conditions while preventing equipment damage caused by factors such as thermal stress and microcracks, thereby improving the safety, reliability, and intelligence of valve operation.

[0019] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0020] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 The present invention is a flowchart of a stop valve control method for a power station boiler according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0023] The present invention provides a stop valve control method for a power station boiler, such as Figure 1 Shown, including: Step 1: Embed an FBG fiber optic sensor array in the key stress area of ​​the valve body to collect the temperature gradient ΔT in real time. At the same time, an AE sensor is placed on the valve housing to capture the characteristic frequency band signal of microcrack propagation. Step 2: Dynamically calculate the real-time thermal stress of the valve body based on the thermal stress prediction model and dynamically analyze the real-time mutation energy of the valve based on the signal analysis model; Step 3: Automatically switch the multi-modal control strategy based on the operating conditions of the power plant boiler and in combination with real-time thermal stress and sudden energy changes, and dynamically adjust the valve operation speed; The arrangement of the FBG optical fiber sensor array satisfies: ≥8 measuring points are evenly distributed on the circumference of the valve seat cone surface, and the sensor diameter is ≤0.5mm; Three layers of annular measuring points are arranged axially on the valve stem neck, and the layer spacing is 0.8-1.2 times the valve stem diameter.

[0024] Preferably, the switching conditions of the multimodal control strategy include: When the energy mutation of the acoustic emission signal in the 20-150kHz frequency band is detected to be ≥5dB, the system is forced to enter the creep mode; When the steam temperature change rate dT / dt>300℃ / min, the safety priority mode is activated; The specific step-by-step deceleration control in creep mode is as follows: Opening 0-30% stage: speed limit ≤ 2% / min, valve body temperature rise rate ≤ 5℃ / min; Opening 30-70% stage: speed limit ≤ 5% / min, valve body temperature rise rate ≤ 10℃ / min; The speed limit is lifted when the opening is >70%.

[0025] Preferably, the thermal stress prediction model is as follows:

[0026] Among them, V1 represents the maximum allowable speed of the valve; 1.8 represents the basic speed coefficient; represents the stress change damping coefficient; 0.6 represents the temperature compensation coefficient; 120 represents the temperature sensitive characteristic value; represents the temperature gradient; Indicates the real-time thermal stress of the valve body Differentiable function with respect to time t.

[0027] In this embodiment, the temperature gradient ΔT refers to the rate of change of temperature in space. For example, different parts of the valve body may have temperature differences due to heat transfer, self-heating, etc. ΔT is a physical quantity that describes the speed and direction of this temperature change. For example, near a boiler valve, there is a significant temperature gradient between the high-temperature steam side and the low-temperature shell side.

[0028] Sudden energy is the abnormal energy released during valve operation due to microcrack expansion, sudden changes in operating conditions, etc. It is obtained through the AE sensor signal analysis model. For example, before metal fatigue fracture, the rapid expansion of microcracks will release sudden energy, which can be detected and analyzed.

[0029] The multimodal control strategy is a set of multiple control modes that are automatically switched according to different operating conditions and monitoring parameters.

[0030] Creep mode is a control mode that takes into account the creep characteristics of materials (slow plastic deformation of materials at high temperatures). During long-term operation in high-temperature, high-pressure pipelines, the material will creep. In this mode, valve operation speed is limited to avoid accelerated creep damage.

[0031] Safety priority mode is a control mode that prioritizes equipment and system safety when parameters such as the steam temperature change rate trigger safety risks. For example, in a chemical plant, when medium parameters are abnormal, the equipment automatically switches to a safe state for operation.

[0032] The beneficial effects of this technical solution are as follows: by placing FBG fiber optic sensor arrays and AE sensors in key locations on the valve body, they respectively collect temperature gradients and microcrack acoustic emission signals. A thermal stress prediction model is then used to calculate real-time thermal stress, and a signal analysis model is used to analyze mutation energy. Based on the power plant boiler's operating conditions, combined with thermal stress and mutation energy, a multimodal control strategy is automatically switched to dynamically adjust the valve's actuation speed. This ensures valve operation efficiency under complex operating conditions while preventing equipment damage caused by factors such as thermal stress and microcracks, thereby improving the safety, reliability, and intelligence of valve operation.

[0033] The present invention provides a stop valve control method for a power station boiler, wherein an FBG optical fiber sensor array is embedded in a key stress area of ​​the valve body, comprising: Use sequential coupling to simulate the thermal-structural transient response of the valve seat cone, define transient thermal boundaries and structural constraints, and cover the actual operating conditions of the valve seat; The stress linearization tool is used to define the surface path at the geometric mutation of the cone surface, decompose it into membrane stress and bending stress, and then calculate the stress according to the stress linearization tool. filter The grid area is marked, where represents the maximum stress on the path, represents membrane stress; represents the stress concentration factor; The mesh of the cone surface is encrypted and verified through mesh convergence analysis; Perform a local search on the marked area and combine it with transient time history analysis to determine whether the stress extreme point distribution is 8 local peaks; If yes, the corresponding marked area will be retained; otherwise, the judgment analysis will continue for the next marked area; The FBG fiber optic sensor array is arranged based on all the reserved areas.

[0034] The beneficial effects of the above technical solution are: The present invention provides a stop valve control method for a power station boiler, which performs mesh encryption on a cone surface and verifies the mesh convergence through mesh convergence analysis, including: Defining a grid density gradient along the normal and circumferential directions of the cone surface, and determining the geometric features of each grid in the cone surface, wherein the geometric features are gradient geometric features, which are divided into core geometric features, transition geometric features, and peripheral geometric features; Match the verification conditions based on each geometric feature from the geometry-mesh verification comparison table, and control the corresponding mesh to converge towards the matching verification condition direction; Among them, when it is a core geometric feature, the corresponding grid size is set to L0 / 5~L0 / 3, where L0 represents the basic grid size of the peripheral geometric feature; When it is a transitional geometric feature, the corresponding grid is gradually transitioned with a preset geometric scaling factor; When it is a peripheral geometric feature, the corresponding mesh will be deleted.

[0035] Preferably, the mesh type under the core geometric feature is a hexahedron-dominated hybrid mesh, and the mesh type under the transition geometric feature is a tetrahedron mesh.

[0036] In this example, sequential coupling is a step-by-step approach to multi-physics coupled analysis. First, thermal analysis is performed to obtain temperature field results, which are then used as loads in structural analysis. For example, when analyzing an engine turbine blade, the heat distribution generated by combustion is first calculated (thermal analysis), and then the blade stress and deformation are calculated using the temperature as a load (structural analysis). This is used here to simulate the thermal-structural transient response of the valve seat cone.

[0037] Transient thermal boundaries refer to heat transfer boundary conditions that vary over time. For example, when a valve is started and shut down, the surface in contact with the high-temperature medium changes temperature over time, and the boundary conditions for heat convection and radiation also change accordingly. This is a transient thermal boundary.

[0038] Structural constraints are restrictions placed on the valve structure to simulate the mounting and support conditions encountered during actual operation. For example, limiting the displacement of the flange connecting the valve to the pipe to ensure that the analysis conforms to the actual forces is a structural constraint.

[0039] Stress Linearization Tool: This analysis tool decomposes complex stress states into membrane stress and bending stress. Membrane stress is a uniformly distributed stress (such as the hoop stress in a pressure vessel shell), while bending stress is a linearly distributed stress (such as the stress in a bent beam). This tool can be used to decompose stress at geometrically abrupt locations (such as the edge of a valve seat cone), facilitating evaluation.

[0040] Membrane stress (Nm): Stress uniformly distributed across a cross section, caused by a uniform load, which results in uniform deformation of the material. For example, when a thin-walled container is subjected to internal pressure, the hoop stress on the wall approximates the membrane stress.

[0041] Bending stress is caused by bending loads. It is a stress distributed linearly along the cross section, causing the material to bend and deform. For example, when a beam is subjected to a transverse force, the stress in the upper section is compressed and the lower section is tensile.

[0042] Mesh refinement refines the model's local mesh for more accurate calculations. In areas of stress concentration with large gradients, a coarse mesh can't accurately calculate them. Mesh refinement, for example, improves stress calculation accuracy by refining the mesh at sudden changes in the valve's conical surface.

[0043] Mesh convergence analysis tests the extent to which the mesh can be refined to the point where the calculated results no longer change significantly. By continuously refining the mesh, we can see whether the stress, deformation, and other results are stable. If they are stable, the mesh is dense enough and the results are reliable.

[0044] Transient time history analysis analyzes the response of a structure over time. For example, when a valve opens and closes, the stress changes over time, allowing us to identify the time and location of extreme stress values.

[0045] The beneficial effects of this technical solution are: through a sequential coupled simulation of thermal-structural response → stress linearization to identify high-risk areas → mesh refinement and convergence verification → transient history analysis to confirm extreme value distributions → screening of critical areas → sensor placement, it accurately locates high-risk areas of stress concentration on the valve seat cone surface, ensuring that subsequent FBG fiber optic sensor arrays are placed in truly critical locations. This allows for precise monitoring of parameters such as stress and temperature during valve operation, identifying potential failure risks in advance, improving valve design reliability and intelligent operation and maintenance, and ensuring safe and stable valve operation under complex operating conditions.

[0046] The present invention provides a stop valve control method for a power station boiler, which dynamically analyzes the real-time sudden change energy of the valve according to a signal analysis model, including: Establish a finite element model of the valve with an initial crack, simulate the crack propagation process, and extract the theoretical spectrum of the acoustic emission signal; Prefabricate microcracks in valve specimens, apply cyclic loads to simulate actual working conditions, and collect the experimental spectrum of acoustic emission signals; The cross-correlation-denoising algorithm is used to select characteristic frequency bands (f1, f2) whose overlap between the theoretical spectrum and the experimental spectrum is greater than or equal to the preset threshold; Wavelet packet decomposition is used to extract the energy time series within the characteristic frequency band; Use long short-term memory network to learn the dynamic law of energy sequence to predict the energy at the next moment and calculate the deviation ,in, For the next moment The predicted energy and actual energy; Set the sliding time window to calculate the energy mutation rate within the window; When the energy mutation rate is greater than or equal to a first threshold and the deviation is greater than or equal to a second threshold, it is determined to be a microcrack propagation energy mutation.

[0047] Preferably, setting a sliding time window to calculate the energy mutation rate within the window includes:

[0048] in, represents the energy mutation rate; Represents the signal mean under the corresponding sliding time window; Represents the noise mean under the corresponding sliding time window; Indicates the maximum signal amplitude among all signals in the corresponding sliding time window; Indicates the minimum signal amplitude among all signals in the corresponding sliding time window; Represents the average amplitude of all signals in the corresponding sliding time window; It represents the maximum signal amplitude among all signals in the corresponding sliding time window after removing outliers. It represents the minimum signal amplitude among all signals in the corresponding sliding time window after removing outliers. It represents the average amplitude of the remaining signal after removing outliers from all signals in the corresponding sliding time window.

[0049] In this embodiment, microcracks are first prefabricated in valve specimens and experimental spectra are collected to simulate actual operating conditions. This spectrum is then combined with the theoretical spectrum and a cross-correlation-denoising algorithm is used to select characteristic frequency bands with high overlap. In practical applications, the acoustic emission signals generated by valve microcrack expansion are complex and subject to interference from environmental noise, valve body vibration, and other factors. Relying solely on theoretically derived spectra cannot accurately reflect the actual situation, while experimental spectra alone contain errors due to noise. Combining these two methods and processing them with an algorithm can select characteristic frequency bands that more accurately represent microcrack expansion, laying a solid foundation for subsequent analysis and conforming to the logic of signal processing from theory to practical verification and denoising.

[0050] In this embodiment, a sliding time window is used to calculate the energy mutation rate. Valve operation is a continuous dynamic process, and the sliding window can continuously analyze energy changes in real time, adapting to the real-time monitoring needs of industrial sites. Microcrack growth varies at different times, and the sliding window can flexibly capture energy mutations at each stage, avoiding the lag and limitations of fixed-time analysis and ensuring timely monitoring of microcrack growth.

[0051] In this embodiment, a microcrack propagation energy mutation is determined when the energy mutation rate is greater than or equal to a first threshold and the deviation is greater than or equal to a second threshold. Judgment based on a single parameter (such as the energy mutation rate or the deviation alone) is susceptible to interference and misjudgment. Industrial scenarios are complex due to factors such as noise and operating condition fluctuations. Fusion of multiple parameters can improve the accuracy and reliability of judgments. For example, a high energy mutation rate with a small deviation may indicate normal operating fluctuations; a large deviation with a low energy mutation rate may indicate an algorithm error. Combining these two parameters can more accurately identify true energy mutations in microcrack propagation, ensuring the effectiveness of valve monitoring and meeting the requirements for multiple determinations and reducing false alarms and missed alarms in industrial equipment fault diagnosis.

[0052] In this embodiment, the formula incorporates the signal mean and noise mean, distinguishing between parameters such as the signal amplitude extreme value and mean before and after outlier removal to calculate the energy mutation rate. This method considers the influence of noise interference and outliers (such as sudden strong noise and abnormal signal spikes) in the actual signal. By processing the original signal and the signal after outlier removal separately and performing weighted fusion, it can more accurately reflect the energy mutation situation. In valve monitoring, environmental noise, transient equipment vibration, and other factors are prone to outliers and noise. This formula effectively filters these interferences, accurately calculates the energy mutation rate, and improves the rationality and accuracy of the judgment. It is an adaptive design tailored to the characteristics of complex industrial environment signals.

[0053] Microcrack growth is a dynamic, gradual process, and energy changes also evolve dynamically over time. Monitoring and prediction require methods that can process time series and learn dynamic patterns, such as LSTM. Traditional static analysis methods are not adaptable to these dynamic characteristics. This solution, using time series analysis and dynamic prediction, is essential for accurately understanding the microcrack growth process and promptly detecting anomalies.

[0054] The present invention provides a stop valve control method for a power station boiler, further comprising: Construct a damage evolution model for valve body materials; Acquire the current surface image of the valve body and quantitatively analyze it to obtain the surface vector; Inputting the surface vector into the valve body material damage evolution model to predict the damage factor; When the predicted damage factor is greater than a preset factor, the valve action of the corresponding valve body is triggered to be suspended.

[0055] Preferably, obtaining a current surface image of the valve body and quantitatively analyzing to obtain a surface vector includes: A structured light 3D scanner and a short-wave infrared camera are used to synchronously collect 3D point cloud data and thermal imaging images of material defects on the valve body surface; Extract the local surface curvature pi, normal vector deviation Δni, and corrosion depth hj from the 3D point cloud. At the same time, perform multi-scale wavelet transform on the material defect thermal image to obtain the 3D feature vector W3, and obtain the surface vector Vk=[pi,Δni,hj,W3]; The calculation formula of corrosion depth hj is as follows:

[0056] in, Indicates the number of points within the localized corrosion area; Indicates a localized corrosion area; represents the pth sub-region in the local corrosion region; They represent the measured depth after and before corrosion of the p-th sub-region respectively; represents the corrosion volume of the jth local corrosion area; The surface area of ​​the jth localized corrosion region.

[0057] In this embodiment, the valve body material damage evolution model is a mathematical model that describes how damage (corrosion, cracks, wear, etc.) develops over time and under varying operating conditions. For example, it simulates how material corrosion depth and crack propagation length change over time when a valve is subjected to long-term high temperature, high pressure, and media erosion. This model can predict, for example, whether the corrosion depth will exceed a safety threshold after 10,000 hours of operation.

[0058] A point cloud is a visualization of the three-dimensional coordinates (X, Y, Z) of countless points on a valve's surface. For example, scanning a valve yields millions of coordinate points, which can be pieced together to form a 3D outline of the valve, useful for analyzing surface flatness and corrosion pit depth.

[0059] The beneficial effect of this technical solution is that it uses dual-modal acquisition to capture both geometric shape (3D point cloud) and material defects (thermal imaging). It then extracts features such as curvature, normal deviation, corrosion depth, and thermal texture. Using surface vectors Vk, it comprehensively quantifies the valve surface, resolving the difficulty of accurately identifying damage using the naked eye or a single device. This surface vector is input into a damage evolution model, which simulates the progression of material damage over time and operating conditions. When the predicted damage factor exceeds a preset value (e.g., insufficient remaining material life), a valve operation pause is automatically triggered, preventing accidents such as leaks and breakages caused by damaged operation.

[0060] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A stop valve control method for a power station boiler, characterized in that: include: An FBG fiber optic sensor array is embedded in the key stress area of ​​the valve body to collect the temperature gradient ΔT in real time. At the same time, an AE sensor is arranged on the valve housing to capture the characteristic frequency band signal of microcrack extension. Dynamically calculate the real-time thermal stress of the valve body based on the thermal stress prediction model and dynamically analyze the real-time sudden energy of the valve based on the signal analysis model; Automatically switch multi-modal control strategies and dynamically adjust valve actuation speed based on the operating conditions of power plant boilers and in combination with real-time thermal stress and sudden energy changes; The arrangement of the FBG optical fiber sensor array satisfies: ≥8 measuring points are evenly distributed on the circumference of the valve seat cone surface, and the sensor diameter is ≤0.5mm; Three layers of annular measuring points are arranged axially on the valve stem neck, and the layer spacing is 0.8-1.2 times the valve stem diameter.

2. The stop valve control method for a power station boiler according to claim 1, characterized in that: The switching conditions of the multi-modal control strategy include: When the energy mutation of the acoustic emission signal in the 20-150kHz frequency band is detected to be ≥5dB, the system is forced to enter the creep mode; When the steam temperature change rate dT / dt>300℃ / min, the safety priority mode is activated; The specific step-by-step deceleration control in creep mode is as follows: Opening 0-30% stage: speed limit ≤ 2% / min, valve body temperature rise rate ≤ 5℃ / min; Opening 30-70% stage: speed limit ≤ 5% / min, valve body temperature rise rate ≤ 10℃ / min; The speed limit is lifted when the opening is >70%.

3. The stop valve control method for a power station boiler according to claim 1, characterized in that: The heat stress prediction model is as follows: Among them, V1 represents the maximum allowable speed of the valve; 1.8 represents the basic speed coefficient; represents the stress change damping coefficient; 0.6 represents the temperature compensation coefficient; 120 represents the temperature sensitive characteristic value; represents the temperature gradient; Indicates the real-time thermal stress of the valve body Differentiable function with respect to time t.

4. The stop valve control method for a power station boiler according to claim 1, characterized in that: Embed FBG fiber optic sensor arrays in key stress areas of the valve body, including: Use sequential coupling to simulate the thermal-structural transient response of the valve seat cone, define transient thermal boundaries and structural constraints, and cover the actual operating conditions of the valve seat; The stress linearization tool is used to define the surface path at the geometric mutation of the cone surface, decompose it into membrane stress and bending stress, and then calculate the stress according to the stress linearization tool. filter The grid area is marked, where represents the maximum stress on the path, represents membrane stress; represents the stress concentration factor; The mesh of the cone surface is encrypted and verified through mesh convergence analysis; Perform a local search on the marked area and combine it with transient time history analysis to determine whether the stress extreme point distribution is 8 local peaks; If yes, the corresponding marked area will be retained; otherwise, the judgment analysis will continue for the next marked area; The FBG fiber optic sensor array is arranged based on all the reserved areas.

5. The stop valve control method for a power station boiler according to claim 4, characterized in that: The mesh of the cone surface is refined and verified through mesh convergence analysis, including: Defining a grid density gradient along the normal and circumferential directions of the cone surface, and determining the geometric features of each grid in the cone surface, wherein the geometric features are gradient geometric features, which are divided into core geometric features, transition geometric features, and peripheral geometric features; Match the verification conditions based on each geometric feature from the geometry-mesh verification comparison table, and control the corresponding mesh to converge towards the matching verification condition direction; Among them, when it is a core geometric feature, the corresponding grid size is set to L0 / 5~L0 / 3, where L0 represents the basic grid size of the peripheral geometric feature; When it is a transitional geometric feature, the corresponding grid is gradually transitioned with a preset geometric scaling factor; When it is a peripheral geometric feature, the corresponding mesh will be deleted.

6. The stop valve control method for a power station boiler according to claim 5, characterized in that: The mesh type under the core geometric feature is a hexahedron-dominated hybrid mesh, and the mesh type under the transition geometric feature is a tetrahedron mesh.

7. The stop valve control method for a power station boiler according to claim 1, characterized in that: Dynamic analysis of valve real-time mutation energy based on signal analysis model, including: Establish a finite element model of the valve with an initial crack, simulate the crack propagation process, and extract the theoretical spectrum of the acoustic emission signal; Prefabricate microcracks in valve specimens, apply cyclic loads to simulate actual working conditions, and collect the experimental spectrum of acoustic emission signals; The cross-correlation-denoising algorithm is used to select characteristic frequency bands (f1, f2) whose overlap between the theoretical spectrum and the experimental spectrum is greater than or equal to the preset threshold; Wavelet packet decomposition is used to extract the energy time series within the characteristic frequency band; Use long short-term memory network to learn the dynamic law of energy sequence to predict the energy at the next moment and calculate the deviation ,in, For the next moment The predicted energy and actual energy; Set the sliding time window to calculate the energy mutation rate within the window; When the energy mutation rate is greater than or equal to a first threshold and the deviation is greater than or equal to a second threshold, it is determined to be a microcrack propagation energy mutation.

8. The stop valve control method for a power station boiler according to claim 1, characterized in that: Set the sliding time window to calculate the energy mutation rate within the window, including: in, represents the energy mutation rate; Represents the signal mean under the corresponding sliding time window; Represents the noise mean under the corresponding sliding time window; Indicates the maximum signal amplitude among all signals in the corresponding sliding time window; Indicates the minimum signal amplitude among all signals in the corresponding sliding time window; Represents the average amplitude of all signals in the corresponding sliding time window; It represents the maximum signal amplitude among all signals in the corresponding sliding time window after removing outliers. It represents the minimum signal amplitude among all signals in the corresponding sliding time window after removing outliers. It represents the average amplitude of the remaining signal after removing outliers from all signals in the corresponding sliding time window.

9. The stop valve control method for a power station boiler according to claim 1, characterized in that: Also includes: Construct a damage evolution model for valve body materials; Acquire the current surface image of the valve body and quantitatively analyze it to obtain the surface vector; Inputting the surface vector into the valve body material damage evolution model to predict the damage factor; When the predicted damage factor is greater than a preset factor, the valve action of the corresponding valve body is triggered to be suspended.

10. The stop valve control method for a power station boiler according to claim 1, characterized in that: Obtain the current surface image of the valve body and quantitatively analyze it to obtain surface vectors, including: A structured light 3D scanner and a short-wave infrared camera are used to synchronously collect 3D point cloud data and thermal imaging images of material defects on the valve body surface; Extract the local surface curvature pi, normal vector deviation Δni, and corrosion depth hj from the 3D point cloud. At the same time, perform multi-scale wavelet transform on the material defect thermal image to obtain the 3D feature vector W3, and obtain the surface vector Vk=[pi,Δni,hj,W3]; The calculation formula of corrosion depth hj is as follows: in, Indicates the number of points within the localized corrosion area; Indicates a localized corrosion area; represents the pth sub-region in the local corrosion region; They represent the measured depth after and before corrosion of the p-th sub-region respectively; represents the corrosion volume of the jth local corrosion area; The surface area of ​​the jth localized corrosion region.

Images