Boiler four-tube leakage risk early warning method and system based on oxide skin evolution

Abstract

The present application belongs to the technical field of boiler four-pipe leakage, and discloses a boiler four-pipe leakage risk early warning method and system based on oxide skin evolution, which comprises the following steps: constructing an oxide skin generation model and an oxide skin peeling determination model, encapsulating them into the same model group in a correlated combination manner to obtain an oxide skin evolution model set; obtaining a state feature vector, mapping the state feature vector to a spatial position, obtaining an oxide skin growth thickness prediction value and an oxide skin peeling risk index based on the oxide skin generation model and the oxide skin peeling determination model; weighting and fusing the temperature change rate and the pressure change rate in the state feature vector to obtain an oxide skin evolution comprehensive risk index; comparing the oxide skin evolution comprehensive risk index with a preset risk threshold to obtain the risk grade of each spatial position unit, and obtaining the boiler four-pipe leakage risk early warning according to the risk grade and the oxide skin evolution comprehensive risk index, thereby avoiding four-pipe leakage, pipe explosion and other faults caused by oxide skin growth and peeling in advance, and reducing the probability of unplanned shutdown.

Description

Technical Field

[0001] This invention belongs to the field of boiler four-tube leakage technology, and relates to a boiler four-tube leakage risk early warning method and system based on oxide scale evolution. Background Technology

[0002] Leaks in the four tubes of a boiler (water-cooled wall, superheater, reheater, and economizer) are a common equipment failure in thermal power generating units, and a cause of unplanned unit shutdowns, reduced equipment reliability, and increased operation and maintenance costs. Boiler heating surface pipes are subjected to harsh conditions of high temperature, high pressure, and complex flue gas scouring over long periods, making them prone to multi-dimensional deterioration problems such as oxide scale growth, oxide scale peeling, pipe creep, and pipe wall wear. Continuous accumulation of these problems can lead to pipe cracking and leakage failure, affecting the safety and economy of the thermal power unit.

[0003] Existing boiler four-tube risk warning systems generally adopt a single over-temperature threshold alarm mode, which relies on real-time tube wall temperature exceeding the limit to trigger the alarm. This can only achieve passive alarm after the pipeline shows obvious over-temperature anomalies, and cannot detect and predict the slow deterioration trend that occurs during long-term pipeline operation. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a method and system for early warning of boiler four-tube leakage risks based on oxide scale evolution.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: This invention provides a boiler four-tube leakage risk early warning method based on oxide scale evolution, comprising the following steps: Data from boiler heating surface design drawings, installation data, historical maintenance records, and oxidation kinetic parameters of pipe metal were collected to construct a three-dimensional digital model. Based on the three-dimensional digital model, the boiler heating surface was divided into several spatial units. An oxide scale generation model was constructed based on historical pipe wall temperature data and oxidation kinetic parameters. An oxide scale peeling judgment model was constructed based on the temperature change rate and oxide scale thickness threshold. The oxide scale generation model and the oxide scale peeling judgment model were encapsulated into the same model group in an associated combination manner to obtain the oxide scale evolution model set corresponding to each spatial unit. Multi-source operating data from various measuring points of the boiler were collected, and state feature vectors were obtained based on the multi-source operating data. The state feature vector is mapped to the corresponding spatial location according to the spatial location in the three-dimensional digital model. Based on the oxide scale generation model and oxide scale peeling judgment model, the predicted value of oxide scale growth thickness and the oxide scale peeling risk index of each spatial location unit are obtained. The predicted value of oxide scale growth thickness and the oxide scale peeling risk index are weighted and fused with the temperature change rate and pressure change rate in the state feature vector to obtain the comprehensive risk index of oxide scale evolution of each spatial location unit. Based on the comparison between the comprehensive risk index of oxide scale evolution and the preset risk threshold, the risk level of each spatial location unit is obtained. The boiler four-tube leakage risk warning is obtained according to the risk level and the comprehensive risk index of oxide scale evolution.

[0006] Furthermore, the process of collecting boiler heating surface design drawings, installation data, historical maintenance records, and oxidation kinetic parameters of the pipe metal to construct a three-dimensional digital model includes: collecting boiler heating surface design drawings, extracting the overall external dimensions of the boiler heating surface, furnace tube spatial layout data, and tube panel arrangement, and obtaining the geometric structure data of the boiler heating surface; collecting boiler heating surface installation data, determining the spatial coordinates and connection relationships of each tube panel, header, weld, hanging component, and connector, and obtaining the spatial topology data of the boiler heating surface components; collecting historical maintenance records of the boiler heating surface, marking the location information and defect type of repaired parts, and obtaining maintenance history annotation data; collecting oxidation kinetic parameters of the boiler heating surface pipe metal, establishing a material property database containing oxidation rate constants, oxidation activation energy, and gas constants, and obtaining the set of pipe oxidation kinetic parameters; and registering and fusing the boiler heating surface geometric structure data, boiler heating surface component spatial topology data, maintenance history annotation data, and the set of pipe oxidation kinetic parameters to obtain a three-dimensional digital model.

[0007] Furthermore, the oxide scale generation model is used to calculate the oxide scale growth thickness based on historical pipe wall temperature data, and the oxide scale peeling judgment model is used to determine the peeling risk based on the temperature change rate and the oxide scale thickness threshold.

[0008] Furthermore, At any given time, the thickness of the oxide layer is :

[0009] in, Here is the oxidation rate constant. It is the activation energy for oxidation. The gas constant is for The tube wall temperature at any given time.

[0010] Furthermore, the oxide scale peeling determination model includes: calculating the temperature change rate of each spatial location unit at the current time and the previous time, comparing the predicted value of oxide scale growth thickness with the preset critical oxide scale peeling thickness, and when the absolute value of the temperature change rate exceeds the preset maximum temperature change rate threshold and the predicted value of oxide scale growth thickness exceeds the critical oxide scale peeling thickness, there is a risk of oxide scale peeling.

[0011] Furthermore, the multi-source operating data includes: pipe wall temperature data, flue gas temperature data, steam flow rate data, pressure data, and operating time data; The method for obtaining state feature vectors based on multi-source operational data includes: denoising, filling in missing values, and normalizing the multi-source operational data to obtain state feature vectors.

[0012] Furthermore, the state feature vector is mapped to the corresponding spatial position according to the spatial position in the three-dimensional digital model, including: establishing a correspondence table between the coordinates of the measuring points and the spatial position units based on the coordinate information of each measuring point in the three-dimensional digital model, and assigning the standardized state feature vector of each measuring point to its corresponding spatial position unit.

[0013] Furthermore, the comprehensive risk index for the evolution of the oxide scale is: :

[0014]

[0015] in, The weighting coefficients for the predicted thickness of oxide scale growth are: This is the predicted value for oxide scale growth thickness. This is the preset critical thickness for oxide scale peeling. The weighting coefficients for the oxide scale peeling risk index are as follows: The risk index for oxide scale peeling. The weighting coefficient for the rate of temperature change. The absolute value of the rate of temperature change. The preset maximum temperature change rate threshold. The weighting coefficient for the rate of change of pressure. The absolute value of the rate of change of pressure. This is the preset maximum pressure change rate threshold.

[0016] Furthermore, the preset risk thresholds include a first risk threshold and a second risk threshold, wherein the first risk threshold is greater than the second risk threshold; based on the comparison between the comprehensive risk index of oxide scale evolution and the preset risk thresholds, the risk level of each spatial location unit is obtained, including: comparing the comprehensive risk index of oxide scale evolution of each spatial location unit with the second risk threshold and the first risk threshold respectively; if the comprehensive risk index of oxide scale evolution is less than the second risk threshold, it is a normal level; if the comprehensive risk index of oxide scale evolution is greater than or equal to the second risk threshold and less than the first risk threshold, it is a level one warning; if the comprehensive risk index of oxide scale evolution is greater than the first risk threshold, it is a level two warning.

[0017] This invention also provides a boiler four-tube leakage risk early warning method based on oxide scale evolution, including: Construction module: Used to collect boiler heating surface design drawing data, installation data, historical maintenance records, and oxidation kinetic parameters of pipe metal to construct a three-dimensional digital model; The assembly module is used to divide the boiler heating surface into several spatial units based on a 3D digital model. It constructs an oxide scale generation model based on historical tube wall temperature data and oxidation kinetic parameters, and an oxide scale peeling determination model based on temperature change rate and oxide scale thickness threshold. The oxide scale generation model and oxide scale peeling determination model are encapsulated into a single model group through a correlation combination, resulting in an oxide scale evolution model set corresponding to each spatial unit. The acquisition module is used to collect multi-source operating data from various measuring points in the boiler and obtain state feature vectors based on this data. The mapping module maps the state feature vectors according to their spatial positions in the 3D digital model. The system maps the scale to the corresponding spatial location and, based on the scale generation model and scale peeling judgment model, obtains the predicted scale growth thickness and scale peeling risk index for each spatial location unit. The fusion module weightedly fuses the predicted scale growth thickness and scale peeling risk index with the temperature change rate and pressure change rate in the state feature vector to obtain the comprehensive scale evolution risk index for each spatial location unit. The early warning module compares the comprehensive scale evolution risk index with a preset risk threshold to obtain the risk level of each spatial location unit, and obtains a boiler four-tube leakage risk warning based on the risk level and the comprehensive scale evolution risk index.

[0018] Compared with the prior art, the present invention has the following beneficial technical effects: This invention provides a boiler four-tube leakage risk early warning method based on oxide scale evolution. It constructs an oxide scale generation model and an oxide scale peeling judgment model, respectively. Through encapsulation, it simulates and analyzes the evolution law of oxide scale growth and peeling, conforming to the evolution law of oxide scale. Based on multi-source boiler operating data, it extracts state feature vectors to map operating conditions to equipment spatial location, acquiring real-time and dynamic predicted values ​​of oxide scale growth thickness and peeling risk index for each heating surface unit, ensuring real-time monitoring and coverage. It weights and integrates operating parameters such as temperature change rate and pressure change rate with oxide scale-related risk indicators to construct a comprehensive oxide scale evolution risk index, taking into account both the inherent risks of equipment material aging and the dynamic risks induced by operating condition fluctuations. This avoids the one-sidedness of judging risk with a single indicator, making the risk assessment results more consistent with the actual complex operating scenarios of the boiler. By setting a preset risk threshold to classify unit risk levels and generating graded early warning results, it accurately locates high-risk leakage points in the boiler's four tubes, proactively avoiding equipment failures such as four-tube leakage and tube rupture caused by excessive oxide scale growth and peeling, and reducing the probability of unplanned boiler shutdowns. Attached Figure Description

[0019] Figure 1 This is a flowchart of a boiler four-tube leakage risk early warning method based on oxide scale evolution according to the present invention. Detailed Implementation

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

[0021] Example 1 This invention discloses a boiler four-tube leakage risk early warning method based on oxide scale evolution, comprising the following steps: collecting boiler heating surface design drawing data, installation data, historical maintenance records, and oxidation kinetic parameters of the tube metal to construct a three-dimensional digital model; based on the three-dimensional digital model, dividing the boiler heating surface into several spatial location units, constructing an oxide scale generation model based on historical tube wall temperature data and oxidation kinetic parameters, constructing an oxide scale peeling judgment model based on temperature change rate and oxide scale thickness threshold, and encapsulating the oxide scale generation model and oxide scale peeling judgment model into the same model group in an associated combination manner to obtain an oxide scale evolution model set corresponding to each spatial location unit; collecting multi-source operating data from various measuring points of the boiler, based on... State feature vectors are obtained from multi-source operational data. These vectors are then mapped to corresponding spatial locations within a 3D digital model. Based on the scale generation model and scale peeling judgment model, predicted scale growth thickness and scale peeling risk index are obtained for each spatial location unit. The predicted scale growth thickness and scale peeling risk index are weighted and fused with the temperature and pressure change rates in the state feature vectors to obtain a comprehensive scale evolution risk index for each spatial location unit. Based on a comparison between the comprehensive scale evolution risk index and a preset risk threshold, the risk level of each spatial location unit is obtained. Based on the risk level and the comprehensive scale evolution risk index, a boiler four-tube leakage risk warning is obtained, such as… Figure 1 As shown.

[0022] Specifically, firstly, boiler heating surface design drawing data (including CAD drawings, 3D design models, etc.) is collected from boiler design units or archives departments. Using computer-aided design (CAD) analysis technology and image recognition algorithms, the overall external dimensions (such as width, depth, and height), furnace tube spatial layout data (such as tube diameter, wall thickness, bending radius, and spacing of each tube panel), and tube panel arrangement method (such as horizontal surrounding, vertical rising, etc.) of the boiler heating surface are extracted to obtain the geometric structure data of the boiler heating surface.

[0023] By combining the boiler equipment installation completion data and the on-site three-dimensional measurement data, the three-dimensional spatial coordinates of each pipe panel, header, pipe weld, hanging support component and connecting fastener are accurately located, the connection relationship, stress structure and relative spatial position of each component are clarified, and the spatial topology data of the boiler heating surface components are constructed.

[0024] Retrieve historical maintenance records of boiler heating surfaces, including water pressure test leak points, wall thickness reduction areas, oxide scale accumulation detection reports, and metallographic changes, extract key information from the text records, or manually mark the location information and defect types (such as creep cracks, corrosion pits, and mechanical wear) of the repaired parts in the 3D model to obtain maintenance history annotation data.

[0025] Finally, oxidation kinetic parameters of the metal materials used in each section of the boiler's heating surface were collected and derived from material handbooks, high-temperature oxidation experiments, or long-term operating data. A material property database was established, containing oxidation rate constants, oxidation activation energies, and gas constants for different grades of steel, thus constructing a dedicated material property database.

[0026] The boiler's heating surface geometry data, component spatial topology data, maintenance history annotation data, and pipe oxidation kinetic parameter set are spatially registered and fused using the boiler design coordinate system as a reference. Point cloud registration algorithms, such as the Iterative Closest Point (ICP) algorithm, are used to align the geometric and topological data. Through attribute mapping, material parameters and historical defect information are attached to the corresponding geometric entities to generate a three-dimensional digital model containing physical properties (material, wall thickness), historical status (repair markers), and spatial relationships (coordinates, connectivity).

[0027] Based on a 3D digital model, the pipeline is meshed along its axial and circumferential directions. Axially, each pipeline is divided into segments of fixed length (e.g., 0.5 to 2 meters, adaptively adjusted according to the curvature of the pipe screen). Circumferentially, it is divided into sectors (e.g., fire-facing and back-facing sides) based on heat flux density distribution. A spatial unit is defined, and a corresponding set of oxide scale evolution models is independently constructed for each spatial unit.

[0028] The oxide scale evolution model set is composed of an oxide scale generation model and an oxide scale peeling judgment model. The oxide scale generation model is used to calculate the oxide scale growth thickness based on historical pipe wall temperature data. The oxide scale peeling judgment model is used to assess whether the generated oxide scale will peel off. There are two reasons for oxide scale peeling: one is that the thickness of the oxide scale itself reaches a critical value, and the internal stress increases; the other is that rapid temperature changes cause a mismatch between the thermal expansion coefficients of the oxide scale and the base metal, resulting in huge thermal stress.

[0029] At any given time, the thickness of the oxide layer is :

[0030] in, Here is the oxidation rate constant. It is the activation energy for oxidation. The gas constant is for The tube wall temperature at any given time.

[0031] The system calculates the rate of change of pipe wall temperature between the current and previous acquisition times for each spatial unit, while simultaneously retrieving the real-time predicted growth thickness from the oxide scale formation model. When the absolute value of the temperature change rate of a spatial unit exceeds a preset maximum temperature change rate threshold, and the predicted oxide scale growth thickness exceeds the critical peeling thickness, that unit is considered to have an oxide scale peeling risk, and a corresponding peeling risk index is output. Units that do not meet both conditions are deemed to have no peeling risk, and their risk index is set to zero. This approach avoids the limitations of relying on a single parameter and improves the accuracy of peeling risk identification.

[0032] The multi-source real-time monitoring data includes five core data categories: pipe wall temperature data of each heated surface, flue gas temperature data of the furnace and flue, steam flow data of the pipeline, medium pressure data, and cumulative operating time data of the equipment. The data acquisition frequency is set to 1 second / time. Wavelet denoising algorithm is used to eliminate abnormal noise caused by electromagnetic interference and equipment vibration. Instantaneous missing data is supplemented by adjacent data interpolation method. Then, extreme value normalization algorithm is used to unify the dimensions of all data, mapping all operating data to the 0-1 interval, and finally generating a standardized, anomaly-free, and directly computable state feature vector. Each spatial measuring point corresponds to a set of independent state feature vectors.

[0033] First, based on the coordinate system of the 3D digital model, the 3D coordinate information of all field monitoring points is sorted out, and a one-to-one correspondence table between the coordinates of the monitoring points and each spatial location unit is established, clarifying the unique spatial grid unit corresponding to each monitoring point. The preprocessed state feature vectors are assigned to the corresponding spatial location units one by one, realizing the precise binding of "monitoring point data - spatial unit - model parameters", so that each independent spatial location unit has its own exclusive real-time operating feature data, solving the problem that traditional overall assessment methods cannot locate local hidden dangers.

[0034] By integrating four core risk factors—oxidized scale growth status, peeling risk, temperature fluctuation, and pressure fluctuation—through a multi-parameter weighted fusion method, a comprehensive risk index for oxide scale evolution is obtained.

[0035]

[0036]

[0037] in, The weighting coefficients for the predicted thickness of oxide scale growth are: This is the predicted value for oxide scale growth thickness. This is the preset critical thickness for oxide scale peeling. The weighting coefficients for the oxide scale peeling risk index are as follows: The risk index for oxide scale peeling. The weighting coefficient for the rate of temperature change. The absolute value of the rate of temperature change. The preset maximum temperature change rate threshold. The weighting coefficient for the rate of change of pressure. The absolute value of the rate of change of pressure. The preset maximum pressure change rate threshold. It is 0.35. It is 0.30. It is 0.20. It is 0.15.

[0038] Risk level assessment and early warning are output through a tiered threshold comparison method. Two risk thresholds are established: a first risk threshold and a second risk threshold. The first risk threshold must be greater than the second risk threshold, corresponding to different risk levels and maintenance strategies. All boiler heating surface spatial units are traversed, and the comprehensive risk index of each unit is compared with the two thresholds to complete the tiered assessment. When the comprehensive risk index is less than the second risk threshold, the spatial unit is classified as having a normal risk level; the oxide scale growth is stable, with no peeling or leakage risks, and routine inspections are sufficient. When the comprehensive risk index is greater than or equal to the second risk threshold but less than the first risk threshold, it is classified as a Level 1 early warning; the oxide scale growth rate in this area is accelerated, posing a potential peeling risk, and inspection frequency should be increased to track parameter changes. When the comprehensive risk index is greater than the first risk threshold, it is classified as a Level 2 early warning; the oxide scale thickness is close to or reaches a critical value, and the probability of oxide scale peeling, high-temperature corrosion of pipelines, and leakage caused by pipe wall thinning is extremely high, requiring immediate maintenance and investigation.

[0039] Example 2 This invention provides a boiler four-tube leakage risk early warning method based on oxide scale evolution, comprising a construction module, an aggregation module, an acquisition module, a mapping module, a fusion module, and an early warning module.

[0040] The construction module is used to collect boiler heating surface design drawings, installation data, historical maintenance records, and oxidation kinetic parameters of pipe metal to construct a three-dimensional digital model. The assembly module is used to divide the boiler heating surface into several spatial units based on the three-dimensional digital model, construct an oxide scale generation model based on historical pipe wall temperature data and oxidation kinetic parameters, construct an oxide scale peeling judgment model based on temperature change rate and oxide scale thickness threshold, and encapsulate the oxide scale generation model and oxide scale peeling judgment model into the same model group in an associated combination manner to obtain the oxide scale evolution model set corresponding to each spatial unit. The acquisition module is used to collect multi-source operating data from various measuring points of the boiler and obtain state feature vectors based on the multi-source operating data. The mapping module maps the state feature vector to the corresponding spatial location in the 3D digital model. Based on the oxide scale generation model and oxide scale peeling judgment model, it obtains the predicted value of oxide scale growth thickness and the oxide scale peeling risk index for each spatial location unit. The fusion module weightedly fuses the predicted value of oxide scale growth thickness and the oxide scale peeling risk index with the temperature change rate and pressure change rate in the state feature vector to obtain the comprehensive risk index of oxide scale evolution for each spatial location unit. The early warning module compares the comprehensive risk index of oxide scale evolution with a preset risk threshold to obtain the risk level of each spatial location unit, and obtains a boiler four-tube leakage risk warning based on the risk level and the comprehensive risk index of oxide scale evolution.

[0041] It should be noted that the boiler four-tube leakage risk early warning system based on oxide scale evolution provided by the present invention can achieve the same method steps as the above method, so it will not be described again.

[0042] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

Claims

1. A method for early warning of boiler four-tube leakage risk based on oxide scale evolution, characterized in that, Includes the following steps: Collect boiler heating surface design drawings, installation data, historical maintenance records, and oxidation kinetic parameters of pipe metal to construct a three-dimensional digital model; Based on the three-dimensional digital model, the boiler heating surface is divided into several spatial units. An oxide scale generation model is constructed based on historical tube wall temperature data and oxidation kinetic parameters. An oxide scale peeling judgment model is constructed based on temperature change rate and oxide scale thickness threshold. The oxide scale generation model and oxide scale peeling judgment model are encapsulated into the same model group in an associated combination manner to obtain the oxide scale evolution model set corresponding to each spatial unit. Collect multi-source operating data from various measuring points of the boiler, and obtain state feature vectors based on the multi-source operating data; The state feature vector is mapped to the corresponding spatial location according to the spatial location in the three-dimensional digital model. Based on the oxide scale generation model and the oxide scale peeling judgment model, the predicted value of oxide scale growth thickness and oxide scale peeling risk index of each spatial location unit are obtained. The predicted value of oxide scale growth thickness and the oxide scale peeling risk index are weighted and integrated with the temperature change rate and pressure change rate in the state feature vector to obtain the comprehensive risk index of oxide scale evolution for each spatial unit. Based on the comparison between the comprehensive risk index of oxide scale evolution and the preset risk threshold, the risk level of each spatial location unit is obtained, and the risk warning of leakage of the four tubes of the boiler is obtained according to the risk level and the comprehensive risk index of oxide scale evolution.

2. The boiler four-tube leakage risk early warning method based on oxide scale evolution according to claim 1, characterized in that, The process involves collecting boiler heating surface design drawings, installation data, historical maintenance records, and oxidation kinetic parameters of the pipe metal to construct a three-dimensional digital model, including: Collect boiler heating surface design drawing data, extract the overall external dimensions of the boiler heating surface, furnace tube space layout data and tube panel arrangement, and obtain the geometric structure data of the boiler heating surface. Collect installation data of boiler heating surfaces, determine the spatial coordinates and connection relationships of each tube panel, header, weld, hanging component and connector, and obtain spatial topology data of boiler heating surface components; Collect historical maintenance records of boiler heating surfaces, mark the location information and defect type of repaired parts, and obtain maintenance history annotation data; Oxidation kinetic parameters of boiler heating surface tubes are collected, a material property database containing oxidation rate constant, oxidation activation energy and gas constant is established, and a set of tube oxidation kinetic parameters is obtained. The geometric data of the boiler heating surface, the spatial topology data of the boiler heating surface components, the maintenance history annotation data, and the set of tube oxidation kinetic parameters are registered and fused to obtain a three-dimensional digital model.

3. The boiler four-tube leakage risk early warning method based on oxide scale evolution according to claim 1, characterized in that: The oxide scale generation model is used to calculate the oxide scale growth thickness based on historical pipe wall temperature data, and the oxide scale peeling judgment model is used to determine the peeling risk based on the temperature change rate and the oxide scale thickness threshold.

4. The boiler four-tube leakage risk early warning method based on oxide scale evolution according to claim 3, characterized in that: At any given time, the thickness of the oxide layer is : in, Here is the oxidation rate constant. It is the activation energy for oxidation. The gas constant is for The tube wall temperature at any given time.

5. The boiler four-tube leakage risk early warning method based on oxide scale evolution according to claim 3, characterized in that: The oxide scale peeling determination model includes: Calculate the temperature change rate of each spatial location unit between the current time and the previous time, compare the predicted value of the oxide scale growth thickness with the preset critical oxide scale peeling thickness, and when the absolute value of the temperature change rate exceeds the preset maximum temperature change rate threshold and the predicted value of the oxide scale growth thickness exceeds the critical oxide scale peeling thickness, there is a risk of oxide scale peeling.

6. The boiler four-tube leakage risk early warning method based on oxide scale evolution according to claim 1, characterized in that: The multi-source operating data includes: pipe wall temperature data, flue gas temperature data, steam flow rate data, pressure data, and operating time data; The method for obtaining state feature vectors based on multi-source operational data includes: denoising, filling in missing values, and normalizing the multi-source operational data to obtain state feature vectors.

7. The boiler four-tube leakage risk early warning method based on oxide scale evolution according to claim 1, characterized in that, Mapping the state feature vector to its corresponding spatial location according to its spatial location in the three-dimensional digital model includes: Based on the coordinate information of each measuring point in the three-dimensional digital model of the boiler, a correspondence table between the measuring point coordinates and the spatial position unit is established, and the standardized state feature vector of each measuring point is assigned to its corresponding spatial position unit.

8. The boiler four-tube leakage risk early warning method based on oxide scale evolution according to claim 7, characterized in that: The comprehensive risk index for the evolution of the oxide scale is: : in, The weighting coefficients for the predicted thickness of oxide scale growth are: This is the predicted value for oxide scale growth thickness. This is the preset critical thickness for oxide scale peeling. The weighting coefficients for the oxide scale peeling risk index are as follows: The risk index for oxide scale peeling. The weighting coefficient for the rate of temperature change. The absolute value of the rate of temperature change. The preset maximum temperature change rate threshold. The weighting coefficient for the rate of change of pressure. The absolute value of the rate of change of pressure. This is the preset maximum pressure change rate threshold.

9. The boiler four-tube leakage risk early warning method based on oxide scale evolution according to claim 1, characterized in that: The preset risk threshold includes a first risk threshold and a second risk threshold, wherein the first risk threshold is greater than the second risk threshold. Based on a comparison between the comprehensive risk index of oxide scale evolution and a preset risk threshold, the risk level of each spatial location unit is obtained, including: The comprehensive risk index of oxide scale evolution for each spatial location unit is compared with the second risk threshold and the first risk threshold. If the comprehensive risk index of oxide scale evolution is less than the second risk threshold, it is considered normal. If the comprehensive risk index of oxide scale evolution is greater than or equal to the second risk threshold and less than the first risk threshold, it is considered a first-level warning. If the comprehensive risk index of oxide scale evolution is greater than the first risk threshold, it is considered a second-level warning.

10. A boiler four-tube leakage risk early warning system based on oxide scale evolution, characterized in that, include: Construction module: Used to collect boiler heating surface design drawing data, installation data, historical maintenance records, and oxidation kinetic parameters of pipe metal to construct a three-dimensional digital model; The collection module is used to divide the boiler heating surface into several spatial units based on a three-dimensional digital model. It constructs an oxide scale generation model based on historical tube wall temperature data and oxidation kinetic parameters, and constructs an oxide scale peeling judgment model based on temperature change rate and oxide scale thickness threshold. The oxide scale generation model and oxide scale peeling judgment model are encapsulated into the same model group in an associated combination manner to obtain the oxide scale evolution model set corresponding to each spatial unit. Acquisition module: Used to collect multi-source operating data from various measuring points of the boiler, and obtain state feature vectors based on the multi-source operating data; Mapping module: used to map the state feature vector to the corresponding spatial position according to the spatial position in the three-dimensional digital model, and obtain the predicted value of oxide scale growth thickness and oxide scale peeling risk index of each spatial position unit based on the oxide scale generation model and oxide scale peeling judgment model. Fusion module: used to weight and fuse the predicted value of oxide scale growth thickness and oxide scale peeling risk index with the temperature change rate and pressure change rate in the state feature vector to obtain the comprehensive risk index of oxide scale evolution for each spatial unit; Early warning module: It is used to obtain the risk level of each spatial location unit by comparing the comprehensive risk index of oxide scale evolution with the preset risk threshold, and to obtain the early warning of leakage risk of boiler four tubes based on the risk level and the comprehensive risk index of oxide scale evolution.

Images