Detection method and analysis system applied to corrosion problem of Texaco gasification device

Abstract

The invention relates to the technical field of pipeline corrosion detection, in particular to a detection method and analysis system applied to the corrosion problem of a Texaco gasification device, and the method comprises the steps: selecting key monitoring points to monitor the corrosion condition of a target device, installing intelligent sensors at the points, collecting operation data in real time, and transmitting the operation data to a central processing unit; analyzing the data, identifying an area possibly having a corrosion risk and an abnormal change trend, and determining a suspected corrosion area; carrying out deep detection on the areas to obtain detection data; detection data, equipment operation data and maintenance records are encrypted and stored, the authenticity and traceability of the data are ensured, and a reliable basis is provided for subsequent analysis and decision making; and in combination with an intelligent manufacturing technology, operation parameters and maintenance plans of the device are optimized according to a corrosion analysis result, and accurate analysis and effective prevention and control of corrosion problems are realized. By means of the scheme, accurate analysis and effective prevention and control of the corrosion problem of the Texaco gasification device are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline corrosion detection, and particularly to a detection method and analysis system for corrosion problems in Texaco gasification plants. Background Art

[0002] The Texaco gasifier is a commonly used gasification device, which is widely used in the coal gasification process. It makes a water-coal slurry by mixing coal with water, and then undergoes a gasification reaction under high temperature and high pressure conditions to generate gas products. The inside of the gasifier is usually lined with refractory bricks to protect the furnace body from high temperature and gasification medium erosion.

[0003] Texaco gasification corrosion mainly refers to the corrosion phenomenon of the materials inside the Texaco gasifier due to high temperature, high pressure and a specific gasification medium environment. This corrosion phenomenon is an inevitable problem during the long-term operation of the gasifier, and has an important impact on the performance, life and safety of the gasifier.

[0004] The reasons for Texaco gasification corrosion include high temperature and high pressure environment: the temperature inside the gasifier is as high as several hundred to over a thousand degrees Celsius, and the pressure is relatively high. This extreme environment accelerates the corrosion rate of the materials inside the furnace; gasification medium: the gases and molten slag generated during the gasification process have a corrosive effect on the materials inside the furnace such as refractory bricks. Gas flow erosion: the gas flow velocity inside the gasifier is relatively fast, especially in the upper part of the cylinder and the conical bottom, where the erosion by the gas-entrained molten slag is relatively serious.

[0005] Existing pipeline corrosion detection methods lack the link of analyzing and evaluating the detection results, which means that the detection data cannot be deeply interpreted and quantitatively evaluated. This will lead to an inability to accurately understand the corrosion status of the pipeline, including key information such as the location, degree and type of corrosion. Without the step of analyzing and evaluating the detection results, it is impossible to formulate a targeted maintenance plan according to the actual corrosion situation, resulting in blindness and inefficiency of the maintenance work. To solve the above problems, we propose a detection method and analysis system for corrosion problems in Texaco gasification plants. Summary of the Invention

[0006] The purpose of the present invention is to solve the problems existing in the prior art, and to propose a detection method and analysis system for corrosion problems in Texaco gasification plants.

[0007] To achieve the above object, the present invention adopts the following technical solutions: A detection method applied to the corrosion problem of a Texaco gasification device, which is applied to a coal slurry preparation system. The damaged parts are the coal mill, the coal slurry tank and the attached pipelines. The material property is carbon steel. Due to medium impact or relative movement, mechanical spalling occurs on the surface layer of the material, forming pits, grooves, sharp grooves, holes and corrugated morphologies, and having a certain directionality, which is classified as erosion-corrosion;

[0008] S1: Select the parts made of carbon steel in the coal mill, the coal slurry tank and their attached pipelines that are vulnerable to erosion-corrosion damage due to medium impact or relative movement; clean the surface to be detected, and remove attachments, dirt and oxide layers;

[0009] S2: Use the detection system of an intelligent automatic diagnosis probe to comprehensively scan the area to be detected, identify and record the morphological characteristics formed by corrosion, calculate the parameters based on the corrosion mechanism model through an algorithm, calculate what corrosion mechanism exists at the thinning position, and give the corresponding monitoring and detection plan according to the calculated corrosion mechanism;

[0010] S3: According to the detection data, analyze the severity, distribution law and development trend of erosion-corrosion, and combine the mechanical properties and service conditions of the material to evaluate its impact on the safe operation of the system; then sort out the detection results, including the position, size, morphology and evaluation conclusion of erosion-corrosion, and form a detection report to provide a scientific basis for subsequent maintenance, repair or replacement.

[0011] As a further technical solution of the present invention, it is also applied to a high-temperature gasification system. The damaged part is the gasification chamber of the gasifier. The material properties are carbon steel, low-alloy steel and stainless steel. First, due to medium impact or relative movement, mechanical spalling occurs on the surface layer of the material, forming pits, grooves, sharp grooves, holes and corrugated morphologies, and having a certain directionality, which is classified as erosion-corrosion; second, when contacting with sulfide under the condition of a temperature exceeding 260°C and in the presence of hydrogen, uniform corrosion occurs on the metal, forming a multi-layer FeS scale structure with a firmly bonded and grayish luster, which is classified as high-temperature sulfide corrosion (hydrogen environment); third, when the operating temperature is greater than 595°C, the volume of the component increases, forming fine brittle cracks with a convex, oxide scale and "fish-tail pattern" appearance, which is classified as carburization; fourth, in a high-temperature hydrogen environment, surface decarburization develops inward, forming bulges in the gaps at grain boundaries or inclusion interfaces, and then connecting along the grain boundaries to form microcracks. Ultrasonic thickness measurement shows abnormal "thickening", and decarburization or cracks are seen under a microscope, which is classified as high-temperature hydrogen corrosion; fifth, in a high-temperature oxygen environment, carbon steel and low-alloy steel form oxide films causing uniform thinning, and stainless steel forms dark oxide films, which is classified as high-temperature oxidation; sixth, metal surface deposition and melting occur, and the generated slag melts the surface oxide film, and an obvious smooth interface can be seen between the vitreous hard scale layer of the ash layer and the base metal, which is classified as ash corrosion.

[0012] As a further technical solution of the present invention, it is also applied to the lock hopper slag discharge system. The damaged parts are the lock hopper, the flushing water tank and the auxiliary pipelines, and the slag pond. The material properties are carbon steel and low alloy steel. First, in an acidic water environment containing ammonium hydrosulfide, local corrosion forms at the parts where the medium flow direction changes or in the turbulent flow area where the NH4HS concentration exceeds 2%, and scaling occurs in the low flow rate area, resulting in local corrosion under the scale, which is classified as acidic water corrosion. Second, ammonium chloride crystallizes and forms scale, which absorbs moisture and deliquesces. There are often white, green or brown salt-like deposits, which is classified as ammonium chloride corrosion. Third, due to medium impact or relative movement, mechanical spalling of the material surface layer occurs, forming pits, grooves, sharp grooves, holes and corrugated morphologies with a certain directionality, which is classified as erosion-corrosion.

[0013] As a further technical solution of the present invention, it is also applied to the sour gas system. The damaged parts are the top pipelines of the high, low pressure and vacuum flash tanks, the top pipelines of the flash separators, and the gas phase outlet pipelines at the top of the deaerator. The material properties are carbon steel, low alloy steel and stainless steel. First, there is a liquid water phase containing H2S and the H2S content is greater than 50 ppm. Independent small bubbles form on the steel surface, and the small bubbles do not merge with each other, which is classified as wet H2S damage - hydrogen blistering. Step-shaped cracks parallel to the surface and extending along the rolling direction form inside the steel, which is classified as wet H2S damage - hydrogen-induced cracking. Hydrogen-induced cracking cracks at different depths on the base metal in the heat affected zone of the weld are connected along the thickness direction, which is classified as wet H2S damage - stress-oriented hydrogen-induced cracking. Cracks initiate on the surface of the heat affected zone of the weld and the high hardness zone and extend along the thickness direction, which is classified as wet H2S damage - stress corrosion cracking. Second, due to contact with hydrochloric acid, it is usually uniform corrosion, local corrosion or under-deposit corrosion during local concentration or dew point corrosion, and pitting corrosion occurs on the stainless steel, which is classified as hydrochloric acid corrosion. Third, in an acidic water environment containing hydrogen sulfide and with a pH value of 4.5 - 7.0, it is generally uniform corrosion, and local corrosion or local corrosion under deposited scale is likely to occur in the presence of oxygen. Pitting corrosion occurs on 300SS, which is classified as acidic water corrosion. Fourth, in a humid carbon dioxide environment below 140 °C, it occurs in the turbulent flow and liquid impact areas and at the root of the weld, resulting in wall thickness reduction, corrosion pits or holes. Deeper pitting corrosion pits and grooves may form when carbon steel corrodes in the turbulent flow area, which is classified as carbon dioxide corrosion.

[0014] As a further technical solution of the present invention, it is also applied to the black water and grey water systems. The damaged parts are in the quench chamber of the gasifier, Venturi scrubber, scrubbing tower, scrubbing water system, liquid phase parts of the high, low pressure and vacuum flashing systems, and the liquid phase outlet pipeline at the bottom of the deaerator. When the material property is carbon steel or low alloy steel, first, in an acidic water environment with ammonium hydrosulfide, local corrosion forms at the parts where the medium flow direction changes or in the turbulent flow area where the NH4HS concentration exceeds 2%, and scaling occurs in the low flow rate area, resulting in local corrosion under the scale, which is classified as acidic water corrosion; second, due to medium impact or relative movement, mechanical peeling of the material surface layer occurs, forming pits, grooves, sharp grooves, holes and corrugated morphologies with a certain directionality, which is classified as erosion corrosion; third, due to the crystallization and scaling of ammonium chloride and its hygroscopic deliquescence, white, green or brown salt-like deposits often exist, which is classified as ammonium chloride corrosion; fourth, when the gas phase CO2 concentration > 2% and the temperature > 93°C, the liquid phase pH value > 9.0 and the carbonate concentration > 100 ppm, or the pH value is between 8.0 and 9.0 and the carbonate concentration > 400 ppm, cracks that expand along the direction parallel to the weld appear in the base metal near the welded joint. The cracks mainly propagate along the grain boundaries, and the interior is filled with oxide cracks that are fine and spider-web shaped, which is classified as carbonate stress corrosion cracking.

[0015] When the material property is 300 series stainless steel or 400 series stainless steel, first, under the combined action of tensile stress, temperature > 38°C, and chloride aqueous solution environment with pH value > 2, dendritic and branched transgranular expansion cracks form on the surface. The fracture surface is usually brittle and there is no obvious thinning, which is classified as chloride stress corrosion cracking; second, under medium impact or relative movement, mechanical peeling of the material surface layer occurs, forming pits, grooves, sharp grooves, holes and corrugated morphologies with a certain directionality, which is classified as erosion corrosion; third, ammonium chloride crystallizes and scales, and hygroscopic deliquescence occurs. White, green or brown salt-like deposits that are easily removed exist, and local corrosion usually occurs under the scale layer, which is classified as ammonium chloride corrosion.

[0016] As a further technical solution of the present invention, S2 includes:

[0017] System deployment: Deploy intelligent automatic diagnosis probes at key pipeline parts. The probes have data acquisition functions and are used to directly contact the inner wall of the pipeline for scanning the corrosion morphology.

[0018] Start scanning: Start the probes to comprehensively scan the inner wall of the pipeline along a preset trajectory or a trajectory adjusted according to the real-time pipeline shape, and collect corrosion morphology data in real time.

[0019] Data preprocessing: Input the collected corrosion morphology data into the data processing and analysis unit for data preprocessing operations.

[0020] Feature extraction and recognition: Using image processing technology and machine learning algorithms, extract the features of the corrosion morphology from the preprocessed data, and compare the extracted features with a preset corrosion morphology database to identify the type and degree of corrosion;

[0021] Recording and display: Record the identified corrosion morphology features in the form of graphics or data, and display them through a display system. At the same time, export the data to a report generation module to generate a detailed detection report;

[0022] Intelligent optimization: The detection system has a self-learning function, can optimize the recognition algorithm according to the continuously accumulated detection data, and automatically trigger an alarm mechanism when severe corrosion is detected.

[0023] As a further technical solution of the present invention, the probe dynamically adjusts the scanning trajectory according to the actual shape of the inner wall of the pipeline to fully cover the detection area, and real-time monitors the contact force between the probe and the inner wall of the pipeline.

[0024] As a further technical solution of the present invention, the dynamic scanning trajectory adjustment includes the following steps:

[0025] Before the detection starts, the system first plans a preliminary scanning path according to the shape data of the pipeline; when the probe starts to scan, the laser rangefinder and ultrasonic sensor equipped on it will sense the pipeline shape in real time, and the system will compare the real-time sensed shape data with the initial scanning path and calculate the deviation;

[0026] According to the calculated deviation, the system dynamically adjusts the movement trajectory of the probe;

[0027] The system uses a particle swarm trajectory optimization algorithm to find the optimal scanning trajectory during the real-time sensing and adjustment process. The algorithm considers factors such as scanning speed, contact force between the probe and the inner wall of the pipeline, and detection accuracy;

[0028] Among them, the algorithm steps include the following steps:

[0029] Initializing the particle swarm:

[0030] Determine the size N of the particle swarm, that is, the number of candidates for the trajectory,

[0031] Randomly assign an initial position and velocity to each particle. Here, the position can be represented as a parameter or coordinate sequence of the scanning trajectory, and the velocity represents the change rate of the trajectory parameters or coordinates;

[0032] Initialize the personal best position (pbest) of each particle to its current position, and the global best position (gbest) to the optimal position among all particles;

[0033] Fitness function design and iterative update:

[0034] According to the factors of scanning speed, the contact force between the probe and the inner wall of the pipeline, and detection accuracy, the fitness is designed to evaluate the quality of each particle. A comprehensive evaluation index is obtained by means of weighted summation or multiplication. In each iteration, the fitness value of each particle is calculated according to the fitness function;

[0035] Update the personal best position (pbest) of each particle. If the current fitness value is better than the previous personal best fitness value, update the global best position (gbest) if the fitness value of a certain particle is better than the previous global best fitness value;

[0036] Update the velocity and position of each particle according to the following formula:

[0037] Velocity update formula: Vi(t + 1) = w * Vi(t) + c1 * r1 * (pbesti - Xi(t)) + c2 * r2 * (gbest - Xi(t))

[0038] where w is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers, Xi(t) is the position of particle i at time t, and Vi(t) is the velocity of particle i at time t;

[0039] Position update formula: Xi(t + 1) = Xi(t) + Vi(t + 1)

[0040] Termination condition: When the preset maximum number of iterations is reached, the fitness value of the global best position reaches the preset threshold, and the change amount of the fitness value is less than the preset threshold, indicating that the algorithm has converged;

[0041] Output the optimal solution:

[0042] When the termination condition is met, output the global best position as the optimal scanning trajectory;

[0043] A contact force sensor is integrated on the probe to monitor the contact force between the probe and the inner wall of the pipeline in real time. The contact force sensor can sense tiny force changes and convert these changes into electrical signals for transmission and processing. The system judges the real-time monitored contact force according to the preset contact force threshold. If the contact force exceeds the preset threshold, the system will automatically adjust the moving speed or direction of the probe to reduce the contact force and prevent additional damage to the pipeline.

[0044] An analysis system applied to the corrosion problem of the Texaco gasification device selects key monitoring points for monitoring the corrosion situation of the target device and installs corresponding intelligent sensors at the key monitoring points;

[0045] The operating data of each key monitoring point of the target device are collected in real time through the intelligent sensors and transmitted to the central processing unit;

[0046] Use artificial intelligence algorithms to analyze and process the collected operation data, compare with the preset normal operation data range, identify areas with possible corrosion risks and abnormal data change trends, and obtain suspected corrosion areas;

[0047] For the suspected corrosion areas, use intelligent detection technology for detection to obtain detection data;

[0048] Encrypt and store the detection data, equipment operation data, and maintenance records to ensure the authenticity, non-tampering, and traceability of the data, providing a reliable basis for subsequent analysis and decision-making;

[0049] Combined with intelligent manufacturing technology, optimize the operation parameters and maintenance plan of the device according to the corrosion analysis results, and achieve precise analysis and effective prevention and control of the corrosion problems of the target device;

[0050] Select key monitoring points for monitoring the corrosion situation of the target device, and install corresponding intelligent sensors at the key monitoring points, including:

[0051] Obtain the three-dimensional data of the target device, and establish a three-dimensional model of the target device according to the three-dimensional data of the device;

[0052] Obtain the corrosion data of the same type of products of the target device, and establish a corrosion state prediction model of the target device according to the corrosion data;

[0053] According to the three-dimensional model of the device and the corrosion state prediction model, select key monitoring points for monitoring the corrosion situation of the target device, and install corresponding intelligent sensors at the key monitoring points in combination with sensor attribute data;

[0054] As a further technical solution of the present invention, obtaining the corrosion data of the same type of products of the target device, and establishing a corrosion state prediction model of the target device according to the corrosion data, includes:

[0055] Determine the scope of the same type of products, including products identical to the target device and products similar to the target device in terms of materials, structures, working environments, and usage conditions;

[0056] Collect the corrosion data of the same type of products, and the corrosion data includes corrosion rate, corrosion type, environmental factors, and maintenance records;

[0057] Sort out and analyze the collected corrosion data, identify the key factors affecting the corrosion state, and perform data cleaning to remove outliers and noise;

[0058] Based on the sorted corrosion data, a corrosion state prediction model is established using statistical analysis methods or machine learning algorithms. Specifically: Analyze the correlation between process parameters and corrosion rate, establish a corrosion reaction kinetics model, combine with material science theory, construct a corrosion damage evolution equation, and verify the physical rationality of the model to obtain a corrosion mechanism model; Use long short-term memory networks to process time series data, use convolutional neural networks to extract spatial features, and combine with attention mechanisms to improve prediction accuracy for the training of deep learning models;

[0059] Integrate the mechanism model and the deep learning model, use ensemble learning methods to improve prediction stability, and use Bayesian optimization algorithms to adjust model hyperparameters;

[0060] Set the model update period to achieve online learning;

[0061] Verify the corrosion state prediction model, use historical data to evaluate the accuracy of the model, and adjust and optimize the model according to the evaluation results;

[0062] The steps of selecting key monitoring points for monitoring the corrosion condition of the target device and installing corresponding intelligent sensors at the key monitoring points in combination with sensor attribute data include:

[0063] Analyze the three-dimensional model of the device to identify the first areas where corrosion risks may exist;

[0064] Combined with the corrosion state prediction model, evaluate the corrosion risk levels of each first area to determine the key monitoring points that need to be monitored with emphasis;

[0065] Obtain sensor attribute data related to the key monitoring points;

[0066] According to the corrosion type and monitoring requirements, select intelligent sensors corresponding to the key monitoring points, and formulate the installation location and method of the sensors;

[0067] Install the selected intelligent sensors at the key monitoring points to ensure that the installation of the sensors complies with technical specifications and can effectively monitor the corrosion condition.

[0068] A detection method and analysis system for corrosion problems applied to Texaco gasification plants proposed by the present invention have the beneficial effects that:

[0069] The detection method and analysis system for corrosion problems applied to Texaco gasification units can comprehensively scan and accurately identify the corrosion morphology on the inner wall of pipelines through intelligent automatic diagnostic probes and advanced algorithms, including key information such as the location, type, and degree of corrosion. This not only improves the accuracy of detection but also provides reliable data support for subsequent quantitative evaluation. Based on the detailed detection results and evaluation conclusions, a more targeted maintenance plan can be formulated. This avoids the blindness and inefficiency of traditional maintenance work, ensures the effective utilization of maintenance resources, and extends the service life of equipment.

[0070] The detection method and analysis system for corrosion problems applied to Texaco gasification units also have a self-learning function, which can continuously optimize the recognition algorithm and automatically trigger an alarm mechanism when severe corrosion is detected. This helps to timely detect and handle potential corrosion problems and prevent them from further deteriorating. Through the accurate detection and evaluation of corrosion problems, possible safety hazards can be timely discovered and solved. The system can also dynamically adjust the scanning trajectory according to the pipeline shape data sensed in real time, ensuring full coverage of the detection area. This detection method is applicable not only to multiple scenarios such as coal slurry preparation systems, high-temperature gasification systems, and lock hopper slag discharge systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 FIG. is a schematic diagram of the working process of a detection method and analysis system for corrosion problems applied to Texaco gasification units proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0072] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0073] A detection method for corrosion problems applied to Texaco gasification units, referring to Figure 1 As shown, it is applied to a coal slurry preparation system, the damaged parts are coal mills, coal slurry tanks, and their attached pipelines, the material property is carbon steel, due to medium impact or relative movement, mechanical peeling of the material surface layer occurs, forming pits, grooves, sharp grooves, holes, and corrugated morphologies, and having a certain directionality, which is classified as erosion corrosion;

[0074] S1: Select the parts made of carbon steel in coal mills, coal slurry tanks, and their attached pipelines that are vulnerable to erosion corrosion damage due to medium impact or relative movement; clean the surface to be detected, removing attachments, dirt, and oxide layers;

[0075] S2: Use the detection system with intelligent automatic diagnosis probes to comprehensively scan the area to be detected, identify and record the morphological features formed by corrosion. Calculate the parameters based on the corrosion mechanism model through algorithms, calculate what corrosion mechanism exists at the thinning position, and give corresponding monitoring and detection plans according to the calculated corrosion mechanism;

[0076] S3: Analyze the severity, distribution law and development trend of erosion-corrosion according to the detection data, and evaluate its impact on the safe operation of the system in combination with the mechanical properties and service conditions of the material. Then sort out the detection results, including the location, size, morphology of erosion-corrosion and the evaluation conclusion, and form a detection report to provide a scientific basis for subsequent maintenance, repair or replacement.

[0077] It is also applied to the high-temperature gasification system. The damaged part is the gasification chamber of the gasifier, and the material properties are carbon steel, low alloy steel and stainless steel. First, due to medium impact or relative movement, mechanical spalling of the material surface layer occurs, forming pits, grooves, sharp grooves, holes and corrugated morphologies, and having a certain directionality, which is classified as erosion-corrosion. Second, when contacting with sulfide under the condition of more than 260 °C and hydrogen environment, the metal shows uniform corrosion, forming a multi-layer FeS scale structure with strong adhesion and gray luster, which is classified as high-temperature sulfide corrosion (hydrogen environment). Third, when the operating temperature is greater than 595 °C, the volume of the component increases, forming fine brittle cracks with convexities, scale and "fish tail pattern" appearance, which is classified as carburization. Fourth, in the high-temperature hydrogen environment, surface decarburization develops inward, forming bulges in the gaps at grain boundaries or inclusion interfaces, and then connecting along the grain boundaries to form microcracks. Ultrasonic thickness measurement shows abnormal "thickening", and decarburization or cracks are seen under the microscope, which is classified as high-temperature hydrogen corrosion. Fifth, in the high-temperature oxygen environment, carbon steel and low alloy steel form oxide films causing uniform thinning, and stainless steel forms dark oxide films, which is classified as high-temperature oxidation. Sixth, metal surface deposition and melting occur, and the generated slag melts the surface oxide film. An obvious smooth interface can be seen between the vitreous hard scale layer of the ash layer and the base metal, which is classified as ash corrosion.

[0078] It is also applied to the lock hopper slag discharge system. The damaged parts are the lock hopper, the flushing water tank and its attached pipelines, and the slag pond. The material properties are carbon steel and low alloy steel. First, in the acidic water environment with ammonium hydrosulfide, local corrosion occurs at the parts where the medium flow direction changes or in the turbulent flow area where the concentration of NH4HS exceeds 2%, and scaling occurs in the low-flow area, resulting in under-scale local corrosion, which is classified as acidic water corrosion. Second, ammonium chloride crystallizes and forms scale, which absorbs moisture and deliquesces, and there are mostly white, green or brown salt-like deposits, which is classified as ammonium chloride corrosion. Third, due to medium impact or relative movement, mechanical spalling of the material surface layer occurs, forming pits, grooves, sharp grooves, holes and corrugated morphologies, and having a certain directionality, which is classified as erosion-corrosion.

[0079] It is also applied to the sour gas system. The damaged parts are the pipelines at the tops of the high-pressure, low-pressure and vacuum flash drums, the pipelines at the tops of the flash separators, and the gas-phase outlet pipelines at the tops of the deaerators. The material properties are carbon steel, low alloy steel, and stainless steel. First, there is a liquid water phase containing H2S with an H2S content greater than 50 ppm, forming independent small bubbles on the steel surface. The small bubbles do not merge with each other and are classified as wet H2S damage - hydrogen blistering. Step-shaped cracks parallel to the surface and extending along the rolling direction are formed inside the steel and are classified as wet H2S damage - hydrogen-induced cracking. Hydrogen-induced cracking cracks at different depths on the base metal in the heat-affected zone of the weld are connected along the thickness direction and are classified as wet H2S damage - stress-oriented hydrogen-induced cracking. Cracks initiate on the surface of the heat-affected zone of the weld and the high-hardness zone and extend along the thickness direction, and are classified as wet H2S damage - stress corrosion cracking. Second, due to contact with hydrochloric acid, it is usually uniform corrosion, and local corrosion or under-deposit corrosion occurs during local concentration or dew point corrosion. Pitting corrosion appears on the stainless steel and is classified as hydrochloric acid corrosion. Third, in an acidic water environment containing hydrogen sulfide with a pH value of 4.5 - 7.0, it is generally uniform corrosion, and local corrosion or local under-deposit corrosion is likely to occur in the presence of oxygen. Pitting corrosion appears on 300SS and is classified as acidic water corrosion. Fourth, in a humid carbon dioxide environment below 140°C, it occurs in the turbulent flow and liquid impact areas and at the root of the weld, causing wall thickness reduction, corrosion pits or holes. Deeper pitting corrosion pits and grooves may be formed when carbon steel corrodes in the turbulent flow area and are classified as carbon dioxide corrosion.

[0080] It is also applied to the black water and grey water systems. The damaged parts are the quench chamber of the gasifier, the Venturi scrubber, the scrubbing tower, the scrubbing water system, the liquid phase parts of the high-pressure, low-pressure and vacuum flash systems, and the liquid-phase outlet pipelines at the bottom of the deaerator. When the material properties are carbon steel and low alloy steel, first, in an acidic water environment containing ammonium hydrosulfide, local corrosion forms at the parts where the medium flow direction changes or in the turbulent flow area where the NH4HS concentration exceeds 2%. Scaling occurs in the low-flow rate area, and under-deposit local corrosion occurs, which is classified as acidic water corrosion. Second, due to medium impact or relative movement, mechanical peeling of the material surface layer occurs, forming pits, grooves, sharp grooves, holes and corrugated morphologies with a certain directionality, which is classified as erosion corrosion. Third, due to the crystallization and scaling of ammonium chloride and its hygroscopic deliquescence, white, green or brown salt-like deposits are mostly present, which is classified as ammonium chloride corrosion. Fourth, when the gas-phase CO2 concentration > 2% and the temperature > 93°C, the liquid-phase pH value > 9.0 and the carbonate concentration > 100 ppm, or the pH value is in the range of 8.0 - 9.0 and the carbonate concentration > 400 ppm, cracks extending along the direction parallel to the weld appear on the base metal near the welded joint. The cracks mainly propagate along the grain boundaries, and the inside is filled with oxide cracks that are fine and spider-web-like, which is classified as carbonate stress corrosion cracking.

[0081] When the material properties are 300 series stainless steel and 400 series stainless steel, first, under the combined action of tensile stress, temperature > 38 °C, and chloride aqueous solution environment with pH > 2, dendritic and branched transgranular propagation cracks are formed on the surface, and the fracture surface is usually brittle and without obvious thinning, which is classified as chloride stress corrosion cracking; second, under the impact of the medium or relative movement, mechanical spalling of the material surface layer occurs, forming pits, grooves, sharp grooves, holes, and corrugated morphologies with a certain directionality, which is classified as erosion corrosion; third, ammonium chloride crystallizes and forms scale, which absorbs moisture and deliquesces, and there are white, green, or brown salt-like deposits that are easily removed. Usually, local corrosion occurs under the scale layer, which is classified as ammonium chloride corrosion.

[0082] Among them, S2 includes:

[0083] System deployment: Intelligent automatic diagnostic probes are deployed at key pipeline parts. The probes have data acquisition functions and are used to directly contact the inner wall of the pipeline for scanning the corrosion morphology.

[0084] Start scanning: Start the probe to comprehensively scan the inner wall of the pipeline along a preset trajectory or a trajectory adjusted according to the real-time pipeline shape, and collect corrosion morphology data in real time.

[0085] Data preprocessing: Input the collected corrosion morphology data into the data processing and analysis unit for data preprocessing operations.

[0086] Feature extraction and recognition: Use image processing technology and machine learning algorithms to extract the features of the corrosion morphology from the preprocessed data, and compare the extracted features with the preset corrosion morphology database to identify the type and degree of corrosion.

[0087] Recording and display: Record the identified corrosion morphology features in the form of graphics or data and display them through the display system. At the same time, export the data to the report generation module to generate a detailed detection report.

[0088] Intelligent optimization: The detection system has a self-learning function, can optimize the recognition algorithm according to the continuously accumulated detection data, and automatically trigger the alarm mechanism when severe corrosion is detected.

[0089] The probe dynamically adjusts the scanning trajectory according to the actual shape of the inner wall of the pipeline to fully cover the detection area, and monitors the contact force between the probe and the inner wall of the pipeline in real time.

[0090] The dynamic scanning trajectory adjustment includes the following steps:

[0091] Before the detection starts, the system first plans a preliminary scanning path based on the shape data of the pipeline. When the probe starts scanning, the laser rangefinder and ultrasonic sensor it is equipped with will perceive the pipeline shape in real time. The system will compare the shape data perceived in real time with the initial scanning path and calculate the deviation. According to the calculated deviation, the system dynamically adjusts the movement trajectory of the probe.

[0092] The system adopts a particle swarm trajectory optimization algorithm to find the optimal scanning trajectory during the real-time perception and adjustment process. The algorithm considers factors such as scanning speed, the contact force between the probe and the inner wall of the pipeline, and detection accuracy.

[0093] The algorithm steps include the following steps:

[0094] Initialize the particle swarm:

[0095] Determine the size N of the particle swarm, that is, the number of candidates for the trajectory.

[0096] Randomly assign an initial position and velocity to each particle. Here, the position can be represented as a parameter or coordinate sequence of the scanning trajectory, and the velocity represents the change rate of the trajectory parameters or coordinates.

[0097] Initialize the personal best position (pbest) of each particle to its current position, and the global best position (gbest) to the optimal position among all particles.

[0098] Fitness function design and iterative update:

[0099] According to factors such as scanning speed, the contact force between the probe and the inner wall of the pipeline, and detection accuracy, design a fitness to evaluate the quality of each particle, and obtain a comprehensive evaluation index through weighted summation or multiplication. In each iteration, calculate the fitness value of each particle according to the fitness function.

[0100] Update the personal best position (pbest) of each particle. If the current fitness value is better than the previous personal best fitness value, update the global best position (gbest) if the fitness value of the current particle is better than the previous global best fitness value.

[0101] Update the velocity and position of each particle according to the following formula:

[0102] Velocity update formula: Vi(t + 1) = w * Vi(t) + c1 * r1 * (pbesti - Xi(t)) + c2 * r2 * (gbest - Xi(t))

[0103] Where, w is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers, Xi(t) is the position of particle i at time t, and Vi(t) is the velocity of particle i at time t.

[0104] Position update formula: Xi(t + 1) = Xi(t) + Vi(t + 1)

[0105] Termination conditions: When the preset maximum number of iterations is reached, the fitness value of the global best position reaches the preset threshold, and the change in the fitness value is less than the preset threshold, indicating that the algorithm has converged;

[0106] Output the optimal solution:

[0107] When the termination conditions are met, output the global best position as the optimal scanning trajectory.

[0108] A contact force sensor is integrated on the probe to monitor the contact force between the probe and the inner wall of the pipeline in real time. The contact force sensor can sense minute force changes and convert these changes into electrical signals for transmission and processing. The system judges the real-time monitored contact force according to the preset contact force threshold. If the contact force exceeds the preset threshold, the system will automatically adjust the moving speed or direction of the probe to reduce the contact force and prevent additional damage to the pipeline.

[0109] The probe is also integrated with any one of the detection technologies such as ultrasonic, eddy current, or magnetic particle detection. After the detection, the probe performs three-dimensional reconstruction on the corrosion morphology data to visually display the corrosion situation.

[0110] Automatically trigger the alarm mechanism when severe corrosion is detected, including the following steps:

[0111] The system needs to preset a threshold related to the corrosion degree, which is based on historical data, industry standards, safety specifications, or expert experience. When the detected corrosion degree exceeds the preset threshold, the system will trigger an alarm; the alarm mechanism is manifested as an audible and visual alarm, which attracts the operator's attention by emitting a sound and flashing a light. At the same time, the intelligent automatic diagnosis probe performs a screen display, and alarm information is displayed on the display screen, including the corrosion location, degree, and recommended countermeasures.

[0112] An analysis system applied to the corrosion problem of the Texaco gasification device selects key monitoring points for monitoring the corrosion situation of the target device and installs corresponding intelligent sensors at the key monitoring points;

[0113] The intelligent sensors collect the operation data of each key monitoring point of the target device in real time and transmit the operation data to the central processing unit;

[0114] Use artificial intelligence algorithms to analyze and process the collected operation data, compare with the preset normal operation data range, identify the areas and abnormal data change trends that may have corrosion risks, and obtain suspected corrosion areas;

[0115] For suspected corrosion areas, use intelligent detection technology to conduct detection and obtain detection data;

[0116] Encrypt and store the detection data, equipment operation data, and maintenance records to ensure the authenticity, immutability, and traceability of the data, providing a reliable basis for subsequent analysis and decision-making;

[0117] Combined with intelligent manufacturing technology, optimize the operation parameters and maintenance plan of the device according to the corrosion analysis results to achieve precise analysis and effective prevention and control of the corrosion problems of the target device;

[0118] Select key monitoring points for monitoring the corrosion situation of the target device and install corresponding intelligent sensors at the key monitoring points, including:

[0119] Obtain the three-dimensional data of the target device and establish a three-dimensional model of the target device based on the three-dimensional data of the device;

[0120] Obtain the corrosion data of the same type of products of the target device and establish a corrosion state prediction model of the target device based on the corrosion data;

[0121] According to the three-dimensional model of the device and the corrosion state prediction model, select key monitoring points for monitoring the corrosion situation of the target device, and combine the sensor attribute data to install corresponding intelligent sensors at the key monitoring points;

[0122] Obtain the corrosion data of the same type of products of the target device and establish a corrosion state prediction model of the target device based on the corrosion data, including:

[0123] Determine the scope of the same type of products, including products identical to the target device and products similar to the target device in terms of materials, structure, working environment, and usage conditions;

[0124] Collect the corrosion data of the same type of products, and the corrosion data includes corrosion rate, corrosion type, environmental factors, and maintenance records;

[0125] Sort out and analyze the collected corrosion data, identify the key factors affecting the corrosion state, and perform data cleaning to remove outliers and noise;

[0126] Based on the sorted corrosion data, use statistical analysis methods or machine learning algorithms to establish a corrosion state prediction model. Specifically: analyze the correlation between process parameters and corrosion rate, establish a corrosion reaction kinetics model, combine material science theory, construct a corrosion damage evolution equation, and verify the physical rationality of the model to obtain a corrosion mechanism model; use long short-term memory networks to process time series data, use convolutional neural networks to extract spatial features, and combine attention mechanisms to improve prediction accuracy for training deep learning models;

[0127] Integrate the mechanism model with the deep learning model, adopt the ensemble learning method to improve the prediction stability, and use the Bayesian optimization algorithm to adjust the model hyperparameters;

[0128] Set the model update period to achieve online learning;

[0129] Verify the corrosion state prediction model, use historical data to evaluate the accuracy of the model, and adjust and optimize the model according to the evaluation results;

[0130] Steps to select key monitoring points for monitoring the corrosion situation of the target device and install corresponding intelligent sensors at the key monitoring points in combination with sensor attribute data, including:

[0131] Analyze the three-dimensional model of the device to identify the first areas where corrosion risks may exist;

[0132] Combine the corrosion state prediction model to evaluate the corrosion risk levels of each first area and determine the key monitoring points that need to be focused on;

[0133] Obtain the sensor attribute data related to the key monitoring points;

[0134] According to the corrosion type and monitoring requirements, select the intelligent sensors corresponding to the key monitoring points and formulate the installation positions and methods of the sensors;

[0135] Install the selected intelligent sensors at the key monitoring points to ensure that the installation of the sensors complies with the technical specifications and can effectively monitor the corrosion situation.

[0136] Analysis methods for the corrosion problems of Texaco gasification devices, including:

[0137] Select key monitoring points for monitoring the corrosion situation of the target device and install corresponding intelligent sensors at the key monitoring points;

[0138] Real-time collect the operation data (including but not limited to temperature, pressure, medium composition, flow rate, image and other data) of each key monitoring point of the target device through the intelligent sensors, and transmit the operation data to the central processing unit (the intelligent sensors include temperature sensors, pressure sensors, chemical composition sensors, flow sensors, etc., which are respectively installed at the key parts of the Texaco gasification device, such as gasifiers, pipelines, heat exchangers, etc., to comprehensively obtain the device operation status information);

[0139] Use artificial intelligence algorithms to analyze and process the collected operation data, compare with the preset normal operation data range, identify areas that may have corrosion risks (i.e., areas where the area data meets the preset first condition) and abnormal data change trends, and obtain suspected corrosion areas (the artificial intelligence algorithms include neural network algorithms, deep learning algorithms, etc. Through learning and training on a large amount of historical data, they can accurately identify abnormal data patterns related to corrosion and predict the development trend of corrosion);

[0140] For the suspected corrosion areas, use intelligent detection technologies for detection (such as using pulsed eddy current detection, ultrasonic detection, ray detection, etc. to determine the wall thickness change, internal structure damage, etc. of the equipment or pipeline; the pulsed eddy current detection equipment used in the intelligent detection technology can accurately measure the wall thickness of the pipe, with a detection accuracy of ±0.1mm, and has the function of automatically identifying and positioning corrosion defects; the ultrasonic detection equipment can detect microcracks and internal structure damage, and the minimum detectable crack length is 0.5mm; the ray detection equipment can perform imaging analysis on the internal structure of the device, and the image resolution reaches 0.1mm), and obtain detection data;

[0141] Encrypt and store the detection data, equipment operation data, and maintenance records to ensure the authenticity, non-tampering, and traceability of the data, and provide a reliable basis for subsequent analysis and decision-making (the encryption algorithm used in the blockchain technology is the Advanced Encryption Standard (AES) algorithm to encrypt the data and ensure the security of the data during transmission and storage; at the same time, through the distributed ledger technology, data sharing and synchronization are realized to ensure the consistency and integrity of the data);

[0142] Combined with intelligent manufacturing technology, optimize the operation parameters and maintenance plan of the device according to the corrosion analysis results, and achieve accurate analysis and effective prevention and control of the corrosion problem of the target device (the intelligent manufacturing technology realizes the real-time adjustment of the operation parameters of the Texaco gasification device through an automatic control system, such as automatically adjusting parameters such as the medium flow rate and temperature according to the corrosion risk to slow down the corrosion process; at the same time, use robot technology for equipment maintenance and repair work to improve maintenance efficiency and safety).

[0143] Adopting the technical solution of this embodiment, by installing intelligent sensors at key monitoring points, the operation data of the target device can be collected in real time, making the understanding of the equipment status more timely and accurate, and helping to quickly identify potential corrosion risks; using artificial intelligence algorithms to analyze the collected operation data can effectively identify the deviations from the normal operation data range, not only improving the efficiency of data processing, but also accurately locating the areas and abnormal data change trends that may have corrosion risks, so as to achieve early warning of corrosion problems; for suspected corrosion areas, using intelligent detection technology for in-depth detection can obtain more detailed detection data, and this precise detection ability provides a scientific basis for subsequent maintenance decisions, ensuring the pertinence and effectiveness of maintenance work; encrypting and storing the detection data, equipment operation data and maintenance records to ensure the authenticity, immutability and traceability of the data, not only protecting the enterprise's core data assets, but also providing a reliable basis for subsequent analysis and decision-making; combining the corrosion analysis results can optimize the operation parameters and maintenance plan of the device, not only improving the operation efficiency of the equipment, but also effectively reducing the risks brought by corrosion and extending the service life of the equipment; through the comprehensive application of the above measures, accurate analysis and effective prevention and control of the corrosion problems of the target device are achieved, and the overall equipment management level and safety are improved.

[0144] In some possible implementation manners of the present invention, the step of selecting key monitoring points for monitoring the corrosion situation of the target device and installing corresponding intelligent sensors at the key monitoring points includes:

[0145] Obtaining the three-dimensional data of the device of the target device and establishing a three-dimensional model of the target device according to the three-dimensional data of the device;

[0146] In this step, the target device is scanned omnidirectionally by a 3D laser scanner to obtain device point cloud data; the device point cloud data is preprocessed (including: removing noise points through a spatial filtering algorithm; simplifying the point cloud using a voxel grid downsampling method; registering multi-station point cloud data using an iterative closest point algorithm); based on the preprocessed point cloud data, a region growing algorithm is used for point cloud segmentation to separate different components of the target device; for each segmented component point cloud, geometric feature parameters (including but not limited to flatness, roundness, cylindricity, etc.) are extracted, standard geometric bodies (including but not limited to planes, cylindrical surfaces, spherical surfaces, conical surfaces, etc.) are identified, and surface equations are fitted; the identified geometric bodies and fitted surfaces are combined with the engineering design specifications of the Texaco gasification device to construct a parametric 3D solid model; the accuracy of the established 3D solid model is verified (including: comparing the model with the original point cloud data; calculating the key dimension error, which should be controlled within ±0.5 mm; verifying the geometric feature parameters of key components); a multi-level device information model containing the following information is established using a feature hierarchical storage method (including: a geometric feature layer (storing geometric information such as the device's shape and dimensions), a material feature layer (recording the material information of each component), a process feature layer (containing information such as process parameters and medium flow direction), a detection feature layer (recording historical detection data and detection point location information)); regular tracking scans are performed to obtain real-time deformation data of the device; the deformation data is compared and analyzed with the reference model; the geometric parameters and detection data in the 3D model are updated. Through this step, the high-precision data obtained using 3D scanning technology can accurately reflect the actual shape and dimensions of the target device, reducing human error; through the processing and analysis of 3D data, the geometric features and surface conditions of the target device can be comprehensively obtained, providing basic data for subsequent corrosion monitoring and analysis; through the verification and optimization of the 3D model, the accuracy and practicality of the model are ensured, thereby improving the effectiveness of subsequent analysis and monitoring.

[0147] Obtain the corrosion data of the same type of products of the target device, and establish a corrosion state prediction model of the target device based on the corrosion data;

[0148] According to the device 3D model and the corrosion state prediction model, select key monitoring points for monitoring the corrosion situation of the target device, and install corresponding intelligent sensors (corresponding to the corrosion type) at the key monitoring points in combination with sensor attribute data.

[0149] In some possible implementation manners of the present invention, the step of obtaining the corrosion data of the same type of products of the target device and establishing a corrosion state prediction model of the target device based on the corrosion data includes:

[0150] Determine the scope of similar products, including products identical to the target device and products similar to the target device in terms of materials, structure, working environment, usage conditions, etc.;

[0151] Collect the corrosion data of the similar products, where the corrosion data includes corrosion rate, corrosion type, environmental factors (such as temperature, humidity, medium composition, etc.), and maintenance records;

[0152] Sort out and analyze the collected corrosion data, identify the key factors affecting the corrosion state, and perform data cleaning to remove outliers and noise;

[0153] Based on the sorted corrosion data, establish a corrosion state prediction model using statistical analysis methods or machine learning algorithms (the model can describe the relationship between corrosion rate and environmental factors). Specifically: analyze the correlation between process parameters and corrosion rate, establish a corrosion reaction kinetics model, combine with material science theory, construct a corrosion damage evolution equation, and verify the physical rationality of the model to obtain a corrosion mechanism model; use Long Short-Term Memory Network (LSTM) to process time series data, use Convolutional Neural Network (CNN) to extract spatial features, and combine with an attention mechanism to improve the prediction accuracy for the training of the deep learning model;

[0154] Integrate the mechanism model and the deep learning model, use ensemble learning methods to improve the prediction stability, and use the Bayesian optimization algorithm to adjust the model hyperparameters;

[0155] Set the model update period to achieve online learning;

[0156] Verify the corrosion state prediction model, use historical data to evaluate the accuracy of the model, and adjust and optimize the model according to the evaluation results.

[0157] In this embodiment, by obtaining the corrosion data of similar products, it is possible to provide a data-based analysis basis for the corrosion state of the target device and enhance the scientific nature of decision-making; the established corrosion state prediction model can effectively predict the corrosion behavior of the target device under specific environmental conditions, helping to formulate corresponding maintenance and protection measures; through the accurate prediction of the corrosion state, potential corrosion risks can be identified in advance, thereby reducing the occurrence probability of equipment failures and safety accidents.

[0158] In some possible implementation manners of the present invention, the step of selecting key monitoring points for monitoring the corrosion situation of the target device according to the three-dimensional model of the device and the corrosion state prediction model, and installing corresponding intelligent sensors at the key monitoring points in combination with sensor attribute data includes:

[0159] Analyze the three-dimensional model of the device to identify the first area where corrosion risks may exist;

[0160] In this step, first analyze the three-dimensional model of the device to identify the first areas where corrosion risks may exist, such as high-stress areas, fluid flow areas, and temperature change areas, etc. Then further analyze these areas to ensure accurate detection while saving resources.

[0161] Combined with the corrosion state prediction model, evaluate the corrosion risk levels of each first area and determine the key monitoring points that need to be focused on;

[0162] In this step, based on the three-dimensional model of the device and the corrosion state prediction model, identify the risk areas, mark the predicted corrosion risk levels in the three-dimensional model, identify the geometric features of the high-corrosion-risk areas, determine the key parts in combination with the process flow, and establish a priority ranking for the risk areas.

[0163] Obtain the sensor attribute data related to the key monitoring points (including information such as the type, measurement range, accuracy, and response time of the sensors);

[0164] According to the corrosion type and monitoring requirements, select the intelligent sensors corresponding to the key monitoring points and formulate the installation positions and methods of the sensors;

[0165] In this step, the intelligent sensor matching process includes: selecting the corresponding detection principle according to the corrosion type, considering the influence of environmental conditions on the sensors, evaluating the measurement range and accuracy requirements of the sensors, and analyzing the interference effects between the sensors. Determining the monitoring point layout plan includes: using the grid division method to spatially discretize the high-risk areas, determining the stress concentration positions based on finite element analysis, determining the medium residence areas in combination with the flow field analysis, screening the monitoring points considering the detection accessibility, optimizing the spatial distribution of the monitoring points, and achieving effective coverage of the risk areas.

[0166] Install the selected intelligent sensors at the key monitoring points to ensure that the installation of the sensors complies with the technical specifications and can effectively monitor the corrosion situation.

[0167] In this embodiment, by scientifically selecting the key monitoring points, accurate monitoring of the corrosion situation of the target device can be achieved, improving the effectiveness and pertinence of the monitoring; combined with the three-dimensional model and the corrosion state prediction model, potential corrosion risk areas can be better identified and protective measures can be taken in advance; selecting appropriate sensors according to the sensor attribute data can ensure the flexibility and adaptability of the monitoring system and improve the accuracy of data acquisition; by real-time monitoring of the corrosion situation, problems can be discovered and maintained in a timely manner, reducing the equipment failure rate and improving the overall operation efficiency.

[0168] In some possible embodiments of the present invention, the step of analyzing and processing the collected operation data using an artificial intelligence algorithm, comparing with a preset normal operation data range, identifying areas with possible corrosion risks and abnormal data change trends, and obtaining suspected corrosion areas includes:

[0169] Perform noise reduction and standardization processing on the operation data, decompose the time series data through wavelet transform, extract time-domain features (such as mean, variance, peak value, etc.) and frequency-domain features, and construct a multi-dimensional feature vector (including process parameters, material properties, environmental factors, etc.);

[0170] Establish a multi-level anomaly detection model. Among them, the first layer combines the preset normal operation data range and uses statistical process control methods to establish a parameter reference range. The second layer uses the isolation forest algorithm to identify abnormal data points. The third layer applies an autoencoder for data reconstruction to detect abnormal patterns;

[0171] Establish an anomaly scoring mechanism in the multi-level anomaly detection model to output multi-level detection results;

[0172] Obtain the anomaly detection result according to the multi-dimensional feature vector and the multi-level anomaly detection model;

[0173] Use a long short-term memory network (LSTM) to analyze the time series correlation, use a graph neural network (GNN) to model the association relationship between device components, and combine an expert rule system to construct a hybrid inference model, and integrate multiple models to obtain a deep learning model;

[0174] Combine the deep learning model and real corrosion data (including device attributes of real corrosion conditions, corrosion area data, corrosion process development data, device environment data, etc.) to establish a risk assessment model based on fuzzy logic, and define multi-level risk thresholds and warning levels;

[0175] According to the anomaly detection result and the risk assessment model, calculate the corrosion risk data of each key monitoring point and generate a heat map of the risk area;

[0176] Establish a time series prediction model, analyze the influence degree of parameter changes on corrosion, predict the evolution trend of key parameters, and evaluate the corrosion development data;

[0177] According to the corrosion risk data and the corrosion development data, conduct data analysis and comparison to obtain suspected corrosion areas (that is, areas where the area data reaches a preset first condition, and the first condition can be set first image data or first substance composition data, etc.), and generate a corresponding report.

[0178] In this embodiment, the use of artificial intelligence algorithms can quickly and accurately identify potential corrosion risk areas, improving the timeliness and accuracy of monitoring; through learning a large amount of historical data, it can effectively identify abnormal data patterns related to corrosion, enhancing the prediction ability of corrosion development; the analysis results provide data support for subsequent maintenance and management decisions, helping to formulate more scientific maintenance strategies; by identifying corrosion risks in advance, it can reduce equipment failures and downtime, thereby reducing maintenance costs and improving the overall reliability of the equipment.

[0179] In some possible embodiments of the present invention, the step of using intelligent detection technology to detect the suspected corrosion area to obtain detection data includes:

[0180] Extract location features from the suspected corrosion area;

[0181] Select a suitable intelligent detection method according to the location features of the suspected corrosion area, establish a priority ranking for location point detection, plan the detection path, and formulate detection quality control standards (intelligent detection methods include pulsed eddy current detection, ultrasonic detection, and ray detection, ensuring that the selected technology can meet the detection requirements) to obtain a detection plan;

[0182] According to the detection plan, use pulsed eddy current detection equipment to detect the pipeline or equipment (set detection parameters to ensure that the detection accuracy reaches ±0.1 mm), automatically identify and locate corrosion defects, and record wall thickness change data;

[0183] According to the detection plan, use ultrasonic detection equipment to scan the suspected corrosion area, set the detection frequency and waveform to ensure that small cracks and internal structure damage can be detected (such as ensuring that the minimum detectable crack length is 0.5 mm), and record the detection results;

[0184] According to the detection plan, use ray detection equipment to perform imaging analysis on the internal structure of the device, set the imaging parameters to ensure that the image resolution reaches the preset value (such as 0.1 mm), record the imaging results, and analyze the integrity of the internal structure;

[0185] According to the detection plan, use a camera device to collect image data of the suspected corrosion area;

[0186] Comprehensively analyze all the above detection data to form a detection analysis report, and output the specific location, type, and severity of the corrosion defects.

[0187] In this step, establish a multi-source detection data registration method, adopt the Bayesian fusion algorithm, generate a comprehensive detection report, and evaluate the reliability of the detection results. It also includes: extracting defect feature parameters, establishing a defect classification model, estimating the defect hazard degree, and providing repair suggestions.

[0188] In this embodiment, a variety of intelligent detection technologies such as pulsed eddy current detection, ultrasonic detection, and ray detection are adopted, which can accurately measure the wall thickness and internal structure damage, ensuring the accuracy of the detection results; through the combination of different detection methods, the state of the suspected corrosion area can be comprehensively evaluated, and potential safety hazards can be discovered in a timely manner; the automatic identification and positioning function of the pulsed eddy current detection equipment improves the detection efficiency and reduces the need for manual intervention.

[0189] In some possible embodiments of the present invention, the step of encrypting and storing the detection data, equipment operation data, and maintenance records to ensure the authenticity, non-tampering, and traceability of the data, and providing a reliable basis for subsequent analysis and decision-making, includes:

[0190] Collect the data to be stored, including detection data, equipment operation data, and maintenance records, to ensure the integrity and accuracy of the data;

[0191] Adopt the Advanced Encryption Standard (AES) algorithm, select the key length (128 bits, 192 bits, or 256 bits), and generate an encryption key according to the security requirements;

[0192] Divide the data to be encrypted into data blocks of a fixed size (128 bits) and encrypt each data block;

[0193] Adopt a suitable encryption mode (such as CBC or GCM) for encryption operations to ensure the security and efficiency of the encryption process;

[0194] Store the encrypted data in the blockchain system, and use the distributed ledger technology of the blockchain to achieve data sharing and synchronization (ensure the authenticity, non-tampering, and traceability of the data through blockchain technology, and record the modification and access history of each data).

[0195] In this embodiment, the construction of the blockchain system includes: designing the data block structure, using the SHA-256 hash algorithm to generate block links, using the Merkle tree structure to organize data, and implementing intelligent contract management of data access rights; adopting a multi-node storage architecture to achieve data sharding storage, establishing a data synchronization mechanism between nodes to achieve load balancing; establishing a data integrity verification method to achieve data version control, recording data operation logs, and establishing a data traceability mechanism. This embodiment also includes: implementing regular data backup, establishing a multi-level backup strategy, designing a data recovery process, and conducting regular recovery drills; establishing a role-based access control system, implementing multi-factor identity authentication, recording access operation logs, and conducting regular security audits.

[0196] In this embodiment, the AES algorithm is used to encrypt data to ensure the security of data during transmission and storage, prevent data leakage and unauthorized access; through the characteristics of blockchain technology, ensure that the stored data is true and reliable, and any modification to the data can be traced and verified; utilize distributed ledger technology to achieve data sharing and synchronization, and ensure the consistency and integrity of data between different nodes.

[0197] In some possible embodiments of the present invention, the step of combining intelligent manufacturing technology to optimize the operating parameters and maintenance plan of the device according to the corrosion analysis results to achieve precise analysis and effective prevention and control of the corrosion problem of the target device includes:

[0198] Collect the corrosion analysis results of the target device and identify the corrosion risk areas and their influencing factors;

[0199] Establish a relationship model between the device operating parameters and the corrosion risk, and determine the influence of key operating parameters (such as medium flow rate, temperature, pressure, etc.) on the corrosion process;

[0200] Real-time monitor the operating status of the device through an automated control system to obtain the current operating parameter data;

[0201] According to the corrosion analysis results and the current operating parameter data, automatically adjust the operating parameters of the device (if it is detected that the corrosion risk increases, automatically adjust the medium flow rate and temperature to slow down the corrosion process);

[0202] Formulate and optimize the maintenance plan, and determine the priority and frequency of maintenance in combination with the corrosion analysis results;

[0203] Utilize robotics technology to perform equipment maintenance and repair work, automatically execute routine maintenance tasks, and improve maintenance efficiency and safety;

[0204] Evaluate the optimized operating parameters and maintenance plan, continuously monitor its effectiveness, and make dynamic adjustments according to the feedback.

[0205] In this embodiment, by combining the corrosion analysis results and intelligent manufacturing technology, it is possible to achieve precise analysis and effective prevention and control of the corrosion problem of the target device, reduce the risks brought by corrosion; the application of the automated control system enables the real-time adjustment of the device operating parameters, quickly respond to changes in corrosion risks, and slow down the corrosion process; using robotics technology for equipment maintenance can reduce manual intervention, improve maintenance efficiency and safety, and reduce labor costs.

[0206] In some possible embodiments of the present invention, the key monitoring points include: the acid gas outlet pipeline of the high-pressure flash condenser, the acid gas outlet pipeline of the vacuum condenser, the acid gas outlet pipeline at the top of the deaerator, the raw syngas outlet pipeline at the top of the scrubber, the outlet pipeline of the coal mill discharge pump, the outlet pipeline of the coal slurry feed pump, the outlet pipeline of the high-pressure ash water pump, the outlet pipeline of the low-pressure ash water pump, the bottom outlet pipeline of the high-pressure flash separator, the bottom outlet pipeline of the high-pressure flash tank, the bottom outlet pipeline of the low-pressure flash tank, the bottom outlet pipeline of the vacuum flash tank, the black water outlet pipeline of the gasifier, the pipeline from the scrubber to the high-pressure flash tank, the outlet pipeline of the quench water pump, the outlet pipeline of the lock hopper circulation water pump, the outlet pipeline of the relief line at the top of the lock hopper, the pipeline from the bottom of the gasifier to the inlet of the lock hopper (select a suitable monitoring system according to the corrosion mechanism of different monitoring points, such as corrosion probes, on-line thickness measurement, etc., to monitor the corrosion conditions of each point in real time and feed the monitoring data back to the intelligent monitoring system).

[0207] In some possible embodiments of the present invention, the specific monitoring requirements for the key monitoring points are as follows:

[0208] For the acid gas related pipelines such as the acid gas outlet pipeline of the high-pressure flash condenser, the acid gas outlet pipeline of the vacuum condenser, and the acid gas outlet pipeline at the top of the deaerator, monitor the changes in the acid gas components and the wall thickness changes at the condensation part to determine the first corrosion data of wet hydrogen sulfide corrosion, hydrochloric acid corrosion, acidic acid water corrosion, and carbon dioxide corrosion (the operating conditions are 130°C, 0.2 Mpa; 40°C, -0.06 MPa; 40°C, -0.08 MPa; 120°C, 0.15 MPa, etc. Select appropriate monitoring systems and monitoring frequencies according to the corrosion mechanism and operating conditions to ensure timely detection of corrosion problems); for the crude syngas outlet pipeline at the top of the scrubber, monitor the changes in the crude syngas components and the wall thickness changes to determine the second corrosion data of ammonium chloride corrosion and chloride stress corrosion cracking (the operating conditions are 240°C, 6.0 MPa, and conduct real-time monitoring through online thickness measurement and other methods to provide data support for the operation of the device); for the outlet pipelines of the coal mill discharge pump and the coal slurry feed pump, monitor the wall thickness changes of the coal grinding feed system to determine the erosion corrosion data (the operating conditions are 50°C, 6.5 MPa; 50°C, 1.0 MPa, and select appropriate monitoring means according to the actual situation to ensure the stable operation of the system); for the outlet pipelines of the high-pressure ash water pump, low-pressure ash water pump, the bottom outlet pipeline of the high-pressure flash separator, the bottom outlet pipeline of the high-pressure flash tank, the bottom outlet pipeline of the low-pressure flash tank, the bottom outlet pipeline of the vacuum flash tank, the black water outlet pipeline of the gasifier, the pipeline from the scrubber to the high-pressure flash tank, the outlet pipeline of the quench water pump, the outlet pipeline of the lock hopper circulation water pump, the outlet pipeline of the lock hopper top pressure relief line, and the pipeline from the bottom of the gasifier to the inlet of the lock hopper, monitor the wall thickness changes of the corresponding systems to determine the third corrosion data of basic acidic water corrosion, ammonium chloride corrosion, erosion corrosion, chloride stress corrosion cracking, and carbonate stress corrosion cracking (adopt corresponding monitoring equipment and technologies according to different operating conditions such as temperature and pressure to comprehensively master the corrosion conditions of each system).

[0209] The detection method and analysis system for corrosion problems applied to Texaco gasification plants in this application can comprehensively scan and accurately identify the corrosion morphology on the inner wall of the pipeline, including key information such as the location, type, and degree of corrosion, through intelligent automatic diagnostic probes and advanced algorithms. This not only improves the accuracy of detection but also provides reliable data support for subsequent quantitative evaluation. Based on the detailed detection results and evaluation conclusions, a more targeted maintenance plan can be formulated. This avoids the blindness and inefficiency of traditional maintenance work, ensures the effective utilization of maintenance resources, and extends the service life of the equipment.

[0210] The detection method and analysis system for corrosion problems applied to Texaco gasification units in this application also have a self-learning function, which can continuously optimize the recognition algorithm and automatically trigger an alarm mechanism when severe corrosion is detected. This helps to timely discover and handle potential corrosion problems, prevent their further deterioration, and thus reduce the risk of equipment failures. Through the accurate detection and evaluation of corrosion problems, possible safety hazards can be timely discovered and solved, improving the overall safety and reliability of the equipment. This is of great significance for ensuring the smooth progress of industrial production and the safety of personnel. The detection system adopts intelligent and automated technologies, reducing manual intervention and errors, and improving the detection efficiency and accuracy. At the same time, the system can also dynamically adjust the scanning trajectory according to the pipeline shape data sensed in real time, ensuring full coverage of the detection area. This detection method is not only applicable to multiple scenarios such as coal slurry preparation systems, high-temperature gasification systems, and lock hopper slag discharge systems, but also can accurately identify and evaluate different types of corrosion problems (such as erosion corrosion, high-temperature sulfide corrosion, acidic water corrosion, etc.). This enhances the flexibility and practicality of the detection method.

[0211] Adopt the technical solution of the present invention, select the key monitoring points for monitoring the corrosion situation of the target device, and install corresponding intelligent sensors at the key monitoring points; collect the operation data of each key monitoring point of the target device in real time through the intelligent sensors, and transmit the operation data to the central processing unit; use artificial intelligence algorithms to analyze and process the collected operation data, compare with the preset normal operation data range, identify the areas that may have corrosion risks and abnormal data change trends, and obtain suspected corrosion areas; for the suspected corrosion areas, use intelligent detection technology for detection to obtain detection data; encrypt and store the detection data, equipment operation data, and maintenance records to ensure the authenticity, non-tampering, and traceability of the data, providing a reliable basis for subsequent analysis and decision-making; combine intelligent manufacturing technology, optimize the operation parameters and maintenance plan of the device according to the corrosion analysis results, and achieve precise analysis and effective prevention and control of the corrosion problem of the target device. The solution of the present invention can, by installing intelligent sensors at key monitoring points, collect the operation data of the target device in real time, making the understanding of the equipment status more timely and accurate, and helping to quickly identify potential corrosion risks; using artificial intelligence algorithms to analyze the collected operation data can effectively identify the deviation from the normal operation data range, not only improving the efficiency of data processing, but also accurately locating the areas that may have corrosion risks and abnormal data change trends, so as to achieve early warning of corrosion problems; for the suspected corrosion areas, using intelligent detection technology for in-depth detection can obtain more detailed detection data, and this precise detection ability provides a scientific basis for subsequent maintenance decisions, ensuring the pertinence and effectiveness of maintenance work; encrypting and storing the detection data, equipment operation data, and maintenance records to ensure the authenticity, non-tampering, and traceability of the data not only protects the core data assets of the enterprise, but also provides a reliable basis for subsequent analysis and decision-making; combining the corrosion analysis results can optimize the operation parameters and maintenance plan of the device, not only improving the operation efficiency of the equipment, but also effectively reducing the risks brought by corrosion and extending the service life of the equipment; through the comprehensive application of the above measures, precise analysis and effective prevention and control of the corrosion problem of the target device are achieved, and the overall equipment management level and safety are improved.

[0212] As mentioned above, the above is only the preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent replacements or changes, and should be covered by the protection scope of the present invention.

Claims

1. A detection method applied to the corrosion problem of a Texaco gasification device, characterized in that, Applied to the coal slurry preparation system, the damaged parts are the coal mill, the coal slurry tank and the attached pipelines. The material property is carbon steel. Due to medium impact or relative movement, mechanical spalling occurs on the surface layer of the material, forming pits, grooves, sharp grooves, holes and corrugated morphologies, and having a certain directionality, which is classified as erosion-corrosion; S1: Select the parts made of carbon steel materials in the coal mill, the coal slurry tank and their attached pipelines that are vulnerable to erosion-corrosion damage caused by medium impact or relative movement; clean the surface to be detected and remove attachments, dirt and oxide layers; S2: Use the detection system of the intelligent automatic diagnosis probe to comprehensively scan the area to be detected, identify and record the morphological characteristics formed by corrosion, calculate the parameters based on the corrosion mechanism model through algorithms, calculate what corrosion mechanism exists in the thinning position, and give the corresponding monitoring and detection plan according to the calculated corrosion mechanism; S3: According to the detection data, analyze the severity, distribution law and development trend of erosion-corrosion, and evaluate its impact on the safe operation of the system in combination with the mechanical properties and service conditions of the material; then sort out the detection results, including the location, size, shape and evaluation conclusion of erosion-corrosion, and form a detection report to provide a scientific basis for subsequent maintenance, repair or replacement.

2. The detection method for corrosion problems applied to Texaco gasification units according to claim 1, characterized in that It is also applied to the high-temperature gasification system. The damaged part is the gasification chamber of the gasifier. The material properties are carbon steel, low-alloy steel and stainless steel. First, due to medium impact or relative movement, mechanical spalling occurs on the surface layer of the material, forming pits, grooves, sharp grooves, holes and corrugated morphologies, and having a certain directionality, which is classified as erosion-corrosion; second, when contacting with sulfide under the condition of over 260°C and in the presence of hydrogen, uniform corrosion occurs on the metal, forming a multi-layer FeS scale structure with firm adhesion and a grayish luster, which is classified as high-temperature sulfide corrosion (hydrogen environment); third, when the operating temperature is greater than 595°C, the volume of the component increases, forming fine brittle cracks with a raised, scale and "fish-tail pattern" appearance, which is classified as carburization; fourth, in a high-temperature hydrogen environment, surface decarburization develops inward, forming bulges in the gaps at grain boundaries or inclusion interfaces, and then connecting along the grain boundaries to form microcracks. Ultrasonic thickness measurement shows abnormal "thickening", and decarburization or cracks are seen under the microscope, which is classified as high-temperature hydrogen corrosion; fifth, in a high-temperature oxygen environment, carbon steel and low-alloy steel form oxide films resulting in uniform thinning, and stainless steel forms dark oxide films, which is classified as high-temperature oxidation; sixth, metal surface deposition and melting occur, and the generated slag melts the surface oxide film. An obvious smooth interface can be seen between the vitreous hard scale layer of the ash layer and the base metal, which is classified as ash corrosion.

3. The detection method for corrosion problems applied to Texaco gasification units according to claim 2, wherein It is also applied to the lock hopper slag discharge system. The damaged parts are the lock hopper, the flushing water tank and its attached pipelines, and the slag pond. The material properties are carbon steel and low alloy steel. First, in an acidic water environment with ammonium hydrosulfide, local corrosion forms at the parts where the medium flow direction changes or in the turbulent flow area where the NH4HS concentration exceeds 2%, and scaling occurs in the low flow rate area, resulting in local corrosion under the scale, which is classified as acidic water corrosion. Second, ammonium chloride crystallizes and forms scale, which absorbs moisture and deliquesces. There are usually white, green or brown salt-like deposits, which is classified as ammonium chloride corrosion. Third, due to medium impact or relative movement, mechanical spalling occurs on the surface layer of the material, forming pits, grooves, sharp grooves, holes and corrugated morphologies with a certain directionality, which is classified as erosion corrosion.

4. The detection method for corrosion problems applied to Texaco gasification units according to claim 3, characterized in that, It is also applied to the sour gas system. The damaged parts are the top pipelines of high, low pressure and vacuum flash tanks, the top pipelines of flash separators, and the gas phase outlet pipelines at the top of deaerators. The material properties are carbon steel, low alloy steel and stainless steel. First, when there is a liquid water phase containing H2S and the H2S content is greater than 50 ppm, independent small bubbles form on the steel surface, and the small bubbles do not merge with each other, which is classified as wet H2S damage - hydrogen blistering. Step-like cracks parallel to the surface and extending along the rolling direction form inside the steel, which is classified as wet H2S damage - hydrogen induced cracking. Hydrogen induced cracking cracks at different depths on the base metal in the heat affected zone of the weld are connected along the thickness direction, which is classified as wet H2S damage - stress oriented hydrogen induced cracking. Cracks initiate on the surface of the heat affected zone of the weld and the high hardness zone and extend along the thickness direction, which is classified as wet H2S damage - stress corrosion cracking. Second, due to contact with hydrochloric acid, it is usually uniform corrosion, local corrosion or under-deposit corrosion during local concentration or dew point corrosion. Stainless steel shows pitting corrosion, which is classified as hydrochloric acid corrosion. Third, in an acidic water environment containing hydrogen sulfide and with a pH value of 4.5 - 7.0, it is generally uniform corrosion, and local corrosion or local corrosion under the deposited scale is likely to occur in the presence of oxygen. 300SS shows pitting corrosion, which is classified as acidic water corrosion. Fourth, in a humid carbon dioxide environment below 140 °C, it occurs in the turbulent flow and liquid impact areas and the root of the weld, resulting in wall thickness reduction, corrosion pits or holes. In the turbulent flow area, deeper pitting corrosion pits and grooves may form when carbon steel corrodes, which is classified as carbon dioxide corrosion.

5. The detection method for corrosion problems applied to Texaco gasification units according to claim 4, characterized in that, It is also applied to the black water and grey water systems. The damaged parts are in the quench chamber of the gasifier, Venturi scrubber, scrubbing tower, scrubbing water system, the liquid phase parts of the high, low pressure and vacuum flashing systems, and the liquid phase outlet pipeline at the bottom of the deaerator. When the material property is carbon steel or low alloy steel, first, in an acidic water environment with ammonium hydrosulfide, local corrosion occurs at the parts where the medium flow direction changes or in the turbulent flow area where the NH4HS concentration exceeds 2%, scaling occurs in the low flow rate area, and under-deposit local corrosion occurs, which is classified as acidic water corrosion; second, due to medium impact or relative movement, mechanical peeling of the material surface layer occurs, forming pits, grooves, sharp grooves, holes and corrugated morphologies with a certain directionality, which is classified as erosion corrosion; third, due to the crystallization and scaling of ammonium chloride and its hygroscopic deliquescence, white, green or brown salt-like deposits are mostly present, which is classified as ammonium chloride corrosion; fourth, when the gas phase CO2 concentration > 2% and the temperature > 93°C, the liquid phase pH value > 9.0 and the carbonate concentration > 100 ppm, or the pH value is between 8.0 - 9.0 and the carbonate concentration > 400 ppm, cracks extending along the direction parallel to the weld appear in the base metal near the welded joint. The cracks mainly propagate along the grain boundaries, the interior is filled with oxide, the cracks are fine and spider-web shaped, which is classified as carbonate stress corrosion cracking. When the material property is 300 series stainless steel or 400 series stainless steel, first, under the combined action of tensile stress, temperature > 38°C, and chloride aqueous solution environment with pH value > 2, dendritic and branched transgranular cracks are formed on the surface. The fracture surface is usually brittle and there is no obvious thinning, which is classified as chloride stress corrosion cracking; second, due to medium impact or relative movement, mechanical peeling of the material surface layer occurs, forming pits, grooves, sharp grooves, holes and corrugated morphologies with a certain directionality, which is classified as erosion corrosion; third, ammonium chloride crystallizes and scales, and hygroscopic deliquescence occurs. White, green or brown salt-like deposits that are easily removed are present. Under-deposit corrosion usually occurs under the scale layer, which is classified as ammonium chloride corrosion.

6. The detection method applied to the corrosion problem of the Texaco gasification device according to claim 5, wherein Among them, S2 includes: System deployment: Deploy intelligent automatic diagnosis probes at key pipeline parts. The probes have data acquisition functions and are used to directly contact the inner wall of the pipeline for scanning the corrosion morphology. Start scanning: Start the probes to comprehensively scan the inner wall of the pipeline along a preset trajectory or a trajectory adjusted according to the real-time pipeline shape, and collect corrosion morphology data in real time. Data preprocessing: Input the collected corrosion morphology data into the data processing and analysis unit for data preprocessing operations. Feature extraction and recognition: Use image processing technology and machine learning algorithms to extract the features of the corrosion morphology from the preprocessed data, and compare the extracted features with the preset corrosion morphology database to identify the type and degree of corrosion. Recording and display: Record the identified corrosion morphology features in the form of graphics or data and display them through the display system. At the same time, export the data to the report generation module to generate a detailed detection report. Intelligent optimization: The detection system has a self-learning function, can optimize the recognition algorithm according to the continuously accumulated detection data, and automatically trigger the alarm mechanism when serious corrosion is detected.

7. The detection method for corrosion problems applied to Texaco gasification units according to claim 6, characterized in that, The probe dynamically adjusts the scanning trajectory according to the actual shape of the inner wall of the pipeline to comprehensively cover the detection area, and real-time monitors the contact force between the probe and the inner wall of the pipeline.

8. The detection method for corrosion problems applied to Texaco gasification units according to claim 7, characterized in that, The dynamic scanning trajectory adjustment includes the following steps: Before the detection starts, the system first plans a preliminary scanning path according to the shape data of the pipeline; when the probe starts scanning, the laser rangefinder and ultrasonic sensor equipped on it will sense the pipeline shape in real time, and the system will compare the real-time sensed shape data with the initial scanning path to calculate the deviation. According to the calculated deviation, the system dynamically adjusts the moving trajectory of the probe. The system adopts a particle swarm trajectory optimization algorithm to find the optimal scanning trajectory during the real-time sensing and adjustment process. The algorithm takes into account factors such as scanning speed, contact force between the probe and the inner wall of the pipeline, and detection accuracy. Among them, the algorithm steps include the following steps: Initializing the particle swarm: Determine the size N of the particle swarm, that is, the number of candidates for the trajectory. Randomly assign an initial position and velocity to each particle. Here, the position can be represented as a parameter or coordinate sequence of the scanning trajectory, and the velocity represents the change rate of the trajectory parameters or coordinates. Initialize the personal best position (pbest) of each particle to its current position, and the global best position (gbest) to the optimal position among all particles. Fitness function design and iterative update: According to factors such as scanning speed, contact force between the probe and the inner wall of the pipeline, and detection accuracy, design a fitness to evaluate the quality of each particle, and obtain a comprehensive evaluation index through weighted summation or multiplication. In each iteration, calculate the fitness value of each particle according to the fitness function. Update the personal best position (pbest) of each particle. If the current fitness value is better than the previous personal best fitness value, update the global best position (gbest). If the fitness value of a current particle is better than the previous global best fitness value. Update the velocity and position of each particle according to the following formula: Velocity update formula: Vi(t + 1) = w * Vi(t) + c1 * r1 * (pbesti - Xi(t)) + c2 * r2 * (gbest - Xi(t)) Among them, w is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers, Xi(t) is the position of particle i at time t, and Vi(t) is the velocity of particle i at time t. Position update formula: Xi(t + 1) = Xi(t) + Vi(t + 1) Termination condition: Reach the preset maximum number of iterations, the fitness value of the global best position reaches the preset threshold, and the change amount of the fitness value is less than the preset threshold, indicating that the algorithm has converged. Output the optimal solution: When the termination condition is met, output the global best position as the optimal scanning trajectory. A contact force sensor is integrated on the probe to real-time monitor the contact force between the probe and the inner wall of the pipeline. The contact force sensor can sense tiny force changes and convert these changes into electrical signals for transmission and processing. The system judges the real-time monitored contact force according to the preset contact force threshold. If the contact force exceeds the preset threshold, the system will automatically adjust the moving speed or direction of the probe to reduce the contact force and prevent additional damage to the pipeline.

9. An analysis system applied to the corrosion problem of the Texaco gasification device. The system applied to the corrosion problem of the Texaco gasification device adopts the detection method as described in claim 8, and is characterized in that, Select key monitoring points for monitoring the corrosion condition of the target device, and install corresponding intelligent sensors at the key monitoring points; Collect the operation data of each key monitoring point of the target device in real time through the intelligent sensor, and transmit the operation data to the central processing unit; Use artificial intelligence algorithms to analyze and process the collected operation data, compare with the preset normal operation data range, identify areas that may have corrosion risks and abnormal data change trends, and obtain suspected corrosion areas; For the suspected corrosion areas, use intelligent detection technology for detection to obtain detection data; Encrypt and store the detection data, device operation data, and maintenance records to ensure the authenticity, immutability, and traceability of the data, providing a reliable basis for subsequent analysis and decision-making; Combined with intelligent manufacturing technology, optimize the operation parameters and maintenance plan of the device according to the corrosion analysis results, and achieve precise analysis and effective prevention and control of the corrosion problem of the target device; Select key monitoring points for monitoring the corrosion condition of the target device, and install corresponding intelligent sensors at the key monitoring points, including: Obtain the three-dimensional data of the device of the target device, and establish a three-dimensional model of the device of the target device according to the three-dimensional data of the device; Obtain the corrosion data of the same type of products of the target device, and establish a corrosion state prediction model of the target device according to the corrosion data; According to the three-dimensional model of the device and the corrosion state prediction model, select key monitoring points for monitoring the corrosion condition of the target device, and install corresponding intelligent sensors at the key monitoring points in combination with sensor attribute data.

10. The analysis system for corrosion problems applied to Texaco gasification units according to claim 9, characterized in that, Obtain the corrosion data of the same type of products of the target device, and establish a corrosion state prediction model of the target device according to the corrosion data, including: Determine the scope of the same type of products, including products identical to the target device and products similar to the target device in terms of materials, structures, working environments, and usage conditions; Collect the corrosion data of the same type of products, and the corrosion data includes corrosion rate, corrosion type, environmental factors, and maintenance records; Sort out and analyze the collected corrosion data, identify the key factors affecting the corrosion state, and perform data cleaning to remove outliers and noise; Based on the sorted corrosion data, use statistical analysis methods or machine learning algorithms to establish a corrosion state prediction model. Specifically: analyze the correlation between process parameters and corrosion rate, establish a corrosion reaction kinetics model, combine material science theory, construct a corrosion damage evolution equation, and verify the physical rationality of the model to obtain a corrosion mechanism model; use long short-term memory networks to process time series data, use convolutional neural networks to extract spatial features, and combine attention mechanisms to improve prediction accuracy for training deep learning models; Integrate the mechanism model and the deep learning model, use ensemble learning methods to improve prediction stability, and use Bayesian optimization algorithms to adjust model hyperparameters; Set the model update period to achieve online learning; Verify the corrosion state prediction model, evaluate the accuracy of the model using historical data, and adjust and optimize the model according to the evaluation results; The steps of selecting key monitoring points for monitoring the corrosion condition of the target device and installing corresponding intelligent sensors at the key monitoring points in combination with sensor attribute data include: Analyze the three-dimensional model of the device to identify the first areas where corrosion risks may exist; Combine the corrosion state prediction model to evaluate the corrosion risk levels of each first area and determine the key monitoring points that need to be monitored with emphasis; Obtain sensor attribute data related to the key monitoring points; Select an intelligent sensor corresponding to the key monitoring point according to the corrosion type and monitoring requirements, and formulate the installation position and method of the sensor; Install the selected intelligent sensor at the key monitoring point to ensure that the installation of the sensor complies with technical specifications and can effectively monitor the corrosion condition.

Images