Method and device for monitoring corrosion of water vapor pipeline of supercritical unit

Abstract

The invention provides a supercritical unit water vapor pipeline corrosion monitoring method and device, and relates to the technical field of pipeline corrosion monitoring. A corrosion model is trained through historical working conditions and corrosion amount of a water vapor pipeline, and real-time working condition data of the water vapor pipeline to be monitored are input into the corrosion model to obtain predicted corrosion amount; the method comprises the following steps: acquiring water quality data of a water inlet and a water outlet of a pipeline, forming a water quality difference value, generating a water quality score and a current corrosion amount, acquiring initial vibration data and current vibration data of each single-section pipeline, forming a vibration variation of each single-section pipeline and a vibration variation standard deviation of all pipelines, and setting a vibration variation standard deviation threshold value; the corrosion threshold value I and the corrosion threshold value II are set, and the corrosion condition is analyzed in combination with the vibration variable quantity standard deviation of all the pipelines, so that the corrosion condition of all the single pipelines is analyzed, and a basis is provided for specific pipeline replacement.

Description

Technical Field

[0001] The invention relates to the technical field of pipeline corrosion monitoring, and in particular to a method and a monitoring device for water vapor pipeline corrosion of a supercritical unit. Background Art

[0002] In the industrial field, the safe operation of water vapor pipelines is of vital importance. However, during long-term use, water vapor pipelines are faced with multiple impacts such as corrosion and vibration. Traditional monitoring methods often rely on manual inspections and regular testing, and cannot detect potential corrosion problems and equipment abnormalities in a timely manner, which not only increases safety hazards but also causes economic losses. Existing technologies are insufficient in data collection, real-time monitoring and corrosion prediction capabilities. Especially when dealing with complex variables and dynamic changes, it is often difficult to form an effective basis for evaluation and decision-making, resulting in inefficient pipeline maintenance management and increased risk of production suspension.

[0003] In the prior art, publication number CN117969388A discloses a supercritical unit water vapor pipeline corrosion monitoring system, including an inductance probe module, a data acquisition conversion module, and a data analysis and processing module; the inductance probe module is connected to the data acquisition conversion module, and the data acquisition conversion module is connected to the data analysis and processing module; the two inductance probe modules are respectively installed at the measuring point behind the secondary gate of the boiler feed water sampling pipe and the measuring point behind the secondary gate of the superheated steam sampling pipe, for collecting pipeline corrosion information; the data acquisition conversion module is used to convert the corrosion information collected by the inductance probe into a digital signal and transmit it to the data analysis and processing module; the data analysis and processing module is used to obtain the corrosion condition of the supercritical unit water vapor pipeline by calculation.

[0004] According to public technical documents, although the monitoring of water vapor pipelines has been achieved, the method of using a bypass pipeline inductance probe module has inaccurate results. Because of the influence of the fluid model, the corrosion effects occurring at different positions of the pipeline are different. The inductance probe module installed in the bypass pipeline cannot directly reflect the most realistic corrosion effect of the pipeline, which is prone to errors.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The object of the present invention is to provide a method and a device for monitoring corrosion of a water vapor pipeline of a supercritical unit, so as to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions: A method for monitoring corrosion of a water vapor pipeline of a supercritical unit, comprising the following specific steps: Obtain the historical working condition data and corrosion amount of the scrapped water vapor pipeline, wherein the working condition data includes the number of water vapor pipeline transportation, maximum pressure, maximum temperature, maximum flow rate and working time, and summarize the historical working conditions into a historical working data set, and the corresponding corrosion amount of the water vapor pipeline is used as a label of the historical working data set; Construct a model based on convolutional neural network, use the working data set as input feature, and the corresponding corrosion amount of water vapor pipeline as target variable to train the model. After training, output the corrosion model, collect the working condition data of the water vapor pipeline to be monitored in real time to construct a real-time working data set, input it into the corrosion model, and obtain the predicted corrosion amount of the water vapor pipeline. Obtain water quality data of the current water inlet and outlet of the water vapor pipeline to be monitored, the water quality data including conductivity, ion concentration and dissolved solids, obtain water quality difference, generate a single corrosion score, and obtain the current corrosion amount; Obtaining initial vibration data and current vibration data of each single-section pipeline, wherein the initial vibration data and current vibration data are the vibration amplitude of the single-section pipeline, obtaining the vibration variation of each single-section pipeline, obtaining the standard deviation of the vibration variation of all single-section pipelines, and setting a threshold value of the standard deviation of the vibration variation; Set corrosion threshold I and corrosion threshold II to determine the relationship between the current corrosion amount and corrosion threshold I and corrosion threshold II respectively. Combine the standard deviation of the vibration change of a single section of pipeline with the standard deviation threshold of the vibration change to determine the corrosion condition of the water vapor pipeline to be monitored.

[0008] Furthermore, the historical operating data of the scrapped water vapor pipeline is obtained through the work log, and the historical operating data includes the number of transportation times of the water vapor pipeline, and the maximum pressure, maximum temperature, maximum flow rate and working time during each transportation. The corrosion amount of the water vapor pipeline is obtained through the maintenance log, and the historical operating data is summarized into a historical working data set, and the corrosion amount is used as a label.

[0009] Furthermore, the working data set is input into the convolutional neural network model. The convolutional neural network consists of an input layer, a convolutional layer, a pooling layer, a fully connected layer and an output layer. The activation function of the convolutional layer is set according to the following formula:

[0010]

[0011] in, is the rectified linear unit activation function, is the input of the neuron, For the The weight of the input, For the The activation value of the input, is the bias term; Set the number of input layer nodes to 7, the number of fully connected layer neurons to 32, the initial neural network learning rate to 0.001, and the output layer to 1 neuron; The trained model is calibrated as a corrosion model, and the output result is the predicted corrosion amount; The corrosion model is deployed, and the operating data of the water vapor pipeline to be monitored is collected in real time to build a real-time working data set. The real-time working data set is mapped to the historical working data set and input into the corrosion model to obtain the current predicted corrosion amount of the water vapor pipeline.

[0012] Furthermore, the water quality data of the current water inlet and outlet of the water vapor pipeline to be monitored is obtained, and the water quality data includes conductivity, ion concentration and solid dissolution, wherein the ion concentration is the divalent iron ion concentration, and the water quality difference between the water inlet and the water outlet is obtained, and the formula is as follows:

[0013]

[0014]

[0015] in, is the conductivity difference, is the inlet conductivity, is the outlet conductivity, is the ion concentration difference, is the ion concentration at the water inlet, is the outlet ion concentration, is the difference in solid dissolution, is the amount of dissolved solids at the water inlet, is the amount of dissolved solids at the outlet; A single corrosion score is generated based on the difference in water quality between the inlet and outlet, based on the following formula:

[0016] in, To score a single corrosion, is the conductivity difference, is the ion concentration difference, is the difference in solid dissolution, , , are the scoring coefficients, .

[0017] Furthermore, the current corrosion amount is obtained by predicting the corrosion amount and the single corrosion score, and the formula is as follows:

[0018] in, is the current corrosion amount, To score a single corrosion, To predict the amount of corrosion.

[0019] Furthermore, a vibration collector is arranged in each single section of the water vapor pipeline to be monitored, and the initial vibration data and the current vibration data of each single section of the pipeline are obtained respectively. The initial vibration data and the current vibration data record the vibration amplitude of the single section of the pipeline, and the vibration change of each single section of the pipeline is obtained respectively, according to the following formula:

[0020] in, For the The vibration change of the pipe section, For the The vibration data of the pipeline. For the Initial vibration data of the pipeline, Retrieve variable for pipe number, , , is the number of pipelines; The standard deviation of the vibration variation of all single-section pipelines is obtained according to the following formula:

[0021] in, is the standard deviation of the vibration variation of a single pipe section, For the The vibration change of the pipe section, is the average value of the vibration variation of all pipelines, Retrieve variable for pipe number, , , is the number of pipelines; Set the standard deviation threshold of vibration variation.

[0022] Furthermore, the corrosion threshold I and the corrosion threshold II are set respectively, where the corrosion threshold I is smaller than the corrosion threshold II, and the current corrosion amount and the standard deviation of the vibration change of a single section of the pipeline are analyzed. The analysis logic is as follows: When the current corrosion amount is less than the corrosion threshold I, the monitored water vapor pipeline has slight corrosion, but it does not affect the operation, and the single section of the pipeline does not need to be replaced; When the current corrosion amount is greater than the corrosion threshold I, the standard deviation of the vibration change of a single section of the pipeline is determined. If the standard deviation of the vibration change of a single section of the pipeline is less than the standard deviation threshold, all pipelines are corroded, and the overall corrosion has reached a certain level. However, due to the combined effect of all pipelines, each single section of the pipeline has not reached the corrosion scrap effect, and it will not be replaced. If the standard deviation of the vibration change of a single section of the pipeline is greater than the standard deviation threshold, it means that the individual pipeline is severely corroded, and the individual pipeline is replaced. The replacement logic is as follows: Determine the pipeline with the largest vibration change, replace the vibration change of the pipeline with the largest vibration change with 0, and obtain the standard deviation of the vibration change of the single-section pipeline again. If the standard deviation of the vibration change of the single-section pipeline is still greater than the standard deviation threshold, replace the vibration change of the pipeline with the largest and second largest vibration change with 0 at the same time, and obtain the standard deviation of the vibration change of the single-section pipeline again, and so on, until the standard deviation of the vibration change of the single-section pipeline is less than the standard deviation threshold, and the single-section pipeline involved in the vibration change replacement is replaced; When the current corrosion amount is greater than the corrosion threshold I and the standard deviation of the vibration change of a single pipeline section is always less than the standard deviation threshold, it is determined whether the current corrosion amount is greater than the corrosion threshold II. If the current corrosion amount is less than the corrosion threshold II, the pipeline will not be replaced. If the current corrosion amount is greater than the corrosion threshold II, all pipelines will be replaced.

[0023] A supercritical unit steam pipeline corrosion monitoring device, the device is used to perform the above-mentioned supercritical unit steam pipeline corrosion monitoring method, comprising: The working condition data acquisition module is used to obtain the historical working condition data of the scrapped water vapor pipeline, wherein the historical working condition data includes the number of water vapor pipeline transportation, the maximum pressure, the maximum temperature, the maximum flow rate and the working time, and summarizes them into a historical working data set, obtains the corrosion amount of the pipeline in the historical working condition data set and uses it as a label of the working data set, obtains the real-time working data of the water vapor pipeline to be monitored to form a real-time working data set and maps it to the historical working data set; A model building module is used to input the historical working data set into the convolutional neural network model for training, obtain the corrosion model, and output the predicted corrosion amount; A water quality data acquisition module is used to obtain the water quality data of the current water inlet and outlet of the water vapor pipeline to be monitored, wherein the water quality data includes conductivity, ion concentration and dissolved solids, and obtain the water quality difference between the water inlet and the water outlet and obtain the current corrosion amount; A vibration data acquisition module is used to acquire the initial vibration data and current vibration data of each single-section pipeline, wherein the initial vibration data and current vibration data are the vibration amplitude of the single-section pipeline, generate the vibration variation of each single-section pipeline and the standard deviation of the vibration variation of all single-section pipelines, and set the standard deviation threshold; The single-section pipeline replacement selection module is used to set corrosion threshold I and corrosion threshold II, respectively determine the relationship between the current corrosion amount and corrosion threshold I and corrosion threshold II, and analyze the standard deviation of the vibration change of the single-section pipeline and the standard deviation threshold of the vibration change to determine the corrosion condition of the water vapor pipeline to be monitored.

[0024] Compared with the prior art, the present invention has the following beneficial effects: The present invention obtains a corrosion model based on the historical operating data and corrosion amount of the water vapor pipeline, forms a predicted corrosion amount based on the actual operating data of the water vapor pipeline, obtains the overall corrosion condition of the water vapor pipeline based on the water quality data of the water inlet and the water outlet, obtains the vibration change of each single section of the pipeline and the standard deviation of the vibration change of all single sections of the pipeline through the vibration data of each single section of the pipeline, and obtains the corrosion condition of the monitored water vapor pipeline through analysis. The present invention obtains the real-time corrosion condition of each single section of the pipeline through the overall corrosion condition and the vibration condition of each single section of the pipeline, thereby improving the accuracy and effectiveness of the corrosion of the water vapor pipeline. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the overall method flow of the present invention; Figure 2 It is a schematic diagram of the module structure of the detection device of the present invention. DETAILED DESCRIPTION

[0026] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0027] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0028] Example:

[0029] See also Figure 1 , the present invention provides a technical solution: A method for monitoring corrosion of a water vapor pipeline of a supercritical unit, comprising the following specific steps: Step 1: Obtain the historical working condition data and corrosion amount of the scrapped water vapor pipeline, the working condition data includes the number of water vapor pipeline transportation, maximum pressure, maximum temperature, maximum flow rate and working time, and summarize the historical working conditions into a historical working data set, and the corresponding water vapor pipeline corrosion amount is used as the label of the historical working data set; The step 1 includes the following contents: Step 101: Obtain historical operating data of the scrapped water vapor pipeline through the work log, wherein the historical operating data includes the number of transportation times of the water vapor pipeline, and the maximum pressure, maximum temperature, maximum flow rate and working time during each transportation. Obtain the corrosion amount of the water vapor pipeline through the maintenance log, summarize the historical operating data into a historical working data set, and use the corrosion amount as a label.

[0030] By collecting historical operating data, a comprehensive data foundation is established. This process not only ensures the integrity and accuracy of the data, but also provides a reliable basis for subsequent analysis and model training. As the number of steam pipeline transportation increases, the corrosion of the steam pipeline will increase. Similarly, the maximum pressure, maximum temperature, maximum flow rate and working hours will increase the corrosion of the steam pipeline. By obtaining the corrosion of the steam pipeline through the maintenance log, the most realistic corrosion condition of the steam pipeline can be obtained, the data quality can be improved, and ultimately a solid foundation for subsequent decision-making is laid, ensuring that the subsequent model can operate effectively in a real environment.

[0031] Step 2: Build a model based on convolutional neural network, use the working data set as input features, and the corresponding corrosion amount of the steam pipeline as the target variable to train the model. After training, output the corrosion model, collect the working condition data of the steam pipeline to be monitored in real time to build a real-time working data set, input it into the corrosion model, and obtain the predicted corrosion amount of the steam pipeline; The step 2 includes the following contents: Step 201: Input the working data set into the convolutional neural network model. The convolutional neural network consists of an input layer, a convolutional layer, a pooling layer, a fully connected layer and an output layer. The activation function of the convolutional layer is set according to the following formula:

[0032]

[0033] in, is the rectified linear unit activation function, is the input of the neuron, For the The weight of the input, For the The activation value of the input, is the bias term; Set the number of input layer nodes to 7, the number of fully connected layer neurons to 32, the initial neural network learning rate to 0.001, and the output layer to 1 neuron; The trained model is calibrated as a corrosion model, and the output result is the predicted corrosion amount.

[0034] Using convolutional neural networks (CNN) for data processing enables the model to recognize complex patterns and relationships. CNN performs well when processing multidimensional data and can automatically extract data features, reducing the need for manual feature selection. Through this step, the accuracy of predicting corrosion volume is significantly improved, and the hidden correlations in the data can be mined more deeply. This efficient model training enables the system to maintain a high level of predictive ability in the face of different operating conditions and environmental changes, thereby enhancing the ability to assess pipeline corrosion risks.

[0035] Step 202: deploy the corrosion model, collect the working condition data of the water vapor pipeline to be monitored in real time to build a real-time working data set, map the real-time working data set to the historical working data set and input it into the corrosion model to obtain the current predicted corrosion amount of the water vapor pipeline.

[0036] Step 3: Obtain the water quality data of the current water inlet and outlet of the water vapor pipeline to be monitored, the water quality data including conductivity, ion concentration and dissolved solids, obtain the water quality difference, generate a single corrosion score, and obtain the current corrosion amount; The step 3 includes the following contents: Step 301: Obtain water quality data of the current water inlet and outlet of the water vapor pipeline to be monitored, wherein the water quality data includes conductivity, ion concentration and solid dissolution, wherein the ion concentration is the divalent iron ion concentration, and obtain the water quality difference between the water inlet and the water outlet, based on the following formula:

[0037]

[0038]

[0039] in, is the conductivity difference, is the inlet conductivity, is the outlet conductivity, is the ion concentration difference, is the ion concentration at the water inlet, is the outlet ion concentration, is the difference in solid dissolution, is the amount of dissolved solids at the water inlet, is the amount of dissolved solids at the outlet; Among them, as the anti-corrosion layer inside the water vapor pipe is destroyed, some soluble substances in the anti-corrosion layer will definitely dissolve in the water vapor. The soluble substances will cause changes in the conductivity and ion concentration of the water vapor. The most important of these is the concentration of divalent iron ions. By comparing the water quality difference between the inlet and outlet, the current corrosion condition in the pipeline can be judged. If no corrosion occurs in the water vapor pipe, the water quality at the inlet and outlet will not change. When corrosion has occurred, due to the high temperature and high pressure environment, the corrosion rate will increase exponentially, so each water quality difference is different and also shows an exponential increase trend. The current corrosion status can be reflected by each specific water quality difference.

[0040] A single corrosion score is generated based on the difference in water quality between the inlet and outlet, based on the following formula:

[0041] in, To score a single corrosion, is the conductivity difference, is the ion concentration difference, is the difference in solid dissolution, , , are the scoring coefficients, .

[0042] Step 302: Obtain the current corrosion amount by predicting the corrosion amount and the single corrosion score, according to the following formula:

[0043] in, is the current corrosion amount, To score a single corrosion, To predict the amount of corrosion.

[0044] Among the three factors of a single corrosion score, conductivity can most directly reflect changes in water quality, so the coefficient of the conductivity difference is greater than the other two, and the ion concentration difference is easier to detect and reflect through data than the solid dissolution difference, so the coefficient of the ion concentration difference is greater than the solid dissolution difference. When the conductivity difference, ion concentration difference and solid dissolution difference are all 0, it means that there is almost no corrosion in the pipeline, and the single corrosion score has no correction effect on the predicted corrosion amount. However, as the conductivity difference, ion concentration difference and solid dissolution difference increase, it means that the water vapor pipeline begins to corrode, and the single corrosion score gradually increases, the correction effect on the predicted corrosion amount becomes greater and greater, and the faster corrosion speed of the water vapor pipeline is offset by increasing the correction coefficient, thereby obtaining a more accurate current corrosion amount.

[0045] Comprehensive analysis of water quality data, taking into account multiple influencing factors such as conductivity, ion concentration and dissolved solids, can comprehensively evaluate the impact of water quality on pipeline corrosion. This step not only provides a scientific basis for the generated single corrosion score, but also provides a deeper understanding of the relationship between water quality changes and corrosion risks. Through multi-dimensional analysis, key indicators of water quality changes can be identified.

[0046] Step 4: Obtain the initial vibration data and current vibration data of each single-section pipeline, wherein the initial vibration data and current vibration data are the vibration amplitude of the single-section pipeline, obtain the vibration variation of each single-section pipeline, obtain the standard deviation of the vibration variation of all single-section pipelines, and set the standard deviation threshold of the vibration variation; The step 4 includes the following contents: A vibration collector is arranged in each single section of the water vapor pipeline to be monitored, and the initial vibration data and the current vibration data of each single section of the pipeline are obtained respectively. The initial vibration data and the current vibration data record the vibration amplitude of the single section of the pipeline, and the vibration change of each single section of the pipeline is obtained respectively, according to the following formula:

[0047] in, For the The vibration change of the pipe section, For the The vibration data of the pipeline. For the Initial vibration data of the pipeline, Retrieve variable for pipe number, , , is the number of pipelines; The standard deviation of the vibration variation of all single-section pipelines is obtained according to the following formula:

[0048] in, is the standard deviation of the vibration variation of a single pipe section, For the The vibration change of the pipe section, is the average value of the vibration variation of all pipelines, Retrieve variable for pipe number, , , is the number of pipelines; Set the standard deviation threshold of vibration variation.

[0049] Among them, the water vapor pipeline will vibrate when transporting the medium, and this vibration is mainly caused by its own state. Therefore, the vibration state of each single section of the pipeline will change after corrosion, and the vibration change of each single section of the pipeline reflects this vibration change from the perspective of amplitude. At the same time, because the water vapor pipeline is connected by multiple single pipelines, but the fluid model analysis process of the internal medium is very complicated, the corrosion state of each single pipeline cannot be directly analyzed. The standard deviation of the vibration change of the single section of the pipeline is used to judge. When the standard deviation of the vibration change of the single section of the pipeline is less than the standard deviation threshold of the vibration change, it means that all pipelines have undergone corrosion to a similar degree. When the standard deviation of the vibration change of the single section of the pipeline is greater than the standard deviation threshold of the vibration change, it means that some single sections of the pipeline have undergone severe corrosion, while other pipelines have not undergone too severe corrosion. The specific corrosion state of each single section of the pipeline is analyzed again by the current corrosion amount.

[0050] Vibration monitoring can be used to quantify the structural health of the pipeline and calculate the standard deviation of the vibration variation to detect potential problems in a timely manner. Vibration monitoring technology can capture abnormal conditions that may occur during the operation of the pipeline and provide an objective and quantifiable basis for detection. The introduction of this strategy has significantly improved the reliability and efficiency of fault detection. Compared with traditional visual inspections, it can detect potential structural damage earlier, providing valuable time for subsequent repairs and replacements, and reducing economic losses caused by sudden failures.

[0051] Step 5: Set corrosion threshold I and corrosion threshold II, respectively determine the relationship between the current corrosion amount and corrosion threshold I and corrosion threshold II, and analyze the standard deviation of the vibration change of a single section of the pipeline and the standard deviation threshold of the vibration change to determine the corrosion condition of the water vapor pipeline to be monitored.

[0052] The step 5 includes the following contents: Corrosion threshold I and corrosion threshold II are set respectively, where corrosion threshold I is smaller than corrosion threshold II. The current corrosion amount and the standard deviation of the vibration change of a single section of pipeline are analyzed. The analysis logic is as follows: When the current corrosion amount is less than the corrosion threshold I, the monitored water vapor pipeline has slight corrosion, but it does not affect the operation, and the single section of the pipeline does not need to be replaced; When the current corrosion amount is greater than the corrosion threshold I, the standard deviation of the vibration change of a single section of the pipeline is determined. If the standard deviation of the vibration change of a single section of the pipeline is less than the standard deviation threshold, all pipelines are corroded, and the overall corrosion has reached a certain level. However, due to the combined effect of all pipelines, each single section of the pipeline has not reached the corrosion scrap effect, and it will not be replaced. If the standard deviation of the vibration change of a single section of the pipeline is greater than the standard deviation threshold, it means that the individual pipeline is severely corroded, and the individual pipeline is replaced. The replacement logic is as follows: Determine the pipeline with the largest vibration change, replace the vibration change of the pipeline with the largest vibration change with 0, and obtain the standard deviation of the vibration change of the single-section pipeline again. If the standard deviation of the vibration change of the single-section pipeline is still greater than the standard deviation threshold, replace the vibration change of the pipeline with the largest and second largest vibration change with 0 at the same time, and obtain the standard deviation of the vibration change of the single-section pipeline again, and so on, until the standard deviation of the vibration change of the single-section pipeline is less than the standard deviation threshold, and the single-section pipeline involved in the vibration change replacement is replaced; When the current corrosion amount is greater than the corrosion threshold I and the standard deviation of the vibration change of a single pipeline section is always less than the standard deviation threshold, it is determined whether the current corrosion amount is greater than the corrosion threshold II. If the current corrosion amount is less than the corrosion threshold II, the pipeline will not be replaced. If the current corrosion amount is greater than the corrosion threshold II, all pipelines will be replaced.

[0053] Set the corrosion threshold and the standard deviation threshold of the vibration change to achieve maintenance decisions based on actual working conditions. Through this analysis, it is possible to formulate corresponding maintenance strategies based on the actual operating conditions to avoid unnecessary maintenance and waste of resources. When necessary, ensure that appropriate replacement measures are taken in time to maximize the operating efficiency and safety of the pipeline. This data-driven decision-making method not only improves economic benefits, but also provides a strong guarantee for the long-term safe operation of the pipeline, and ultimately forms an efficient and intelligent pipeline management system.

[0054] See also Figure 2 The present invention further provides a supercritical unit steam pipeline corrosion monitoring device, the device is used to perform the above-mentioned supercritical unit steam pipeline corrosion monitoring method, comprising: The working condition data acquisition module is used to obtain the historical working condition data of the scrapped water vapor pipeline, wherein the historical working condition data includes the number of water vapor pipeline transportation, the maximum pressure, the maximum temperature, the maximum flow rate and the working time, and summarizes them into a historical working data set, obtains the corrosion amount of the pipeline in the historical working condition data set and uses it as a label of the working data set, obtains the real-time working data of the water vapor pipeline to be monitored to form a real-time working data set and maps it to the historical working data set; A model building module is used to input the historical working data set into the convolutional neural network model for training, obtain the corrosion model, and output the predicted corrosion amount; A water quality data acquisition module is used to obtain the water quality data of the current water inlet and outlet of the water vapor pipeline to be monitored, wherein the water quality data includes conductivity, ion concentration and dissolved solids, and obtain the water quality difference between the water inlet and the water outlet and obtain the current corrosion amount; A vibration data acquisition module is used to acquire the initial vibration data and current vibration data of each single-section pipeline, wherein the initial vibration data and current vibration data are the vibration amplitude of the single-section pipeline, generate the vibration variation of each single-section pipeline and the standard deviation of the vibration variation of all single-section pipelines, and set the standard deviation threshold; The single-section pipeline replacement selection module is used to set corrosion threshold I and corrosion threshold II, respectively determine the relationship between the current corrosion amount and corrosion threshold I and corrosion threshold II, and analyze the standard deviation of the vibration change of the single-section pipeline and the standard deviation threshold of the vibration change to determine the corrosion condition of the water vapor pipeline to be monitored.

[0055] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0056] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0057] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0058] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for monitoring corrosion of water vapor pipelines in a supercritical unit, characterized in that: The specific steps include: Obtain the historical working condition data and corrosion amount of the scrapped water vapor pipeline, wherein the working condition data includes the number of water vapor pipeline transportation, maximum pressure, maximum temperature, maximum flow rate and working time, and summarize the historical working conditions into a historical working data set, and the corresponding corrosion amount of the water vapor pipeline is used as a label of the historical working data set; Construct a model based on convolutional neural network, use the working data set as input feature, and the corresponding corrosion amount of water vapor pipeline as target variable to train the model. After training, output the corrosion model, collect the working condition data of the water vapor pipeline to be monitored in real time to construct a real-time working data set, input it into the corrosion model, and obtain the predicted corrosion amount of the water vapor pipeline. Obtain water quality data of the current water inlet and outlet of the water vapor pipeline to be monitored, the water quality data including conductivity, ion concentration and dissolved solids, obtain water quality difference, generate a single corrosion score, and obtain the current corrosion amount; Obtaining initial vibration data and current vibration data of each single-section pipeline, wherein the initial vibration data and current vibration data are the vibration amplitude of the single-section pipeline, obtaining the vibration variation of each single-section pipeline, obtaining the standard deviation of the vibration variation of all single-section pipelines, and setting a threshold value of the standard deviation of the vibration variation; Set corrosion threshold I and corrosion threshold II to determine the relationship between the current corrosion amount and corrosion threshold I and corrosion threshold II respectively. Combine the standard deviation of the vibration change of a single section of pipeline with the standard deviation threshold of the vibration change to determine the corrosion condition of the water vapor pipeline to be monitored.

2. A method for monitoring corrosion of water vapor pipelines of a supercritical unit according to claim 1, characterized in that: The historical working condition data of the scrapped water vapor pipeline is obtained through the work log, and the historical working condition data includes the number of transportation times of the water vapor pipeline, and the maximum pressure, maximum temperature, maximum flow rate and working time during each transportation. The corrosion amount of the water vapor pipeline is obtained through the maintenance log, and the historical working condition data is summarized into a historical working data set, and the corrosion amount is used as a label.

3. A method for monitoring corrosion of water vapor pipelines of a supercritical unit according to claim 2, characterized in that: The working data set is input into the convolutional neural network model. The convolutional neural network consists of an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer. The activation function of the convolutional layer is set according to the following formula: ， ， in, is the rectified linear unit activation function, is the input of the neuron, For the The weight of the input, For the The activation value of the input, is the bias term; Set the number of input layer nodes to 7, the number of fully connected layer neurons to 32, the initial neural network learning rate to 0.001, and the output layer to 1 neuron; The trained model is calibrated as a corrosion model, and the output result is the predicted corrosion amount; The corrosion model is deployed, and the operating data of the water vapor pipeline to be monitored is collected in real time to build a real-time working data set. The real-time working data set is mapped to the historical working data set and input into the corrosion model to obtain the current predicted corrosion amount of the water vapor pipeline.

4. A method for monitoring corrosion of water vapor pipelines of a supercritical unit according to claim 3, characterized in that: The water quality data of the current water inlet and outlet of the water vapor pipeline to be monitored are obtained. The water quality data includes conductivity, ion concentration and solid dissolved content, where the ion concentration is the divalent iron ion concentration, and the water quality difference between the water inlet and the water outlet is obtained. The formula is as follows: ， ， ， in, is the conductivity difference, is the inlet conductivity, is the outlet conductivity, is the ion concentration difference, is the ion concentration at the water inlet, is the outlet ion concentration, is the difference in solid dissolution, is the amount of dissolved solids at the water inlet, is the amount of dissolved solids at the outlet; A single corrosion score is generated based on the difference in water quality between the inlet and outlet, based on the following formula: ， in, To score a single corrosion, is the conductivity difference, is the ion concentration difference, is the difference in solid dissolution, , , are the scoring coefficients, .

5. A method for monitoring corrosion of water vapor pipelines of a supercritical unit according to claim 4, characterized in that: The current corrosion amount is obtained by predicting the corrosion amount and the single corrosion score, and the formula is as follows: ， in, is the current corrosion amount, To score a single corrosion, To predict the amount of corrosion.

6. A method for monitoring corrosion of water vapor pipelines of a supercritical unit according to claim 5, characterized in that: A vibration collector is arranged in each single section of the water vapor pipeline to be monitored, and the initial vibration data and the current vibration data of each single section of the pipeline are obtained respectively. The initial vibration data and the current vibration data record the vibration amplitude of the single section of the pipeline, and the vibration change of each single section of the pipeline is obtained respectively, according to the following formula: ， in, For the The vibration change of the pipe section, For the The vibration data of the pipeline. For the Initial vibration data of the pipeline, Retrieve variable for pipe number, , , is the number of pipelines; The standard deviation of the vibration variation of all single-section pipelines is obtained according to the following formula: ， in, is the standard deviation of the vibration variation of a single pipe section, For the The vibration change of the pipe section, is the average value of the vibration variation of all pipelines, Retrieve variable for pipe number, , , is the number of pipelines; Set the standard deviation threshold of vibration variation.

7. A method for monitoring corrosion of water vapor pipelines of a supercritical unit according to claim 6, characterized in that: Corrosion threshold I and corrosion threshold II are set respectively, where corrosion threshold I is smaller than corrosion threshold II. The current corrosion amount and the standard deviation of the vibration change of a single section of pipeline are analyzed. The analysis logic is as follows: When the current corrosion amount is less than the corrosion threshold I, the monitored water vapor pipeline has slight corrosion, but it does not affect the operation, and the single section of the pipeline does not need to be replaced; When the current corrosion amount is greater than the corrosion threshold I, the standard deviation of the vibration change of a single section of the pipeline is determined. If the standard deviation of the vibration change of a single section of the pipeline is less than the standard deviation threshold, all pipelines are corroded, and the overall corrosion has reached a certain level. However, due to the combined effect of all pipelines, each single section of the pipeline has not reached the corrosion scrap effect, and it will not be replaced. If the standard deviation of the vibration change of a single section of the pipeline is greater than the standard deviation threshold, it means that the individual pipeline is severely corroded, and the individual pipeline is replaced. The replacement logic is as follows: Determine the pipeline with the largest vibration change, replace the vibration change of the pipeline with the largest vibration change with 0, and obtain the standard deviation of the vibration change of the single-section pipeline again. If the standard deviation of the vibration change of the single-section pipeline is still greater than the standard deviation threshold, replace the vibration change of the pipeline with the largest and second largest vibration change with 0 at the same time, and obtain the standard deviation of the vibration change of the single-section pipeline again, and so on, until the standard deviation of the vibration change of the single-section pipeline is less than the standard deviation threshold, and the single-section pipeline involved in the vibration change replacement is replaced; When the current corrosion amount is greater than the corrosion threshold I and the standard deviation of the vibration change of a single pipeline section is always less than the standard deviation threshold, it is determined whether the current corrosion amount is greater than the corrosion threshold II. If the current corrosion amount is less than the corrosion threshold II, the pipeline will not be replaced. If the current corrosion amount is greater than the corrosion threshold II, all pipelines will be replaced.

8. A supercritical unit steam pipeline corrosion monitoring device, the device is used to perform a supercritical unit steam pipeline corrosion monitoring method according to any one of claims 1 to 7, characterized in that: include: The working condition data acquisition module is used to obtain the historical working condition data of the scrapped water vapor pipeline, wherein the historical working condition data includes the number of water vapor pipeline transportation, the maximum pressure, the maximum temperature, the maximum flow rate and the working time, and summarizes them into a historical working data set, obtains the corrosion amount of the pipeline in the historical working condition data set and uses it as a label of the working data set, obtains the real-time working data of the water vapor pipeline to be monitored to form a real-time working data set and maps it to the historical working data set; A model building module is used to input the historical working data set into the convolutional neural network model for training, obtain the corrosion model, and output the predicted corrosion amount; A water quality data acquisition module is used to obtain the water quality data of the current water inlet and outlet of the water vapor pipeline to be monitored, wherein the water quality data includes conductivity, ion concentration and dissolved solids, and obtain the water quality difference between the water inlet and the water outlet and obtain the current corrosion amount; A vibration data acquisition module is used to acquire the initial vibration data and current vibration data of each single-section pipeline, wherein the initial vibration data and current vibration data are the vibration amplitude of the single-section pipeline, generate the vibration variation of each single-section pipeline and the standard deviation of the vibration variation of all single-section pipelines, and set the standard deviation threshold; The single-section pipeline replacement selection module is used to set corrosion threshold I and corrosion threshold II, respectively determine the relationship between the current corrosion amount and corrosion threshold I and corrosion threshold II, and analyze the standard deviation of the vibration change of the single-section pipeline and the standard deviation threshold of the vibration change to determine the corrosion condition of the water vapor pipeline to be monitored.

Images