System and method for evaluating the corrosion state of nuclear power plant piping

Abstract

The present application relates to the field of nondestructive testing, and in particular to a nuclear power pipeline corrosion state evaluation system and method, the evaluation system comprising: a plurality of fiber grating sensors arranged at a plurality of preset positions of a nuclear power pipeline, for converting the state of the nuclear power pipeline at the preset positions into optical signals; a fiber grating demodulator connected to the plurality of fiber grating sensors, for analyzing the optical signals to obtain pipeline parameters; and an evaluation device for evaluating the corrosion state of the nuclear power pipeline based on a preset pipeline corrosion prediction model according to the pipeline parameters. The present application uses a plurality of fiber grating sensors, uses optical signals to represent pipeline parameters, and uses a fiber grating demodulator to process the optical signals. The optical signals have less dependence on the material of the pipeline and can be adapted to pipelines of different materials and structures.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of non-destructive testing, in particular to a nuclear power pipeline corrosion state evaluation system and method. BACKGROUND

[0002] Nuclear power pipelines are channels for transporting critical fluids in nuclear power plants, and the safety of the pipelines is directly related to the safe operation and personnel safety of the nuclear power plant. Corrosion is one of the main threats faced by pipelines during operation. Corrosion can cause the wall thickness of the pipeline to thin and the strength to decrease, thereby causing serious accidents such as leakage and explosion. These accidents not only cause huge economic losses, but also pose a serious threat to the environment and public safety. Through corrosion monitoring, pipeline corrosion problems can be detected in a timely manner and appropriate repair measures can be taken to prevent accidents.

[0003] Currently, the main method for pipeline corrosion is electromagnetic detection. Electromagnetic detection generates an alternating magnetic field around the nuclear power pipeline. When there are defects on the inside or surface of the pipeline due to corrosion, these defects will change the magnetic permeability and electrical conductivity of the material, thereby affecting the distribution of the magnetic field. The detector analyzes the changes in the magnetic field to identify and quantify the degree and location of corrosion. However, this method has a narrow application range and cannot be used to detect pipelines made of non-ferromagnetic materials. SUMMARY

[0004] To solve the above problems, the present application provides a nuclear power pipeline corrosion state evaluation system and method.

[0005] The first aspect of the present application discloses a nuclear power pipeline corrosion state evaluation system, which comprises:

[0006] A plurality of fiber grating sensors are arranged at a plurality of predetermined positions of the nuclear power pipeline and are used to convert the state of the nuclear power pipeline at the predetermined positions into optical signals;

[0007] A fiber grating demodulator is connected to the plurality of fiber grating sensors and is used to analyze the optical signals to obtain pipeline parameters;

[0008] An evaluation device is used to evaluate the corrosion state of the nuclear power pipeline based on a predetermined pipeline corrosion prediction model according to the pipeline parameters.

[0009] Further, the fiber grating sensor comprises:

[0010] A fiber temperature and humidity sensor is arranged at an outer position of the nuclear power pipeline and is used to convert the temperature and humidity of the nuclear power pipeline at the arranged position into an optical signal;

[0011] A fiber strain sensor is arranged at an outer position of the nuclear power pipeline and is used to convert the strain generated by the inner wall of the nuclear power pipeline at the arranged position into an optical signal;

[0012] The optical fiber PH value sensor is arranged at the inner side of the nuclear power pipeline and is used for converting the PH value of the fluid in the nuclear power pipeline at the arrangement position into an optical signal.

[0013] The optical fiber flow rate sensor is arranged at the inner side of the nuclear power pipeline and is used for converting the flow rate of the fluid in the nuclear power pipeline at the arrangement position into an optical signal.

[0014] Further, the pipeline corrosion prediction model is a stacking ensemble learning model.

[0015] The evaluation device is used for inputting the pipeline parameters into a plurality of different prediction models in a primary learner of the stacking ensemble learning model respectively, calculating an initial prediction result according to the prediction result of each prediction model, and obtaining the corrosion state of the nuclear power pipeline through a secondary learner of the stacking ensemble learning model according to the initial prediction result.

[0016] The second aspect of the present application discloses an evaluation method of a corrosion state of a nuclear power pipeline, which is applied to the evaluation system of the corrosion state of the nuclear power pipeline as any one of the first aspect of the present application.

[0017] The pipeline state of the nuclear power pipeline is collected through the fiber grating sensor, and a corresponding optical signal is generated;

[0018] The pipeline parameters of the nuclear power pipeline are obtained by demodulating the optical signal through the fiber grating demodulator.

[0019] The pipeline parameters are collected regularly through the evaluation device, and all the collected pipeline parameters are used to predict the corrosion state of the nuclear power pipeline based on the preset pipeline corrosion prediction model.

[0020] Further, the pipeline corrosion prediction model is a stacking ensemble learning model, and the step of predicting the corrosion state of the nuclear power pipeline based on the preset pipeline corrosion prediction model by using all the collected pipeline parameters comprises:

[0021] The pipeline parameters are input into a plurality of different prediction models in a primary learner of the stacking ensemble learning model respectively, and an initial prediction result is calculated according to the prediction result of each prediction model.

[0022] The initial prediction result is input into a secondary learner of the stacking ensemble learning model to obtain the corrosion state of the nuclear power pipeline.

[0023] Further, the primary learner comprises a random forest model, an adaptive boosting model, a gradient boosting decision tree model and an extreme gradient boosting model, and the secondary learner is a BP neural network model.

[0024] Further, the pipeline corrosion prediction model training method comprises:

[0025] a plurality of samples are obtained; wherein the samples comprise sample data and corresponding sample labels; the sample data comprises temperature of a nuclear power pipeline, humidity of the nuclear power pipeline, strain generated on an inner wall of the nuclear power pipeline, PH value of a fluid, and flow rate of the fluid; and the sample label is a corrosion state of the nuclear power pipeline;

[0026] According to a K-fold cross-validation algorithm, the plurality of samples are grouped into a plurality of training sets comprising training subsets and verification subsets; wherein the number of samples in each training subset is consistent, the number of samples in each verification subset is consistent, and each verification subset in the training set is different from each other;

[0027] The primary learner is trained based on the training set to obtain a trained primary learner;

[0028] An initial prediction result is obtained through the trained primary learner;

[0029] The secondary learner is trained based on the initial prediction result to obtain a trained secondary learner;

[0030] The trained primary learner and the trained secondary learner are combined to form a pipeline corrosion prediction model.

[0031] Further, the step of training the primary learner based on the training set to obtain a trained primary learner comprises:

[0032] Each prediction model in the primary learner is trained using a training subset in the training set, and a trained prediction model of the primary learner is obtained;

[0033] Each prediction model in the primary learner is verified using a verification subset in the training set:

[0034] A model that passes the verification is a trained prediction model of the primary learner;

[0035] A model that fails the verification is trained again using a training subset;

[0036] When all prediction models are trained, all trained prediction models are combined to form the primary learner.

[0037] Further, the step of training each prediction model in the primary learner using a training subset in the training set to obtain a trained prediction model of the primary learner comprises:

[0038] For each prediction model in the primary learner:

[0039] Input the sample data of the training subset in the training set into the prediction model to obtain the corresponding prediction results;

[0040] Based on the genetic algorithm, the value of the preset fitness function is calculated according to the prediction results and the sample labels in the training subset; the model parameters are adjusted according to the calculation results.

[0041] Furthermore, the fitness function includes a goodness of fit evaluation and a root mean square error; wherein the goodness of fit evaluation represents the variance percentage of the prediction result of a prediction model in the primary learner, and the root mean square error represents the square root of the average of the sum of the squares of the deviations between the prediction result of a prediction model in the primary learner and the sample label.

[0042] The present invention uses a variety of fiber grating sensors to represent pipeline parameters through optical signals, and uses a fiber grating demodulator to process the optical signals. The optical signals are less dependent on the pipeline material and can adapt to pipelines of different materials and structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 This is a schematic structural diagram of a nuclear power pipeline corrosion status assessment system disclosed in an embodiment of the present invention;

[0045] Figure 2 The present invention is a flowchart of a method for evaluating the corrosion status of a nuclear power pipeline disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0047] The terms "first", "second", and the like, in the description and in the claims of the present application, and in the above-described drawings, are used to distinguish between similar objects and are not necessarily used to describe a particular sequential or chronological order. Moreover, the terms "comprises", "comprising", "includes", "including", and the like, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises, includes, or consists of a list of steps or units, without limitation, does not necessarily disclose the only possible configuration of the process, method, article, or apparatus. Other steps or units can optionally be added in the future.

[0048] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase "in an embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. It is expressly understood that any of the embodiments described herein can be incorporated in to other embodiments.

[0049] Reference is made to Figure 1 , shown in the drawings, Figure 1 is a structural schematic diagram of an evaluation system for a corrosion state of a nuclear power pipeline according to an embodiment of the application. As Figure 1 shown, the evaluation system for the corrosion state of the nuclear power pipeline can include the following modules:

[0050] A plurality of fiber grating sensors 101 are arranged at a plurality of preset positions of the nuclear power pipeline, for converting the state of the nuclear power pipeline at the preset positions into optical signals;

[0051] In this alternative embodiment, the nuclear power pipeline refers to a channel for conveying critical fluids in a nuclear power plant, including but not limited to coolant, steam, water, fuel, and waste, etc. For example, the pipeline of the main coolant circulation system is responsible for carrying the high-temperature, high-pressure coolant generated from the reactor, sending it to the steam generator for heat exchange, and returning the cooled coolant to the reactor. These fluids are critical to the operation of the nuclear power plant, and any leakage or contamination can lead to serious consequences.

[0052] In one alternative embodiment, the fiber grating sensor includes:

[0053] A fiber-optic temperature and humidity sensor is arranged at an outer position of the nuclear power pipeline, for converting the temperature and humidity of the nuclear power pipeline at its arranged position into an optical signal;

[0054] A fiber-optic strain sensor is arranged at an outer position of the nuclear power pipeline, for converting the strain generated by the inner wall of the nuclear power pipeline at its arranged position into an optical signal;

[0055] The optical fiber PH value sensor is arranged on the inner side of the nuclear power pipeline and is used for converting the PH value of the fluid in the nuclear power pipeline at the arrangement position into an optical signal.

[0056] The optical fiber flow rate sensor is arranged on the inner side of the nuclear power pipeline and is used for converting the flow rate of the fluid in the nuclear power pipeline at the arrangement position into an optical signal.

[0057] In this optional embodiment, the optical fiber temperature and humidity sensor can convert the temperature and humidity information of the environment into an optical signal. Such a sensor is installed on the outer side of the nuclear power pipeline. By monitoring and analyzing the reflected optical signal, the temperature and humidity of the pipeline surface can be accurately measured.

[0058] The optical fiber strain sensor is a sensor that can convert physical deformation (such as stretching or compression) into an optical signal. Such a sensor is usually installed on the outer side of the nuclear power pipeline. By measuring the change in the reflected optical signal, the strain size of the inner wall of the pipeline can be accurately obtained. Strain is a physical quantity used to describe the degree of shape change of an object, including stretching and compression. When the inner wall of the pipeline is strained, the optical fiber strain sensor installed on the outer side of the pipeline will also be deformed. This deformation will change the reflection spectrum of the optical fiber grating. The optical fiber grating demodulator can obtain the size of the strain by analyzing this change.

[0059] The optical fiber PH value sensor is a sensor that can measure the acidity or alkalinity of a liquid. It converts the PH value into an optical signal and then transmits it to the data processing center through an optical fiber. Such a sensor is usually installed on the inner side of the nuclear power pipeline to monitor the PH value of the fluid in the pipeline. Accurate measurement of the PH value helps to prevent corrosion problems caused by changes in acidity or alkalinity. PH value is a parameter that measures the acidity or alkalinity of a solution, with a value ranging from 0 to 14. A value of 7 indicates neutrality, a value less than 7 indicates acidity, and a value greater than 7 indicates alkalinity. The PH value of the fluid in the nuclear power pipeline is very important for predicting and preventing pipeline corrosion. In general, fluids with strong acidity and alkalinity are more likely to cause pipeline corrosion. By monitoring and adjusting the PH value of the fluid, pipeline corrosion can be effectively prevented or slowed down, thereby prolonging the service life of the pipeline and reducing safety risks.

[0060] The optical fiber flow rate sensor is a sensor that can measure the flow rate of a fluid by converting the flow rate into an optical signal. Such a sensor is usually installed on the inner side of the nuclear power pipeline to monitor the flow rate of the fluid in the pipeline. Monitoring the flow rate helps to prevent problems caused by excessive or insufficient flow rate, such as corrosion, wear and tear, etc.

[0061] The optical fiber grating demodulator 102 is connected to multiple optical fiber grating sensors and is used to analyze the optical signal to obtain pipeline parameters.

[0062] In the optional embodiment, the fiber grating demodulator is a special device for analyzing the optical signal transmitted by the fiber grating sensor. The fiber grating is a special optical fiber with micro gratings engraved on its length direction at a specific interval. When light passes through these gratings, different wavelengths of light will produce different mutual interference to form a unique reflection spectrum. The fiber grating demodulator can obtain the physical or chemical parameters such as temperature, humidity, strain, PH value and flow rate at the position of the fiber grating sensor by analyzing these reflection spectra.

[0063] The evaluation device 103 is configured to evaluate the corrosion state of the nuclear power pipeline based on a preset pipeline corrosion prediction model according to the pipeline parameters.

[0064] In an optional embodiment, the pipeline corrosion prediction model is a stacking ensemble learning model.

[0065] The evaluation device is configured to input the pipeline parameters into a plurality of different prediction models in the primary learners of the stacking ensemble learning model, calculate an initial prediction result according to the prediction results of each prediction model, and obtain the corrosion state of the nuclear power pipeline through the secondary learner of the stacking ensemble learning model according to the initial prediction result.

[0066] The stacking ensemble learning model is an ensemble learning model that takes the outputs of a plurality of base models as a new training set to retrain a meta model. By combining the prediction results of different models, the accuracy and robustness of the overall prediction can be improved, and the risk of overfitting can be reduced. It is a very effective ensemble learning strategy.

[0067] Referring to Figure 2 , a flowchart of a method for evaluating the corrosion state of a nuclear power pipeline is shown. Figure 2 As shown in the figure, the method for evaluating the corrosion state of the nuclear power pipeline can include the following steps: Figure 2

[0068] S201, acquiring the pipeline state of the nuclear power pipeline by the fiber grating sensor and generating corresponding optical signals;

[0069] For the type and setting method of the fiber grating sensor in this embodiment, please refer to the related description in the evaluation system for the corrosion state of the nuclear power pipeline.

[0070] S202, analyzing the optical signals by the fiber grating demodulator to obtain the pipeline parameters of the nuclear power pipeline;

[0071] ​For the introduction of the fiber grating demodulator in the embodiment, please refer to the relevant introduction in the application about the evaluation system of the corrosion state of the nuclear power pipeline.

[0072] S203, periodically collecting pipeline parameters by the evaluation device, and predicting the corrosion state of the nuclear power pipeline based on the preset pipeline corrosion prediction model according to all the collected pipeline parameters.

[0073] In an optional embodiment, the pipeline corrosion prediction model is a Stacking integrated learning model, and the step of predicting the corrosion state of the nuclear power pipeline based on the preset pipeline corrosion prediction model according to all the collected pipeline parameters includes:

[0074] inputting the pipeline parameters into a plurality of different prediction models in the primary learner of the Stacking integrated learning model respectively, and calculating initial prediction results according to the prediction results of each prediction model;

[0075] inputting the initial prediction results into the secondary learner of the Stacking integrated learning model to obtain the corrosion state of the nuclear power pipeline.

[0076] In a further optional embodiment, the primary learner includes a random forest model, an adaptive boosting model, a gradient boosting decision tree model and an extreme gradient boosting model, and the secondary learner is a BP neural network model.

[0077] In this optional embodiment, the random forest is an integrated learning algorithm based on decision trees. It makes the final prediction by constructing multiple decision trees and combining their prediction results. In the training process, the random forest adopts the method of bootstrap sampling to randomly select samples and feature subsets to construct each decision tree, which can improve the diversity and robustness of the model. Random forest has good classification and regression performance, can effectively process high-dimensional data and noise, and has strong robustness to outliers and missing values.

[0078] The adaptive boosting model is an iterative boosting algorithm that constructs a strong learner by combining multiple weak learners. In each iteration, the adaptive boosting model adjusts the weights of the samples according to the performance of the weak learners in the last round. For the samples that are misclassified, increase their weights; for the samples that are correctly classified, reduce their weights. In this way, the subsequent weak learners will pay more attention to those difficult-to-classify samples. Finally, the final prediction is made by weighted combination of the prediction results of all weak learners. The adaptive boosting model is sensitive to noise and outliers, but performs well in many practical applications.

[0079] Gradient Boosting Decision Tree model is a boosting algorithm based on decision trees, which iteratively builds a series of decision trees to optimize the model step by step. In each iteration, the Gradient Boosting Decision Tree model fits a new decision tree according to the residual of the current model and adds it to the model. The residual represents the difference between the predicted value of the current model and the true value. By continuously fitting the residual, the Gradient Boosting Decision Tree model gradually reduces the error of the model and improves the prediction performance. Gradient Boosting Decision Tree model can handle various types of data, including continuous values and category values, and has certain robustness to outliers and noise.

[0080] Extreme Gradient Boosting model is an optimized implementation of Gradient Boosting Decision Tree model, which introduces some improvements and optimization strategies based on Gradient Boosting Decision Tree model. Extreme Gradient Boosting model uses second-order Taylor expansion to approximate the loss function, considering the first-order derivative and second-order derivative, so as to more accurately fit the residual. In addition, Extreme Gradient Boosting model also introduces a regularization term to control the complexity of the model and prevent overfitting. Extreme Gradient Boosting model has significant improvement in training speed and prediction performance, and supports parallel computing, which can effectively handle large-scale data sets. Extreme Gradient Boosting model has achieved excellent results in many data mining competitions

[0081] BP neural network is a supervised learning algorithm based on gradient descent, which trains multi-layer perceptron through backpropagation algorithm. BP neural network is composed of input layer, hidden layer and output layer, each layer is composed of multiple neurons. In the forward propagation process, the input signal is transmitted from the input layer to the output layer through the hidden layer, generating the predicted result. In the backpropagation process, according to the error between the predicted result and the true value, the error term of each neuron is calculated through the chain rule, and the weights and biases of the network are updated using gradient descent method to minimize the loss function. BP neural network has strong non-linear fitting ability and can learn complex patterns and relationships.

[0082] In this optional embodiment, the selected primary learners include different types of models, such as tree-based models (random forest, gradient boosting decision tree model, extreme gradient boosting model) and boosting-based models (adaptive boosting model). These models differ in algorithm principles, feature processing, and decision boundaries, and can capture data features and patterns from different perspectives. The diversity of models helps improve the performance of the Stacking ensemble learning model, because the prediction errors of different models may be uncorrelated, and by combining their prediction results, the overall error rate can be reduced. The BP neural network as a secondary learner has strong nonlinear fitting capability. It can learn the complex nonlinear relationship between the prediction results of the primary learners and the true values, thereby further improving the overall prediction performance. The addition of the BP neural network enables the Stacking ensemble learning model to better handle complex data patterns and nonlinear problems.

[0083] In an optional embodiment, the method for training the pipeline corrosion prediction model comprises:

[0084] Obtaining a plurality of samples; wherein the samples include sample data and corresponding sample labels; the sample data includes the temperature of the nuclear power pipeline, the humidity of the nuclear power pipeline, the strain generated on the inner wall of the nuclear power pipeline, the PH value of the fluid, and the flow rate of the fluid; and the sample label is the corrosion state of the nuclear power pipeline;

[0085] According to the K-fold cross-validation algorithm, a plurality of training sets including training subsets and validation subsets are formed from the plurality of samples; wherein the number of samples in each training subset is consistent, the number of samples in each validation subset is consistent, and each validation subset in the training set is different from each other;

[0086] Training the primary learners based on the training set to obtain trained primary learners;

[0087] Obtaining initial prediction results through the trained primary learners;

[0088] Training the secondary learners based on the initial prediction results to obtain trained secondary learners;

[0089] Forming a pipeline corrosion prediction model by combining the trained primary learners and the trained secondary learners.

[0090] According to the K-fold cross-validation algorithm, a plurality of training sets including training subsets and validation subsets are formed from the plurality of samples; wherein the number of samples in each training subset is consistent, the number of samples in each validation subset is consistent, and each validation subset in the training set is different from each other;

[0091] In the optional embodiment, each prediction model in the primary learner is trained K times, and each time a validation subset prediction result is obtained and a training subset prediction result After m prediction models are trained, m sets of validation subset prediction results are obtained and m sets of training subset prediction results are obtained The training subset prediction results obtained by K times of training are averaged to obtain m sets of training subset prediction results: The validation subset prediction results obtained by K times of training are averaged to obtain m sets of validation subset prediction results Take and as the input of the secondary learner.

[0092] In an optional embodiment, the step of training the primary learner based on the training set to obtain the trained primary learner includes:

[0093] Each prediction model in the primary learner is trained using the training subset in the training set, and the trained prediction model of the primary learner is obtained;

[0094] Each prediction model in the primary learner is verified using the validation subset in the training set:

[0095] The model that passes the verification is the trained prediction model of the primary learner;

[0096] The model that fails the verification is trained again using the training subset;

[0097] When all the prediction models are trained, the trained prediction models are combined to form the primary learner.

[0098] In an optional embodiment, the step of training each prediction model in the primary learner using the training subset in the training set to obtain the trained prediction model of the primary learner includes:

[0099] For each prediction model in the primary learner:

[0100] The sample data in the training subset in the training set is input into the prediction model to obtain a corresponding prediction result;

[0101] Based on the genetic algorithm, the value of the preset fitness function is calculated according to the prediction result and the sample label in the training subset; and the model parameters are adjusted according to the calculation result.

[0102] In this optional embodiment, a genetic algorithm searches based on the principles of natural selection and inheritance, repeatedly modifying and improving the current solution to find the optimal solution. The initial value of the genetic algorithm is a possible combination of model parameters of the prediction model. This initial value can be a randomly generated set of parameter combinations or can be pre-set based on experience.

[0103] In a genetic algorithm, a fitness function is used to evaluate the performance of each candidate solution (i.e., model parameter combination). Solutions with good fitness have a higher probability of being selected as the parent of the next generation. New solutions are then generated through the two main operations of the genetic algorithm: crossover (or pairing) and mutation. Crossover involves exchanging some parameters of two solutions to generate a new solution; mutation involves randomly modifying some of the parameters of a solution. These two steps introduce randomness, ensuring that the algorithm searches globally and avoids being trapped in local optima. After multiple iterations (i.e., multiple generations of population evolution), the algorithm eventually converges to a solution, or a set of solutions, that are optimal in terms of the fitness function. This means that the model parameter configuration optimizes its predictive performance.

[0104] In an optional embodiment, the fitness function includes a goodness of fit evaluation and a root mean square error; wherein the goodness of fit evaluation represents the variance percentage of the prediction result of a prediction model in the primary learner, and the root mean square error represents the square root of the average of the sum of the squares of the deviations between the prediction result of a prediction model in the primary learner and the sample label.

[0105] In this alternative embodiment, R-squared (R 2 , determination coefficient) to measure the goodness of fit. R 2 The value of ranges between 0 and 1. The closer it is to 1, the more consistent the model's predictions are with the actual values, indicating a better fit. The root mean square error (RMSE) is a commonly used metric for forecast error. It represents the square root of the average of the squared deviations between the forecasted and actual values. The smaller the RMSE, the more accurate the forecast model.

[0106] In summary, the present invention discloses a nuclear power pipeline corrosion assessment system and method. This system utilizes multiple fiber Bragg grating (FBG) sensors to represent pipeline parameters through optical signals, which are then processed using a fiber Bragg grating (FBG) interrogator. The optical signals are less dependent on pipeline material and can be adapted to pipelines of varying materials and structures. Therefore, this invention effectively overcomes the shortcomings of existing technologies and possesses high industrial applicability.

[0107] The above embodiments are only illustrative of the principles of the present application and its efficacy, and are not intended to limit the present application. Any modification or change made by any person skilled in the art without departing from the spirit and scope of the present application shall be covered by the claims of the present application.

Claims

1. A nuclear power pipeline corrosion status assessment system, characterized in that: The system comprises: A plurality of fiber grating sensors are arranged at a plurality of preset positions of the nuclear power pipeline, and are used to convert the nuclear power pipeline status at the preset positions into optical signals; A fiber Bragg grating demodulator, connected to a plurality of fiber Bragg grating sensors, for analyzing the optical signal to obtain pipeline parameters; An evaluation device for evaluating the corrosion status of nuclear power pipelines based on the pipeline parameters and a preset pipeline corrosion prediction model; The fiber grating sensor comprises: The optical fiber temperature and humidity sensor is installed outside the nuclear power pipeline and is used to convert the temperature and humidity of the nuclear power pipeline where it is installed into optical signals; The optical fiber strain sensor is installed outside the nuclear power pipeline and is used to convert the strain generated by the inner wall of the nuclear power pipeline where it is installed into an optical signal; The optical fiber pH sensor is installed inside the nuclear power pipeline and is used to convert the pH value of the fluid in the nuclear power pipeline into an optical signal; The optical fiber flow velocity sensor is installed inside the nuclear power pipeline and is used to convert the flow velocity of the fluid in the nuclear power pipeline into an optical signal; The pipeline corrosion prediction model is a Stacking ensemble learning model; The evaluation device is used to input the pipeline parameters into multiple different prediction models in the primary learner of the Stacking ensemble learning model, calculate an initial prediction result based on the prediction results of each prediction model; and obtain the corrosion status of the nuclear power pipeline through the secondary learner of the Stacking ensemble learning model based on the initial prediction result; The primary learner includes a random forest model, an adaptive boosting model, a gradient boosting decision tree model and an extreme gradient boosting model, and the secondary learner is a BP neural network model.

2. A method for evaluating the corrosion status of nuclear power pipelines, characterized in that: The nuclear power pipeline corrosion status assessment system according to claim 1 is applied, wherein the assessment method comprises: The pipeline status of nuclear power pipelines is collected through fiber grating sensors and corresponding optical signals are generated; The optical signal is analyzed by a fiber Bragg grating demodulator to obtain the pipeline parameters of the nuclear power pipeline; Pipeline parameters are collected regularly through evaluation equipment, and all collected pipeline parameters are used to predict the corrosion status of nuclear power pipelines based on the preset pipeline corrosion prediction model.

3. The method for evaluating the corrosion status of nuclear power pipelines according to claim 2, characterized in that: The pipeline corrosion prediction model is a stacking ensemble learning model. The steps of predicting the corrosion status of nuclear power pipelines based on all collected pipeline parameters and the preset pipeline corrosion prediction model include: Inputting the pipeline parameters into a plurality of different prediction models in the primary learner of the stacking ensemble learning model respectively, and calculating an initial prediction result based on the prediction result of each prediction model; The initial prediction result is input into the secondary learner of the Stacking ensemble learning model to obtain the corrosion status of the nuclear power pipeline.

4. A method for evaluating the corrosion status of nuclear power pipelines according to claim 3, characterized in that: The training method of the pipeline corrosion prediction model includes: Acquire multiple samples; wherein the samples include sample data and corresponding sample labels; the sample data includes the temperature of the nuclear power pipeline, the humidity of the nuclear power pipeline, the strain generated by the inner wall of the nuclear power pipeline, the pH value of the fluid, and the flow rate of the fluid; the sample label is the corrosion state of the nuclear power pipeline; According to the K-fold cross-validation algorithm, the plurality of samples are grouped into a plurality of training sets including training subsets and validation subsets; wherein the number of samples in each training subset is the same, the number of samples in each validation subset is the same; and each validation subset in the training set is different from each other; Training the primary learner based on the training set to obtain the trained primary learner; And obtain the initial prediction results through the trained primary learner; Based on the initial prediction result, the secondary learner is trained to obtain the trained secondary learner; The trained primary learner and the trained secondary learner are combined into a pipeline corrosion prediction model.

5. The method for evaluating the corrosion status of nuclear power pipelines according to claim 4, characterized in that: The step of training the primary learner based on the training set to obtain the trained primary learner includes: Using the training subset in the training set to train each prediction model in the primary learner, thereby obtaining a corresponding prediction model of the primary learner after training; Each prediction model in the primary learner is validated using the validation subset in the training set: The model that passes the verification is the prediction model of the trained primary learner; If the model fails the verification, the prediction model is trained again using the training subset; When all prediction models are trained, all trained prediction models are combined into the primary learner.

6. A method for evaluating the corrosion status of nuclear power pipelines according to claim 5, characterized in that: The steps of training each prediction model in the primary learner using a training subset in the training set, and obtaining the prediction model of the primary learner after training include: For each prediction model in the primary learner: Input the sample data of the training subset in the training set into the prediction model to obtain the corresponding prediction results; Based on the genetic algorithm, the value of the preset fitness function is calculated according to the prediction results and the sample labels in the training subset; the model parameters are adjusted according to the calculation results.

7. A method for evaluating the corrosion status of nuclear power pipelines according to claim 6, characterized in that: The fitness function includes a goodness of fit evaluation and a root mean square error; wherein the goodness of fit evaluation represents the variance percentage of the prediction result of a prediction model in the primary learner, and the root mean square error represents the square root of the average of the sum of the squares of the deviations between the prediction result of a prediction model in the primary learner and the sample label.

Images