Safety Monitoring Method, Device, Storage Medium and Computer Equipment for Boiler Pipelines

Abstract

The present application discloses a safety monitoring method, device, storage medium and computer device for boiler pipelines. The method includes: inputting the target inlet parameters and target measurement point parameters of the boiler pipeline into a modal coefficient prediction model to determine the erosion rate distribution modal coefficients of the boiler pipeline at different times; performing a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution of the boiler pipeline to determine the target erosion rate distribution matrix; calculating the second thickness distribution matrix at the current moment based on the first thickness distribution matrix and the target erosion rate distribution matrix of the boiler pipeline obtained in the previous sampling period; if the thickness at any measurement point position in the second thickness distribution matrix is less than or equal to the burst pipe thickness at any measurement point position, an early warning message is output. Thereby, accurate prediction of the erosion rate at different measurement point positions and different times of the boiler pipeline is realized, and the change of the thickness of the boiler pipeline is reflected in real time, and an early warning signal is sent when a burst pipe may occur.

Description

Technical Field

[0001] The present application relates to the technical field of boiler safety monitoring, and in particular, to a safety monitoring method, device, storage medium and computer equipment for boiler pipes. Background Art

[0002] In the related art, prediction is carried out by means of numerical simulation. During the prediction process, it is necessary to perform mesh division on the model and calculate the continuous phase and discrete phase. The time required to predict the wear rate of the furnace tubes throughout the boiler is usually at the hour level. During the calculation process, the working conditions have fluctuated countless times. Therefore, this calculation method cannot predict the wear rate of the boiler furnace tubes in real time, resulting in low prediction accuracy of the wear amount of the furnace tubes. At the same time, if this method is deployed to the production site, it is equivalent to constantly performing numerical simulation calculations, and the prediction process will also consume a large amount of computer computing power. Summary of the Invention

[0003] In view of this, the present application provides a safety monitoring method, device, storage medium and computer equipment for boiler pipes, which can reduce the amount of calculation while ensuring the prediction accuracy of the erosion rate distribution, and can monitor the wall thickness after erosion.

[0004] According to one aspect of the present application, a safety monitoring method for boiler pipes is provided, including:

[0005] Input the target inlet parameters and target measuring point parameters of the boiler pipe into the modal coefficient prediction model to determine the erosion rate distribution modal coefficients of the boiler pipe at different times, where the modal coefficient prediction model is trained based on the sample data of the boiler pipe at different times, and the sample data includes the corresponding erosion rate distribution modal coefficient samples, inlet parameter samples and measuring point parameter samples of the boiler pipe;

[0006] Perform a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution mode of the boiler pipe to determine the target erosion rate distribution matrix;

[0007] Based on the first thickness distribution matrix of the boiler pipe obtained in the previous sampling period and the target erosion rate distribution matrix, calculate the second thickness distribution matrix at the current time;

[0008] If the thickness at any measuring point position in the second thickness distribution matrix is less than or equal to the burst pipe thickness at the any measuring point position, output a warning message.

[0009] Optionally, the safety monitoring method for the boiler pipe further includes:

[0010] Based on the simulation model of the boiler pipe, calculate the measuring point parameter samples and erosion rate samples at different times after the boiler pipe is stable under the inlet parameter samples;

[0011] Based on the erosion rate samples, the moments, and the measuring point parameter samples, establish a sample of the erosion rate distribution matrix;

[0012] Perform modal decomposition on the sample of the erosion rate distribution matrix to determine the erosion rate distribution modes and the sample of the erosion rate distribution mode coefficients;

[0013] Based on the sample of the erosion rate distribution mode coefficients, the inlet parameter samples, and the measuring point parameter samples, train the modal coefficient prediction model.

[0014] Optionally, the training of the modal coefficient prediction model based on the sample of the erosion rate distribution mode coefficients, the inlet parameter samples, and the measuring point parameter samples includes:

[0015] If the modal order of the sample of the erosion rate distribution mode coefficients is greater than the first quantity threshold, group the sample of the erosion rate distribution mode coefficients to form multiple coefficient sample sets;

[0016] Input the coefficient sample sets and their corresponding inlet parameter samples and measuring point parameter samples into a deep learning model for iterative training to obtain the modal coefficient prediction model for the coefficient sample sets;

[0017] Based on the inlet parameter samples and measuring point parameter samples corresponding to the coefficient sample sets, determine the inlet parameter intervals and the measuring point parameter intervals;

[0018] Based on the inlet parameter intervals and the measuring point parameter intervals, associate the modal coefficient prediction models.

[0019] Optionally, the inputting the target inlet parameters and target measuring point parameters of the boiler pipeline into the modal coefficient prediction model to determine the erosion rate distribution mode coefficients of the boiler pipeline at different moments includes:

[0020] Match the target inlet parameters and the target measuring point parameters with the inlet parameter intervals and the measuring point parameter intervals associated with different modal coefficient prediction models respectively to screen the modal coefficient prediction models;

[0021] Input the target inlet parameters and the target measuring point parameters into the screened modal coefficient prediction model to determine the erosion rate distribution mode coefficients;

[0022] If the number of the screened modal coefficient prediction models is multiple, combine the multiple erosion rate distribution mode coefficients obtained by the multiple modal coefficient prediction models.

[0023] Optionally, the linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode of the boiler pipeline to determine the target erosion rate distribution matrix includes:

[0024] If the modal order of the erosion rate distribution modal coefficient is greater than the second quantity threshold, determine the minimum modal order that meets the error requirement;

[0025] Based on the minimum modal order, perform a linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode to determine the target erosion rate distribution matrix.

[0026] Optionally, the determination of the minimum modal order that meets the error requirement includes:

[0027] Based on the modal order of the erosion rate distribution modal coefficient, perform a linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode to determine the first erosion rate distribution matrix;

[0028] Based on a preset modal order, perform a linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode to determine the second erosion rate distribution matrix;

[0029] Compare the first erosion rate distribution matrix and the second erosion rate distribution matrix, and calculate the reconstruction error value;

[0030] If the reconstruction error value is less than or equal to the error threshold, use the preset modal order as the minimum modal order;

[0031] If the reconstruction error value is greater than the error threshold, increase the preset modal order based on a preset step size until the reconstruction error value is less than or equal to the error threshold.

[0032] Optionally, the safety monitoring method for the boiler pipeline further includes:

[0033] If the maximum erosion rate in the target erosion rate distribution matrix is greater than the erosion rate threshold, calculate the extreme value of the operating condition parameters through the maximum erosion rate optimization algorithm, where the maximum erosion rate corresponding to the extreme value of the operating condition parameters is less than or equal to the erosion rate threshold and meets the load requirement of the boiler pipeline;

[0034] Adjust the boiler operating parameters based on the extreme value of the operating condition parameters.

[0035] According to another aspect of the present application, there is provided a safety monitoring device for a boiler pipeline, including:

[0036] An erosion rate prediction module, configured to input the target inlet parameters and target measurement point parameters of a boiler pipeline into a modal coefficient prediction model, and determine the erosion rate distribution modal coefficients of the boiler pipeline at different times, where the modal coefficient prediction model is trained based on sample data of the boiler pipeline at different times, and the sample data includes corresponding erosion rate distribution modal coefficient samples, inlet parameter samples, and measurement point parameter samples of the boiler pipeline; and perform a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution modes of the boiler pipeline to determine a target erosion rate distribution matrix;

[0037] A thickness prediction module, configured to calculate a second thickness distribution matrix at the current time based on the first thickness distribution matrix of the boiler pipeline obtained in the previous sampling period and the target erosion rate distribution matrix;

[0038] An early warning module, configured to output an early warning message if the thickness at any measurement point position in the second thickness distribution matrix is less than or equal to the burst pipe thickness at the any measurement point position.

[0039] Optionally, the safety monitoring device of the boiler pipeline further includes:

[0040] A sample acquisition module, configured to calculate measurement point parameter samples and erosion rate samples at different times after the boiler pipeline stabilizes under the inlet parameter samples based on the simulation model of the boiler pipeline; and establish an erosion rate distribution matrix sample based on the erosion rate samples, the times, and the measurement point parameter samples; and perform modal decomposition on the erosion rate distribution matrix sample to determine the erosion rate distribution modes and the erosion rate distribution modal coefficient samples;

[0041] A training module, configured to train the modal coefficient prediction model based on the erosion rate distribution modal coefficient samples, the inlet parameter samples, and the measurement point parameter samples.

[0042] Optionally, the safety monitoring device of the boiler pipeline further includes:

[0043] A grouping module, configured to group the erosion rate distribution modal coefficient samples to form multiple coefficient sample sets if the modal order of the erosion rate distribution modal coefficient samples is greater than a first quantity threshold;

[0044] The training module is specifically configured to input the coefficient sample sets and their corresponding inlet parameter samples and measurement point parameter samples into a deep learning model for iterative training to obtain the modal coefficient prediction model of the coefficient sample sets;

[0045] An association module, configured to determine an inlet parameter interval and a measuring point parameter interval based on the coefficient sample set corresponding to the inlet parameter sample and the measuring point parameter sample; and, based on the inlet parameter interval and the measuring point parameter interval, associate the modal coefficient prediction model.

[0046] Optionally, the safety monitoring device for the boiler pipeline further includes:

[0047] A model matching module, configured to match the target inlet parameter and the target measuring point parameter with the inlet parameter interval and the measuring point parameter interval associated with different modal coefficient prediction models respectively, so as to screen the modal coefficient prediction models;

[0048] The erosion rate prediction module is specifically configured to input the target inlet parameter and the target measuring point parameter into the screened modal coefficient prediction model to determine the erosion rate distribution modal coefficient; and, if the number of the screened modal coefficient prediction models is multiple, combine the multiple erosion rate distribution modal coefficients obtained by the multiple modal coefficient prediction models.

[0049] Optionally, the safety monitoring device for the boiler pipeline further includes:

[0050] A modal order updating module, configured to determine the minimum modal order that meets the error requirement if the modal order of the erosion rate distribution modal coefficient is greater than a second quantity threshold;

[0051] The erosion rate prediction module is specifically configured to perform a linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode based on the minimum modal order to determine a target erosion rate distribution matrix.

[0052] Optionally, the modal order updating module is specifically configured to perform a linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode based on the modal order of the erosion rate distribution modal coefficient to determine a first erosion rate distribution matrix; perform a linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode based on a preset modal order to determine a second erosion rate distribution matrix; compare the first erosion rate distribution matrix and the second erosion rate distribution matrix, and calculate a reconstruction error value; if the reconstruction error value is less than or equal to an error threshold, use the preset modal order as the minimum modal order; if the reconstruction error value is greater than the error threshold, increase the preset modal order based on a preset step size until the reconstruction error value is less than or equal to the error threshold.

[0053] Optionally, the safety monitoring device for the boiler pipeline further includes:

[0054] A control module, configured to calculate extreme values of operating condition parameters through a maximum erosion rate optimization algorithm if the maximum value of the erosion rate in the target erosion rate distribution matrix is greater than an erosion rate threshold, where the maximum value of the erosion rate corresponding to the extreme values of the operating condition parameters is less than or equal to the erosion rate threshold and meets the load requirement of the boiler pipeline; and adjust the boiler operating parameters based on the extreme values of the operating condition parameters.

[0055] According to another aspect of the present application, there is provided a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the above-mentioned safety monitoring method for boiler pipelines are implemented.

[0056] According to yet another aspect of the present application, there is provided a computer device, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, and when the processor executes the program, the steps of the above-mentioned safety monitoring method for boiler pipelines are implemented.

[0057] By means of the above technical solution, taking the target inlet parameters and target measuring point parameters of the boiler pipeline as inputs, the erosion rate distribution modal coefficients of the boiler pipeline at different times are predicted through a trained modal coefficient prediction model. A linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution of the boiler pipeline is generated to form a target erosion rate distribution matrix containing the possible erosion rates at different measuring point positions of the boiler pipeline at different times. Using the known first thickness distribution matrix, the time step of data acquisition, and the erosion rate predicted by the model in the target erosion rate distribution matrix, the second thickness distribution matrix at the current moment is calculated. And monitor and give early warnings to the wall thicknesses at different measuring point positions in the second thickness distribution matrix. Thus, through the combination of modal decomposition and machine learning, the erosion rate distribution of the whole-field boiler tubes is calculated in real time according to the on-site operating condition parameters and on-site measuring point parameters, which can fully consider the influence of the time-series change during the stable operation of the boiler on the erosion rate distribution, and realize the accurate prediction of the erosion rates at different measuring point positions and different times of the boiler pipeline. And it can reflect the change of the thickness of the boiler pipeline in real time, send out early warning signals when a pipe burst may occur for further on-site inspection and taking measures in advance, realizing timely pipeline safety early warning, effectively avoiding safety accidents such as leakage and explosion caused by pipeline erosion, reducing energy loss and system failures caused by pipeline erosion, and ensuring the safety of personnel and boiler operation.

[0058] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the specific embodiments of the present application. Brief Description of the Drawings

[0059] The accompanying drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:

[0060] Figure 1 A schematic flow chart of a safety monitoring method for boiler pipelines provided by an embodiment of the present application is shown;

[0061] Figure 2 A schematic flow chart of modal coefficient grouping provided by an embodiment of the present application is shown;

[0062] Figure 3 A structural block diagram of a safety monitoring device for boiler pipelines provided by an embodiment of the present application is shown;

[0063] Figure 4 A schematic electronic structure diagram of a computer device provided by an embodiment of the present application is shown. Detailed implementation manners

[0064] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other.

[0065] Embodiments of the present application will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are only used to explain the present application and should not be construed as a limitation of the present application.

[0066] Those skilled in the art of the present technology can understand that, unless specifically stated otherwise, the singular forms "a", "an", "the", and "said" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of the present application means the presence of the described features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or their groups. It should be understood that when we say that an element is "connected" or "joined" to another element, it can be directly connected or joined to other elements, or there may also be intermediate elements. In addition, the "connection" or "joining" used herein may include wireless connection or wireless coupling. The phrase "and / or" used herein includes all or any unit and all combinations of one or more related listed items.

[0067] Now, exemplary embodiments according to the present application will be described in more detail with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many different forms and should not be construed as being limited only to the embodiments set forth herein. It should be understood that these embodiments are provided so that the disclosure of the present application is thorough and complete, and the concept of these exemplary embodiments is fully conveyed to those of ordinary skill in the art.

[0068] In this embodiment, a safety monitoring method for boiler pipelines is provided. This method can be applied to a terminal, a server, or software running on a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the safety monitoring method for boiler pipelines, etc., but is not limited to the above forms.

[0069] As Figure 1 shown, the safety monitoring method for boiler pipelines includes:

[0070] Step 101: Input the target inlet parameters and target measurement point parameters of the boiler pipeline into the modal coefficient prediction model to determine the erosion rate distribution modal coefficients of the boiler pipeline at different times.

[0071] Among them, the modal coefficient prediction model is trained based on the sample data of the boiler pipeline at different times. The sample data includes the corresponding erosion rate distribution modal coefficient samples, inlet parameter samples, and measurement point parameter samples of the boiler pipeline.

[0072] Specifically, the target inlet parameters and target measurement point parameters are the inlet parameters and measurement point parameters sampled according to the system response prediction instruction. The inlet parameters are the operating parameters at the pipeline inlet, and the measurement point parameters are the operating parameters at different measurement point positions of the boiler pipeline. The operating parameters include, but are not limited to: flue gas temperature, flow rate, pressure, fly ash particle content in the flue gas, fly ash particle size, temperature, inlet angle, etc.

[0073] In this embodiment, the modal coefficient prediction model predicts the erosion rate distribution modal coefficient of the current operating state through the target inlet parameters and target measurement point parameters in the current operating state. It can compress the high-dimensional erosion rate field into low-dimensional modal coefficients, retain the main energy characteristics, and avoid the computational burden of directly processing high-dimensional field data, thereby reducing the time and computational resources consumed in the prediction process.

[0074] In one embodiment, training a modal coefficient prediction model includes the following steps:

[0075] Step 201: Based on the simulation model of the boiler pipeline, calculate the measured point parameter samples and erosion rate samples at different moments after the boiler pipeline stabilizes under the inlet parameter samples.

[0076] Among them, the simulation model is used to simulate the numerical values of the flow of boiler flue gas in the flue gas pipeline, the heat exchange between the flue gas and the water wall, superheater, evaporator, and economizer, the flow and heat transfer of fly ash particles in the flue gas, and the interaction between the fly ash and the furnace tube wall. Specifically, the calculation of the flow and heat transfer process of the flue gas in the boiler flue gas pipeline uses the mass conservation equation, momentum conservation equation, and energy conservation equation. The process of the fly ash particles in the flue gas flowing with the flue gas and transferring heat with the flue gas and furnace tubes uses the DPM model. The erosion model during the impact of the fly ash particles on the boiler furnace tubes uses the Oka model.

[0077] It can be understood that the simulation model can be established based on the historical inlet parameters of the boiler and the corresponding historical measured point parameters. After establishing the simulation model, computational fluid dynamics (CFD) simulation can be performed on the simulation model, and the simulation results can be compared with the actual measured point parameters to perform error verification to ensure that the simulation error of the simulation model can be lower than the error value required by the user.

[0078] In this embodiment, the measured point parameter samples and erosion rate samples at different moments are obtained by using the simulation model. Thus, the parameters of each measured point (such as temperature, pressure, flow rate, etc.) and the dynamic change law of the erosion rate over time after the boiler pipeline operates stably can be determined without waiting for detection. This is convenient for quickly sampling sample data and improving the model training efficiency.

[0079] Step 202: Based on the erosion rate samples, time, and measured point parameter samples, establish an erosion rate distribution matrix sample.

[0080] Specifically, the erosion rate distribution matrix is expressed as:

[0081] ;

[0082] ;

[0083] ;

[0084] In the formula, represents the erosion rate distribution vector at the N th moment under the n th working condition parameter, M represents the number of grids on the wall surface of the entire furnace tube, N is the number of working condition parameters, nis the number of moments under specific working conditions.

[0085] Step 203: Perform modal decomposition on the erosion rate distribution matrix sample to determine the erosion rate distribution modes and the erosion rate distribution mode coefficient samples.

[0086] Among them, the erosion rate distribution mode is used to represent the main patterns or characteristics of the erosion rate distribution. The erosion rate distribution mode coefficient is used to represent the proportion of each mode in the overall erosion rate distribution.

[0087] In this embodiment, the erosion rate distribution matrix usually contains a large number of data points. Directly processing and analyzing these data may face problems such as complex calculations and difficulty in extracting information. Through modal decomposition, the erosion rate distribution matrix is decomposed into the erosion rate distribution modes and the mode coefficient samples, which can achieve data dimensionality reduction. Using fewer modes and the corresponding coefficients to represent the original erosion rate distribution information greatly simplifies the data representation and processing difficulty, thereby improving the accuracy and reliability of the prediction results.

[0088] It is worth mentioning that modal decomposition can be performed through algorithms such as the Proper Orthogonal Decomposition (POD) algorithm and the wavelet decomposition algorithm. The embodiments of the present application do not make specific limitations.

[0089] Exemplarily, the proper orthogonal decomposition can be achieved through two mathematical approaches: singular value decomposition (SVD) and eigenvalue decomposition based on the covariance matrix. Taking singular value decomposition as an example, the erosion rate distribution matrix sample is decomposed into three different matrices through the SVD method, that is, ER = BSV ^ T matrix.

[0090] Among them, B ∈ M×M is the left singular vector matrix, and its column vectors are the POD modes (spatial modes) of POD. The modes of POD are expressed as:

[0091] .

[0092] In the formula, .

[0093] S ∈ M×Nn is the singular value matrix, and the diagonal elements are the singular values, representing the energy contributions of each mode.

[0094] V ∈ M×Nn is the right singular vector matrix, and its column vectors are the time coefficients, representing the variation of the modes over time.

[0095] The modal coefficient vector of the erosion rate distribution generated at a specific working condition and a specific moment after the boiler operates stably is expressed as:

[0096] .

[0097] Step 204: Train a modal coefficient prediction model based on the erosion rate distribution modal coefficient samples, inlet parameter samples, and measuring point parameter samples.

[0098] In this embodiment, the inlet parameter samples obtained by numerical simulation and the parameters obtained by simulation are used as the input of the model, and the erosion rate distribution modal coefficient samples are used as the output of the model. The model is trained through machine learning models such as neural networks and support vector machines, enabling the model to automatically learn the complex mapping relationship between the input parameters and the modal coefficients, capture the implicit physical laws, and then generate a modal coefficient prediction model. For subsequent prediction of the erosion rate distribution under the current operating state, first, the working condition parameters and the temperature, flow rate, and pressure data of the on-site measuring points are input into the trained prediction model to generate the POD modal coefficient vector of the current operating state.

[0099] In one embodiment, step 204 specifically includes:

[0100] Step 204-1: If the modal order of the erosion rate distribution modal coefficient samples is greater than the first quantity threshold, group the erosion rate distribution modal coefficient samples to form multiple coefficient sample sets.

[0101] Among them, the first quantity threshold can be reasonably set according to the accuracy requirements of the model, and no specific limitation is made in the embodiments of the present application.

[0102] Step 204-2: Input the coefficient sample sets and their corresponding inlet parameter samples and measuring point parameter samples into a deep learning model for iterative training to obtain a modal coefficient prediction model for the coefficient sample sets.

[0103] In this embodiment, in order to ensure that the reconstruction result meets the requirements of high precision, usually a relatively large number of modal orders are adopted, and the corresponding number of modal coefficients is also large. In the process of regression prediction by machine learning, using the operating condition parameters as the input of the model and all the modal coefficients as the output usually cannot take into account the prediction errors of each modal coefficient, and the more modal coefficients there are, the more obvious this phenomenon becomes. When the modal order of the modal coefficient samples of the erosion rate distribution is large, group the modal coefficient samples. Use the inlet parameter samples and measurement point parameter samples corresponding to the modal coefficient samples belonging to the same group as the input of the deep learning model to which the group of modal coefficient samples belongs, and make the model output the modal coefficients included in the group. Thus, the dimension of the sample data is reduced by grouped training, so that the amount of data to be processed during model training is reduced, which is convenient for accelerating the training speed, reducing the training time and the consumption of computing resources. At the same time, the grouped data set is relatively small and more targeted, and the model is easier to learn the essential features of the data, rather than overfitting the noise and details in the training data. The model can more effectively extract the features related to the erosion rate distribution, and then take into account the calculation errors of each modal coefficient, which helps to improve the prediction accuracy of the modal coefficients.

[0104] Step 204-3: Based on the inlet parameter samples and measurement point parameter samples corresponding to the coefficient sample set, determine the inlet parameter interval and the measurement point parameter interval.

[0105] Step 204-4: Based on the inlet parameter interval and the measurement point parameter interval, correlate the modal coefficient prediction model.

[0106] In this embodiment, after obtaining the modal coefficient prediction model, correlate the inlet parameter interval and the measurement point parameter interval corresponding to the same coefficient sample set with the modal coefficient prediction model, so as to define the effective range of the model input through the inlet parameter interval and the measurement point parameter interval. In the subsequent prediction process, the system can call the available model for prediction according to the actual working conditions, realize the flexible application of the model in multiple scenarios, and improve the prediction accuracy of the modal coefficients.

[0107] In one embodiment, if the modal order of the modal coefficient samples of the erosion rate distribution is greater than the first quantity threshold, then in step 101, that is, input the target inlet parameters and target measurement point parameters of the boiler pipeline into the modal coefficient prediction model to determine the modal coefficients of the erosion rate distribution of the boiler pipeline at different times, which specifically includes: matching the target inlet parameters and target measurement point parameters with the inlet parameter intervals and measurement point parameter intervals associated with different modal coefficient prediction models respectively to screen the modal coefficient prediction models; inputting the target inlet parameters and target measurement point parameters into the screened modal coefficient prediction model to obtain the modal coefficients of the erosion rate distribution; if the number of the screened modal coefficient prediction models is multiple, combine the multiple modal coefficients of the erosion rate distribution obtained by the multiple modal coefficient prediction models.

[0108] In this embodiment, by matching the target inlet parameters and the target measurement point parameters with the inlet parameter intervals and the measurement point parameter intervals of different modal coefficient prediction models, the modal coefficient prediction model suitable for the current target parameters is dynamically screened. Different modal coefficient prediction models may have better prediction performance within different parameter intervals. Thus, the model can be made to fit the actual working conditions, enabling the system to adapt to different parameter situations and improving the prediction accuracy of the erosion rate distribution modal coefficients under different working conditions. This avoids the problem of large prediction errors caused by the limitations of a fixed model itself or its inadaptability to a specific parameter range, thereby improving the performance and reliability of the entire prediction system.

[0109] Exemplarily, as Figure 2 shown, the POD modal coefficients used as samples are divided into L groups, and each group of samples corresponds to a machine learning model. During the process of training the modal coefficient prediction model, the inlet parameter samples and the measurement point parameter samples are iterated for the machine learning models corresponding to the respective POD modal coefficients to form a proprietary model for this group of data. During the subsequent prediction process, the target inlet parameters and the target measurement point parameters are input into the relevant modal coefficient prediction models to obtain L groups of predicted POD modal coefficients. Finally, the POD modal coefficients of each group are combined to form the predicted modal coefficient vector under the current working conditions.

[0110] Step 102: Perform a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution mode of the boiler pipeline to determine the target erosion rate distribution matrix.

[0111] Specifically, the current working condition erosion rate distribution matrix after linear combination reconstruction is:

[0112] ;

[0113] In the formula, represents the vector of erosion rate distribution modal coefficients predicted by the model, represents the erosion rate distribution mode.

[0114] In one embodiment, step 102, that is, performing a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution mode of the boiler pipeline to determine the target erosion rate distribution matrix, specifically includes the following steps:

[0115] Step 102-1: If the modal order of the erosion rate distribution modal coefficients is greater than the second quantity threshold, determine the minimum modal order that meets the error requirement.

[0116] In an actual application scenario, step 102-1 specifically includes: based on the modal order of the erosion rate distribution modal coefficient, linearly combining the erosion rate distribution modal coefficient and the erosion rate distribution mode to determine the first erosion rate distribution matrix; based on a preset modal order, linearly combining the erosion rate distribution modal coefficient and the erosion rate distribution mode to determine the second erosion rate distribution matrix; comparing the first erosion rate distribution matrix and the second erosion rate distribution matrix, and calculating the reconstruction error value; if the reconstruction error value is less than or equal to the error threshold, taking the preset modal order as the minimum modal order; if the reconstruction error value is greater than the error threshold, increasing the preset modal order based on a preset step size until the reconstruction error value is less than or equal to the error threshold.

[0117] In this embodiment, after performing modal decomposition on the erosion rate distribution matrix sample, the first erosion rate distribution matrix is reconstructed first through the modal order of the erosion rate distribution modal coefficient obtained by simulation, and the second erosion rate distribution matrix is reconstructed with a relatively small preset modal order. Then, the reconstructed second erosion rate distribution matrix is compared with the first erosion rate distribution matrix. If the relative error between the two is not greater than the acceptable relative error threshold, the order of the mode is increased to reconstruct the erosion rate distribution again until the error is less than ξ. When the relative error between the two is less than the relative error threshold, the modal order at this time is the modal order used in the subsequent reconstruction and prediction processes. Thus, by continuously comparing the relative error between the first erosion rate distribution matrix and the second erosion rate distribution matrix reconstructed with different modal orders, and gradually adjusting the modal order. While reducing the eigencomplexity of the matrix and the computational amount of the system, it can ensure that the finally determined modal order can make the reconstructed erosion rate distribution matrix have a high similarity with the matrix obtained by simulation, guarantee the reconstruction accuracy, achieve the balance between accuracy and computational efficiency, and enhance the scene adaptability of safety monitoring.

[0118] Step 102-2: Based on the minimum modal order, linearly combine the erosion rate distribution modal coefficient and the erosion rate distribution mode to determine the target erosion rate distribution matrix.

[0119] In this embodiment, the distribution matrix is reconstructed with the minimum modal order that meets the error requirements. Unnecessary high-order modes can be removed, reducing the modal order of the original data and the number of modes participating in the calculation. When linearly combining the erosion rate distribution modal coefficient and the mode, the storage and memory resources required for calculation are greatly reduced, improving the computational efficiency, especially in scenarios involving large-scale data or high real-time requirements. Moreover, it can also make the target erosion rate distribution matrix more clearly show the relationship between the erosion rate distribution and each mode, reduce the noise and interference in the data, and improve the quality and usability of the data.

[0120] Step 103: Calculate the second thickness distribution matrix at the current moment based on the first thickness distribution matrix and the target erosion rate distribution matrix of the boiler pipeline obtained in the previous sampling period.

[0121] Specifically, for the prediction of the real-time thickness of the boiler pipeline, the calculation is carried out through the following formula:

[0122] ;

[0123] In the formula, is the real-time thickness distribution of the boiler pipeline, unit: m; is the thickness distribution of the boiler pipeline at the end of the previous sampling period, unit: m; is the erosion rate predicted by the learning model, unit: kg / (m 2 ·s); is the time step of data acquisition, unit: s; is the density of the boiler pipeline material, unit: kg / m 3 .

[0124] Step 104: If the thickness at any measurement point position in the second thickness distribution matrix is less than or equal to the burst pipe thickness at any measurement point position, output a warning message.

[0125] In the embodiment of the present application, with the target inlet parameters and target measurement point parameters of the boiler pipeline as inputs, the modal coefficients of the erosion rate distribution of the boiler pipeline at different moments are predicted through a trained modal coefficient prediction model. The modal coefficients of the erosion rate distribution and the erosion rate distribution mode of the boiler pipeline are linearly combined to generate a target erosion rate distribution matrix containing the possible erosion rates at different measurement point positions of the boiler pipeline at different moments. Using the known first thickness distribution matrix, the time step of data acquisition, and the erosion rate predicted by the model in the target erosion rate distribution matrix, calculate the second thickness distribution matrix at the current moment for real-time display. And monitor and give early warnings to the wall thickness at different measurement point positions in the second thickness distribution matrix. Thus, through the combination of modal decomposition and machine learning, the erosion rate distribution of the whole-field boiler tubes can be calculated in real time according to the on-site working condition parameters and on-site measurement point parameters, which can fully consider the influence of the time series change during the stable operation of the boiler on the erosion rate distribution, and realize the accurate prediction of the erosion rates at different measurement point positions and different moments of the boiler pipeline. And it can reflect the change of the thickness of the boiler pipeline in real time, send out a warning signal when a burst pipe may occur for further on-site inspection and taking early measures, realizing timely pipeline safety warning, effectively avoiding safety accidents such as leakage and explosion caused by pipeline erosion, reducing energy loss and system failures caused by pipeline erosion, and ensuring the safety of personnel and boiler operation.

[0126] In one embodiment, the safety monitoring method for boiler pipelines further includes: if the maximum value of the erosion rate in the target erosion rate distribution matrix is greater than the erosion rate threshold, calculating the extreme values of operating parameters through the maximum erosion rate optimization algorithm; and adjusting the boiler operating parameters based on the extreme values of operating parameters.

[0127] Among them, the maximum value of the erosion rate corresponding to the extreme values of operating parameters is less than or equal to the erosion rate threshold and meets the load requirements of the boiler pipelines.

[0128] Specifically, the maximum erosion rate optimization algorithm includes, but is not limited to, particle swarm optimization algorithm, genetic algorithm, ant colony algorithm, etc.

[0129] In this embodiment, after obtaining the boiler erosion rate distribution, if the maximum value of the erosion rate is greater than the acceptable erosion rate threshold. The maximum erosion rate optimization algorithm is used to obtain the extreme values of operating parameters that can minimize the maximum erosion rate of the boiler furnace tubes while ensuring the boiler load remains unchanged. Then, the boiler operating parameters are adjusted according to the calculated extreme values of operating parameters. This not only meets the demand for heat energy in the production process but also can optimize the operating parameters to reduce the erosion degree of the furnace tubes, avoid potential safety hazards caused by excessive erosion, extend the service life of the boiler furnace tubes, and ensure the safe and stable operation of the boiler.

[0130] It should be noted that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0131] Furthermore, as Figure 3 shown, as a specific implementation of the above safety monitoring method for boiler pipelines, an embodiment of the present application provides a safety monitoring device 300 for boiler pipelines. The safety monitoring device 300 for boiler pipelines includes: an erosion rate prediction module 301, a thickness prediction module 302, and an early warning module 303.

[0132] Among them, the erosion rate prediction module 301 is configured to input the target inlet parameters and target measurement point parameters of the boiler pipeline into the modal coefficient prediction model to determine the erosion rate distribution modal coefficients of the boiler pipeline at different times, where the modal coefficient prediction model is trained based on the sample data at different times of the boiler pipeline, and the sample data includes the corresponding erosion rate distribution modal coefficient samples, inlet parameter samples, and measurement point parameter samples of the boiler pipeline; and perform a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution mode of the boiler pipeline to determine the target erosion rate distribution matrix.

[0133] The thickness prediction module 302 is configured to calculate the second thickness distribution matrix at the current moment based on the first thickness distribution matrix of the boiler pipeline obtained in the previous sampling period and the target erosion rate distribution matrix.

[0134] An early warning module 303, configured to output an early warning message if the thickness at any measurement point position in the second thickness distribution matrix is less than or equal to the pipe burst thickness at any measurement point position.

[0135] Furthermore, the safety monitoring device 300 for boiler pipes further includes:

[0136] A sample acquisition module (not shown in the figure), configured to calculate sample measurement point parameters and sample erosion rate at different moments after the boiler pipes are stabilized under the inlet parameter samples based on the simulation model of the boiler pipes; and, based on the sample erosion rate, time, and sample measurement point parameters, establish a sample erosion rate distribution matrix; and perform modal decomposition on the sample erosion rate distribution matrix to determine the erosion rate distribution mode and the sample erosion rate distribution mode coefficients;

[0137] A training module (not shown in the figure), configured to train a modal coefficient prediction model based on the sample erosion rate distribution mode coefficients, inlet parameter samples, and measurement point parameter samples.

[0138] Furthermore, the safety monitoring device 300 for boiler pipes further includes:

[0139] A grouping module (not shown in the figure), configured to group the sample erosion rate distribution mode coefficients to form multiple coefficient sample sets if the modal order of the sample erosion rate distribution mode coefficients is greater than the first quantity threshold;

[0140] The training module is specifically configured to input the coefficient sample sets and their corresponding inlet parameter samples and measurement point parameter samples into a deep learning model for iterative training to obtain a modal coefficient prediction model for the coefficient sample sets;

[0141] An association module (not shown in the figure), configured to determine an inlet parameter interval and a measurement point parameter interval based on the inlet parameter samples and measurement point parameter samples corresponding to the coefficient sample sets; and, based on the inlet parameter interval and the measurement point parameter interval, associate the modal coefficient prediction model.

[0142] Furthermore, the safety monitoring device 300 for boiler pipes further includes:

[0143] A model matching module (not shown in the figure), configured to match the target inlet parameters and target measurement point parameters with the inlet parameter intervals and measurement point parameter intervals associated with different modal coefficient prediction models respectively to screen the modal coefficient prediction models;

[0144] The erosion rate prediction module 301 is specifically configured to input the target inlet parameters and target measurement point parameters into the screened modal coefficient prediction model to determine the erosion rate distribution mode coefficients; and, if the number of the screened modal coefficient prediction models is multiple, combine the multiple erosion rate distribution mode coefficients obtained by the multiple modal coefficient prediction models.

[0145] Further, the safety monitoring device 300 for boiler pipes further includes:

[0146] A modal order updating module (not shown in the figure), configured to determine the minimum modal order that meets the error requirement if the modal order of the erosion rate distribution modal coefficient is greater than the second quantity threshold;

[0147] An erosion rate prediction module 301, specifically configured to perform a linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode based on the minimum modal order to determine a target erosion rate distribution matrix.

[0148] Further, the modal order updating module is specifically configured to perform a linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode based on the modal order of the erosion rate distribution modal coefficient to determine a first erosion rate distribution matrix; perform a linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode based on a preset modal order to determine a second erosion rate distribution matrix; compare the first erosion rate distribution matrix and the second erosion rate distribution matrix, and calculate a reconstruction error value; if the reconstruction error value is less than or equal to the error threshold, use the preset modal order as the minimum modal order; if the reconstruction error value is greater than the error threshold, increase the preset modal order based on a preset step size until the reconstruction error value is less than or equal to the error threshold.

[0149] Further, the safety monitoring device 300 for boiler pipes further includes:

[0150] A control module (not shown in the figure), configured to calculate the extreme value of the operating condition parameters through a maximum erosion rate optimization algorithm if the maximum value of the erosion rate in the target erosion rate distribution matrix is greater than the erosion rate threshold, where the maximum value of the erosion rate corresponding to the extreme value of the operating condition parameters is less than or equal to the erosion rate threshold and meets the load requirement of the boiler pipes; adjust the boiler operating parameters based on the extreme value of the operating condition parameters.

[0151] For the specific limitations on the safety monitoring device for boiler pipes, reference can be made to the limitations on the safety monitoring method for boiler pipes in the foregoing text, which will not be elaborated here. Each module in the above-mentioned safety monitoring device for boiler pipes can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned modules can be embedded in the processor in the computer device in the form of hardware or be independent of it, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

[0152] Based on the method as described above Figure 1 shown, correspondingly, an embodiment of the present application further provides a readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the safety monitoring method for boiler pipes as described above Figure 1 shown.

[0153] Based on such understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, a USB flash drive, a mobile hard disk, etc.), and includes several instructions for causing a computer device (such as a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present application.

[0154] Based on the above-mentioned Figure 1 method, as well as Figure 3 the virtual device embodiment shown, in order to achieve the above object, as Figure 4 shown, an embodiment of the present application further provides a computer device. The computer device 400 includes a processor 401 and a memory 402. A program or instruction that can run on the processor 401 is stored on the memory 402. When the program or instruction is executed by the processor 401, it implements the safety monitoring method of the boiler pipeline as Figure 1 shown above.

[0155] The memory 402 can be used to store software programs and various data. The memory 402 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data. Among them, the first storage area can store an operating system, application programs or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 402 can include volatile memory or non-volatile memory, or the memory 402 can include both volatile and non-volatile memory. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDR SDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synch link dynamic random access memory (SLDRAM), and a direct rambus random access memory (DRRAM). The memory 402 in the embodiments of the present application includes but is not limited to these and any other suitable types of memory.

[0156] The processor 401 may include one or more processing units; optionally, the processor 401 integrates an application processor and a modem processor. Among them, the application processor mainly processes operations related to the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communication signals, such as a baseband processor. It can be understood that the above modem processor may not be integrated into the processor 401 either.

[0157] The computer device can specifically be a personal computer, a server, a network device, etc.

[0158] Optionally, the computer device may further include a user interface, a network interface, a camera, a Radio Frequency (RF) circuit, sensors, an audio circuit, a WI-FI module, and so on. The user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), etc. Optionally, the user interface may further include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a Bluetooth interface, a WI-FI interface), etc.

[0159] Those skilled in the art can understand that the structure of a computer device provided in this embodiment does not limit the computer device, and it may include more or fewer components, or combine certain components, or have different component arrangements.

[0160] Through the description of the above embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus a necessary general hardware platform, or can also be implemented by hardware. Input the target inlet parameters and target measuring point parameters of the boiler pipeline into the modal coefficient prediction model to determine the erosion rate distribution modal coefficients of the boiler pipeline at different times. Among them, the modal coefficient prediction model is trained based on the sample data of the boiler pipeline at different times, and the sample data includes the erosion rate distribution modal coefficient samples, inlet parameter samples, and measuring point parameter samples corresponding to the boiler pipeline; perform a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution of the boiler pipeline to determine the target erosion rate distribution matrix; based on the first thickness distribution matrix and the target erosion rate distribution matrix of the boiler pipeline obtained in the previous sampling period, calculate the second thickness distribution matrix at the current time; if the thickness at any measuring point position in the second thickness distribution matrix is less than or equal to the burst pipe thickness at any measuring point position, output a warning message. In the embodiment of this application, through the combination of modal decomposition and machine learning, the erosion rate distribution of the entire field of boiler tubes is calculated in real time according to the on-site working condition parameters and on-site measuring point parameters, which can fully consider the influence of the time-series change during the stable operation of the boiler on the erosion rate distribution, and realize the accurate prediction of the erosion rates at different measuring point positions and different times of the boiler pipeline. And it can reflect the change of the thickness of the boiler pipeline in real time, send a warning signal when a burst pipe may occur for further on-site inspection and taking measures in advance, achieve timely pipeline safety warning, effectively avoid safety accidents such as leakage and explosion caused by pipeline erosion, reduce energy loss and system failures caused by pipeline erosion, and ensure the safety of personnel and boiler operation.

[0161] Those skilled in the art can understand that the attached drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the attached drawings are not necessarily essential for implementing the present application. Those skilled in the art can understand that the modules in the devices in the implementation scenario can be distributed in the devices in the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed and located in one or more devices different from this implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0162] The above serial numbers of the present application are only for description and do not represent the advantages or disadvantages of the implementation scenarios. The above-disclosed are only several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present application.

Claims

1. A safety monitoring method for boiler pipelines, characterized in that, Including: Performing modal decomposition on the erosion rate distribution matrix sample to determine the erosion rate distribution modal coefficient sample, where the erosion rate distribution matrix sample includes the measured point parameter sample and the erosion rate sample at different moments after the boiler pipeline stabilizes under the inlet parameter sample; If the modal order of the erosion rate distribution modal coefficient sample is greater than the first quantity threshold, grouping the erosion rate distribution modal coefficient sample to form multiple coefficient sample sets; Inputting the coefficient sample set, its corresponding inlet parameter sample, and the measured point parameter sample into a deep learning model for iterative training to obtain a modal coefficient prediction model associated with the inlet parameter interval and the measured point parameter interval corresponding to the coefficient sample set; Inputting the target inlet parameter and the target measured point parameter of the boiler pipeline into the modal coefficient prediction model of the corresponding parameter interval to determine the erosion rate distribution modal coefficient of the boiler pipeline at different moments, where the inlet parameter is the working condition parameter at the inlet of the boiler pipeline, and the measured point parameter is the working condition parameter at different measured point positions of the boiler pipeline; Performing a linear combination of the erosion rate distribution modal coefficient and the erosion rate distribution mode of the boiler pipeline to determine the target erosion rate distribution matrix; Calculating the second thickness distribution matrix at the current moment based on the first thickness distribution matrix of the boiler pipeline obtained in the previous sampling period and the target erosion rate distribution matrix; If the thickness at any measured point position in the second thickness distribution matrix is less than or equal to the burst pipe thickness at any measured point position, outputting a warning message.

2. The safety monitoring method for boiler pipelines according to claim 1, characterized in that, The safety monitoring method of the boiler pipeline further includes: Calculating the measured point parameter sample and the erosion rate sample at different moments after the boiler pipeline stabilizes under the inlet parameter sample based on the simulation model of the boiler pipeline; Based on the erosion rate sample, the moment, and the measured point parameter sample, establishing the erosion rate distribution matrix sample; performing modal decomposition on the erosion rate distribution matrix sample to determine the erosion rate distribution mode.

3. The safety monitoring method for boiler pipelines according to claim 1, characterized in that, The method further includes: Determining the inlet parameter interval and the measured point parameter interval based on the coefficient sample set corresponding to the inlet parameter sample and the measured point parameter sample; Associating the modal coefficient prediction model based on the inlet parameter interval and the measured point parameter interval.

4. The safety monitoring method for boiler pipelines according to claim 1, characterized in that, The step of inputting the target inlet parameter and the target measured point parameter of the boiler pipeline into the modal coefficient prediction model of the corresponding parameter interval to determine the erosion rate distribution modal coefficient of the boiler pipeline at different moments includes: Matching the target inlet parameter and the target measured point parameter with the inlet parameter interval and the measured point parameter interval associated with different modal coefficient prediction models respectively to screen the modal coefficient prediction model; Inputting the target inlet parameter and the target measured point parameter into the screened modal coefficient prediction model to obtain the erosion rate distribution modal coefficient; If the number of the screened modal coefficient prediction models is multiple, combining the multiple erosion rate distribution modal coefficients obtained by the multiple modal coefficient prediction models.

5. The safety monitoring method for boiler pipes according to claim 1, characterized in that, Performing a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution mode of the boiler pipeline to determine a target erosion rate distribution matrix includes: If the modal order of the erosion rate distribution modal coefficients is greater than a second quantity threshold, determining the minimum modal order that meets the error requirement; Based on the minimum modal order, performing a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution mode to determine the target erosion rate distribution matrix.

6. The safety monitoring method for boiler pipelines according to claim 5, characterized in that, The determining the minimum modal order that meets the error requirement includes: Based on the modal order of the erosion rate distribution modal coefficients, performing a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution mode to determine a first erosion rate distribution matrix; Based on a preset modal order, performing a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution mode to determine a second erosion rate distribution matrix; Comparing the first erosion rate distribution matrix and the second erosion rate distribution matrix, and calculating a reconstruction error value; If the reconstruction error value is less than or equal to an error threshold, taking the preset modal order as the minimum modal order; If the reconstruction error value is greater than the error threshold, increasing the preset modal order based on a preset step size until the reconstruction error value is less than or equal to the error threshold.

7. The safety monitoring method for boiler pipes according to any one of claims 1 to 6, characterized in that, The safety monitoring method for the boiler pipeline further includes: If the maximum value of the erosion rate in the target erosion rate distribution matrix is greater than an erosion rate threshold, calculating the extreme value of the operating condition parameters through a maximum erosion rate optimization algorithm, where the maximum value of the erosion rate corresponding to the extreme value of the operating condition parameters is less than or equal to the erosion rate threshold and meets the load requirement of the boiler pipeline; Adjusting the boiler operating parameters based on the extreme value of the operating condition parameters.

8. A safety monitoring device for boiler pipes, characterized in that, It includes: A sample acquisition module, configured to perform modal decomposition on an erosion rate distribution matrix sample to determine an erosion rate distribution modal coefficient sample, where the erosion rate distribution matrix sample includes measured point parameter samples and erosion rate samples at different moments after the boiler pipeline is stabilized under an inlet parameter sample; A grouping module, configured to group the erosion rate distribution modal coefficient samples to form a plurality of coefficient sample sets if the modal order of the erosion rate distribution modal coefficient samples is greater than a first quantity threshold; A training module, configured to input the coefficient sample sets and their corresponding inlet parameter samples and measured point parameter samples into a deep learning model for iterative training to obtain a modal coefficient prediction model associated with the inlet parameter interval and the measured point parameter interval corresponding to the coefficient sample sets; An erosion rate prediction module, configured to input the target inlet parameters and target measured point parameters of the boiler pipeline into the modal coefficient prediction model of the corresponding parameter interval to determine the erosion rate distribution modal coefficients of the boiler pipeline at different moments, where the inlet parameters are the operating condition parameters at the inlet of the boiler pipeline, and the measured point parameters are the operating condition parameters at different measured point positions of the boiler pipeline; and Performing a linear combination of the erosion rate distribution modal coefficients and the erosion rate distribution mode of the boiler pipeline to determine a target erosion rate distribution matrix; A thickness prediction module, configured to calculate a second thickness distribution matrix at the current moment based on a first thickness distribution matrix of a boiler pipeline obtained in the previous sampling period and the target erosion rate distribution matrix; An early warning module, configured to output an early warning message if the thickness at any measuring point position in the second thickness distribution matrix is less than or equal to the pipe burst thickness at the any measuring point position.

9. A readable storage medium having a program or instructions stored thereon, characterized in that, When the program or instruction is executed by a processor, the steps of the safety monitoring method for a boiler pipeline as described in any one of claims 1 to 7 are implemented.

10. A computer device, comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the program, the safety monitoring method for a boiler pipeline as described in any one of claims 1 to 7 is implemented.

Images