Pipeline displacement stress prediction method and device, storage medium and equipment

Abstract

The application discloses a pipeline displacement stress prediction method and device, a storage medium and equipment, relates to the technical field of pipeline state monitoring, takes into account full-pipeline system coverage, low-cost deployment and real-time response, and realizes high-precision prediction of pipeline displacement stress. The method comprises the following steps: acquiring historical temperature, historical pressure, measured historical displacement and measured historical stress of a preset node measured on site; calculating simulated historical displacement and simulated historical stress of the preset node of the full-pipeline system under the historical temperature and the historical pressure based on a pipeline model; training an initial pipeline displacement stress prediction model according to the historical temperature, the historical pressure, the simulated historical displacement, the simulated historical stress, the measured historical displacement and the measured historical stress; if the training is not completed, a preset adjustment strategy is executed until the pipeline displacement stress prediction model is obtained; and collecting the temperature and the pressure of pipeline operation in real time, and predicting the displacement and the stress of the preset node of the full-pipeline system according to the pipeline displacement stress prediction model.

Description

Technical Field

[0001] This invention relates to the field of pipeline condition monitoring technology, and in particular to a method, device, storage medium and equipment for predicting pipeline displacement stress. Background Technology

[0002] In thermal power generation systems, the four major pipelines, as core components of energy transmission, have long been operating in harsh environments characterized by high temperatures, high pressures, and continuous vibrations. With the significant increase in the demand for deep peak-shaving capacity in thermal power units, the thermal and mechanical loads borne by these pipelines fluctuate frequently, leading to increasingly complex load conditions. If abnormal displacement or stress accumulation exceeds allowable values, pipelines are prone to weld cracking, pipeline fractures, and even equipment impact damage, seriously threatening the safe and stable operation of the power plant.

[0003] Existing methods for predicting pipeline displacement stress have significant drawbacks: 1. Measurement-based methods are costly and have significant limitations: deploying sensors at numerous points along the pipeline results in high costs due to the sheer number of sensors and the associated wiring expenses, while deploying sensors at a limited number of points makes it difficult to cover the entire piping system (such as complex bends and manifold areas), creating blind spots; 2. Existing prediction models have narrow coverage: due to the inability to obtain sensor data covering the entire piping system, existing prediction models have limited training samples, being limited to sensor data from single pipe bends or localized locations, making it difficult to accurately predict pipeline displacement stress at multiple nodes throughout the entire piping system. Summary of the Invention

[0004] In view of this, the present invention provides a method, device, storage medium and equipment for predicting pipeline displacement stress, which takes into account full pipeline system coverage, low-cost deployment and real-time response, and realizes high-precision prediction of pipeline displacement stress.

[0005] According to one aspect of the present invention, a method for predicting pipeline displacement stress is provided, the method comprising:

[0006] The historical temperature of the pipeline during operation, the historical pressure of the pipeline during operation, the measured historical displacement of the preset nodes in the field, and the measured historical stress of the preset nodes in the field are obtained.

[0007] Based on the constructed pipeline model, the simulated historical displacement and simulated historical stress of the preset nodes of the entire pipeline system are calculated according to the historical temperature and the historical pressure.

[0008] Using the historical temperature, historical pressure, preset nodes of the entire piping system, simulated historical displacement, and simulated historical stress, an initial pipeline displacement stress prediction model is trained until the initial pipeline displacement stress prediction model converges, resulting in a pipeline displacement stress prediction model to be verified. Based on the historical temperature, historical pressure, preset nodes of the entire piping system, simulated historical displacement, simulated historical stress, measured historical displacement, and measured historical stress, it is determined whether the pipeline displacement stress prediction model to be verified has completed training. If not, a preset adjustment strategy is executed until a fully trained pipeline displacement stress prediction model is obtained.

[0009] The temperature and pressure of the pipeline are collected in real time. The time series data of the temperature and pressure, along with the preset nodes of the entire pipeline system, are input into the pipeline displacement and stress prediction model to predict the displacement and stress of the preset nodes of the entire pipeline system.

[0010] Preferably, the method further includes:

[0011] The system acquires preset displacement safety range, displacement warning range, and displacement danger range. If the displacement is within the displacement safety range, it displays the displacement safety of the preset nodes of the entire pipeline system corresponding to the displacement. If the displacement is within the displacement warning range, it displays the displacement warning of the preset nodes of the entire pipeline system corresponding to the displacement. If the displacement is within the displacement danger range, it displays the displacement alarm of the preset nodes of the entire pipeline system corresponding to the displacement, so as to perform online displacement monitoring.

[0012] The system acquires preset stress safety range, stress warning range, and stress danger range. If the stress is within the stress safety range, it displays the stress safety of the preset nodes in the entire piping system corresponding to the stress. If the stress is within the stress warning range, it displays the stress warning of the preset nodes in the entire piping system corresponding to the stress. If the stress is within the stress danger range, it displays the stress alarm of the preset nodes in the entire piping system corresponding to the stress, thereby enabling online stress monitoring.

[0013] Preferably, the acquisition of the historical temperature of the pipeline, the historical pressure of the pipeline, the measured historical displacement of the preset node in the field, and the measured historical stress of the preset node in the field includes:

[0014] Based on a temperature sensor, the historical operating temperature of the pipeline at the reserved pipeline interface is collected; based on a pressure sensor, the historical operating pressure of the pipeline at the reserved pipeline interface is collected.

[0015] The measured historical displacement of preset nodes on site is measured using a binocular stereo vision displacement measurement device.

[0016] Based on the vibrating wire surface strain gauges deployed at the preset nodes in the field, the vibration frequency of the pipeline is collected, and the data acquisition instrument inside the vibrating wire surface strain gauge is used to read the values ​​to obtain the surface temperature of the pipeline at the location of the vibrating wire surface strain gauge.

[0017] Obtain preset calculation parameters, and calculate the measured historical stress of the preset node in the field based on the pipeline vibration frequency, the pipeline surface temperature and the preset calculation parameters. The preset calculation parameters include at least the reference frequency, reference temperature, material elastic modulus, linear expansion coefficient, temperature correction coefficient and strain gauge sensitivity coefficient.

[0018] Preferably, the step of training an initial pipeline displacement stress prediction model using the historical temperature, the historical pressure, the preset nodes of the entire pipeline system, the simulated historical displacement, and the simulated historical stress, until the initial pipeline displacement stress prediction model converges, to obtain the pipeline displacement stress prediction model to be verified, includes:

[0019] Based on the historical temperature and the historical pressure, different historical time series data are determined according to different continuous historical moments. Training historical time series data is divided from all the historical time series data. The training historical time series data and the preset nodes of the entire pipeline system are input into the initial pipeline displacement stress prediction model to obtain the first predicted historical displacement and the first predicted historical stress of the preset nodes of the entire pipeline system output by the initial pipeline displacement stress prediction model.

[0020] The simulated historical displacement corresponding to the training historical time series data is used as the training simulated historical displacement. The first loss function value is calculated based on the training simulated historical displacement of the same preset node of the entire pipeline system and the first predicted historical displacement. The simulated historical displacement of the last historical moment of the training historical time series data is the simulated historical displacement corresponding to the training historical time series data.

[0021] The simulated historical stress corresponding to the training historical time series data is used as the training simulated historical stress. The second loss function value is calculated based on the training simulated historical stress of the same preset node of the entire pipeline system and the first predicted historical stress. The simulated historical stress at the last historical moment of the training historical time series data is the simulated historical stress corresponding to the training historical time series data.

[0022] If at least one of the first loss function values ​​is greater than or equal to a preset displacement threshold, or if at least one of the second loss function values ​​is greater than or equal to a preset stress threshold, then it is determined that the initial pipeline displacement stress prediction model has not converged. The parameters of the initial pipeline displacement stress prediction model are adjusted until the initial pipeline displacement stress prediction model converges, and the pipeline displacement stress prediction model to be verified is obtained.

[0023] Preferably, the step of determining whether the pipeline displacement-stress prediction model to be verified has completed training based on the historical temperature, the historical pressure, the preset nodes of the entire pipeline system, the simulated historical displacement, the simulated historical stress, the measured historical displacement, and the measured historical stress includes:

[0024] The verification historical time series data is divided from all the historical time series data. The verification historical time series data and the preset nodes of the entire pipeline system are input into the pipeline displacement stress prediction model to be verified. The second predicted historical displacement and the second predicted historical stress of the preset nodes of the entire pipeline system are output by the pipeline displacement stress prediction model to be verified.

[0025] The simulated historical displacement corresponding to the verification historical time series data is used as the verification simulated historical displacement. A first displacement deviation rate is calculated based on the verification simulated historical displacement and the second predicted historical displacement of the same preset node of the entire pipeline system. The simulated historical stress corresponding to the verification historical time series data is used as the verification simulated historical stress. A first stress deviation rate is calculated based on the verification simulated historical stress and the second predicted historical stress of the same preset node of the entire pipeline system. A second displacement deviation rate is calculated based on the measured historical displacement and the second predicted historical displacement of the same preset node of the field measurement. A second stress deviation rate is calculated based on the measured historical stress and the second predicted historical stress of the same preset node of the field measurement. The simulated historical displacement at the last historical moment of the verification historical time series data is the simulated historical displacement corresponding to the verification historical time series data, and the simulated historical stress at the last historical moment of the verification historical time series data is the simulated historical stress corresponding to the verification historical time series data.

[0026] The training status of the pipeline displacement stress prediction model to be verified is determined based on the first displacement deviation rate, the first preset displacement threshold, the first stress deviation rate, the first preset stress threshold, the second displacement deviation rate, the second preset displacement threshold, the second stress deviation rate, and the second preset stress threshold.

[0027] Preferably, if not, a preset adjustment strategy is executed until a trained pipeline displacement stress prediction model is obtained, including:

[0028] If at least one of the first displacement deviation rates is greater than or equal to the first preset displacement threshold, or at least one of the first stress deviation rates is greater than or equal to the first preset stress threshold, or at least one of the second displacement deviation rates is greater than or equal to the second preset displacement threshold, or at least one of the second stress deviation rates is greater than or equal to the second preset stress threshold, then it is determined that the pipeline displacement stress prediction model to be verified has not completed training.

[0029] The preset adjustment strategy is to increase the number of training samples. A new sample set is obtained, and the pipeline displacement stress prediction model to be verified is retrained using the new sample set until the retrained pipeline displacement stress prediction model to be verified is completed, thus obtaining the trained pipeline displacement stress prediction model.

[0030] Preferably, the method further includes:

[0031] If all the first displacement deviation rates are less than the first preset displacement threshold, and all the first stress deviation rates are less than the first preset stress threshold, and all the second displacement deviation rates are less than the second preset displacement threshold, and all the second stress deviation rates are less than the second preset stress threshold, then the pipeline displacement stress prediction model to be verified has completed training, and a trained pipeline displacement stress prediction model has been obtained.

[0032] According to another aspect of the present invention, a pipe displacement stress prediction device is provided, the device comprising:

[0033] The acquisition module is used to acquire the historical temperature of the pipeline during operation, the historical pressure of the pipeline during operation, the measured historical displacement of the preset nodes in the field, and the measured historical stress of the preset nodes in the field.

[0034] The calculation module is used to calculate the simulated historical displacement and simulated historical stress of the preset nodes of the entire pipeline system based on the constructed pipeline model and according to the historical temperature and the historical pressure.

[0035] The first training module is used to train an initial pipeline displacement stress prediction model using the historical temperature, the historical pressure, the preset nodes of the entire piping system, the simulated historical displacement, and the simulated historical stress, until the initial pipeline displacement stress prediction model converges to obtain a pipeline displacement stress prediction model to be verified. Based on the historical temperature, the historical pressure, the preset nodes of the entire piping system, the simulated historical displacement, the simulated historical stress, the measured historical displacement, and the measured historical stress, it is determined whether the pipeline displacement stress prediction model to be verified has completed training. If not, a preset adjustment strategy is executed until a fully trained pipeline displacement stress prediction model is obtained.

[0036] The prediction module is used to collect the temperature and pressure of the pipeline in real time, input the time series data of the temperature and pressure, and the preset nodes of the entire pipeline system into the pipeline displacement stress prediction model, and predict the displacement and stress of the preset nodes of the entire pipeline system.

[0037] According to another aspect of the present invention, a storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the above-described pipeline displacement stress prediction method.

[0038] According to another aspect of the present invention, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and running on the processor, wherein the processor executes the program to implement the above-described pipeline displacement stress prediction method.

[0039] By employing the above technical solutions, this invention provides a method, apparatus, storage medium, and device for predicting pipeline displacement stress. Through this invention's technical solution, traditional pipeline displacement stress analysis, based on finite element analysis physical modeling, requires iterative calculation of mechanical equations, consuming a significant amount of time. Under the dynamic operating conditions of deep peak shaving in thermal power units, this speed cannot meet the safety early warning requirements of second-level response. Furthermore, the more sensors deployed, the higher the cost, and some pipeline locations cannot be equipped with sensors. Therefore, only a small number of sensors can be deployed on-site, resulting in a small amount of data and low accuracy during model training. This invention, based on historical temperature and pressure, obtains the measured historical displacement and pressure of each pre-defined node through field measurements. Using the same historical temperature and pressure, and through a pipeline model, it obtains the simulated historical displacement and stress of each pre-defined node in the entire piping system. The advantage of measured historical displacement and pressure is their data accuracy. Since the number of pre-defined nodes in the entire piping system is greater than the number of pre-defined nodes measured in the field, the advantage of simulated historical displacement and stress is their large data volume. A portion of the simulated historical displacement and stress is used to train an initial pipeline displacement and stress prediction model, resulting in a pipeline displacement and stress prediction model to be validated. The accuracy of this model is then validated in two ways: using measured historical displacement and pressure, and using the remaining simulated historical displacement and stress. If at least one validation fails, a pre-defined adjustment strategy is executed until both validations pass, resulting in a highly accurate pipeline displacement and stress prediction model. This pipeline displacement and stress prediction model does not require repeated iterations of the mechanical equations and has a fast response. Therefore, it takes into account full pipeline coverage, low-cost deployment, and real-time response, and achieves high-precision prediction of pipeline displacement stress.

[0040] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0041] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of this application. In the drawings:

[0042] Figure 1 A schematic flowchart of a pipeline displacement stress prediction method provided by an embodiment of the present invention is shown;

[0043] Figure 2 A flowchart illustrating another pipeline displacement stress prediction method provided by an embodiment of the present invention is shown.

[0044] Figure 3 A schematic diagram of a pipeline displacement stress prediction device provided in an embodiment of the present invention is shown.

[0045] Figure 4 A schematic diagram of another pipeline displacement stress prediction device provided in an embodiment of the present invention is shown. Detailed Implementation

[0046] The present invention will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0047] This embodiment provides a method for predicting pipeline displacement stress, such as Figure 1 As shown, the method includes:

[0048] 101. Obtain the historical temperature of the pipeline, the historical pressure of the pipeline, the measured historical displacement of the preset node in the field, and the measured historical stress of the preset node in the field.

[0049] In this embodiment, the four main pipelines are the main steam pipeline, the hot reheat steam pipeline, the cold reheat steam pipeline, and the high-pressure feedwater pipeline. For each pipeline, the historical operating temperature, historical operating pressure, measured historical displacement of the preset nodes, and measured historical stress of the preset nodes are obtained. The measured preset nodes indicate which of the four main pipelines they belong to and their location within that pipeline.

[0050] Taking four pipelines, each with three pre-set on-site measurement nodes, as an example:

[0051] The historical temperature 1 and historical pressure 1 at the reserved interface of the main steam pipeline, the historical temperature 2 and historical pressure 2 at the reserved interface of the main steam pipeline, ..., the historical temperature 10 and historical pressure 10 at the reserved interface of the main steam pipeline are measured to form a historical time series data. The measured historical displacement 1 and measured historical stress 1 at the reserved node 1-main steam pipeline at the reserved node 2-main steam pipeline at the reserved node 3-main steam pipeline at the reserved node 4 are measured to obtain the measured historical displacement 3 and measured historical stress 3 at the reserved node 5-main steam pipeline at the reserved node 6. The historical time 1 to the reserved node 10 are 10 consecutive historical time moments.

[0052] At the reserved interface of the hot reheat steam pipeline, the historical temperature 11 and historical pressure 11 at historical time 1, the historical temperature 12 and historical pressure 12 at historical time 2, ..., the historical temperature 20 and historical pressure 20 at historical time 10 are measured to form a historical time series data. The measured historical displacement 4 and measured historical stress 4 at historical time 10 of the preset node 1-hot reheat steam pipeline are obtained on site. The measured historical displacement 5 and measured historical stress 5 at historical time 10 of the preset node 2-hot reheat steam pipeline are obtained on site. The measured historical displacement 6 and measured historical stress 6 at historical time 10 of the preset node 3-hot reheat steam pipeline are obtained on site.

[0053] At the reserved interface of the cold reheat steam pipeline, the historical temperature 21 and historical pressure 21 at historical time 1, the historical temperature 22 and historical pressure 22 at historical time 2, ..., the historical temperature 30 and historical pressure 30 at historical time 10 are measured to form a historical time series data. The measured historical displacement 7 and measured historical stress 7 at historical time 10 of the preset node 1-cold reheat steam pipeline are obtained at the preset node 2-cold reheat steam pipeline at the preset node, the measured historical displacement 8 and measured historical stress 8 at historical time 10 of the preset node, and the measured historical displacement 9 and measured historical stress 9 at historical time 10 of the preset node 3-cold reheat steam pipeline at the preset node.

[0054] At the reserved interface of the high-pressure water supply pipeline, the historical temperature 31 and historical pressure 31 at historical time 1, the historical temperature 32 and historical pressure 32 at historical time 2, ..., the historical temperature 40 and historical pressure 40 at historical time 10 are measured to form a historical time series data. The measured historical displacement 10 and measured historical stress 10 at historical time 10 of the preset node 1 - high-pressure water supply pipeline are obtained on site. The measured historical displacement 11 and measured historical stress 11 at historical time 10 of the preset node 2 - high-pressure water supply pipeline are obtained on site. The measured historical displacement 12 and measured historical stress 12 at historical time 10 of the preset node 3 - high-pressure water supply pipeline are obtained on site.

[0055] 102. Based on the constructed pipeline model, calculate the simulated historical displacement and simulated historical stress of the preset nodes of the entire pipeline system according to the historical temperature and the historical pressure.

[0056] In this embodiment, due to the limited number of pre-defined nodes obtained from field measurements, the amount of measured historical displacement and stress data is insufficient, making accurate model training impossible. Therefore, pipeline models corresponding to the four major pipelines are constructed based on stress analysis software. Multiple pre-defined nodes for the entire pipeline system are set on each pipeline model. Under the same historical operating temperature and pressure conditions as in step 101 of the embodiment, the simulated historical displacement and simulated historical stress of each pre-defined node for the entire pipeline system are simulated. For each pipeline model, the number of all pre-defined nodes for the entire pipeline system is greater than the number of pre-defined nodes obtained from field measurements, thereby obtaining more data and improving the accuracy of model training. Furthermore, for the same pipeline model, the pre-defined nodes obtained from field measurements are included as part of the pre-defined nodes for the entire pipeline system. For example, if the pre-defined nodes for the main steam pipeline are nodes 1 to 10, then the pre-defined nodes for the pipeline model corresponding to the main steam pipeline should include the same nodes 1 to 10, as well as other nodes, such as nodes 11 to 225. Each pre-defined node for the entire pipeline system indicates which of the four major pipelines it belongs to and its location within that pipeline.

[0057] Taking four pipelines, each with a pre-defined total of 225 nodes in its entire piping system, as an example:

[0058] Similar to step 101 of the embodiment, the historical temperature and pressure of the pipeline during operation are obtained: the historical temperature 1 and historical pressure 1 at the reserved interface of the main steam pipeline, the historical temperature 2 and historical pressure 2 at historical time 1, ..., the historical temperature 10 and historical pressure 10 at historical time 10 are obtained and combined to form a historical time series data.

[0059] The simulated historical displacement 1 and simulated historical stress 1 at the preset node 1 of the entire piping system - the main steam pipeline at historical time 10 are obtained; the simulated historical displacement 2 and simulated historical stress 2 at the preset node 2 of the entire piping system - the main steam pipeline at historical time 10 are obtained; ..., the simulated historical displacement 225 and simulated historical stress 225 at the preset node 225 of the entire piping system - the main steam pipeline at historical time 10 are obtained through field measurement. The historical time 1 to historical time 10 are 10 consecutive historical times.

[0060] Similar to step 101 of the embodiment, the historical temperature and pressure of the pipeline during operation are obtained: the historical temperature 11 and historical pressure 11 at the reserved interface of the hot reheat steam pipeline, the historical temperature 12 and historical pressure 12 at historical time 1, ..., the historical temperature 20 and historical pressure 20 at historical time 10 are obtained and combined to form a historical time series data.

[0061] Obtain the simulated historical displacement 226 and simulated historical stress 226 at historical time 10 for the preset node 1 of the entire pipeline system - hot reheat steam pipeline; the simulated historical displacement 227 and simulated historical stress 227 at historical time 10 for the preset node 2 of the entire pipeline system - hot reheat steam pipeline; ..., and the measured historical displacement 450 and measured historical stress 450 at historical time 10 for the preset node 225 of the entire pipeline system - hot reheat steam pipeline.

[0062] Similar to step 101 of the embodiment, the historical temperature and pressure of the pipeline during operation are obtained: the historical temperature 21 and historical pressure 21 at historical time 1, the historical temperature 22 and historical pressure 22 at historical time 2, ..., the historical temperature 30 and historical pressure 30 at historical time 10 are obtained at the reserved interface of the cold reheat steam pipeline, and a historical time series data is formed.

[0063] Obtain the simulated historical displacement 451 and simulated historical stress 451 at historical time 10 for the preset node 1 of the entire pipeline system - cold reheat steam pipeline; the measured historical displacement 452 and measured historical stress 452 at historical time 10 for the preset node 2 of the entire pipeline system - cold reheat steam pipeline; ..., and the measured historical displacement 675 and measured historical stress 675 at historical time 10 for the preset node 3 of the entire pipeline system - cold reheat steam pipeline.

[0064] Similar to step 101 of the embodiment, the historical temperature and historical pressure of the pipeline during operation are obtained: the historical temperature 31 and historical pressure 31 at historical time 1, the historical temperature 32 and historical pressure 32 at historical time 2, ..., the historical temperature 40 and historical pressure 40 at historical time 10 are measured at the reserved interface of the high-pressure water supply pipeline, and a historical time series data is formed.

[0065] Obtain the measured historical displacement 676 and measured historical stress 676 at historical time 10 for the preset node 1 of the entire pipeline system - high-pressure water supply pipeline; the measured historical displacement 677 and measured historical stress 677 at historical time 10 for the preset node 2 of the entire pipeline system - high-pressure water supply pipeline; ..., the measured historical displacement 900 and measured historical stress 900 at historical time 10 for the preset node 3 of the entire pipeline system - high-pressure water supply pipeline.

[0066] 103. Using the historical temperature, historical pressure, preset nodes of the entire piping system, simulated historical displacement, and simulated historical stress, train an initial pipeline displacement stress prediction model until the initial pipeline displacement stress prediction model converges to obtain a pipeline displacement stress prediction model to be verified. Based on the historical temperature, historical pressure, preset nodes of the entire piping system, simulated historical displacement, simulated historical stress, measured historical displacement, and measured historical stress, determine whether the pipeline displacement stress prediction model to be verified has completed training. If not, execute a preset adjustment strategy until a fully trained pipeline displacement stress prediction model is obtained.

[0067] In this embodiment, as one implementation method, the initial pipeline displacement stress prediction model is a Long Short-Term Memory (LSTM) network.

[0068] Using the increased amount of data obtained in step 102 of the embodiment, the initial pipeline displacement stress prediction model is trained until it converges, resulting in the pipeline displacement stress prediction model to be verified. However, the increased amount of data in step 102 of the embodiment is based on pipeline model simulation, not on-site measurement. Therefore, it is necessary to further verify the prediction accuracy of the pipeline displacement stress prediction model to be verified. The verification of its prediction accuracy includes two aspects: firstly, the relationship between its prediction results and the measured historical displacement and measured historical stress obtained from on-site measurement in step 101 of the embodiment; secondly, the relationship between its prediction results and the simulated historical displacement and simulated historical stress obtained from pipeline model simulation in step 102 of the embodiment. If at least one aspect of the verification fails, a preset adjustment strategy is executed until both aspects of the verification pass, resulting in a pipeline displacement stress prediction model with high accuracy.

[0069] 104. Real-time acquisition of pipeline operating temperature and pipeline operating pressure, inputting the time series data of the temperature and pressure, along with the preset nodes of the entire pipeline system, into the pipeline displacement stress prediction model to predict the displacement and stress of the preset nodes of the entire pipeline system.

[0070] In this embodiment, after obtaining a pipeline displacement stress prediction model with high prediction accuracy, the pipeline displacement stress prediction model is applied.

[0071] Taking a main steam pipeline with 225 pre-set nodes as an example:

[0072] Temperature and pressure at time 1, pressure at time 2, ..., and temperature and pressure at time 10 are measured in real time at the reserved interface of the main steam pipeline. The time series data formed by these data, along with the preset nodes 1-main steam pipeline, 2-main steam pipeline, ..., and 225-main steam pipeline of the entire pipeline system, are input into the pipeline displacement stress prediction model to predict the displacement and pressure at time 10 for each of the preset nodes 1-main steam pipeline, 2-main steam pipeline, ..., and 225-main steam pipeline of the entire pipeline system. Here, time 1 to time 10 are continuous times.

[0073] Thus, the displacement and pressure of the main steam pipeline at each preset node of the entire piping system at time 10 were predicted, where time 10 is the time to be predicted.

[0074] This invention provides a method, apparatus, storage medium, and device for predicting pipeline displacement stress. Through the technical solution of this invention, traditional pipeline displacement stress analysis is based on finite element analysis and physical modeling, requiring iterative calculations of mechanical equations, which is time-consuming. Under the dynamic operating conditions of deep peak shaving in thermal power units, this speed cannot meet the safety early warning requirements of second-level response. Furthermore, the more sensors deployed, the higher the cost, and some pipeline locations cannot be equipped with sensors. Therefore, only a small number of sensors can be deployed on-site, resulting in a small amount of data and low accuracy when training the model. This invention, based on historical temperature and pressure, obtains the measured historical displacement and pressure of each pre-defined node through field measurements. Using the same historical temperature and pressure, and through a pipeline model, it obtains the simulated historical displacement and stress of each pre-defined node in the entire piping system. The advantage of measured historical displacement and pressure is their data accuracy. Since the number of pre-defined nodes in the entire piping system is greater than the number of pre-defined nodes measured in the field, the advantage of simulated historical displacement and stress is their large data volume. A portion of the simulated historical displacement and stress is used to train an initial pipeline displacement and stress prediction model, resulting in a pipeline displacement and stress prediction model to be validated. The accuracy of this model is then validated in two ways: using measured historical displacement and pressure, and using the remaining simulated historical displacement and stress. If at least one validation fails, a pre-defined adjustment strategy is executed until both validations pass, resulting in a highly accurate pipeline displacement and stress prediction model. This pipeline displacement and stress prediction model does not require repeated iterations of the mechanical equations and has a fast response. Therefore, it takes into account full pipeline coverage, low-cost deployment, and real-time response, and achieves high-precision prediction of pipeline displacement stress.

[0075] Furthermore, as a refinement and extension of the specific implementation methods described above, and to fully illustrate the specific implementation process in this embodiment, another method for predicting pipeline displacement stress is provided, such as... Figure 2 As shown, the method includes:

[0076] 201. Obtain the historical temperature of the pipeline, the historical pressure of the pipeline, the historical displacement of the preset node measured on site, and the historical stress of the preset node measured on site.

[0077] In this embodiment, acquiring the historical operating temperature, historical operating pressure, measured historical displacement of the preset node, and measured historical stress of the preset node includes: acquiring the historical operating temperature of the pipeline at the reserved interface based on a temperature sensor, and acquiring the historical operating pressure of the pipeline at the reserved interface based on a pressure sensor; measuring the measured historical displacement of the preset node based on a binocular stereoscopic displacement measurement device; acquiring the pipeline vibration frequency based on a vibrating wire surface strain gauge installed at the preset node, and performing numerical readings on the data acquisition instrument inside the vibrating wire surface strain gauge to obtain the pipeline surface temperature at the location of the vibrating wire surface strain gauge; acquiring preset calculation parameters, and calculating the measured historical stress of the preset node based on the pipeline vibration frequency, the pipeline surface temperature, and the preset calculation parameters, wherein the preset calculation parameters include at least a reference frequency, a reference temperature, the material elastic modulus, the coefficient of linear expansion, a temperature correction coefficient, and a strain gauge sensitivity coefficient.

[0078] Specifically, for each pipeline, a temperature sensor is used to collect the historical operating temperature of the pipeline at the reserved interface, and a pressure sensor is used to collect the historical operating pressure of the pipeline at the reserved interface. In particular, for each pipeline, an integrated temperature and pressure sensor is used and installed at the reserved interface of the pipeline through a threaded connection to collect the historical operating temperature and historical operating pressure of the pipeline.

[0079] Specifically, for measuring the measured historical displacement of preset nodes on-site using a binocular stereo vision displacement measurement device, the device consists of two industrial cameras (e.g., resolution ≥ 2 million pixels, frame rate ≥ 30fps), a target plate, and calibration software. For each pipe, the industrial cameras are fixed to two key pipe nodes (such as elbows and manifold interfaces) using high-temperature resistant brackets. The optical axes of the two cameras are parallel, and the baseline distance is adjusted according to the pipe spacing, for example, 100-300mm. The target plate is attached to the preset nodes on-site. The Zhang Zhengyou calibration method is used to obtain the camera intrinsic parameter matrix (focal length, principal point coordinates) and extrinsic parameter matrix (relative position, attitude). The measured historical displacement of the preset nodes on-site is calculated based on the parallax principle and the similarity theorem of triangles. The measured historical displacement is the three-dimensional displacement of the pipe (Dx, Dy, Dz).

[0080] Specifically, the formula for calculating the measured historical stress of the preset node based on the pipe vibration frequency, the pipe surface temperature, and the preset calculation parameters is as follows:

[0081]

[0082] In the formula, To measure historical stress, The strain gauge sensitivity coefficient, The frequency of pipe vibration. As the reference frequency, The surface temperature of the pipe. As the reference temperature, The elastic modulus of the material. The coefficient of linear expansion is 1 / 3. This is the temperature correction factor.

[0083] 202. Based on the constructed pipeline model, calculate the simulated historical displacement and simulated historical stress of the preset nodes of the entire pipeline system according to the historical temperature and the historical pressure.

[0084] In this embodiment, a pipeline model is constructed based on stress analysis software. Specifically, the pipeline modeling for each pipeline is the same: the pipeline dimensions (diameter, wall thickness) and material parameters (elastic modulus, Poisson's ratio, coefficient of linear expansion) are input according to the construction drawings, and the pipeline is simplified into a beam element model. A second preset number of preset nodes for the entire pipeline system are defined. All preset nodes for the entire pipeline system cover the key areas of the entire pipeline system. For example, the second preset number is 225.

[0085] Then, for each pipeline model, based on the historical temperature and historical pressure imported from the stress analysis software, the simulated historical displacement and simulated historical stress of each preset node of the entire pipeline system are generated in batches.

[0086] 203. Using the historical temperature, historical pressure, preset nodes of the entire piping system, simulated historical displacement, and simulated historical stress, train an initial pipeline displacement stress prediction model until the initial pipeline displacement stress prediction model converges to obtain a pipeline displacement stress prediction model to be verified. Based on the historical temperature, historical pressure, preset nodes of the entire piping system, simulated historical displacement, simulated historical stress, measured historical displacement, and measured historical stress, determine whether the pipeline displacement stress prediction model to be verified has completed training. If not, execute a preset adjustment strategy until a fully trained pipeline displacement stress prediction model is obtained.

[0087] It should be noted that step 201 of the embodiment obtained the historical operating temperature, historical operating pressure, measured historical displacement of the preset nodes in the field, and measured historical stress of the preset nodes in the field. Step 202 of the embodiment obtained the simulated historical displacement of the preset nodes of the entire pipeline system under the historical operating temperature and historical operating pressure of the four major pipelines, as well as the simulated historical stress of the preset nodes in the entire pipeline system. The field-measured preset nodes indicate which specific pipeline among the four major pipelines they belong to, and also their location within that pipeline. Similarly, the preset nodes of the entire pipeline system indicate which specific pipeline among the four major pipelines they belong to, and also their location within that pipeline.

[0088] In this embodiment, the step of training an initial pipeline displacement stress prediction model using the historical temperature, historical pressure, preset nodes of the entire pipeline system, simulated historical displacement, and simulated historical stress until the initial pipeline displacement stress prediction model converges to obtain a pipeline displacement stress prediction model to be verified includes: determining different historical time series data according to different continuous historical moments based on the historical temperature and historical pressure; dividing training historical time series data from all the historical time series data; inputting the training historical time series data and the preset nodes of the entire pipeline system into the initial pipeline displacement stress prediction model; obtaining the first predicted historical displacement and the first predicted historical stress of the preset nodes of the entire pipeline system output by the initial pipeline displacement stress prediction model; using the simulated historical displacement corresponding to the training historical time series data as the training simulated historical displacement; and calculating the training simulated historical displacement and the simulated historical stress of the same preset node of the entire pipeline system... The first predicted historical displacement is calculated using a first loss function value, wherein the simulated historical displacement at the last historical moment of the training historical time series data is the simulated historical displacement corresponding to the training historical time series data; the simulated historical stress corresponding to the training historical time series data is used as the training simulated historical stress, and a second loss function value is calculated based on the training simulated historical stress of the same preset node in the entire pipeline system and the first predicted historical stress, wherein the simulated historical stress at the last historical moment of the training historical time series data is the simulated historical stress corresponding to the training historical time series data; if at least one of the first loss function values ​​is greater than or equal to a preset displacement threshold, or at least one of the second loss function values ​​is greater than or equal to a preset stress threshold, then it is determined that the initial pipeline displacement stress prediction model has not converged, and the parameters of the initial pipeline displacement stress prediction model are adjusted until the initial pipeline displacement stress prediction model converges, thus obtaining the pipeline displacement stress prediction model to be verified.

[0089] Each historical moment and each pipeline corresponds to a historical temperature and a historical pressure. The historical temperature and pressure of each pipeline under a first preset number of consecutive historical moments are combined to form a historical time series data set. Therefore, all pipelines under multiple sets of the first preset number of consecutive historical moments correspond to multiple historical time series data sets. The last historical moment of each historical time series data set corresponds to the simulated historical displacement and simulated historical stress of each preset node in the entire pipeline system. For example, if the first preset number is 10, historical moments 1 to 10 correspond to one historical time series data set 1, corresponding to the simulated historical displacement and simulated historical stress of preset node 1 in the entire pipeline system, up to the simulated historical displacement and simulated historical stress of preset node 225 in the entire pipeline system. Historical moments 11 to 20 correspond to one historical time series data set 2, corresponding to the simulated historical displacement and simulated historical stress of preset node 1 in the entire pipeline system, up to the simulated historical displacement and simulated historical stress of preset node 225 in the entire pipeline system. Until historical time 980 to historical time 1000, corresponding to a historical time series data 100, corresponding to the simulated historical displacement and simulated historical stress of the whole pipeline preset node 1, and to the simulated historical displacement and simulated historical stress of the whole pipeline preset node 225.

[0090] Based on the historical temperature and historical pressure, different historical time series data are determined according to different continuous historical moments. Training historical time series data is divided from all the historical time series data, and validation historical time series data is divided from all the historical time series data. Specifically, all historical time series data are divided into training historical time series data and validation historical time series data according to different historical moments. For example, historical time series data 1 to historical time series data 80 are divided into training historical time series data, and historical time series data 81 to historical time series data 100 are divided into validation historical time series data.

[0091] The training historical time series data and the preset nodes of the entire pipeline system are input into the initial pipeline displacement and stress prediction model, and the first predicted historical displacement and the first predicted historical stress of each preset node of the entire pipeline system are output. The first loss function value is calculated based on the training simulated historical displacement and the first predicted historical displacement of the same preset node of the entire pipeline system. For example, the first loss function value is calculated based on the training simulated historical displacement of the preset node 2 of the entire pipeline system - the main steam pipeline and the first predicted historical displacement of the preset node 2 of the entire pipeline system - the main steam pipeline. The first loss function value is calculated based on the training simulated historical displacement of the preset node 1 of the entire pipeline system - the hot reheat steam pipeline and the first predicted historical displacement of the preset node 1 of the entire pipeline system - the hot reheat steam pipeline. The first loss function value is calculated based on the training simulated historical displacement of the preset node 5 of the entire pipeline system - the cold reheat steam pipeline and the first predicted historical displacement of the preset node 20 of the entire pipeline system - the high-pressure water supply pipeline and the first predicted historical displacement of the preset node 20 of the entire pipeline system - the high-pressure water supply pipeline. The first loss function value is calculated until the first loss function value corresponding to a total of 4 times 225 preset nodes of the entire pipeline system for the four major pipelines is calculated. All 4 times 225 first loss function values ​​are greater than or equal to the preset displacement threshold. Similarly, when all 4 multiplied by 225 values ​​of the second loss function are greater than or equal to the preset stress threshold, the initial pipeline displacement stress prediction model converges, and the pipeline displacement stress prediction model to be verified is obtained.

[0092] The step of determining whether the pipeline displacement stress prediction model to be verified has completed training based on the historical temperature, historical pressure, preset nodes of the entire pipeline system, simulated historical displacement, simulated historical stress, measured historical displacement, and measured historical stress includes: dividing verification historical time series data from all the historical time series data; inputting the verification historical time series data and the preset nodes of the entire pipeline system into the pipeline displacement stress prediction model to be verified; obtaining the second predicted historical displacement and the second predicted historical stress of the preset nodes of the entire pipeline system output by the pipeline displacement stress prediction model to be verified; using the simulated historical displacement corresponding to the verification historical time series data as the verification simulated historical displacement; calculating the first displacement deviation rate based on the verification simulated historical displacement and the second predicted historical displacement of the same preset node of the entire pipeline system; using the simulated historical stress corresponding to the verification historical time series data as the verification simulated historical stress; and calculating the first displacement deviation rate based on the same preset node of the entire pipeline system. The first stress deviation rate is calculated based on the verified simulated historical stress and the second predicted historical stress of the preset nodes of the entire pipeline system. The second displacement deviation rate is calculated based on the measured historical displacement and the second predicted historical displacement of the same preset node in the field. The second stress deviation rate is also calculated based on the measured historical stress and the second predicted historical stress of the same preset node in the field. The simulated historical displacement at the last historical moment of the verified historical time series data is the simulated historical displacement corresponding to the verified historical time series data, and the simulated historical stress at the last historical moment of the verified historical time series data is the simulated historical stress corresponding to the verified historical time series data. The training status of the pipeline displacement stress prediction model to be verified is determined based on the first displacement deviation rate, the first preset displacement threshold, the first stress deviation rate, the first preset stress threshold, the second displacement deviation rate, the second preset displacement threshold, the second stress deviation rate, and the second preset stress threshold.

[0093] The first displacement deviation rate is the absolute value of the result of subtracting the second predicted historical displacement from the verified simulated historical displacement, divided by the verified simulated historical displacement; the first stress deviation rate is the absolute value of subtracting the second predicted historical stress from the verified simulated historical stress, divided by the verified simulated historical stress.

[0094] The second displacement deviation rate is the absolute value of the result of subtracting the second predicted historical displacement from the measured historical displacement, divided by the measured historical displacement. The second stress deviation rate is the absolute value of subtracting the second predicted historical stress from the measured historical stress, divided by the measured historical stress.

[0095] In this embodiment, the step of executing a preset adjustment strategy until a trained pipeline displacement stress prediction model is obtained includes: if at least one of the first displacement deviation rates is greater than or equal to the first preset displacement threshold, or at least one of the first stress deviation rates is greater than or equal to the first preset stress threshold, or at least one of the second displacement deviation rates is greater than or equal to the second preset displacement threshold, or at least one of the second stress deviation rates is greater than or equal to the second preset stress threshold, then it is determined that the pipeline displacement stress prediction model to be verified has not completed training; adding training samples is determined as a preset adjustment strategy, a new sample set is obtained, and the pipeline displacement stress prediction model to be verified is retrained using the new sample set until the retrained pipeline displacement stress prediction model to be verified completes training, thus obtaining a trained pipeline displacement stress prediction model.

[0096] The method further includes: if all the first displacement deviation rates are less than the first preset displacement threshold, and all the first stress deviation rates are less than the first preset stress threshold, and all the second displacement deviation rates are less than the second preset displacement threshold, and all the second stress deviation rates are less than the second preset stress threshold, then the pipeline displacement stress prediction model to be verified has completed training, and a trained pipeline displacement stress prediction model is obtained.

[0097] As another implementation, the preset adjustment strategy also includes updating the weights of the pipeline displacement stress prediction model to be verified and continuing to train using the original training samples until a fully trained pipeline displacement stress prediction model is obtained.

[0098] 204. Real-time acquisition of pipeline operating temperature and pipeline operating pressure, inputting the time series data of the temperature and pressure, along with the preset nodes of the entire pipeline system, into the pipeline displacement stress prediction model to predict the displacement and stress of the preset nodes of the entire pipeline system.

[0099] For this embodiment, the specific implementation method is the same as step 104 of the embodiment, and will not be repeated here.

[0100] 205. Obtain preset displacement safety range, displacement warning range, and displacement danger range. If the displacement is in the displacement safety range, display the displacement safety of the preset nodes of the entire pipeline system corresponding to the displacement. If the displacement is in the displacement warning range, display the displacement warning of the preset nodes of the entire pipeline system corresponding to the displacement. If the displacement is in the displacement danger range, display the displacement alarm of the preset nodes of the entire pipeline system corresponding to the displacement, so as to perform online displacement monitoring.

[0101] 206. Obtain preset stress safety range, stress warning range, and stress danger range. If the stress is in the stress safety range, display the stress safety of the preset nodes of the entire piping system corresponding to the stress. If the stress is in the stress warning range, display the stress warning of the preset nodes of the entire piping system corresponding to the stress. If the stress is in the stress danger range, display the stress alarm of the preset nodes of the entire piping system corresponding to the stress, so as to perform online stress monitoring.

[0102] For steps 205 and 206 of the embodiment, following the example of step 104 of the embodiment, after predicting the displacement and pressure of the main steam pipeline at each preset node of the entire pipeline system at time 10, it is determined which interval of the displacement safety interval, displacement warning interval, and displacement danger interval the displacement falls into, and the corresponding interval is displayed. Similarly, it is determined which interval of the stress safety interval, stress warning interval, and stress danger interval the stress falls into, and the corresponding interval is displayed.

[0103] It should be noted that this also includes: dynamic display: loading the pipeline model, associating the output data of the pipeline displacement stress prediction model (that is, the displacement of the preset nodes of the entire pipeline system and the stress of the preset nodes of the entire pipeline system in step 204 of the embodiment), determining the displacement direction based on the displacement value and the original displacement, marking the displacement direction with colored arrows, the arrow length mapping the displacement value, and performing gradient coloring according to the stress value, for example, 0 to 100MPa gradually corresponding to blue to red.

[0104] This invention provides a method, apparatus, storage medium, and device for predicting pipeline displacement stress. Through the technical solution of this invention, traditional pipeline displacement stress analysis is based on finite element analysis and physical modeling, requiring iterative calculations of mechanical equations, which is time-consuming. Under the dynamic operating conditions of deep peak shaving in thermal power units, this speed cannot meet the safety early warning requirements of second-level response. Furthermore, the more sensors deployed, the higher the cost, and some pipeline locations cannot be equipped with sensors. Therefore, only a small number of sensors can be deployed on-site, resulting in a small amount of data and low accuracy when training the model. This invention, based on historical temperature and pressure, obtains the measured historical displacement and pressure of each pre-defined node through field measurements. Using the same historical temperature and pressure, and through a pipeline model, it obtains the simulated historical displacement and stress of each pre-defined node in the entire piping system. The advantage of measured historical displacement and pressure is their data accuracy. Since the number of pre-defined nodes in the entire piping system is greater than the number of pre-defined nodes measured in the field, the advantage of simulated historical displacement and stress is their large data volume. A portion of the simulated historical displacement and stress is used to train an initial pipeline displacement and stress prediction model, resulting in a pipeline displacement and stress prediction model to be validated. The accuracy of this model is then validated in two ways: using measured historical displacement and pressure, and using the remaining simulated historical displacement and stress. If at least one validation fails, a pre-defined adjustment strategy is executed until both validations pass, resulting in a highly accurate pipeline displacement and stress prediction model. This pipeline displacement and stress prediction model does not require repeated iterations of the mechanical equations and has a fast response. Therefore, it takes into account full pipeline coverage, low-cost deployment, and real-time response, and achieves high-precision prediction of pipeline displacement stress.

[0105] Furthermore, as Figure 1 and Figure 2 The specific implementation of the method shown in this invention provides a pipe displacement stress prediction device, such as... Figure 3 As shown, the device includes: an acquisition module 31, a calculation module 32, a first training module 33, and a prediction module 34;

[0106] The acquisition module 31 is used to acquire the historical temperature of the pipeline, the historical pressure of the pipeline, the measured historical displacement of the preset node in the field, and the measured historical stress of the preset node in the field.

[0107] The calculation module 32 is used to calculate the simulated historical displacement and simulated historical stress of the preset nodes of the entire pipeline system based on the constructed pipeline model and according to the historical temperature and the historical pressure.

[0108] The first training module 33 is used to train an initial pipeline displacement stress prediction model using the historical temperature, the historical pressure, the preset nodes of the entire pipeline system, the simulated historical displacement, and the simulated historical stress until the initial pipeline displacement stress prediction model converges to obtain a pipeline displacement stress prediction model to be verified. Based on the historical temperature, the historical pressure, the preset nodes of the entire pipeline system, the simulated historical displacement, the simulated historical stress, the measured historical displacement, and the measured historical stress, it is determined whether the pipeline displacement stress prediction model to be verified has completed training. If not, a preset adjustment strategy is executed until a pipeline displacement stress prediction model that has been trained is obtained.

[0109] The prediction module 34 is used to collect the temperature and pressure of the pipeline in real time, input the time series data of the temperature and pressure, as well as the preset nodes of the entire pipeline system, into the pipeline displacement stress prediction model, and predict the displacement and stress of the preset nodes of the entire pipeline system.

[0110] In specific application scenarios, a pipeline displacement stress prediction device, such as Figure 4 As shown, the device further includes: a setting module 35, specifically used to acquire preset displacement safety range, displacement warning range, and displacement danger range; if the displacement is within the displacement safety range, the device displays the displacement safety of the preset nodes of the entire piping system corresponding to the displacement; if the displacement is within the displacement warning range, the device displays the displacement warning of the preset nodes of the entire piping system corresponding to the displacement; if the displacement is within the displacement danger range, the device displays the displacement alarm of the preset nodes of the entire piping system corresponding to the displacement, for online displacement monitoring; and to acquire preset stress safety range, stress warning range, and stress danger range; if the stress is within the stress safety range, the device displays the stress safety of the preset nodes of the entire piping system corresponding to the stress; if the stress is within the stress warning range, the device displays the stress warning of the preset nodes of the entire piping system corresponding to the stress; if the stress is within the stress danger range, the device displays the stress alarm of the preset nodes of the entire piping system corresponding to the stress, for online stress monitoring.

[0111] Accordingly, in order to obtain the historical operating temperature of the pipeline, the historical operating pressure of the pipeline, the measured historical displacement of the preset node in the field, and the measured historical stress of the preset node in the field, the acquisition module 31 is specifically used to: collect the historical operating temperature of the pipeline at the reserved interface based on a temperature sensor; collect the historical operating pressure of the pipeline at the reserved interface based on a pressure sensor; measure the measured historical displacement of the preset node in the field based on a binocular stereoscopic displacement measurement device; collect the pipeline vibration frequency based on a vibrating wire surface strain gauge installed at the preset node in the field, and perform numerical readings on the data acquisition instrument inside the vibrating wire surface strain gauge to obtain the pipeline surface temperature at the location of the vibrating wire surface strain gauge; acquire preset calculation parameters, and calculate the measured historical stress of the preset node in the field based on the pipeline vibration frequency, the pipeline surface temperature, and the preset calculation parameters. The preset calculation parameters include at least a reference frequency, a reference temperature, the elastic modulus of the material, the coefficient of linear expansion, a temperature correction coefficient, and a strain gauge sensitivity coefficient.

[0112] Accordingly, in order to train an initial pipeline displacement-stress prediction model using the historical temperature, historical pressure, preset nodes of the entire pipeline system, simulated historical displacement, and simulated historical stress, until the initial pipeline displacement-stress prediction model converges to obtain the pipeline displacement-stress prediction model to be verified, the first training module 33 is specifically used to determine different historical time series data according to different continuous historical moments based on the historical temperature and historical pressure, divide training historical time series data from all the historical time series data, input the training historical time series data and the preset nodes of the entire pipeline system into the initial pipeline displacement-stress prediction model, obtain the first predicted historical displacement and the first predicted historical stress of the preset nodes of the entire pipeline system output by the initial pipeline displacement-stress prediction model; use the simulated historical displacement corresponding to the training historical time series data as the training simulated historical displacement, and calculate the training simulated historical displacement based on the same preset node of the entire pipeline system. A first loss function value is calculated based on the displacement and the first predicted historical displacement, wherein the simulated historical displacement at the last historical moment of the training historical time series data is the simulated historical displacement corresponding to the training historical time series data; the simulated historical stress corresponding to the training historical time series data is used as the training simulated historical stress, and a second loss function value is calculated based on the training simulated historical stress and the first predicted historical stress at the same preset node of the entire pipeline system, wherein the simulated historical stress at the last historical moment of the training historical time series data is the simulated historical stress corresponding to the training historical time series data; if at least one of the first loss function values ​​is greater than or equal to a preset displacement threshold, or at least one of the second loss function values ​​is greater than or equal to a preset stress threshold, then it is determined that the initial pipeline displacement stress prediction model has not converged, and the parameters of the initial pipeline displacement stress prediction model are adjusted until the initial pipeline displacement stress prediction model converges, thus obtaining the pipeline displacement stress prediction model to be verified.

[0113] Accordingly, in order to determine whether the pipeline displacement stress prediction model to be verified has completed training based on the historical temperature, the historical pressure, the preset nodes of the entire pipeline system, the simulated historical displacement, the simulated historical stress, the measured historical displacement, and the measured historical stress, the first training module 33 is specifically used to divide the verification historical time series data from all the historical time series data, input the verification historical time series data and the preset nodes of the entire pipeline system into the pipeline displacement stress prediction model to be verified, obtain the second predicted historical displacement and the second predicted historical stress of the preset nodes of the entire pipeline system output by the pipeline displacement stress prediction model to be verified; take the simulated historical displacement corresponding to the verification historical time series data as the verification simulated historical displacement, calculate the first displacement deviation rate based on the verification simulated historical displacement and the second predicted historical displacement of the same preset node of the entire pipeline system, and take the simulated historical stress corresponding to the verification historical time series data as the verification simulated historical displacement. The stress is calculated by: a first stress deviation rate based on the verified simulated historical stress and the second predicted historical stress of the same preset node in the entire piping system; a second displacement deviation rate based on the measured historical displacement and the second predicted historical displacement of the same preset node in the field; and a third stress deviation rate based on the measured historical stress and the second predicted historical stress of the same preset node in the field. The simulated historical displacement at the last historical moment of the verified historical time series data is the simulated historical displacement corresponding to the verified historical time series data, and the simulated historical stress at the last historical moment of the verified historical time series data is the simulated historical stress corresponding to the verified historical time series data. The training status of the pipeline displacement stress prediction model to be verified is determined based on the first displacement deviation rate, the first preset displacement threshold, the first stress deviation rate, the first preset stress threshold, the second displacement deviation rate, the second preset displacement threshold, the second stress deviation rate, and the second preset stress threshold.

[0114] Accordingly, if not, a preset adjustment strategy is executed until a trained pipeline displacement stress prediction model is obtained. Specifically, the first training module 33 is used to determine that the pipeline displacement stress prediction model to be verified has not completed training if at least one of the first displacement deviation rates is greater than or equal to the first preset displacement threshold, or at least one of the first stress deviation rates is greater than or equal to the first preset stress threshold, or at least one of the second displacement deviation rates is greater than or equal to the second preset displacement threshold, or at least one of the second stress deviation rates is greater than or equal to the second preset stress threshold; and to determine that adding training samples is a preset adjustment strategy, obtain a new sample set, and retrain the pipeline displacement stress prediction model to be verified using the new sample set until the retrained pipeline displacement stress prediction model to be verified completes training, thereby obtaining a trained pipeline displacement stress prediction model.

[0115] In specific application scenarios, a pipeline displacement stress prediction device, such as Figure 4 As shown, the device further includes a second training module 36, specifically used to complete the training of the pipeline displacement stress prediction model to be verified if all the first displacement deviation rates are less than the first preset displacement threshold, and all the first stress deviation rates are less than the first preset stress threshold, and all the second displacement deviation rates are less than the second preset displacement threshold, and all the second stress deviation rates are less than the second preset stress threshold, and the trained pipeline displacement stress prediction model is obtained.

[0116] It should be noted that other corresponding descriptions of the functional units involved in the pipeline displacement stress prediction device provided in this embodiment can be found in [reference needed]. Figures 1 to 2 The corresponding description will not be repeated here.

[0117] Based on the above, Figures 1 to 2 Accordingly, this embodiment also provides a storage medium, which may be volatile or non-volatile, storing a computer program that, when executed by a processor, implements the above-described method. Figures 1 to 2 The method for predicting pipeline displacement stress is shown.

[0118] Based on this understanding, the technical solution of the present invention 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, USB flash drive, portable hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of the present invention.

[0119] Based on the above, Figures 1 to 2 The method shown and Figure 3 , Figure 4To achieve the above objectives, the present application also provides a computer device, specifically a personal computer, server, network device, etc., as shown in the illustrated embodiment. This computer device includes a storage medium and a processor; the storage medium stores a computer program; the processor executes the computer program to achieve the above-described objectives. Figure 1 and Figure 2 The method for predicting pipeline displacement stress is shown.

[0120] Optionally, the computer device may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0121] Those skilled in the art will understand that the computer device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0122] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned computer device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the non-volatile storage medium, as well as communication with other hardware and software in the information processing entity device.

[0123] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general-purpose hardware platform, or it can be implemented by hardware.

[0124] This invention provides a method, apparatus, storage medium, and device for predicting pipeline displacement stress. Through the technical solution of this invention, traditional pipeline displacement stress analysis is based on finite element analysis and physical modeling, requiring iterative calculations of mechanical equations, which is time-consuming. Under the dynamic operating conditions of deep peak shaving in thermal power units, this speed cannot meet the safety early warning requirements of second-level response. Furthermore, the more sensors deployed, the higher the cost, and some pipeline locations cannot be equipped with sensors. Therefore, only a small number of sensors can be deployed on-site, resulting in a small amount of data and low accuracy when training the model. This invention, based on historical temperature and pressure, obtains the measured historical displacement and pressure of each pre-defined node through field measurements. Using the same historical temperature and pressure, and through a pipeline model, it obtains the simulated historical displacement and stress of each pre-defined node in the entire piping system. The advantage of measured historical displacement and pressure is their data accuracy. Since the number of pre-defined nodes in the entire piping system is greater than the number of pre-defined nodes measured in the field, the advantage of simulated historical displacement and stress is their large data volume. A portion of the simulated historical displacement and stress is used to train an initial pipeline displacement and stress prediction model, resulting in a pipeline displacement and stress prediction model to be validated. The accuracy of this model is then validated in two ways: using measured historical displacement and pressure, and using the remaining simulated historical displacement and stress. If at least one validation fails, a pre-defined adjustment strategy is executed until both validations pass, resulting in a highly accurate pipeline displacement and stress prediction model. This pipeline displacement and stress prediction model does not require repeated iterations of the mechanical equations and has a fast response. Therefore, it takes into account full pipeline coverage, low-cost deployment, and real-time response, and achieves high-precision prediction of pipeline displacement stress.

[0125] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention. Those skilled in the art will understand that the modules in the apparatus of the embodiment can be distributed within the apparatus of the embodiment as described, or they can be located in one or more apparatuses different from this embodiment, with corresponding changes. The modules of the above-described embodiment can be combined into one module, or further divided into multiple sub-modules.

[0126] The serial numbers used above are for descriptive purposes only and do not represent the superiority or inferiority of the implementation scenarios. The above disclosures are merely a few specific implementation scenarios of the present invention; however, the present invention is not limited thereto, and any variations conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A method for predicting pipeline displacement stress, characterized in that, The method includes: The historical temperature of the pipeline during operation, the historical pressure of the pipeline during operation, the measured historical displacement of the preset nodes in the field, and the measured historical stress of the preset nodes in the field are obtained. Based on the constructed pipeline model, the simulated historical displacement and simulated historical stress of the preset nodes of the entire pipeline system are calculated according to the historical temperature and the historical pressure. Using the historical temperature, historical pressure, preset nodes of the entire piping system, simulated historical displacement, and simulated historical stress, an initial pipeline displacement stress prediction model is trained until the initial pipeline displacement stress prediction model converges, resulting in a pipeline displacement stress prediction model to be verified. Based on the historical temperature, historical pressure, preset nodes of the entire piping system, simulated historical displacement, simulated historical stress, measured historical displacement, and measured historical stress, it is determined whether the pipeline displacement stress prediction model to be verified has completed training. If not, a preset adjustment strategy is executed until a fully trained pipeline displacement stress prediction model is obtained. The temperature and pressure of the pipeline are collected in real time. The time series data of the temperature and pressure, along with the preset nodes of the entire pipeline system, are input into the pipeline displacement and stress prediction model to predict the displacement and stress of the preset nodes of the entire pipeline system. The process of training an initial pipeline displacement and stress prediction model using the historical temperature, historical pressure, preset nodes of the entire pipeline system, simulated historical displacement, and simulated historical stress, until the initial pipeline displacement and stress prediction model converges, yields a pipeline displacement and stress prediction model to be verified, including: Based on the historical temperature and the historical pressure, different historical time series data are determined according to different continuous historical moments. Training historical time series data is divided from all the historical time series data. The training historical time series data and the preset nodes of the entire pipeline system are input into the initial pipeline displacement stress prediction model to obtain the first predicted historical displacement and the first predicted historical stress of the preset nodes of the entire pipeline system output by the initial pipeline displacement stress prediction model. The simulated historical displacement corresponding to the training historical time series data is used as the training simulated historical displacement. The first loss function value is calculated based on the training simulated historical displacement of the same preset node of the entire pipeline system and the first predicted historical displacement. The simulated historical displacement of the last historical moment of the training historical time series data is the simulated historical displacement corresponding to the training historical time series data. The simulated historical stress corresponding to the training historical time series data is used as the training simulated historical stress. The second loss function value is calculated based on the training simulated historical stress of the same preset node of the entire pipeline system and the first predicted historical stress. The simulated historical stress at the last historical moment of the training historical time series data is the simulated historical stress corresponding to the training historical time series data. If at least one of the first loss function values ​​is greater than or equal to a preset displacement threshold, or if at least one of the second loss function values ​​is greater than or equal to a preset stress threshold, then it is determined that the initial pipeline displacement stress prediction model has not converged. The parameters of the initial pipeline displacement stress prediction model are adjusted until the initial pipeline displacement stress prediction model converges, and the pipeline displacement stress prediction model to be verified is obtained.

2. The method according to claim 1, characterized in that, The method further includes: The system acquires preset displacement safety range, displacement warning range, and displacement danger range. If the displacement is within the displacement safety range, it displays the displacement safety of the preset nodes of the entire pipeline system corresponding to the displacement. If the displacement is within the displacement warning range, it displays the displacement warning of the preset nodes of the entire pipeline system corresponding to the displacement. If the displacement is within the displacement danger range, it displays the displacement alarm of the preset nodes of the entire pipeline system corresponding to the displacement, so as to perform online displacement monitoring. The system acquires preset stress safety range, stress warning range, and stress danger range. If the stress is within the stress safety range, it displays the stress safety of the preset nodes in the entire piping system corresponding to the stress. If the stress is within the stress warning range, it displays the stress warning of the preset nodes in the entire piping system corresponding to the stress. If the stress is within the stress danger range, it displays the stress alarm of the preset nodes in the entire piping system corresponding to the stress, thereby enabling online stress monitoring.

3. The method according to claim 1, characterized in that, The acquisition of historical temperature, historical pressure, measured historical displacement of preset nodes, and measured historical stress of preset nodes during pipeline operation includes: Based on a temperature sensor, the historical operating temperature of the pipeline at the reserved pipeline interface is collected; based on a pressure sensor, the historical operating pressure of the pipeline at the reserved pipeline interface is collected. The measured historical displacement of preset nodes on site is measured using a binocular stereo vision displacement measurement device. Based on the vibrating wire surface strain gauges deployed at the preset nodes in the field, the vibration frequency of the pipeline is collected, and the data acquisition instrument inside the vibrating wire surface strain gauge is used to read the values ​​to obtain the surface temperature of the pipeline at the location of the vibrating wire surface strain gauge. Obtain preset calculation parameters, and calculate the measured historical stress of the preset node in the field based on the pipeline vibration frequency, the pipeline surface temperature and the preset calculation parameters. The preset calculation parameters include at least the reference frequency, reference temperature, material elastic modulus, linear expansion coefficient, temperature correction coefficient and strain gauge sensitivity coefficient.

4. The method according to claim 1, characterized in that, The step of determining whether the pipeline displacement-stress prediction model to be verified has completed training based on the historical temperature, historical pressure, preset nodes of the entire pipeline system, simulated historical displacement, simulated historical stress, measured historical displacement, and measured historical stress includes: The verification historical time series data is divided from all the historical time series data. The verification historical time series data and the preset nodes of the entire pipeline system are input into the pipeline displacement stress prediction model to be verified. The second predicted historical displacement and the second predicted historical stress of the preset nodes of the entire pipeline system are output by the pipeline displacement stress prediction model to be verified. The simulated historical displacement corresponding to the verification historical time series data is used as the verification simulated historical displacement. A first displacement deviation rate is calculated based on the verification simulated historical displacement and the second predicted historical displacement of the same preset node of the entire pipeline system. The simulated historical stress corresponding to the verification historical time series data is used as the verification simulated historical stress. A first stress deviation rate is calculated based on the verification simulated historical stress and the second predicted historical stress of the same preset node of the entire pipeline system. A second displacement deviation rate is calculated based on the measured historical displacement and the second predicted historical displacement of the same preset node of the field measurement. A second stress deviation rate is calculated based on the measured historical stress and the second predicted historical stress of the same preset node of the field measurement. The simulated historical displacement at the last historical moment of the verification historical time series data is the simulated historical displacement corresponding to the verification historical time series data, and the simulated historical stress at the last historical moment of the verification historical time series data is the simulated historical stress corresponding to the verification historical time series data. The training status of the pipeline displacement stress prediction model to be verified is determined based on the first displacement deviation rate, the first preset displacement threshold, the first stress deviation rate, the first preset stress threshold, the second displacement deviation rate, the second preset displacement threshold, the second stress deviation rate, and the second preset stress threshold.

5. The method according to claim 4, characterized in that, If not, a preset adjustment strategy is executed until a trained pipeline displacement stress prediction model is obtained, including: If at least one of the first displacement deviation rates is greater than or equal to the first preset displacement threshold, or at least one of the first stress deviation rates is greater than or equal to the first preset stress threshold, or at least one of the second displacement deviation rates is greater than or equal to the second preset displacement threshold, or at least one of the second stress deviation rates is greater than or equal to the second preset stress threshold, then it is determined that the pipeline displacement stress prediction model to be verified has not completed training. The preset adjustment strategy is to increase the number of training samples. A new sample set is obtained, and the pipeline displacement stress prediction model to be verified is retrained using the new sample set until the retrained pipeline displacement stress prediction model to be verified is completed, thus obtaining the trained pipeline displacement stress prediction model.

6. The method according to claim 4, characterized in that, The method further includes: If all the first displacement deviation rates are less than the first preset displacement threshold, and all the first stress deviation rates are less than the first preset stress threshold, and all the second displacement deviation rates are less than the second preset displacement threshold, and all the second stress deviation rates are less than the second preset stress threshold, then the pipeline displacement stress prediction model to be verified has completed training, and a trained pipeline displacement stress prediction model has been obtained.

7. A device for predicting pipeline displacement stress, characterized in that, The device includes: The acquisition module is used to acquire the historical temperature of the pipeline during operation, the historical pressure of the pipeline during operation, the measured historical displacement of the preset nodes in the field, and the measured historical stress of the preset nodes in the field. The calculation module is used to calculate the simulated historical displacement and simulated historical stress of the preset nodes of the entire pipeline system based on the constructed pipeline model and according to the historical temperature and the historical pressure. The first training module is used to train an initial pipeline displacement stress prediction model using the historical temperature, the historical pressure, the preset nodes of the entire piping system, the simulated historical displacement, and the simulated historical stress, until the initial pipeline displacement stress prediction model converges to obtain a pipeline displacement stress prediction model to be verified. Based on the historical temperature, the historical pressure, the preset nodes of the entire piping system, the simulated historical displacement, the simulated historical stress, the measured historical displacement, and the measured historical stress, it is determined whether the pipeline displacement stress prediction model to be verified has completed training. If not, a preset adjustment strategy is executed until a fully trained pipeline displacement stress prediction model is obtained. The prediction module is used to collect the temperature and pressure of the pipeline in real time, input the time series data of the temperature and pressure, and the preset nodes of the entire pipeline system into the pipeline displacement stress prediction model to predict the displacement and stress of the preset nodes of the entire pipeline system. The first training module is configured to determine different historical time series data according to the historical temperature and historical pressure, based on different consecutive historical moments; divide training historical time series data from all the historical time series data; input the training historical time series data and the preset nodes of the entire pipeline system into an initial pipeline displacement-stress prediction model; obtain the first predicted historical displacement and the first predicted historical stress of the preset nodes of the entire pipeline system output by the initial pipeline displacement-stress prediction model; use the simulated historical displacement corresponding to the training historical time series data as the training simulated historical displacement; calculate a first loss function value based on the training simulated historical displacement and the first predicted historical displacement of the same preset node of the entire pipeline system; wherein, the simulated historical displacement of the last historical moment of the training historical time series data... The displacement is the simulated historical displacement corresponding to the training historical time series data; the simulated historical stress corresponding to the training historical time series data is used as the training simulated historical stress, and a second loss function value is calculated based on the training simulated historical stress of the same preset node of the entire pipeline system and the first predicted historical stress, wherein the simulated historical stress at the last historical moment of the training historical time series data is the simulated historical stress corresponding to the training historical time series data; if at least one of the first loss function values ​​is greater than or equal to a preset displacement threshold, or at least one of the second loss function values ​​is greater than or equal to a preset stress threshold, then it is determined that the initial pipeline displacement stress prediction model has not converged, and the parameters of the initial pipeline displacement stress prediction model are adjusted until the initial pipeline displacement stress prediction model converges to obtain the pipeline displacement stress prediction model to be verified.

8. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the pipeline displacement stress prediction method according to any one of claims 1 to 6.

9. A computer device comprising a memory, a processor, and a computer program stored on a storage medium and running on the processor, characterized in that, When the processor executes the program, it implements the pipeline displacement stress prediction method according to any one of claims 1 to 6.

Images