A system for simulating performance and designing structure of an anticorrosion and heat preservation pipeline product

Abstract

The application discloses an anti-corrosion and heat-preservation pipeline product performance simulation and structure design system, and relates to the technical field of pipeline engineering design.The system comprises a multi-dimensional parameter database module used for storing basic design parameters of the pipeline, wherein the basic design parameters include pipeline number, material, medium temperature, pressure, heat-preservation grade, anti-corrosion layer type and environmental conditions; and an intelligent parameter checking module connected with the multi-dimensional parameter database module.The anti-corrosion and heat-preservation pipeline product performance simulation and structure design system avoids parameter omission, input error or format non-standard problems caused by human negligence in the design process, ensures the accuracy and reliability of the parameter set input to the subsequent module, saves the complicated operation of manual modeling and repeated input of parameters, greatly shortens the modeling cycle, provides an accurate model basis, guarantees the numerical credibility of the simulation calculation result, and provides solid data support for subsequent optimization decision.

Description

Technical Field

[0001] This invention relates to the field of pipeline engineering design technology, specifically to a performance simulation and structural design system for corrosion-resistant and heat-insulating pipeline products. Background Technology

[0002] Corrosion-resistant and heat-insulated pipelines are widely used in industrial fields such as petroleum, chemical, thermal, and nuclear power. Especially in high-temperature, high-pressure, and corrosive media transportation environments, the design of their anti-corrosion and insulation layers directly affects operational safety, energy consumption control, and service life. With the increasing complexity and standardization requirements of industrial pipeline systems, the design accuracy and simulation analysis of corrosion-resistant and heat-insulating structures have become crucial aspects of engineering design.

[0003] Traditional corrosion-resistant and thermal insulation pipeline design relies heavily on designers' experience and manual modeling, requiring comprehensive consideration of multiple parameters such as pipeline material, insulation layer thickness, corrosion protection layer type, medium temperature, and environmental conditions. Taking a nuclear power plant's nuclear island building as an example, the sheer number and complex layout of pipelines mean that corrosion-resistant and thermal insulation modeling not only affects design drawings but also directly impacts subsequent collision checks, stress analysis, and construction and installation. Existing design methods typically employ general-purpose 3D design platforms (such as PDMS and E3D) for manual modeling. Designers must manually input or select parameters such as insulation level, heat tracing level, temperature, and corrosion protection layer type, resulting in tedious and repetitive operations.

[0004] Patent CN115795753A discloses "An Automatic Modeling Method and System for 3D Design of Industrial Pipeline Insulation Layers." This patent proposes a scheme that automatically retrieves and assigns values ​​to pipeline insulation layers during pipeline modeling by analyzing pipeline insulation layer modeling rules, establishing a pipeline design parameter database, and developing a program to establish the logical relationship between pipeline 3D model attributes and database data. This achieves automated batch modeling of pipeline insulation layers, avoiding human error issues such as missing insulation design parameters and input errors to a certain extent. However, this technical solution mainly focuses on the automation of the 3D modeling process and still has the following shortcomings: First, it lacks an effective verification mechanism for input design parameters. If the parameters have logical errors or abnormal combinations, the basic modeling data will be unreliable. Second, it lacks a unified parameter management and retrieval mechanism. Design parameters are scattered in different files or tables, making it difficult to ensure data consistency and traceability. Parameter transfer and version control issues are prominent in multi-disciplinary collaboration. Third, simulation analysis and structural design are disconnected. This solution does not integrate performance simulation and optimization feedback. Traditional methods require repeated switching of parameters between modeling and simulation, making it difficult to achieve integrated design-simulation closed-loop optimization.

[0005] To address the aforementioned issues, there is an urgent need to develop a performance simulation and structural design system for anti-corrosion and thermal insulation pipeline products. This system would integrate intelligent verification, automatic modeling, and performance simulation of pipeline anti-corrosion and insulation layer parameters, thereby improving design efficiency and accuracy, reducing human error rates, and ensuring the safety and economy of pipeline systems during the design, construction, and operation phases. Summary of the Invention

[0006] The purpose of this invention is to provide a performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products to solve the problems mentioned in the background art.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products, comprising: The multidimensional parameter database module is used to store the basic design parameters of the pipeline, including pipeline number, material, medium temperature, pressure, insulation level, anti-corrosion layer type, and environmental conditions. The intelligent parameter verification module, connected to the multidimensional parameter database module, is used to perform at least two rounds of verification on the input original design parameters using different mechanisms, and generate a set of parameters that have passed the verification. An automatic modeling and attribute mapping module, connected to the intelligent parameter verification module, is used to automatically generate a three-dimensional model of the pipeline and anti-corrosion insulation layer in the three-dimensional design platform based on the parameter set that has passed verification, and dynamically assign the parameters in the parameter set to the corresponding attributes of the three-dimensional model. The multiphysics coupling simulation module is connected to the automatic modeling and attribute mapping module and is used to perform heat conduction simulation, stress field simulation and corrosion life simulation of the generated three-dimensional model to obtain simulation results. The optimization feedback and closed-loop control module is connected to both the multiphysics coupled simulation module and the intelligent parameter verification module. It compares the simulation results with the preset design objectives. If the design objectives are not met, the optimization algorithm is triggered to automatically adjust the design parameters, and the adjusted parameters are re-input into the intelligent parameter verification module, initiating a new round of modeling and simulation until the final design parameters that meet the design objectives are obtained. Each cycle retains complete parameters, models, and simulation data, achieving end-to-end traceability of the design process.

[0008] Preferably, the intelligent parameter verification module includes: The first verification unit has a built-in industry standard rule library, which is used to automatically review the completeness and compliance of the original design parameters. If the parameters violate the preset rules in the rule library, a modification prompt will be output. The second verification unit is connected to the first verification unit and has a built-in anomaly detection machine learning model trained based on historical project data. It is used to evaluate the probability of reasonableness of the parameters that pass the first verification unit and output the confidence value of the parameter combination. If the confidence value is lower than the preset threshold, the manual review process is triggered. The first verification unit and the second verification unit sequentially perform verification on the original design parameters, jointly constituting the at least two rounds of verification using different mechanisms. After the two rounds of verification are completed, a verification report containing the results of all verification items is generated and synchronously transferred to subsequent modules along with the parameter set.

[0009] Preferably, the intelligent parameter verification module further includes a third verification unit connected to the second verification unit. This third verification unit incorporates a simplified physical model and is used to cross-verify key parameters that have undergone the first two rounds of verification. These key parameters include the insulation layer thickness or anti-corrosion layer type corresponding to the pipe where the medium temperature exceeds a set temperature threshold. The output of the third verification unit serves as additional criteria for determining the reasonableness of the parameters. The temperature threshold can be customized according to the pipe material and design specifications to adapt to different pipe design requirements under various operating conditions.

[0010] Preferably, the multiphysics coupling simulation module includes: The heat conduction simulation unit is used to calculate the temperature field distribution of a three-dimensional model under specified operating conditions. The stress field simulation unit is connected to the heat conduction simulation unit and is used to analyze the stress field distribution of the pipeline caused by thermal expansion based on the temperature field distribution results. The corrosion life simulation unit is connected to the heat conduction simulation unit and the stress field simulation unit. It is used to predict the failure time of the anti-corrosion coating and the remaining life of the pipeline based on the corrosion model and the simulation results of the temperature field and stress field. The heat conduction simulation unit, stress field simulation unit, and corrosion life simulation unit are coupled and simulated using parallel computing. This parallel computing mode allows for simultaneous numerical solutions to the three physical fields, reducing the overall simulation computation time.

[0011] Preferably, the multiphysics coupled simulation module further includes a simulation verification unit, which is connected to the heat conduction simulation unit, stress field simulation unit, and corrosion lifetime simulation unit, respectively. This verification unit is used to cross-compare the simulation results of the same physical field using at least two different numerical solution methods. If the deviation between the output results of the two solution methods exceeds a preset deviation threshold, the mesh density or solver parameters are adjusted, and the simulation is recalculated until the deviation is reduced to within the preset deviation threshold, thus completing a secondary verification of the simulation results. The deviation threshold can be customized according to the design accuracy requirements. The adjusted mesh is preferentially refined in regions with drastic changes in physical quantity gradients.

[0012] Preferably, the optimized feedback and closed-loop control module includes: The target comparison unit is used to compare the simulation results with the preset design targets one by one. The design targets include the maximum allowable heat loss, the minimum safe life, and the maximum allowable stress. The parameter optimization unit is connected to the target comparison unit and has a built-in multi-objective genetic optimization algorithm. When the simulation results do not meet the design objectives, it automatically generates at least one set of adjusted design parameters based on the difference data of the target comparison unit. The cyclic control unit, connected to both the parameter optimization unit and the intelligent parameter verification module, re-inputs the adjusted design parameters into the intelligent parameter verification module and triggers iterative processing by the automatic modeling and attribute mapping module and the multiphysics coupling simulation module until the target comparison unit determines that the simulation results meet the design objectives and outputs the final design parameters. This unit can record the input parameters and output results of each iteration, forming a complete iterative process log.

[0013] Preferably, the loop control unit includes a convergence determination subunit. This subunit calculates the change in the objective function between the current iteration and the previous iteration after each iteration. If the change in the objective function is lower than a preset convergence threshold twice consecutively, the iteration is determined to have converged, the loop is terminated, and the design parameters for the current iteration are output as the final design parameters. The convergence threshold can be set according to the design optimization accuracy requirements to ensure the stability of the output parameters. Preferably, the automatic modeling and attribute mapping module includes: The interface calling unit is used to call the secondary development interface of the 3D design platform, which includes the PDMS platform or the E3D platform. The model generation unit, connected to the interface calling unit, is used to automatically generate three-dimensional geometric models of pipes, anti-corrosion layers and insulation layers in the three-dimensional design platform based on the parameter set that has passed the verification. The attribute mapping unit, connected to both the multidimensional parameter database module and the model generation unit, establishes a dynamic mapping relationship between parameters in the parameter set and attribute fields of the 3D model. It automatically writes parameter values ​​into the corresponding attribute fields, which include insulation level, heat tracing level, temperature, anti-corrosion layer type, and thickness. This unit automatically converts parameter data types to the platform's attribute field requirements, ensuring that the written parameters can be correctly recognized and used by the platform.

[0014] Preferably, the automatic modeling and attribute mapping module further includes a batch processing unit connected to the model generation unit. This batch processing unit, upon receiving a design task containing multiple pipelines, automatically identifies and distinguishes main pipes and branch pipes based on the corresponding pipeline numbers in the validated parameter set, and generates corresponding 3D models for each pipeline and its associated fittings, thus achieving batch automatic modeling of corrosion-resistant and heat-insulating pipelines. This unit can simultaneously record the modeling results of a single pipeline; failure to model a single pipeline does not affect the modeling process of other pipelines.

[0015] Preferably, the system further includes a data bus module. The multidimensional parameter database module, intelligent parameter verification module, automatic modeling and attribute mapping module, multiphysics coupling simulation module, and optimization feedback and closed-loop control module all interact and transmit commands through the data bus module, forming a closed-loop data flow from parameter input, parameter verification, model generation, simulation analysis to optimization feedback. The multidimensional parameter database module is also used to store the final design parameters and their corresponding simulation results in a historical project database after the optimization feedback and closed-loop control module outputs the final design parameters, for subsequent use in training machine learning models. The data bus module can add timestamps and unique identifiers to the transmitted data packets to ensure the traceability of the data flow process.

[0016] This invention provides a system for simulating the performance and designing the structure of corrosion-resistant and heat-insulating pipeline products. It offers the following advantages: This performance simulation and structural design system for corrosion-resistant and thermally insulated pipeline products avoids issues such as parameter omissions, input errors, or non-standard formats caused by human negligence during the design process, ensuring the accuracy and reliability of the parameter sets input to subsequent modules. Simultaneously, it eliminates the tedious manual modeling and repetitive parameter input by designers in the 3D design platform, significantly shortening the modeling cycle and providing an accurate model foundation for multiphysics coupling simulation. The multiphysics coupling simulation module performs coupled analysis of heat conduction, stress field, and corrosion life through parallel computing, and sets up a simulation verification unit to call at least two different numerical solution methods to cross-compare and perform secondary verification of the simulation results, ensuring the numerical credibility of the simulation calculation results and providing solid data support for subsequent optimization decisions.

[0017] This anti-corrosion and thermal insulation pipeline product performance simulation and structural design system automatically compares simulation results with preset design targets through an optimization feedback and closed-loop control module. When targets are not met, a multi-objective genetic optimization algorithm is triggered to automatically adjust design parameters. The adjusted parameters are then re-inputted into the verification, modeling, and simulation process through a loop control unit, forming a closed-loop self-optimization mechanism from parameter input to final design output. The convergence judgment subunit determines iterative convergence by two consecutive objective function changes falling below a preset threshold, ensuring the stability and optimality of the output results. The data bus module coordinates data interaction between modules and stores the final design parameters and all process data in a historical project database for incremental training of subsequent machine learning models. This enables the system to self-evolve, continuously improving the accuracy of parameter rationality assessment and optimization recommendations, providing an automated, highly reliable, and traceable integrated solution for industrial pipeline anti-corrosion and thermal insulation design. Attached Figure Description

[0018] Figure 1 This is an optimization and iterative closed-loop flowchart of a performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to the present invention. Figure 2 This is a state machine diagram of a performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please see Figure 1 and Figure 2 This invention provides a technical solution: a performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products, comprising: The multidimensional parameter database module is used to store the basic design parameters of the pipeline, including pipeline number, material, medium temperature, pressure, insulation level, anti-corrosion layer type, and environmental conditions. The intelligent parameter verification module, connected to the multidimensional parameter database module, is used to perform at least two rounds of verification on the input original design parameters using different mechanisms, and generate a set of parameters that have passed the verification. The automatic modeling and attribute mapping module, connected to the intelligent parameter verification module, is used to automatically generate a 3D model of the pipeline and anti-corrosion insulation layer in the 3D design platform based on the parameter set that has passed verification, and dynamically assign the parameters in the parameter set to the corresponding attributes of the 3D model. The multiphysics coupling simulation module, connected to the automatic modeling and attribute mapping module, is used to perform heat conduction simulation, stress field simulation, and corrosion life simulation on the generated 3D model to obtain simulation results. The optimization feedback and closed-loop control module is connected to the multiphysics coupling simulation module and the intelligent parameter verification module, respectively. It is used to compare the simulation results with the preset design target. If the design target is not met, the optimization algorithm is triggered to automatically adjust the design parameters and re-input the adjusted design parameters into the intelligent parameter verification module to start a new round of modeling and simulation cycle until the final design parameters that meet the design target are obtained.

[0021] It should be further explained that the system first stores the basic design parameters required for pipeline design in advance through a multi-dimensional parameter database module. These basic design parameters specifically include pipeline number, material, medium temperature, pressure, insulation level, anti-corrosion layer type, and environmental conditions. The multi-dimensional parameter database module has a built-in standard specification library, historical project database, and material property library to provide data support for subsequent parameter verification and optimization.

[0022] The pipeline foundation design parameters stored in the multi-dimensional parameter database module are classified according to the insulation level standards of GB50264-2013 "Code for Design of Thermal Insulation Engineering for Industrial Equipment and Pipelines". Specifically, they are divided into three levels: Level I insulation is suitable for working conditions with medium temperature of 0℃~120℃, Level II insulation is suitable for working conditions with medium temperature of 120℃~350℃, and Level III insulation is suitable for working conditions with medium temperature above 350℃. Each insulation level corresponds to a clear insulation material type, thermal conductivity limit, and minimum thickness requirement. The range of anti-corrosion coating types available includes single-layer epoxy powder coating, epoxy coal tar coating, polyethylene tape coating, epoxy powder + polyethylene composite coating, three-layer polyethylene structural coating, and polyurea elastomer coating. Among them, single-layer epoxy powder coating is suitable for weak corrosion and normal temperature conditions; epoxy coal tar coating is suitable for weak to moderate corrosion conditions of buried pipelines; polyethylene tape coating is suitable for moderate corrosion conditions of overhead pipelines; epoxy powder + polyethylene composite coating is suitable for moderate to strong corrosion conditions; three-layer polyethylene structural coating is suitable for strong corrosion and buried long-distance pipeline conditions; and polyurea elastomer coating is suitable for strong corrosion and high wear conditions. The specific parameters of the environmental conditions include annual average ambient temperature, extreme maximum ambient temperature, extreme minimum ambient temperature, ambient wind speed, soil corrosivity level, atmospheric corrosivity level, soil moisture content of buried pipelines, and annual solar radiation of overhead pipelines. All environmental parameters are determined based on meteorological data and geological survey data of the project site. The specific format specifications for the parameters are as follows: the pipe number adopts the fixed format of "system code-pipe number-medium code", the material parameter adopts the standard steel grade format, the temperature parameter is in °C and retains 1 decimal place, the pressure parameter is in MPa and retains 2 decimal places, the thickness parameter is in mm and retains 1 decimal place, and the value range of all parameters complies with the limit requirements of the corresponding industry standards to ensure the standardization and consistency of the parameters.

[0023] After the designer inputs the original design parameters, the intelligent parameter verification module performs at least two rounds of verification using different mechanisms. The first round of verification is automatically reviewed by the first verification unit, which has a built-in industry standard rule library, to determine whether the parameters are complete and conform to preset rules, such as the matching relationship between medium temperature and insulation level, and the corresponding constraints between anti-corrosion layer type and medium corrosivity. If the parameters violate the rules, a modification prompt is output. The second round of verification is performed by the second verification unit, which uses an anomaly detection machine learning model trained on historical project data, to conduct a probability assessment of reasonableness. This anomaly detection machine learning model uses an isolated forest or autoencoder algorithm to identify whether the parameter combination deviates from the normal distribution of historical data and outputs the corresponding confidence value. If the confidence value is lower than a preset threshold, a manual review process is triggered. After these two rounds of verification using different mechanisms, a parameter set that has passed the verification is generated.

[0024] The automatic modeling and attribute mapping module receives the validated parameter set and calls the secondary development interface of the 3D design platform through the interface calling unit. The 3D design platform includes the PDMS platform or the E3D platform. The model generation unit automatically generates 3D geometric models of pipelines, anti-corrosion layers and insulation layers in the 3D design platform according to the parameter set. At the same time, the attribute mapping unit establishes a dynamic mapping relationship between each parameter in the parameter set and the attribute fields of the 3D model, and automatically writes the parameter values ​​into the corresponding attribute fields. The attribute fields include insulation level, heat tracing level, temperature, anti-corrosion layer type and thickness.

[0025] The multiphysics coupled simulation module performs heat conduction simulation, stress field simulation, and corrosion life simulation on the generated 3D model. The heat conduction simulation unit calculates the temperature field distribution of the 3D model under specified operating conditions. The stress field simulation unit analyzes the stress field distribution caused by thermal expansion of the pipeline based on the temperature field distribution results. The corrosion life simulation unit predicts the failure time of the anti-corrosion layer and the remaining life of the pipeline based on the corrosion model and the simulation results of the temperature field and stress field. The heat conduction simulation unit, stress field simulation unit, and corrosion life simulation unit use parallel computing to perform coupled simulation. During the simulation process, the simulation verification unit calls at least two different numerical solution methods to cross-compare the simulation results of the same physical field. For example, the temperature field is calculated using both the finite element method and the finite volume method. If the output results of the two solution methods deviate from the preset deviation threshold, the mesh density or solver parameters are adjusted and the simulation calculation is repeated until the deviation is reduced to within the preset deviation threshold. This completes the secondary verification of the simulation results and obtains the simulation results.

[0026] The optimization feedback and closed-loop control module compares the simulation results with the preset design objectives, which include the maximum allowable heat loss, minimum safe life, and maximum allowable stress. If the simulation results do not meet the design objectives, the objective comparison unit outputs the difference data. The parameter optimization unit has a built-in multi-objective genetic optimization algorithm that automatically generates at least one set of adjusted design parameters based on the difference data. The loop control unit re-inputs the adjusted design parameters into the intelligent parameter verification module and triggers the automatic modeling and attribute mapping module and the multi-physics coupling simulation module for iterative processing. After each iteration, the convergence judgment subunit calculates the change value of the objective function between the current round and the previous round. If the change value of the objective function is lower than the preset convergence threshold twice in a row, the iteration is judged to have converged and the loop is terminated. The design parameters of the current round are output as the final design parameters. At the same time, the multi-dimensional parameter database module stores the final design parameters and their corresponding simulation results into the historical project database for subsequent training of machine learning models.

[0027] The entire system uses a data bus module to enable data interaction and command transmission between the multidimensional parameter database module, intelligent parameter verification module, automatic modeling and attribute mapping module, multiphysics coupling simulation module, and optimization feedback and closed-loop control module, forming a closed-loop data flow from parameter input, parameter verification, model generation, simulation analysis to optimization feedback.

[0028] The intelligent parameter verification module includes: The first verification unit has a built-in industry standard rule library, which is used to automatically review the completeness and compliance of the original design parameters. If the parameters violate the preset rules in the rule library, a modification prompt will be output. The second verification unit is connected to the first verification unit and has a built-in anomaly detection machine learning model trained based on historical project data. It is used to evaluate the probability of reasonableness of the parameters that pass the first verification unit and output the confidence value of the parameter combination. If the confidence value is lower than the preset threshold, the manual review process is triggered. The first verification unit and the second verification unit perform verification on the original design parameters in sequence, which together constitute at least two rounds of verification with different mechanisms.

[0029] It should be further explained that the intelligent parameter verification module first receives the original design parameters from the multidimensional parameter database module or external input. These original design parameters include basic data such as pipe number, medium temperature, pressure, insulation level, and anti-corrosion layer type.

[0030] The first verification unit has a built-in industry standard rule library that pre-loads and updates multiple design standards and specifications in real time, such as GB / T4272-2024 "General Technical Rules for Thermal Insulation of Equipment and Piping", GB50264-2013 "Code for Design of Thermal Insulation Engineering for Industrial Equipment and Piping", or mandatory clauses in ASME B31.3 process piping specifications. When the original design parameters enter the first verification unit, the unit automatically parses the parameters one by one and matches them with the corresponding rules in the rule library. The specific review content includes, but is not limited to: whether the medium temperature value is within the temperature range allowed by the selected insulation level, whether the anti-corrosion layer type and the medium corrosivity level meet the preset correspondence in the rule library, and whether the pipeline pressure level and wall thickness meet the minimum strength value required by the specification. If any parameter violates the preset rules in the rule library, such as the medium temperature exceeding the upper limit allowed by the insulation level, the first verification unit immediately generates a prompt message containing specific violations and modification suggestions, and outputs it to the designer through the human-computer interaction interface. At the same time, the subsequent process is suspended until the parameters are corrected and resubmitted.

[0031] The rule base pre-loads and updates in real time the mandatory clauses of GB / T4272-2024 "General Technical Rules for Thermal Insulation of Equipment and Piping", GB50264-2013 "Code for Design of Thermal Insulation Engineering for Industrial Equipment and Piping", and ASME B31.3-2024 "Code for Process Piping". The core verification rules include the matching rules between medium temperature and insulation level. Specifically, when the medium temperature is in the range of 0℃ to 120℃, the corresponding insulation level is I, and the thermal conductivity of the insulation material shall not exceed 0.044W / (m·K); when the medium temperature is in the range of 120℃ to 350℃, the corresponding insulation level is II, and the thermal conductivity of the insulation material shall not exceed 0.048W / (m·K); when the medium temperature is higher than 350℃, the corresponding insulation level is III, and the thermal conductivity of the insulation material shall not exceed 0.052W / (m·K). When the input medium temperature does not match the selected insulation level, the rule verification fails. The specific constraint rules for the correspondence between the type of anti-corrosion layer and the corrosivity of the medium are as follows: when the corrosivity level of the medium is weak corrosion, a single-layer epoxy powder coating with a thickness of not less than 200 μm is selected as the anti-corrosion layer; when the corrosivity level of the medium is moderate corrosion, an epoxy powder + polyethylene composite coating with a total thickness of not less than 600 μm is selected as the anti-corrosion layer; when the corrosivity level of the medium is strong corrosion, a three-layer polyethylene structure coating with a total thickness of not less than 1000 μm is selected as the anti-corrosion layer. When the type of anti-corrosion layer does not match the corrosivity level of the medium, the rule verification will fail. The specific strength verification rules for pressure rating and wall thickness are as follows: based on the pipe wall thickness calculation formula in GB50316-2000 "Code for Design of Industrial Metal Pipelines", calculate the minimum pipe wall thickness at the corresponding design pressure and design temperature. When the input pipe wall thickness is less than the calculated minimum wall thickness, the rule verification fails. The triggering logic of the rules is as follows: when the original design parameters enter the first verification unit, the unit matches the corresponding verification rules one by one according to the parameter type. The verification result is output after each rule verification is completed. When any rule verification fails, the subsequent verification process is immediately suspended, and a modification prompt is generated. The modification prompt includes the specific name of the violating parameter, the source of the violating clause, the parameter range required by the specification, and the recommended modification value. This is output to the designer in real time through the human-computer interaction interface until the parameters are corrected and resubmitted for verification.

[0032] For the original design parameters that have passed the review of the first verification unit, the second verification unit receives the parameters and calls the built-in anomaly detection machine learning model to perform a probability assessment of reasonableness. This machine learning model uses pipeline data of completed designs accumulated in the historical project database as the training set. The training algorithm includes the Isolation Forest algorithm or the Autoencoder algorithm to learn the inherent distribution characteristics of normal parameter combinations. During the evaluation, the model maps the input parameter combination to the feature space, calculates its deviation from the distribution center of the training set, and outputs a confidence value between 0 and 1. The lower the confidence value, the more likely the parameter combination is to be an abnormal or erroneous input. For example, the medium temperature of a high-temperature steam pipeline... The temperature is 350℃ and the pressure is 0.1MPa. Although this combination may not violate any individual rules in the rule base, the machine learning model found based on historical data that high temperature and low pressure combinations are extremely rare. Therefore, it gives a confidence level below the preset threshold. At this point, the second verification unit triggers the manual review process and pushes the parameters to the interface of the design reviewer. The reviewer then judges whether the parameter combination needs to be corrected or confirmed. Only when the original design parameters pass the rule review of the first verification unit and the confidence threshold judgment of the second verification unit in sequence, that is, when the confidence level is not lower than the preset threshold and there is no manual review rejection, is the parameter set marked as a parameter set that has passed the verification and output to the automatic modeling and attribute mapping module.

[0033] The anomaly detection machine learning model built into the second verification unit prioritizes the isolated forest algorithm. The input feature dimension of the model training is 8-dimensional, specifically including pipeline medium temperature, design pressure, pipeline nominal diameter, insulation layer thickness, anti-corrosion layer type code, ambient temperature, pipeline material yield strength, and medium corrosivity level. All input features are normalized to the 0-1 range through min-max normalization. The selection criteria for historical project data are as follows: select industrial anti-corrosion and thermal insulation pipeline project data that have been completed and accepted within the past 5 years and are in stable operation, and remove abnormal project data with corrosion failure, thermal insulation failure, and stress exceeding the standard. The final number of training samples is no less than 10,000 sets to ensure that the training dataset covers pipeline design scenarios with different working conditions, different materials, and different environmental conditions. The core parameters of the Isolation Forest algorithm are as follows: the number of trees is set to 100, the maximum depth of a single tree is set to 10, the subsample size is set to 256, and the subsamples are randomly selected from the training dataset without repetition. After the model is trained, anomaly scores are calculated for the input parameter combinations. The anomaly score ranges from 0 to 1. The confidence value is calculated by subtracting the anomaly score from 1. The closer the confidence value is to 1, the higher the rationality of the parameter combination. The normalization rule for confidence scores is as follows: based on the distribution of abnormal ratings in the training dataset, the minimum value of the abnormal rating corresponds to a confidence score of 1, and the maximum value corresponds to a confidence score of 0. The standardization of confidence scores is completed by using linear normalization. The confidence threshold is set based on the abnormal rating distribution of the training dataset. The confidence value corresponding to 95% of the samples in the training dataset is taken as the default threshold, specifically 0.85. When the confidence value of the input parameter combination is lower than 0.85, it is determined that the parameter combination has an abnormal risk and the manual review process is immediately triggered. The parameter combination, confidence calculation result and historical similar project data are pushed to the interface of the design reviewer, and the reasonableness of the parameters are reviewed and confirmed manually. Only when the confidence value is not lower than 0.85 and there is no manual review rejection can the parameter combination pass the test of the second verification unit. If an autoencoder algorithm is used for anomaly detection, the core parameters of the autoencoder are set as follows: the encoder contains 3 fully connected layers with a network layer count of 8-5-3, the decoder contains 3 fully connected layers with a network layer count of 3-5-8, the activation function is ReLU, the output layer uses Sigmoid, the training batch size is set to 64, the training epochs are set to 200, the optimizer is Adam, and the learning rate is set to 0.001. The mean squared error between the input parameters and the reconstructed output is used as the anomaly score. The calculation of the confidence value, the normalization rules, and the threshold setting standards are consistent with those of the Isolation Forest algorithm.

[0034] The sequential connection and collaborative operation of the first and second verification units constitute at least two rounds of verification with different mechanisms. The first round is a deterministic verification based on explicit rules, and the second round is a probabilistic verification based on data-driven mechanisms. The two verification mechanisms are independent of each other and complementary, jointly ensuring the accuracy and reliability of the output parameter set.

[0035] The intelligent parameter verification module also includes a third verification unit, which is connected to the second verification unit. It has a built-in simplified physical model and is used to cross-verify the key parameters that have been verified in the first two rounds. The key parameters include the insulation layer thickness or anti-corrosion layer type corresponding to the pipeline where the medium temperature exceeds the set temperature threshold. The output of the third verification unit serves as an additional basis for judging the rationality of the parameters.

[0036] It should be further explained that after the first verification unit performs rule verification and the second verification unit performs machine learning model verification, the intelligent parameter verification module adds a third verification unit for cross-validation of key parameters.

[0037] The input of the third verification unit is connected to the output of the second verification unit. It is used to receive the parameter set that has passed the first two rounds of verification. This parameter set contains detailed design data such as the medium temperature, insulation layer thickness, and anti-corrosion layer type of each pipeline.

[0038] The third verification unit is pre-configured with a simplified physical model, which is a one-dimensional steady-state calculation model built based on the basic equations of heat transfer and empirical formulas of corrosion kinetics. Specifically, it includes a Fourier thermal conductivity model for calculating the temperature on the outside of the insulation layer and an Arrhenius formula model for estimating the corrosion rate of the anti-corrosion layer.

[0039] The activation of the third verification unit is triggered by both parameter type and parameter value. The system has a preset temperature threshold, which is set to, for example, 350℃ according to the pipe material and design specifications. When the medium temperature value of a certain pipe in the parameter set exceeds the preset temperature threshold, the pipe is automatically marked as a high-temperature pipe, and its corresponding insulation layer thickness and anti-corrosion layer type are identified as key parameters.

[0040] For the identified key parameters, the third verification unit automatically extracts input data such as the medium temperature, ambient temperature, pipe diameter, insulation layer thickness, and anti-corrosion layer material properties of the pipeline, and substitutes them into the simplified physical model for rapid verification. The thermal conductivity model calculates the theoretical value of the outer surface temperature of the insulation layer, and the corrosion model estimates the theoretical corrosion depth of the anti-corrosion layer within a specified number of years based on this temperature value.

[0041] The third verification unit compares the calculation results of the simplified physical model with the corresponding design values ​​in the parameter set. For example, it compares the theoretical corrosion depth with the design thickness of the anti-corrosion layer. If the theoretical corrosion depth exceeds the allowable range of the design thickness, it determines that there is a potential risk in the key parameter.

[0042] The output of the third verification unit serves as an additional criterion for determining the rationality of the parameters. If the verification results of the simplified physical model are consistent with the results of the first two rounds of verification, the reliability of the parameters is enhanced. If the verification results conflict with the results of the first two rounds, for example, if both the rule verification and machine learning verification are deemed qualified but the simplified physical model shows that the corrosion depth exceeds the standard, the third verification unit generates a warning message and returns this message along with the parameter set to the designer's interface, requiring manual review or forced correction of the parameters. Only the parameter set that has been manually confirmed can be finally marked as a parameter set that has passed the verification and output to the subsequent modules.

[0043] By introducing a third verification unit based on a simplified physical model, cross-validation of key parameters of high-risk pipelines at the physical mechanism level is achieved. This mechanism is independent of the rule base and the data-driven model, and constitutes a third round of verification of the original design parameters by different mechanisms.

[0044] The simplified physical models built into the third verification unit include a one-dimensional steady-state Fourier heat conduction model of a multi-layer cylindrical wall and an Arrhenius corrosion rate model. The specific calculation formula for the one-dimensional steady-state Fourier heat conduction model of the multi-layer cylindrical wall is q = 2πλ(t n -t w ) / ln(d n / d w In the formula, q is the heat flow rate per unit pipe length, in W / m, λ is the thermal conductivity of the insulation material, in W / (m·K), and t n Temperature of the medium inside the pipeline, in Kelvin (K) and t. w The temperature of the outer surface of the insulation layer is expressed in Kelvin (K) and d. n The outer diameter of the pipe is expressed in meters (m) and d. w The outer diameter of the insulation layer is in meters (m). The specific boundary conditions for the model are: the inner wall temperature of the pipe equals the medium temperature; and the convective heat transfer coefficient between the outer surface of the insulation layer and the environment is taken as 10 W / (m²). 2 ·K), the ambient temperature is taken as the annual average ambient temperature of the design input, ignoring the axial heat conduction of the pipe, and only considering the radial one-dimensional steady-state heat conduction; The specific form of the Arrhenius corrosion rate formula is v = A × exp(-Ea / (R × T)), where v is the corrosion rate of the anti-corrosion layer in mm / a, and A is the pre-exponential factor. For epoxy powder anti-corrosion coatings, A is taken as 1.2 × 10⁻⁶. 8 For a polyethylene anti-corrosion coating, A is taken as 3.5 × 10 mm / a. 7mm / a, Ea is the activation energy of the reaction. For epoxy powder anti-corrosion coating, Ea is 65000 J / mol, and for polyethylene anti-corrosion coating, Ea is 58000 J / mol. R is the universal gas constant, which is 8.314 J / (mol·K). T is the thermodynamic temperature of the anti-corrosion layer, in K, which is determined by the temperature at the corresponding location calculated by the thermal conductivity model. The specific criteria for model cross-validation are as follows: for high-temperature pipelines with a medium temperature exceeding 350℃, the outer surface temperature of the insulation layer calculated by the thermal conductivity model shall not exceed 50℃. If it exceeds this temperature, the insulation layer thickness parameter is deemed to have potential risks. The cumulative corrosion depth of the anti-corrosion layer within the design service life calculated by the corrosion rate model shall not exceed 80% of the design thickness of the anti-corrosion layer. If it exceeds this proportion, the anti-corrosion layer type or thickness parameter is deemed to have potential risks. The application rules for the model output results are as follows: when the simplified physical model verification results are consistent with the results of the first two rounds of verification, the parameter set directly passes through the third verification unit. When the verification results show that the parameters have potential risks, the third verification unit immediately generates a warning message, clearly indicating the name of the risk parameter, the difference between the verification result and the specification limit, and pushes the warning message and the parameter set together to the designer's interface, requiring manual review or forced correction of the parameters. Only when the manual review confirms that the parameters are reasonable or the parameter correction is passed again can the parameter set complete the entire verification process.

[0045] The multiphysics coupling simulation module includes: The heat conduction simulation unit is used to calculate the temperature field distribution of a three-dimensional model under specified operating conditions. The stress field simulation unit, connected to the heat conduction simulation unit, is used to analyze the stress field distribution caused by thermal expansion of the pipeline based on the temperature field distribution results. The corrosion life simulation unit, connected to the heat conduction simulation unit and the stress field simulation unit, is used to predict the failure time of the anti-corrosion coating and the remaining life of the pipeline based on the corrosion model and the simulation results of the temperature field and stress field. The heat conduction simulation unit, stress field simulation unit, and corrosion life simulation unit are coupled and simulated using parallel computing.

[0046] It should be further explained that after receiving the 3D model output by the automatic modeling and attribute mapping module, the multiphysics coupling simulation module starts the internally integrated heat conduction simulation unit, stress field simulation unit, and corrosion life simulation unit to work together.

[0047] The heat conduction simulation unit first obtains the geometric dimensions of the pipe and the anti-corrosion insulation layer, the thermal conductivity of the material, the temperature of the medium, the ambient temperature, and the surface convective heat transfer coefficient in the three-dimensional model as input boundary conditions. It then uses the finite element method or the finite volume method to mesh the model and solve the steady-state or transient heat conduction equations. The unit calculates the temperature field distribution data of the entire pipeline system from the inner wall to the outer wall and the surrounding space. This temperature field data is stored in the shared memory area of ​​the simulation module in the form of a node temperature matrix.

[0048] The stress field simulation unit is connected to the heat conduction simulation unit, automatically reads the temperature field distribution data mentioned above, and establishes a thermoelastic mechanics equation by combining the elastic modulus, Poisson's ratio, thermal expansion coefficient of the pipe material and the constraint conditions of the pipe. By solving this equation, the thermal stress field distribution generated by the pipe under the action of temperature gradient is obtained. The thermal stress field distribution is specifically manifested as the equivalent stress value and deformation of various parts of the pipe.

[0049] The corrosion life simulation unit is connected to the heat conduction simulation unit and the stress field simulation unit, respectively, and reads temperature field distribution data and stress field distribution data simultaneously. The corrosion life simulation unit has a pre-built library of various corrosion models, including the Faraday law model for uniform corrosion, the Paris formula model for stress corrosion cracking, and the Arrhenius model for high-temperature oxidation. The corrosion life simulation unit automatically selects the matching corrosion model according to the material type of the pipeline anti-corrosion layer and the corrosion characteristics of the medium. It uses temperature and stress values ​​as model input parameters to calculate the corrosion rate of the anti-corrosion layer in the service environment. Then, based on the corrosion rate and the design thickness of the anti-corrosion layer, it predicts the failure time of the anti-corrosion layer. At the same time, it predicts the remaining life of the pipeline body based on the relationship between the corrosion rate and time.

[0050] During the simulation process, the heat conduction simulation unit, stress field simulation unit, and corrosion life simulation unit are not executed sequentially, but rather coupled in parallel computing. That is, the three units perform their respective calculation tasks within the same time step and exchange intermediate data in real time through shared memory or message passing interfaces. For example, when the stress field simulation unit needs to update the mesh node coordinates during the calculation, it will notify the heat conduction simulation unit to adjust the corresponding boundary conditions in real time. When the corrosion life simulation unit detects an abnormal local corrosion rate during the calculation, it will trigger the stress field simulation unit to refine the mesh and recalculate the region. Through this bidirectional data interaction, true coupling of multiple physics fields is achieved, and the final output is a multi-dimensional simulation result dataset containing temperature field, stress field, corrosion rate field, and life prediction results.

[0051] The corrosion lifetime simulation unit of the multiphysics coupling simulation module has a built-in corrosion model library that includes the Faraday law model for uniform corrosion, the Paris formula model for stress corrosion cracking, and the Arrhenius model for high-temperature oxidation. The specific calculation formula for the Faraday law model is v. corr =(M×i corr ) / (n×F×ρ), where v corr The corrosion rate is expressed as a uniform rate in mm / a, where M is the molar mass of the pipe metal material in g / mol, and for carbon steel it is taken as 55.85 g / mol. corr Corrosion current density, in A / cm² 2 n is the number of electrons gained or lost in the metal corrosion reaction, which is taken as 2 for the iron oxidation reaction of carbon steel; F is the Faraday constant, which is taken as 96485 C / mol; and ρ is the density of the metallic material, in g / cm³. 3 For carbon steel, the value is 7.85 g / cm³. 3 This model is applicable to uniform corrosion conditions where the inner wall of a pipeline is in contact with corrosive media. The parameter selection criteria are based on the pH value, chloride ion concentration, and temperature of the medium to determine the corresponding corrosion current density. The specific calculation formula for the Paris formula model is d. a / d N =C×(ΔK) m In the formula d a / d N ΔK represents the fatigue crack propagation rate in m / cycle, and ΔK represents the stress intensity factor amplitude in MPa·m. 0.5 C and m are the fatigue performance constants of the material. For carbon steel, C is taken as 3.5 × 10⁻⁶. -10 The value of m is 3.2. This model is applicable to the stress corrosion cracking condition of pipelines caused by thermal stress cycle. The parameter selection standard is to determine the corresponding fatigue performance constant based on the tensile test data of the pipeline material. The specific calculation formula of the Arrhenius model is consistent with the corrosion rate formula built into the third verification unit. It is applicable to the thermo-oxidative aging corrosion of the anti-corrosion layer under high temperature environment. The parameter selection standard is consistent with the value selection rule of the third verification unit. The mapping rules between temperature field and stress field node data and corrosion model input parameters are as follows: After the heat conduction simulation unit and stress field simulation unit complete the calculation, the temperature values ​​and equivalent stress values ​​of each grid node of the pipeline and anti-corrosion layer are directly mapped to the temperature input parameters and stress correction coefficients at the corresponding positions in the corrosion model through the one-to-one correspondence of grid node numbers. The stress correction coefficient is set as follows: when the equivalent stress is less than 50% of the material yield strength, the correction coefficient is 1.0; when the equivalent stress is in the range of 50% to 80% of the material yield strength, the correction coefficient is 1.5; and when the equivalent stress exceeds 80% of the material yield strength, the correction coefficient is 2.5. The implementation of the thermal-stress-corrosion multiphysics coupling is a two-way coupling. The collaborative logic of parallel computing is as follows: parallel computing is performed using the same time step, which is set to 1 hour according to the simulation conditions. Within each time step, the heat conduction simulation unit, stress field simulation unit, and corrosion lifetime simulation unit start calculations simultaneously. After the heat conduction simulation unit completes the temperature field calculation, it synchronizes the node temperature data to the stress field simulation unit and corrosion lifetime simulation unit in real time through shared memory. After the stress field simulation unit completes the stress field calculation based on the updated temperature field, it synchronizes the node stress data to the corrosion lifetime simulation unit in real time. After the corrosion lifetime simulation unit completes the corrosion rate calculation based on the updated temperature and stress data, if the calculated local corrosion depth exceeds 10% of the anti-corrosion layer thickness, it triggers the heat conduction simulation unit and stress field simulation unit in reverse to refine the mesh of the area and recalculate the temperature and stress fields of the corresponding area, thereby completing the two-way coupling parallel computing of multiphysics.

[0052] The multiphysics coupled simulation module also includes a simulation verification unit, which is connected to the heat conduction simulation unit, stress field simulation unit, and corrosion life simulation unit, respectively. It is used to call at least two different numerical solution methods to cross-compare the simulation results of the same physical field. If the output results of the two solution methods deviate from the preset deviation threshold, the mesh density or solver parameters are adjusted and the simulation calculation is re-performed until the deviation is reduced to within the preset deviation threshold, thus completing the secondary verification of the simulation results.

[0053] It should be further explained that the simulation verification unit set up inside the multiphysics coupling simulation module is automatically activated after the heat conduction simulation unit, stress field simulation unit, or corrosion life simulation unit completes the preliminary simulation calculation, and is used to cross-verify the reliability of the simulation results.

[0054] Taking the temperature field output by the heat conduction simulation unit as an example, the simulation verification unit first reads the mesh generation scheme, solver type and convergence conditions used in the calculation of the unit, and then calls at least two different numerical solution methods to recalculate the same physical model and the same boundary conditions. The first solution method uses the same finite element method as the heat conduction simulation unit but adjusts the element type or mesh density. The second solution method uses the finite volume method or boundary element method, which are completely different from the finite element method algorithm. The two solution methods run independently and generate their own temperature field distribution data respectively.

[0055] The simulation verification unit compares the results output by the two solution methods point by point at the same spatial node location, calculates the temperature difference at each node, and counts the root mean square or maximum absolute value of the differences at all nodes. If the maximum absolute value or root mean square exceeds the preset deviation threshold, for example, the preset deviation threshold is set to 5 Kelvin or the relative deviation is 5%, then the simulation result is determined to have unacceptable numerical error.

[0056] After the deviation exceeds the limit, the simulation verification unit automatically triggers the adjustment mechanism, which includes two parallel measures. First, it sends an instruction to the heat conduction simulation unit, requiring it to refine the mesh of the original model, especially in areas with drastic temperature gradient changes. Second, it adjusts the solver parameters, such as lowering the residual convergence criterion of the finite element method or increasing the upper limit of the iteration steps of the finite volume method. After the adjustment is completed, the simulation verification unit drives the heat conduction simulation unit to re-perform the simulation calculation and performs a cross-comparison of the above two solution methods on the re-calculated output results.

[0057] The above adjustment and recalculation process is repeated until the deviation between the output results of the two solution methods is reduced to within the preset deviation threshold. At this time, the simulation verification unit marks the final temperature field data as having passed the second verification and allows the data to be output as a valid simulation result to the optimization feedback and closed-loop control module or for other subsequent processing.

[0058] For stress field simulation and corrosion life simulation, the simulation verification unit uses the same mechanism for cross-verification. For stress field, the two solution methods can be the finite element method and the finite difference method. For corrosion life simulation, the analytical solution based on the Paris formula and the numerical solution based on finite element damage accumulation can be compared to ensure that the simulation results of all physical fields are double-verified before being adopted.

[0059] The optimized feedback and closed-loop control module includes: The target comparison unit is used to compare the simulation results with the preset design targets one by one. The design targets include the maximum allowable heat loss, the minimum safe life, and the maximum allowable stress. The parameter optimization unit is connected to the target comparison unit and has a built-in multi-objective genetic optimization algorithm. When the simulation results do not meet the design objectives, it automatically generates at least one set of adjusted design parameters based on the difference data of the target comparison unit. The cyclic control unit is connected to the parameter optimization unit and the intelligent parameter verification module respectively. It is used to re-input the adjusted design parameters into the intelligent parameter verification module and trigger the automatic modeling and attribute mapping module and the multiphysics coupling simulation module to perform iterative processing until the target comparison unit determines that the simulation results meet the design objectives and outputs the final design parameters.

[0060] It should be further explained that after receiving the simulation result dataset output by the multiphysics coupling simulation module, the optimization feedback and closed-loop control module first uses the target comparison unit to compare the simulation results with the preset design target item by item.

[0061] The target comparison unit pre-stores the design objectives determined in the design task phase. These objectives include the maximum allowable heat loss, minimum safe service life, and maximum allowable equivalent stress for the pipeline. The target comparison unit automatically parses the heat flux density data in the simulation results and converts it into a heat loss value per unit length. It then compares this heat loss value with the maximum allowable heat loss value. Simultaneously, it extracts the corrosion failure time of the anti-corrosion layer output by the corrosion life simulation unit as the predicted safe service life value and compares it with the minimum safe service life. It also extracts the maximum equivalent stress value output by the stress field simulation unit and compares it with the maximum allowable equivalent stress value. If any of the three indicators fails to meet the corresponding design objective—for example, if the actual heat loss value is higher than the maximum allowable heat loss value or the predicted service life is lower than the minimum safe service life—the target comparison unit generates difference data containing specific difference values ​​and exceedance items, and sends this difference data to the parameter optimization unit.

[0062] After receiving the difference data, the parameter optimization unit calls the built-in multi-objective genetic optimization algorithm to start parameter adjustment. The initial population of this multi-objective genetic optimization algorithm consists of the currently verified parameter set and the historical parameters of similar working conditions stored in the multi-dimensional parameter database module. The fitness function of the algorithm aims to reduce the gap between the simulation results and the design target. Specifically, it aims to make the heat loss value approach below the maximum allowable heat loss value, make the life approach above the minimum safe lifespan, and make the stress approach below the maximum allowable equivalent stress value. Through selection, crossover, and mutation operations, at least one set of adjusted design parameters is generated iteratively. The adjusted design parameters include the increase or decrease in the thickness of the insulation layer, the replacement scheme of the anti-corrosion layer material type, or the adjustment value of the heat tracing level.

[0063] The parameter optimization unit sends at least one set of adjusted design parameters to the loop control unit. The loop control unit first determines the current iteration round and uses the adjusted design parameters as the input data for the new round. It then resends the adjusted parameters to the intelligent parameter verification module via the data bus module. The intelligent parameter verification module re-executes rule checks, machine learning model checks, and optional simplified physical model cross-validation on the adjusted design parameters to ensure that the adjusted parameters themselves meet the specifications and have physical rationality. The verified parameter set then enters the automatic modeling and attribute mapping module and the multiphysics coupling simulation module for a new round of model reconstruction and simulation analysis. The generated new simulation results are then compared by the target comparison unit. The loop control unit records the difference data of each iteration and compares the new simulation results with the simulation results of the previous round. Through this loop feedback mechanism, the system automatically completes the iterative optimization of the design parameters until the target comparison unit determines that the simulation results meet all design objectives.

[0064] The multi-objective genetic optimization algorithm built into the parameter optimization unit of the optimized feedback and closed-loop control module has three sub-objective functions, namely the heat loss objective function f1=|q real -q max | / q max Lifetime objective function f2=|L min -L real | / L min The objective function for stress is f3 = |σ real -σ max | / σ max In the formula q real The actual heat loss per unit pipe length obtained from simulation is expressed in W / m, q max The maximum allowable heat loss for the design objective, expressed in W / m, L. real The actual safe life of the pipeline obtained from the simulation is expressed in units of a and L. min The minimum safe life required by the design objectives, expressed in units of a and σ. real The maximum equivalent stress in the pipeline obtained from the simulation is expressed in MPa, σ. max The maximum equivalent stress allowed for the design objective, expressed in MPa; The overall objective function is F = w1 × f1 + w2 × f2 + w3 × f3, where w1, w2, and w3 are the weight coefficients of the three sub-objective functions. The weight coefficients are set based on the core requirements of the design project. When the core requirement of the project is energy consumption control, w1 is 0.5, w2 is 0.3, and w3 is 0.2. When the core requirement of the project is operational safety life, w1 is 0.2, w2 is 0.6, and w3 is 0.2. When the core requirement of the project is structural safety, w1 is 0.2, w2 is 0.2, and w3 is 0.6. Under the default operating conditions, the weight coefficients are all set to 1 / 3. The core operating parameters of the genetic algorithm are as follows: population size is set to 50, crossover probability is set to 0.8, mutation probability is set to 0.05, roulette wheel selection is used as the selection operator, single-point crossover is used as the crossover operator, uniform mutation is used as the mutation operator, and the maximum number of iterations is set to 100. The specific range and constraints for the optimizable parameters are as follows: the insulation layer thickness ranges from 20mm to 200mm, with the constraint that it must not be less than the minimum insulation thickness required by the specification under the corresponding medium temperature; the anti-corrosion layer thickness ranges from 200μm to 2000μm, with the constraint that it must not be less than the minimum anti-corrosion layer thickness required by the specification under the corresponding medium corrosion level; and the thermal conductivity of the insulation material ranges from 0.03W / (m·K) to 0.06W / (m·K), with the constraint that it must meet the thermal conductivity limit for the corresponding insulation level. The specific calculation rule for the fitness function is: fitness value = 1 / (1+F), where F is the comprehensive objective function value. The closer the fitness value is to 1, the closer the set of parameters is to the design goal. In each iteration, the algorithm prioritizes retaining the top 30% of individuals with the highest fitness values ​​and generates the next generation population through selection, crossover, and mutation operations to complete the iterative optimization of parameters.

[0065] The loop control unit has a convergence determination subunit. After each iteration, the convergence determination subunit calculates the change value of the objective function between the current iteration and the previous iteration. If the change value of the objective function is lower than the preset convergence threshold twice in a row, the iteration is determined to be converged, the loop is terminated, and the design parameters of the current iteration are output as the final design parameters.

[0066] It should be further explained that the convergence determination subunit set inside the loop control unit is automatically activated after each iteration to determine whether the current iteration process has reached the convergence state, thereby terminating the loop.

[0067] At the beginning of each iteration, the loop control unit assigns an iteration number to the current iteration and records the set of design parameters input in this iteration, as well as the difference data between the design target and the output of the target comparison unit. The difference data specifically includes the difference between the heat loss value and the maximum allowable heat loss value, the difference between the predicted life and the minimum safe life, and the difference between the maximum equivalent stress value and the maximum allowable equivalent stress value. The convergence judgment subunit sums these differences according to preset weight coefficients to calculate a scalar objective function value. This objective function value is used to quantify the comprehensive gap between the current design parameter set and the ideal design target.

[0068] After completing all simulation and comparison processes in this round, the convergence determination subunit temporarily stores the calculated objective function value in the internal register and compares it with the objective function value calculated in the previous iteration. The absolute value of the difference between the two is then used as the change value of the objective function.

[0069] If the current iteration is the first iteration, there is no data from the previous iteration for comparison. The convergence determination subunit is marked as non-converged by default, and the loop control unit is allowed to continue executing the next iteration.

[0070] Starting from the second iteration, after obtaining the objective function value of this round, the convergence determination subunit performs the above-mentioned difference calculation with the objective function value of the previous round, and compares the change value of the objective function with the internally preset convergence threshold. The convergence threshold is a positive number set according to engineering experience, such as one-thousandth of the initial objective function value or a fixed decimal.

[0071] When the convergence determination subunit detects that the change in the objective function between the current round and the previous round is lower than the preset convergence threshold, it does not immediately determine convergence, but records the state as the first time the target has been met, and continues to the next round of iteration.

[0072] After the next iteration, the convergence determination subunit recalculates the new objective function change value. If the value is lower than the preset convergence threshold again, it means that the objective function change values ​​of the two consecutive iterations meet the convergence condition. At this time, the convergence determination subunit formally determines that the iteration process has entered the stable convergence region.

[0073] Once convergence is determined, the convergence determination subunit immediately sends a termination command to the loop control unit. Upon receiving the command, the loop control unit stops calling the intelligent parameter verification module, the automatic modeling and attribute mapping module, and the multiphysics coupling simulation module for a new round of iteration, and outputs the design parameter set of the current round as the final design parameters to the multidimensional parameter database module and the user interface.

[0074] By setting a mechanism that determines convergence only after two consecutive values ​​fall below the threshold, misjudgments caused by single-value fluctuations are effectively avoided, ensuring that the final output set of design parameters remains a stable optimization result across multiple iterations.

[0075] The weighted summation formula of the objective function of the convergence judgment subunit is consistent with the comprehensive objective function F of the parameter optimization unit, and the value standard of the weight coefficient is completely consistent with the weight setting rule of the parameter optimization unit, so as to ensure the consistency of the objective function during the iteration process. The preset convergence threshold is set based on the objective function value of the initial iteration round, specifically 0.1% of the comprehensive objective function value of the initial round. When there are no special accuracy requirements for the design objective, the fixed convergence threshold value is 0.001. The specific logic verification rule for determining convergence after two consecutive successful convergences is as follows: After each iteration, the convergence determination subunit calculates the comprehensive objective function value F(n) of the current iteration and calculates the objective function change value ΔF = |F(n) - F(n-1)| with the comprehensive objective function value F(n-1) of the previous iteration. When ΔF is detected to be lower than the preset convergence threshold for the first time, it is marked as the first successful convergence, the iteration is not terminated, and the next iteration continues. After the next iteration, the new objective function change value ΔF(n+1) = |F(n+1) - F(n)| is calculated again. If ΔF(n+1) is still lower than the preset convergence threshold, it is determined that the iteration process has achieved the target twice consecutively and has entered the stable convergence region. A termination command is immediately sent to the loop control unit to terminate the iteration loop. If ΔF(n+1) is higher than the preset convergence threshold, the mark of the first successful convergence is cleared, and the accumulation of the number of successful convergences is restarted. This avoids convergence misjudgment caused by fluctuations in the single value and ensures that the output design parameters are stable convergence optimization results.

[0076] The automatic modeling and attribute mapping module includes: The interface calling unit is used to call the secondary development interface of the 3D design platform, which includes the PDMS platform or the E3D platform. The model generation unit, connected to the interface call unit, is used to automatically generate three-dimensional geometric models of pipes, anti-corrosion layers, and insulation layers in the three-dimensional design platform based on the validated parameter set. The attribute mapping unit is connected to the multidimensional parameter database module and the model generation unit respectively. It is used to establish a dynamic mapping relationship between each parameter in the parameter set and the attribute fields of the three-dimensional model, and automatically write the parameter values ​​into the corresponding attribute fields. The attribute fields include thermal insulation level, heat tracing level, temperature, anti-corrosion layer type and thickness.

[0077] It should be further explained that after receiving the set of parameters that have passed the verification from the intelligent parameter verification module, the automatic modeling and attribute mapping module first identifies the type and version of the 3D design platform running in the current computer system by the interface calling unit. Specifically, the 3D design platform is either the PDMS platform or the E3D platform. The interface calling unit automatically loads the corresponding dynamic link library file or script interface file according to the platform type and establishes a data communication channel with the kernel of the 3D design platform.

[0078] The interface calling unit calls the secondary development function library provided by the 3D design platform through this communication channel. In the PDMS platform, this function library is represented as an executable script set written in PML language, and in the E3D platform, it is represented as a set of API interfaces under the .NET framework.

[0079] The model generation unit sends modeling instructions to the 3D design platform through the interface call unit. Based on the geometric information such as pipe number, pipe diameter, wall thickness, and direction coordinates contained in the validated parameter set, it sequentially creates pipe system nodes, pipe area nodes, main pipe nodes, branch pipe nodes, and fitting nodes in the database hierarchy of the 3D design platform and generates corresponding 3D geometric entities. At the same time, based on the insulation level and anti-corrosion layer type in the parameter set, it calls the insulation layer generation function and the anti-corrosion layer generation function to create a cylindrical shell with a specific thickness as the insulation layer model around the pipe geometric entity, and a coating with specific material properties as the anti-corrosion layer model on the outer surface of the insulation layer or the inner surface of the pipe.

[0080] The attribute mapping unit is activated immediately after the model generation unit creates the entity. The attribute mapping unit first reads the field mapping table stored in the multidimensional parameter database module. This field mapping table predefines the correspondence between each parameter name in the parameter set and the corresponding attribute field name in the 3D design platform. For example, "medium temperature" in the parameter set corresponds to "TEMP" in the platform attribute, "insulation level" corresponds to "INSULATION_LEVEL", and "corrosion protection type" corresponds to "COATING_TYPE". The attribute mapping unit locates the newly generated 3D model entity according to the pipe number, then iterates through each item in the mapping table, and calls the platform's attribute writing function through the interface call unit to automatically write the corresponding parameter values ​​in the parameter set into the specified attribute fields of the 3D model entity. After the writing is completed, the design parameters of the pipe can be directly displayed in the attribute panel of the 3D model, realizing the synchronization between the 3D model and the design parameters.

[0081] During the writing process, the attribute mapping unit is also responsible for data type conversion, such as converting character-type insulation level data in the parameter set into enumerated data required by the platform attributes, or converting floating-point temperature data into double-precision floating-point data required by the platform attributes, to ensure that the data can be correctly recognized and displayed by the platform after it is written.

[0082] The automatic modeling and attribute mapping module also includes a batch processing unit, which is connected to the model generation unit. When a design task containing multiple pipelines is received, the batch processing unit automatically identifies and distinguishes between main pipes and branch pipes based on the corresponding pipeline numbers in the parameter set that has passed verification, and generates corresponding three-dimensional models for each pipeline and its auxiliary fittings, thereby realizing batch automatic modeling of corrosion-resistant and heat-insulating pipelines.

[0083] It should be further explained that the batch processing unit set up inside the automatic modeling and attribute mapping module is activated first after receiving a design task containing multiple pipelines. The design task is specifically manifested as a list file or database query result set containing multiple pipeline numbers and their corresponding validated parameter sets.

[0084] The batch processing unit parses the list file, extracts all pipe numbers to be modeled, and automatically identifies the role of each pipe in the overall process flow based on the naming rules of the pipe numbers or the pipe hierarchy stored in the database. The identification rules specifically include: if a pipe number is from the outlet of an upstream device to the inlet of a downstream device and has no branches, it is marked as a main pipe; if a pipe number is attached to another pipe number and the connection direction is from the main pipe to the steam consumption point or equipment, it is marked as a branch pipe; if a pipe number has both the characteristics of a main pipe and a branch pipe, it is further judged according to the preset priority rules.

[0085] After identification, the batch processing unit sorts the pipelines according to the hierarchical relationship of the pipeline system, first processing all main pipes, and then processing the branch pipes corresponding to each main pipe.

[0086] For each pipeline to be modeled, the batch processing unit calls the model generation unit and the attribute mapping unit to repeatedly execute the process of creating 3D geometric entities and writing design parameters.

[0087] When creating the main pipe model, the batch processing unit generates a continuous three-dimensional solid of the main pipe based on the starting and ending point coordinates, pipe diameter and direction data, and at the same time generates models of pipe fittings such as elbows, tees and flanges attached to the main pipe.

[0088] When creating the branch pipe model, the batch processing unit first reads the coordinates of the connection point between the branch pipe and the main pipe to ensure that the starting end face of the branch pipe is precisely aligned with the outer wall face of the main pipe or the end face of the tee outlet, and automatically adjusts the length and direction of the branch pipe to avoid geometric interference. The models of valves, joints and other pipe fittings on the branch pipe are also generated simultaneously.

[0089] During the processing, the batch processing unit monitors the response status of the 3D design platform in real time. If a pipeline fails to model due to parameter errors or geometric conflicts, the pipeline is skipped and the error information is recorded in the log file. At the same time, the subsequent pipelines are processed. After all pipelines are processed, the batch processing unit summarizes the records of success and failure and generates a modeling report.

[0090] For each pipeline that is successfully modeled, the batch processing unit ensures that all fields in its parameter set have been written into the corresponding attributes through the attribute mapping unit. The subordinate relationship between the main pipe and the branch pipe is also clearly recorded in the attributes by means of links or references, ultimately realizing the batch automatic modeling of all anti-corrosion and heat-insulating pipelines in the entire design task.

[0091] The field mapping table in the automatic modeling and attribute mapping module predefines the specific correspondence rules between parameters in the parameter set and attribute fields in the PDMS / E3D platform. Specifically, the "Pipe Number" in the parameter set corresponds to the "NAME" attribute field in the PDMS platform and the "ItemCode" attribute field in the E3D platform; the "Medium Temperature" in the parameter set corresponds to the "TEMP" attribute field in the PDMS platform and the "DesignTemperature" attribute field in the E3D platform; the "Design Pressure" in the parameter set corresponds to the "PRES" attribute field in the PDMS platform and the "DesignPressure" attribute field in the E3D platform; and the "Insulation Rating" in the parameter set corresponds to the "..." attribute field in the PDMS platform. The "INSULATION_LEVEL" attribute field corresponds to the "InsulationGrade" attribute field in the E3D platform. The "Coating Type" attribute field in the parameter set corresponds to the "COATING_TYPE" attribute field in the PDMS platform and the "CoatingType" attribute field in the E3D platform. The "Insulation Thickness" attribute field in the parameter set corresponds to the "INS_THK" attribute field in the PDMS platform and the "InsulationThickness" attribute field in the E3D platform. The "Coating Thickness" attribute field in the parameter set corresponds to the "COAT_THK" attribute field in the PDMS platform and the "CoatingThickness" attribute field in the E3D platform. The implementation logic of dynamic mapping is as follows: the attribute mapping unit establishes a one-to-one correspondence between the parameter set and the attribute fields of the 3D model based on the field mapping table. When the parameter value in the parameter set is updated, the attribute writing instruction is automatically triggered, and the updated parameter value is synchronously written into the attribute field of the corresponding 3D model, so as to realize the dynamic linkage between parameters and model attributes. The specific data type conversion rules are as follows: string type parameters in the parameter set are converted into enumeration type data required by the platform attributes; numeric type parameters in the parameter set are converted into double-precision floating-point type data required by the platform attributes; and boolean type parameters in the parameter set are converted into logical type data required by the platform attributes, so as to ensure that the parameters can be correctly recognized and called by the platform after they are written. The specific rules for identifying main pipes and branches during batch modeling are as follows: based on the naming rules of pipe numbers, pipe numbers containing the main pipe code "ZS" are marked as main pipes, and pipe numbers containing the branch code "ZS-XX" are marked as branches of the corresponding main pipes. At the same time, auxiliary identification is based on the flow parameters and pipe diameter parameters of the pipes. When two pipes are connected to each other, the pipe with a larger diameter and a larger design flow rate is marked as the main pipe, and the pipe with a smaller diameter and a smaller design flow rate is marked as a branch pipe. The geometric interference avoidance logic works as follows: when generating the branch pipe model, the coordinates of the connection point between the main pipe and the branch pipe are first read to ensure that the starting end face of the branch pipe is completely in contact with the outer wall of the main pipe or the end face of the tee outlet. At the same time, based on the outer diameter of the pipe, a minimum safety distance of 1.5 times the outer diameter of the pipe is set. When the distance between the generated branch pipe model and the surrounding pipes and equipment is less than the minimum safety distance, the direction and turning radius of the branch pipe are automatically adjusted to avoid the interference area. After the adjustment is completed, interference is checked again until there are no geometric interference problems, ensuring that the batch-generated 3D models meet the engineering design requirements.

[0092] The system also includes a data bus module, a multidimensional parameter database module, an intelligent parameter verification module, an automatic modeling and attribute mapping module, a multiphysics coupling simulation module, and an optimization feedback and closed-loop control module. All of these modules interact and transmit commands through the data bus module, forming a closed-loop data flow from parameter input, parameter verification, model generation, simulation analysis to optimization feedback. The multidimensional parameter database module is also used to store the final design parameters and their corresponding simulation results into a historical project database after the optimization feedback and closed-loop control module outputs the final design parameters, for use in subsequent training of machine learning models.

[0093] It should be further explained that the data bus module set up inside the system serves as the central channel for data interaction and command transmission between all modules. This data bus module adopts a service-oriented architecture and realizes real-time communication between the multidimensional parameter database module, intelligent parameter verification module, automatic modeling and attribute mapping module, multiphysics coupling simulation module, and optimization feedback and closed-loop control module through message queues or shared memory mechanisms.

[0094] When designers input raw design parameters through the user interface, the parameters are first sent to the data bus module. The data bus module forwards the parameters to the intelligent parameter verification module for processing based on the target address identifier in the message header.

[0095] After completing one or more rounds of verification, the intelligent parameter verification module packages the generated set of parameters that have passed verification and the corresponding verification logs into a data packet, and sends it back to the multidimensional parameter database module for temporary storage via the data bus module. At the same time, the data bus module routes the parameter set to the automatic modeling and attribute mapping module.

[0096] After the automatic modeling and attribute mapping module completes the modeling and attribute assignment, it sends a modeling completion signal to the data bus module and pushes the identification information of the 3D model and the parameter set association data to the multiphysics coupling simulation module.

[0097] The multiphysics coupling simulation module starts the simulation calculation and reads the physical property parameters stored in the multidimensional parameter database module in real time through the data bus module as supplementary input during the simulation. After the simulation is completed, the simulation result dataset is sent to the data bus module.

[0098] The optimization feedback and closed-loop control module subscribes to simulation results from the data bus module, performs target comparison and optimization iteration, and the adjusted design parameters generated in each iteration are resent to the intelligent parameter verification module through the data bus module to start a new cycle. The entire closed-loop data flow flows in an orderly manner under the coordination of the data bus module, ensuring that the input and output relationship between each module is clear and traceable.

[0099] After optimizing the feedback and closed-loop control module to output the final design parameters, these final design parameters, along with all intermediate parameters, verification logs, and simulation results generated throughout the iteration process, are packaged by the data bus module and written into the historical project database in the multidimensional parameter database module. This historical project database is indexed by project number, pipeline number, and timestamp, and stores the original design parameters, verification confidence values ​​for each round, temperature and stress field data for each simulation, objective function values ​​for each iteration, and the final design parameters. This data is used to train the anomaly detection machine learning model in the second verification unit. When a new design task is started, the system can call the data in the historical project database to incrementally train the machine learning model, enabling the model to learn the distribution characteristics of more successful cases, thereby continuously improving the accuracy of parameter rationality assessment.

[0100] This system employs an intelligent parameter verification module to perform at least two rounds of verification on the original design parameters using different mechanisms. These include deterministic review based on a rule base and probabilistic evaluation based on a machine learning model. Combined with a simplified physical model in the third verification unit, key parameters are cross-validated. This effectively avoids parameter omissions, input errors, or non-standard formatting issues caused by human negligence during the design process, ensuring the accuracy and reliability of the parameter sets input to subsequent modules. Simultaneously, the automatic modeling and attribute mapping module enables batch generation of 3D models and automatic parameter assignment, eliminating the tedious manual modeling and repetitive parameter input by designers in the 3D design platform. This significantly shortens the modeling cycle and provides an accurate model foundation for multiphysics coupling simulation. The multiphysics coupling simulation module performs coupled analysis of heat conduction, stress field, and corrosion life through parallel computing. A simulation verification unit calls at least two different numerical solution methods to cross-compare and perform secondary verification of the simulation results, ensuring the numerical credibility of the simulation calculation results and providing solid data support for subsequent optimization decisions.

[0101] Furthermore, this system automatically compares simulation results with preset design targets through an optimized feedback and closed-loop control module. When targets are not met, a multi-objective genetic optimization algorithm is triggered to automatically adjust design parameters. The adjusted parameters are then re-input into the verification, modeling, and simulation process via a loop control unit, forming a closed-loop self-optimization mechanism from parameter input to final design output. The convergence judgment subunit determines iterative convergence by two consecutive objective function changes falling below a preset threshold, ensuring the stability and optimality of the output results. The data bus module coordinates data interaction between modules and stores the final design parameters and all process data in a historical project database for incremental training of subsequent machine learning models. This enables the system to self-evolve, continuously improving the accuracy of parameter rationality assessment and optimization recommendations, providing an automated, highly reliable, and traceable integrated solution for industrial pipeline corrosion protection and insulation design.

[0102] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0103] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A performance simulation and structural design system for corrosion-resistant and heat-insulating pipeline products, characterized in that, include: The multidimensional parameter database module is used to store the basic design parameters of the pipeline, including pipeline number, material, medium temperature, pressure, insulation level, anti-corrosion layer type, and environmental conditions. The intelligent parameter verification module, connected to the multidimensional parameter database module, is used to perform at least two rounds of verification on the input original design parameters using different mechanisms, and generate a set of parameters that have passed the verification. An automatic modeling and attribute mapping module, connected to the intelligent parameter verification module, is used to automatically generate a three-dimensional model of the pipeline and anti-corrosion insulation layer in the three-dimensional design platform based on the parameter set that has passed verification, and dynamically assign the parameters in the parameter set to the corresponding attributes of the three-dimensional model. The multiphysics coupling simulation module is connected to the automatic modeling and attribute mapping module and is used to perform heat conduction simulation, stress field simulation and corrosion life simulation of the generated three-dimensional model to obtain simulation results. The optimization feedback and closed-loop control module is connected to the multiphysics coupling simulation module and the intelligent parameter verification module, respectively. It is used to compare the simulation results with the preset design target. If the design target is not met, the optimization algorithm is triggered to automatically adjust the design parameters, and the adjusted design parameters are re-input into the intelligent parameter verification module to start a new round of modeling and simulation cycle until the final design parameters that meet the design target are obtained.

2. The performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to claim 1, characterized in that: The intelligent parameter verification module includes: The first verification unit has a built-in industry standard rule library, which is used to automatically review the completeness and compliance of the original design parameters. If the parameters violate the preset rules in the rule library, a modification prompt will be output. The second verification unit is connected to the first verification unit and has a built-in anomaly detection machine learning model trained based on historical project data. It is used to evaluate the probability of reasonableness of the parameters that pass the first verification unit and output the confidence value of the parameter combination. If the confidence value is lower than the preset threshold, the manual review process is triggered. The first verification unit and the second verification unit perform verification on the original design parameters in sequence, which together constitute the at least two rounds of verification with different mechanisms.

3. The performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to claim 2, characterized in that: The intelligent parameter verification module also includes a third verification unit, which is connected to the second verification unit and has a built-in simplified physical model. It is used to cross-verify the key parameters that have undergone the first two rounds of verification. The key parameters include the insulation layer thickness or anti-corrosion layer type corresponding to the pipe where the medium temperature exceeds the set temperature threshold. The output of the third verification unit serves as an additional basis for judging the rationality of the parameters.

4. The performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to claim 1, characterized in that: The multiphysics coupling simulation module includes: The heat conduction simulation unit is used to calculate the temperature field distribution of a three-dimensional model under specified operating conditions. The stress field simulation unit is connected to the heat conduction simulation unit and is used to analyze the stress field distribution of the pipeline caused by thermal expansion based on the temperature field distribution results. The corrosion life simulation unit is connected to the heat conduction simulation unit and the stress field simulation unit. It is used to predict the failure time of the anti-corrosion coating and the remaining life of the pipeline based on the corrosion model and the simulation results of the temperature field and stress field. The heat conduction simulation unit, stress field simulation unit, and corrosion life simulation unit are coupled and simulated using parallel computing.

5. The performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to claim 4, characterized in that: The multiphysics coupled simulation module also includes a simulation verification unit, which is connected to the heat conduction simulation unit, stress field simulation unit, and corrosion lifetime simulation unit respectively. It is used to call at least two different numerical solution methods to cross-compare the simulation results of the same physical field. If the output results of the two solution methods deviate from the preset deviation threshold, the mesh density or solver parameters are adjusted and the simulation calculation is re-performed until the deviation is reduced to within the preset deviation threshold, thus completing the secondary verification of the simulation results.

6. The performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to claim 1, characterized in that: The optimized feedback and closed-loop control module includes: The target comparison unit is used to compare the simulation results with the preset design targets one by one. The design targets include the maximum allowable heat loss, the minimum safe life, and the maximum allowable stress. The parameter optimization unit is connected to the target comparison unit and has a built-in multi-objective genetic optimization algorithm. When the simulation results do not meet the design objectives, it automatically generates at least one set of adjusted design parameters based on the difference data of the target comparison unit. The cyclic control unit is connected to the parameter optimization unit and the intelligent parameter verification module, respectively. It is used to re-input the adjusted design parameters into the intelligent parameter verification module and trigger the automatic modeling and attribute mapping module and the multiphysics coupling simulation module to perform iterative processing until the target comparison unit determines that the simulation results meet the design target and outputs the final design parameters.

7. The performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to claim 6, characterized in that: The loop control unit is equipped with a convergence determination subunit. After each iteration, the convergence determination subunit calculates the change value of the objective function between the current iteration and the previous iteration. If the change value of the objective function is lower than the preset convergence threshold twice in a row, the iteration is determined to be converged, the loop is terminated, and the design parameters of the current iteration are output as the final design parameters.

8. The performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to claim 1, characterized in that: The automatic modeling and attribute mapping module includes: The interface calling unit is used to call the secondary development interface of the 3D design platform, which includes the PDMS platform or the E3D platform. The model generation unit, connected to the interface calling unit, is used to automatically generate three-dimensional geometric models of pipes, anti-corrosion layers and insulation layers in the three-dimensional design platform based on the parameter set that has passed the verification. The attribute mapping unit is connected to the multidimensional parameter database module and the model generation unit respectively. It is used to establish a dynamic mapping relationship between each parameter in the parameter set and the attribute fields of the three-dimensional model, and automatically write the parameter values ​​into the corresponding attribute fields. The attribute fields include thermal insulation level, heat tracing level, temperature, anti-corrosion layer type and thickness.

9. The performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to claim 8, characterized in that: The automatic modeling and attribute mapping module also includes a batch processing unit, which is connected to the model generation unit. When a design task containing multiple pipelines is received, the batch processing unit automatically identifies and distinguishes main pipes and branch pipes based on the pipeline numbers corresponding to the verified parameter set, and generates corresponding three-dimensional models for each pipeline and its auxiliary fittings, thereby realizing batch automatic modeling of corrosion-resistant and heat-insulating pipelines.

10. The performance simulation and structural design system for anti-corrosion and heat-insulating pipeline products according to claim 1, characterized in that: The system also includes a data bus module. The multidimensional parameter database module, intelligent parameter verification module, automatic modeling and attribute mapping module, multiphysics coupling simulation module, and optimization feedback and closed-loop control module all interact and transmit commands through the data bus module, forming a closed-loop data flow from parameter input, parameter verification, model generation, simulation analysis to optimization feedback. The multidimensional parameter database module is also used to store the final design parameters and their corresponding simulation results into a historical project database after the optimization feedback and closed-loop control module outputs the final design parameters, for use in subsequent training of machine learning models.

Images