A welding prediction method and device suitable for a new energy natural gas pipeline

Abstract

The application belongs to the technical field of pipeline welding, and specifically discloses a welding prediction method and device suitable for new energy natural gas pipelines. The method comprises the following steps: obtaining a load spectrum containing pressure fluctuation, temperature cycle and hydrogen partial pressure change; setting a target fatigue life threshold; calculating a three-dimensional residual stress field through a thermal-mechanical coupling finite element model; superimposing the load spectrum and calculating the fatigue crack initiation life by using the critical plane method combined with the Smith-Watson-Topper damage model, and introducing a hydrogen environment correction factor to quantify the influence of hydrogen; taking the standard fatigue life as a constraint, optimizing the welding parameters by using an improved non-dominated sorting genetic algorithm; and outputting the optimal parameters to a welding system. The application solves the problem that the traditional welding parameter design lacks dynamic load and hydrogen environment coupling prediction ability, realizes reverse design of process parameters from service requirements, and significantly improves the reliability of welding quality and the efficiency of process development.

Description

Technical Field

[0001] This invention belongs to the field of welding technology, specifically a welding prediction method and device applicable to new energy natural gas pipelines. Background Technology

[0002] New energy natural gas (such as hydrogen-blended natural gas with a hydrogen volume fraction of 5% to 20%) is widely used as a transitional energy source in urban gas supply and long-distance pipeline networks. However, its complex composition poses a severe challenge to the reliability of pipeline welded joints. Traditional welding process design mainly relies on experience or static mechanical performance indicators, such as tensile strength and impact energy, neglecting the cumulative damage effects of dynamic loads on the pipeline during its 30-year service life, such as pressure fluctuations, temperature cycling, and changes in hydrogen partial pressure, on the fatigue performance of welded joints.

[0003] In existing technologies, welding parameter optimization often employs a forward prediction model:

[0004] This involves predicting residual stress or deformation through thermo-mechanical coupling simulation, given parameters such as welding current, voltage, and speed. More importantly, existing methods lack a reverse design approach that starts with the end in mind. In engineering practice, pipeline design specifications explicitly require that the fatigue life of welded joints under a specific load spectrum not be less than 10... 6 The loop continues.

[0005] However, the current process involves welding first and then testing, meaning welding is done according to empirical parameters first, followed by fatigue testing for verification. If the parameters are not met, rework is required, resulting in significant cost waste and project delays. Furthermore, hydrogen in natural gas can exacerbate hydrogen-induced delayed cracking and fatigue acceleration effects in welded joints, and existing models do not incorporate the coupling of the hydrogen environment and dynamic loads into their prediction framework. Summary of the Invention

[0006] The purpose of this invention is to provide a welding prediction method and device suitable for new energy natural gas pipelines, which solves the technical problems of the lack of dynamic load and hydrogen environment coupling prediction capability in the design of welding parameters in the prior art, and the high process rework rate caused by the reliance on experience in traditional methods.

[0007] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0008] A welding prediction method applicable to new energy natural gas pipelines includes the following steps:

[0009] Step 1: Obtain the load spectrum of the target pipeline during its future service life. The load spectrum includes pressure fluctuation time series, temperature cycle data, and hydrogen partial pressure change curve.

[0010] Step 2: Set the target fatigue life threshold for the welded joint.

[0011] Step 3: Construct a thermo-mechanical coupled finite element model of the welded joint, input the initial welding parameter combination, which includes heat input per unit length, interpass temperature and welding speed, and output the three-dimensional residual stress field distribution of the weld and heat-affected zone.

[0012] Step four: Superimpose the residual stress field distribution with the load spectrum, calculate the fatigue crack initiation life of the welded joint using the critical plane method combined with the Smith-Watson-Topper damage model, and introduce a hydrogen environment correction factor, which is expressed as: ,in The hydrogen concentration in the heat-affected zone. The hydrogen sensitivity coefficient of the material.

[0013] Step 5: Using the constraint that the fatigue crack initiation life is greater than or equal to the target fatigue life threshold, minimize the objective function. For the goal, among which and These are the weighting coefficients. For heat input per unit length, For welding speed, meet An improved non-dominated sorting genetic algorithm is used to iteratively optimize the welding parameter combinations, generating the Pareto optimal solution set; in the improved non-dominated sorting genetic algorithm, the weight coefficients... The value range is 0.3-0.7. The value range is 0.3-0.7.

[0014] The objective function is related to fatigue life constraints, and an improved non-dominated sorting genetic algorithm is used to ensure... .

[0015] Step 6: Select the final welding parameter combination from the Pareto optimal solution set. The final welding parameter combination includes the optimized heat input per unit length, the optimized interpass temperature, and the optimized welding speed.

[0016] Step 7: Output the final welding parameter combination to the welding control system or process card generation module through a digital interface.

[0017] In one embodiment of the present invention, the load spectrum in step one is generated by fusing historical operating data of the data acquisition and monitoring control system, meteorological model prediction data and hydrogen source fluctuation model output results. The data fusion adopts a time series alignment algorithm to superimpose the pressure fluctuation, temperature cycle and hydrogen partial pressure change curves on a unified time axis, and generates a multidimensional dynamic load spectrum based on the weighted average method.

[0018] In one embodiment of the present invention, the thermo-mechanical coupled finite element model in step three uses a double ellipsoidal heat source model to describe the welding heat source and sets adiabatic boundary conditions.

[0019] In one embodiment of the present invention, the hydrogen concentration in the heat-affected zone in step four is calculated based on the hydrogen partial pressure change curve in the loading spectrum using Fick's diffusion law. The calculation formula is as follows: ;

[0020] in : The surface equilibrium hydrogen concentration based on changes in hydrogen partial pressure; calculated using Sieverts' law: , This is the hydrogen solubility constant (calibrated through materials experiments). This represents the hydrogen partial pressure in the loading spectrum;

[0021] The hydrogen diffusion coefficient of the material was determined by electrochemical permeation.

[0022] : This represents the hydrogen diffusion time, corresponding to the time series of the loading spectrum;

[0023] The hydrogen sensitivity coefficient k of the material was obtained through preliminary experimental calibration.

[0024] In one embodiment of the present invention, the improved non-dominated sorting genetic algorithm in step five employs adaptive crossover probability and adaptive mutation probability. and mutation probability ,in For the current algebra, This represents the maximum algebraic value. The adaptive probability mechanism can dynamically adjust the search strategy based on the optimization process, avoiding getting trapped in local optima.

[0025] Furthermore, in this embodiment, the improved non-dominated sorting genetic algorithm (NSGA-II), in addition to employing the aforementioned adaptive crossover probability and adaptive mutation probability, also makes targeted adjustments to the population initialization strategy, elite retention mechanism, or crowding distance calculation method for the welding parameter optimization problem involved in this invention. For example, by introducing a population initialization method based on historical convergence trends, the search blind zone is effectively reduced, and the quality of the initial population is improved. These improvements work together to improve the convergence speed of the algorithm compared to the traditional NSGA-II algorithm when dealing with the multi-objective optimization problem of welding parameters.

[0026] In one embodiment of the present invention, the optimization process in step five supports an expanded variable space, which includes at least one of welding current, welding voltage, and wire feed speed.

[0027] In one embodiment of the present invention, the Pareto optimal solution set in step six includes multiple candidate solutions, each of which satisfies the constraint that the fatigue crack initiation life is greater than or equal to the target fatigue life threshold.

[0028] In one embodiment of this invention, the digital interface in step seven uses the Industrial Ethernet protocol to connect with the welding control system. Specifically, the digital interface in step seven uses the Modbus TCP / IP Industrial Ethernet protocol, and transmits the optimized welding parameters to the PLC register of the welding control system through a preset register address mapping table.

[0029] In one embodiment of the present invention, the method supports the deployment of a sensor network through an online monitoring system to collect the actual stress state of the welded joint during its service life, and dynamically corrects the fatigue life prediction model by combining digital twin technology.

[0030] In one embodiment of the present invention, the method supports a distributed computing architecture, enabling parallel processing of large-scale finite element simulation tasks through a cloud computing platform.

[0031] In addition, this invention also discloses a welding prediction device suitable for new energy natural gas pipelines, comprising:

[0032] The target life setting module is used to set the target fatigue life threshold of the welded joint according to industry standards.

[0033] The stress field calculation module is used to construct a thermo-mechanical coupled finite element model of the welded joint and output a three-dimensional residual stress field distribution.

[0034] The fatigue life calculation module is used to calculate the fatigue crack initiation life of welded joints using the critical plane method combined with the Smith-Watson-Topper damage model, and introduces a hydrogen environment correction factor.

[0035] The parameter optimization module is used to iteratively optimize the welding parameter combination using an improved non-dominated sorting genetic algorithm, with the fatigue crack initiation life being greater than or equal to the target fatigue life threshold as a constraint, to generate the Pareto optimal solution set.

[0036] The improved non-dominated sorting genetic algorithm has a population size of 100, a maximum number of generations G of 200, and a crossover probability of... and mutation probability Adaptively adjusts with the iteration algebra g;

[0037] The parameter selection module is used to select the final welding parameter combination from the Pareto optimal solution set;

[0038] The output control module is used to output the final welding parameter combination to the welding control system or process card generation module through a digital interface.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] This invention breaks away from the traditional welding process design approach of welding first and then testing, constructing a reverse design system that deduces process parameters from the service failure mechanism. Traditional methods treat welding parameters and service performance as loosely correlated, while this invention decomposes the problem of hydrogen-induced fatigue damage into a quantifiable physicochemical process: based on Fick's diffusion law, it accurately calculates the hydrogen concentration field, couples the SWT damage model with the critical plane method, and for the first time achieves precise multi-physics coupling of dynamic load, residual stress, and the accelerating effect of the hydrogen environment within a predictive framework. Through the deterministic relationship of thermo-mechanical-hydrogen coupling, the ambiguous "hydrogen embrittlement" effect is transformed into a concrete correction factor. , to optimize the objective , It becomes a computable rigid constraint.

[0041] Meanwhile, the process optimization problem is reconstructed into a constrained solution problem that meets service reliability requirements through the improved NSGA-II algorithm. An adaptive probabilistic mechanism dynamically balances global search and local convergence, ensuring efficient locking of the Pareto solution set that satisfies both fatigue life thresholds and efficiency within a vast parameter space. This reverse design chain, starting from service requirements, enables a leap from empirical trial and error to scientific computation in welding processes, providing a quantifiable and reusable deterministic design method for the field. Attached Figure Description

[0042] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating the method of the present invention.

[0044] Figure 2 This is a schematic diagram of the structure of a thermo-mechanical coupled finite element model.

[0045] Figure 3 This is a schematic diagram of the optimization process of the improved NSGA-II algorithm. Detailed Implementation

[0046] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the embodiments of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0047] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0048] Example 1:

[0049] See Figures 1-3 This embodiment discloses a welding prediction method applicable to new energy natural gas pipelines. By combining the load spectrum under service conditions with the prediction of fatigue performance of welded joints, intelligent optimization of welding parameters is achieved.

[0050] In addition, this embodiment also discloses a welding prediction device suitable for new energy natural gas pipelines, including:

[0051] The target life setting module is used to set the target fatigue life threshold of the welded joint according to industry standards.

[0052] The stress field calculation module is used to construct a thermo-mechanical coupled finite element model of the welded joint and output a three-dimensional residual stress field distribution.

[0053] The fatigue life calculation module is used to calculate the fatigue crack initiation life of welded joints using the critical plane method combined with the Smith-Watson-Topper damage model, and introduces a hydrogen environment correction factor.

[0054] The parameter optimization module is used to iteratively optimize the welding parameter combination using an improved non-dominated sorting genetic algorithm, with the fatigue crack initiation life being greater than or equal to the target fatigue life threshold as a constraint, to generate the Pareto optimal solution set.

[0055] The parameter selection module is used to select the final welding parameter combination from the Pareto optimal solution set;

[0056] The output control module is used to output the final welding parameter combination to the welding control system or process card generation module through a digital interface.

[0057] The specific welding prediction method is as follows:

[0058] To generate the load spectrum for the target pipeline during its future service life, multi-source data fusion is required. In practice, historical operational data is sourced from the SCADA system, meteorological model prediction data is provided by meteorological analysis software, and the hydrogen source fluctuation model outputs results through a combination of experimental measurements and theoretical calculations. These data are integrated by the data processing module to form a multidimensional dynamic load spectrum, which includes pressure fluctuation time series, temperature cycle data, and hydrogen partial pressure variation curves. The data processing module connects to external data sources through a standardized interface to ensure the real-time performance and accuracy of data transmission.

[0059] In practical applications, the load spectrum serves as the basic input for subsequent fatigue life prediction. The generation process relies on the collaborative work between various data sources. For example, real-time monitoring data from the SCADA system provides the basis for the pressure fluctuation time series, while the hydrogen partial pressure change curve in the hydrogen source fluctuation model completes the quantitative description of hydrogen diffusion behavior through thermodynamic equations.

[0060] The target fatigue life threshold for welded joints is set according to industry standards. Based on ASME B31.8 specifications, the minimum fatigue life requirement for welded joints under a specific load spectrum was determined. In this process, the target fatigue life threshold was determined. The settings need to be adjusted according to actual engineering conditions to meet the needs of different application scenarios. For example, in high-pressure natural gas transmission pipelines, the target fatigue life is usually set to no less than 100,000 cycles to ensure that the welded joint has sufficient reliability during long-term service.

[0061] Next, a thermo-mechanical coupled finite element model of the weld joint is constructed to simulate heat conduction, phase transformation, and mechanical response during the welding process, thereby obtaining the three-dimensional residual stress field distribution of the weld and heat-affected zone. The input parameters of the thermo-mechanical coupled finite element model include the initial welding parameter combination. ,in For heat input per unit length, Interlayer temperature, This refers to the welding speed.

[0062] In the model, the welding area is divided into several mesh elements, and the thermophysical properties and mechanical properties of each element are defined using a material database. The welding heat source is described using a double ellipsoidal heat source model, and the trajectory of the heat source is determined by the welding speed. control.

[0063] In addition, the model boundary conditions were set to adiabatic boundaries to reduce the influence of the external environment on the welding process. After solving the problem using finite element analysis software, the three-dimensional residual stress field distribution of the weld and heat-affected zone can be output. This distribution provides a key input for subsequent fatigue crack initiation life calculation.

[0064] Subsequently, the residual stress field distribution 3 was superimposed with the load spectrum, and the fatigue crack initiation life of the welded joint was calculated using the critical plane method combined with the Smith-Watson-Topper (SWT) damage model. In this process, a hydrogen environment correction factor is introduced. ,in Indicates the hydrogen concentration in the heat-affected zone. This represents the hydrogen sensitivity coefficient of the material. The hydrogen environment correction factor, calibrated through small-scale diffusion tests, is used to quantify the impact of hydrogen on the fatigue performance of welded joints.

[0065] In practice, hydrogen concentration Based on the hydrogen partial pressure data in the loading spectrum, the diffusion coefficient D was obtained by numerical solution using Fick's second law. The hydrogen sensitivity coefficient of the material was pre-calibrated using the electrochemical permeation method. The result is obtained through experimental fitting. The calculation process of the critical plane method includes determining the principal stress direction, calculating the equivalent stress amplitude, and assessing the damage accumulation, ultimately obtaining the fatigue crack initiation life. This was achieved using fatigue analysis software, which integrated the SWT model and related algorithms for hydrogen environment correction factors.

[0066] Furthermore, with As constraints, minimize heat input Or maximize welding speed Using the objective function, an improved NSGA-II multi-objective genetic algorithm is used to iteratively optimize the welding parameter combination.

[0067] like Figure 3 As shown, the optimization process of the improved NSGA-II algorithm includes adaptive crossover probability. and mutation probability ,in For the current algebra, This represents the maximum algebraic value. The adaptive probability mechanism can dynamically adjust the search strategy based on the optimization process, avoiding getting trapped in local optima.

[0068] During the optimization process, the algorithm first randomly generates an initial population, and then generates a new generation population through selection, crossover, and mutation operations. Each individual in each generation corresponds to a set of welding parameter combinations, and its fitness value is determined by the fatigue crack initiation life. The objective function and the solution are jointly determined. After multiple iterations, the algorithm outputs the Pareto optimal solution set, which contains multiple candidate solutions, each of which meets the fatigue life requirement and achieves a balance between heat input and welding speed.

[0069] Finally, the optimal parameter combination in the Pareto optimal solution set is output to the welding control system or process card generation module to guide the actual welding operation.

[0070] Optimized parameters are transmitted to automated welding equipment or manual operating guidelines via a digital interface to ensure the precise implementation of welding parameters. The digital interface uses the industrial Ethernet protocol, supporting seamless integration with various welding control systems.

[0071] To further improve welding quality, the online monitoring system can deploy a sensor network to collect the actual stress state of the welded joint during its service life, and combine it with digital twin technology to dynamically correct the prediction model, thereby improving prediction accuracy.

[0072] Supported by a distributed computing architecture, the above methods can process large-scale finite element simulation tasks in parallel through a cloud computing platform, significantly shortening the optimization cycle. The computing nodes in the cloud computing platform allocate tasks through load balancing algorithms, ensuring maximum resource utilization efficiency. Furthermore, the welding parameter optimization method can be integrated into a smart factory management system for collaborative optimization with other production processes. For example, by linking with the pipe supply chain management module, consistency between welding parameters and pipe material can be ensured. In practical engineering, users can flexibly adjust the priority weights between fatigue life, heat input, and welding speed according to their needs, thereby achieving multi-objective trade-off analysis.

[0073] To facilitate a better understanding of the present invention by those skilled in the art, the present invention will be further described below in conjunction with specific embodiments.

[0074] In the implementation of welding process optimization for new energy natural gas pipelines, historical operating data of the target pipeline, including pressure fluctuation time series and temperature cycle data, is first acquired through the SCADA system. Simultaneously, meteorological analysis software provides environmental prediction data for the future service life, such as temperature change trends and extreme climate conditions; the hydrogen source fluctuation model outputs hydrogen partial pressure change curves based on experimental measurements and theoretical calculations. These data are integrated by the data processing module to generate load spectrum 1, which accurately reflects the actual operating conditions of the pipeline during its future service life. Through a standardized interface, the above multi-source data is transmitted to the data processing module in real time, ensuring the dynamics and accuracy of load spectrum 1. For example, real-time monitoring data from the SCADA system provides the basic input for the pressure fluctuation time series, while the hydrogen source fluctuation model, combined with thermodynamic equations, completes a quantitative description of hydrogen diffusion behavior, thus forming a multidimensional dynamic load spectrum 1.

[0075] Subsequently, the target fatigue life threshold for the welded joint was set according to the requirements of ASME B31.8. Taking high-pressure natural gas transmission pipelines as an example, the target fatigue life is typically set at no less than 100,000 cycles to meet long-term service requirements. During this process, the target fatigue life threshold... The settings need to be adjusted based on the actual service conditions of the pipeline. For example, in a high hydrogen partial pressure environment, the target fatigue life may need to be further improved to cope with the risk of hydrogen-induced delayed cracking.

[0076] Next, a thermo-mechanical coupled finite element model of the weld joint was constructed to simulate heat conduction, phase transformation, and mechanical response during the welding process. The welding region was divided into several mesh elements, and the thermophysical properties and mechanical properties of each element were defined by a material database. The welding heat source was described using a double ellipsoidal heat source model, and its movement trajectory was determined by the welding speed. Control measures were implemented. The model boundary conditions were set to adiabatic boundaries to minimize the impact of the external environment on the welding process. After solving the problem using finite element analysis software, the three-dimensional residual stress field distribution of the weld and heat-affected zone was output. This distribution provides crucial input for subsequent fatigue crack initiation life calculations. For example, the moving speed of the welding heat source directly affects the heat input distribution, thereby altering the residual stress state in the weld region.

[0077] Subsequently, the residual stress field distribution was superimposed with the load spectrum, and the fatigue crack initiation life of the welded joint was calculated using the critical plane method combined with the Smith-Watson-Topper (SWT) damage model. In this process, a hydrogen environment correction factor is introduced. ,in Indicates the hydrogen concentration in the heat-affected zone. This represents the hydrogen sensitivity coefficient of the material. The calculation process of the critical plane method includes determining the principal stress direction, calculating the equivalent stress amplitude, and assessing the damage accumulation, ultimately obtaining the fatigue crack initiation life. This process is achieved using fatigue analysis software, which integrates the SWT model and related algorithms for hydrogen environment correction factors. For example, when the hydrogen partial pressure is high, the hydrogen environment correction factor increases significantly, leading to a longer fatigue crack initiation life. Significantly reduced.

[0078] Furthermore, with As constraints, minimize heat input Or maximize welding speed Using the objective function, an improved NSGA-II multi-objective genetic algorithm is used to iteratively optimize the welding parameter combination.

[0079] The optimization process of the improved NSGA-II algorithm includes adaptive crossover probability. and mutation probability ,in For the current algebra, It represents the maximum algebraic value. The adaptive probability mechanism can dynamically adjust the search strategy based on the optimization process, avoiding getting trapped in local optima.

[0080] During the optimization process, the algorithm first randomly generates an initial population, and then generates a new generation population through selection, crossover, and mutation operations. Each individual in each generation corresponds to a set of welding parameter combinations, and its fitness value is determined by the fatigue crack initiation life. The objective function and the heat input are jointly determined. After multiple iterations, the algorithm outputs a Pareto optimal solution set, which contains multiple candidate solutions, each of which satisfies the fatigue life requirement and achieves a balance between heat input and welding speed. For example, in one candidate solution, the welding speed v is increased by 10%, but the heat input Q is reduced by 5%, thus achieving a comprehensive optimization of welding speed and heat input.

[0081] Finally, the optimal parameter combination from the Pareto optimal solution set is output to the welding control system or process card generation module to guide the actual welding operation. The digital interface adopts the industrial Ethernet protocol, supporting seamless integration with various welding control systems.

[0082] In practical applications, the welding control system automatically adjusts process parameters such as welding current, voltage, and wire feed speed based on the received optimized parameters, thereby achieving intelligent control of the welding process. For example, in a specific welding scenario, the welding current is adjusted to 180A, the welding voltage to 22V, and the wire feed speed to 5m / min to meet the optimized welding parameter requirements.

[0083] To further improve welding quality, the online monitoring system can deploy a sensor network to collect the actual stress state of the welded joint during service. This data, combined with digital twin technology, can dynamically correct the prediction model, thereby improving prediction accuracy. For example, if stress data collected by the online monitoring system reveals that the stress amplitude under actual service conditions is 5% higher than the predicted value, digital twin technology can be used to dynamically correct the prediction model, ensuring that the prediction results more closely reflect actual working conditions.

[0084] With the support of a distributed computing architecture, the above method can process large-scale finite element simulation tasks in parallel through a cloud computing platform, significantly shortening the optimization cycle. The computing nodes in the cloud computing platform allocate tasks through a load balancing algorithm, ensuring maximum resource utilization efficiency.

[0085] In a specific optimization task, the cloud computing platform distributed the finite element simulation task across 10 computing nodes. Each node was responsible for handling simulation calculations with different combinations of welding parameters, thereby reducing the overall optimization cycle from 48 hours to 6 hours. For X70 steel pipes, after optimization using the above method, the welding parameters were obtained: heat input per unit length. Interlayer temperature welding speed Predicting fatigue life The loop continues until the requirement is met.

[0086] This invention establishes a complete closed-loop technical solution from front-end design to back-end implementation by combining dynamic load spectra under service conditions with fatigue performance prediction of welded joints. Specifically, load spectra are generated through multi-source data fusion, providing high-fidelity input for fatigue life prediction; an experimentally calibrated hydrogen environment correction factor is introduced, quantifying for the first time in the prediction model the accelerating effect of hydrogen on the fatigue performance of welded joints; an improved non-dominated sorting genetic algorithm with an adaptive probability mechanism is employed to achieve multi-objective collaborative optimization of heat input per unit length and welding speed while ensuring that the fatigue life is not lower than the target threshold; finally, the optimized parameters are directly output to the welding execution system through a standardized digital interface.

[0087] This method reduces the rework rate of welding processes, improves the accuracy of fatigue life prediction, and enhances the efficiency of optimization calculations, providing a full life-cycle reliability guarantee for the welding of new energy natural gas pipelines.

[0088] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0089] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A welding prediction method suitable for a new energy natural gas pipeline, characterized by, The method comprises the following steps: Step 1: obtaining a load spectrum of the target pipeline in a future service period, the load spectrum comprising a pressure fluctuation time series, temperature cycle data and a hydrogen partial pressure change curve; Step 2: setting a target fatigue life threshold of the welded joint; Step 3: constructing a thermal-mechanical coupled finite element model of the welded joint, inputting an initial welding parameter combination comprising unit length heat input, interpass temperature and welding speed, and outputting a three-dimensional residual stress field distribution of the weld and heat-affected zone; Step four, superimposing the residual stress field distribution with the load spectrum, using the critical plane method combined with the Smith-Watson-Topper damage model to calculate the fatigue crack initiation life of the welded joint, and introducing a hydrogen environment correction factor, which is expressed as wherein is the hydrogen concentration of the heat-affected zone, is the hydrogen sensitive coefficient of the material; Step five, taking the fatigue crack initiation life greater than or equal to the target fatigue life threshold as a constraint condition, minimizing the objective function For the purpose, And The weight coefficient is The heat input per unit length is The welding speed is , the improved non-dominated sorting genetic algorithm is used to iteratively optimize the welding parameter combination, and a Pareto optimal solution set is generated; Step 6: selecting a final welding parameter combination from the Pareto optimal solution set, the final welding parameter combination comprising optimized unit length heat input, optimized interpass temperature and optimized welding speed; Step 7: outputting the final welding parameter combination to a welding control system or a process card generation module through a digital interface.

2. The welding prediction method for new energy natural gas pipelines according to claim 1, characterized in that: The load spectrum in step 1 is generated by fusing historical operation data of a data acquisition and monitoring control system, meteorological model prediction data and output results of a hydrogen source fluctuation model.

3. The method for welding prediction suitable for new energy natural gas pipeline according to claim 1, characterized in that, The thermal-mechanical coupled finite element model in step 3 uses a double-ellipsoid heat source model to describe the welding heat source and sets an adiabatic boundary condition.

4. The welding prediction method suitable for new energy natural gas pipelines according to claim 1, characterized in that, The hydrogen concentration of the heat-affected zone in step four is calculated based on the hydrogen partial pressure change curve in the load spectrum by Fick diffusion law, and the calculation formula is: ; wherein : is the surface equilibrium hydrogen concentration based on the change in hydrogen partial pressure; : material hydrogen diffusion coefficient, calibrated by electrochemical permeation method; : is the hydrogen diffusion time; : the distance from the calculated point in the heat-affected zone to the center of the weld; the hydrogen-sensitive coefficient k of the material is obtained by pre-experiment calibration.

5. The method for welding prediction suitable for new energy natural gas pipeline according to claim 1, characterized in that, The improved non-dominated sorting genetic algorithm in step five adopts adaptive crossover probability and adaptive mutation probability, the adaptive crossover probability ; the mutation probability: , wherein is the current generation number, is the maximum generation number, the adaptive probability mechanism is used for dynamically adjusting the search strategy according to the optimization process, and avoiding falling into local optimum.

6. The method for welding prediction suitable for new energy natural gas pipeline according to claim 1, characterized in that, The optimization process in step 5 supports an extended variable space comprising at least one of welding current, welding voltage and wire feeding speed.

7. The method for welding prediction suitable for new energy natural gas pipeline according to claim 1, characterized in that, The Pareto optimal solution set in step 6 contains multiple candidate schemes, each of which satisfies the constraint condition that the fatigue crack initiation life is greater than or equal to the target fatigue life threshold.

8. The method for welding prediction suitable for new energy natural gas pipeline according to claim 1, characterized in that, The digital interface in step 7 uses an industrial Ethernet protocol to realize connection with the welding control system.

9. The method for welding prediction suitable for new energy natural gas pipeline according to claim 1, characterized in that, In the improved non-dominated sorting genetic algorithm, the value range of the weight coefficient is 0.3-0.7, the value range of the weight coefficient is 0.3-0.

7.

10. A welding prediction device suitable for use in a new energy natural gas pipeline, characterized by: A welding prediction method suitable for new energy natural gas pipelines according to any one of claims 1-9, comprises: a target life setting module for setting a target fatigue life threshold of the welded joint according to industry standards; a stress field calculation module for constructing a thermal-mechanical coupled finite element model of the welded joint and outputting a three-dimensional residual stress field distribution; a fatigue life calculation module for calculating the fatigue crack initiation life of the welded joint using the critical plane method combined with the Smith-Watson-Topper damage model and introducing a hydrogen environment correction factor; a parameter optimization module for iteratively optimizing the welding parameter combination using an improved non-dominated sorting genetic algorithm with the constraint condition that the fatigue crack initiation life is greater than or equal to the target fatigue life threshold, to generate a Pareto optimal solution set; a parameter selection module for selecting a final welding parameter combination from the Pareto optimal solution set; an output control module for outputting the final welding parameter combination to a welding control system or a process card generation module through a digital interface.

Images