Prediction model construction method, system and equipment for heat affected zone and prediction method

Abstract

The invention relates to the field of pipeline welding, in particular to a prediction model construction method, system and equipment of a heat affected zone and a prediction method. The method comprises the steps that phase change parameters of a metal microstructure phase change model are corrected based on a thermal simulation test; building a heat affected zone microstructure finite element prediction model according to the circumferential weld simulation calculation finite element model; and correction is carried out based on a heat affected zone hardness test and a small punch test correlation parameter correction method, and the heat affected zone strength prediction model is trained by using the expansion database to obtain the pipeline welding joint heat affected zone strength prediction model. Based on a material performance determination test, a finite element model construction method and a finite element model expansion database mode, the accuracy of the heat affected zone strength prediction model is greatly improved, the research cost is reduced, the prediction from a microstructure to macroscopic performance is realized, and the method is suitable for popularization and application. And the defect that the softening mechanism of the hot area material cannot be clarified by traditional macroscopic mechanical property test analysis is overcome.

Description

Technical Field

[0001] The present invention relates to the field of pipeline welding, and in particular to a prediction model construction method, system and equipment for a heat affected zone, and a prediction method. Background Art

[0002] At present, my country's oil and gas pipelines are in a period of large-scale construction. With the development of some harsh environment oil and gas fields, higher requirements are put forward for the engineering construction of pipelines in the new era. The two major problems facing contemporary pipeline engineering are the economic requirements for improving the long-distance pipeline transportation capacity and the safety requirements for coping with harsh environments. In the past decade, with the large-scale construction and commissioning of high-grade steel large-diameter pipelines, many girth weld cracking and leakage accidents occurred during the pipeline pressure test stage and the initial operation. More than 70% of them were caused by girth weld defects. Therefore, the weld is a weak link in long-distance pipelines. During the welding heat process, due to factors such as the nature of the welding heat source, heating time, and cooling rate, the parent metal near the weld area undergoes changes in organization and performance. Because there is a large temperature gradient in the heat-affected zone, the microstructure and material strength of the heat-affected zone are significantly uneven, resulting in low strength matching of the girth weld joint.

[0003] The existing technology mainly analyzes and predicts the strength performance of the heat affected zone of welded joints from the data-driven perspective. However, the existing technology requires a large number of material performance measurement tests from the data-driven perspective, which consumes a lot of manpower and material resources, and the traditional macroscopic mechanical performance test analysis cannot explain the softening mechanism of the heat affected zone material; at the same time, from the traditional material strengthening theory, it is difficult to accurately predict the strength performance of the heat affected zone of the pipeline welded joint. Summary of the invention

[0004] The present invention provides a prediction model construction method, system and equipment for a heat affected zone and a prediction method to solve at least one of the above-mentioned technical problems.

[0005] The technical solution of the present invention to solve the above technical problem is as follows: A method for constructing a prediction model of a heat affected zone comprises:

[0006] S1, based on the thermal simulation test, the phase change parameters of the metal microstructure phase change model are corrected to obtain the finite element model of the metal microstructure evolution;

[0007] S2, constructing a finite element prediction model of the heat affected zone microstructure according to the finite element model of the girth weld simulation calculation and the finite element model of the metal microstructure evolution;

[0008] S3, based on the method of correcting the associated parameters of the heat affected zone hardness test and the small punch test, the heat affected zone microstructure finite element prediction model is corrected to obtain the heat affected zone microstructure finite element prediction correction model;

[0009] S4, determining an expanded database corresponding to the heat-affected zone microstructure finite element prediction and correction model, and using the expanded database to train the heat-affected zone strength prediction model to obtain the heat-affected zone strength prediction model of the pipeline welded joint.

[0010] Based on the above-mentioned heat-affected zone prediction model construction method, the present invention also provides a heat-affected zone prediction model construction system.

[0011] A prediction model building system for a heat affected zone, applied to the above-mentioned prediction model building method for a heat affected zone, comprises:

[0012] A phase change parameter correction module, which is used to correct the phase change parameters of the metal microstructure phase change model based on thermal simulation tests to obtain a finite element model of the metal microstructure evolution;

[0013] A finite element modeling module, which is used to construct a finite element prediction model of the heat-affected zone microstructure based on the girth weld simulation calculation finite element model and the metal microstructure evolution finite element model;

[0014] An associated parameter correction module, which is used to correct the heat-affected zone microstructure finite element prediction model based on a heat-affected zone hardness test and a small punch test to obtain a heat-affected zone microstructure finite element prediction correction model;

[0015] A database expansion module is used to determine an expanded database corresponding to the heat-affected zone microstructure finite element prediction and correction model, and use the expanded database to train the heat-affected zone strength prediction model to obtain the heat-affected zone strength prediction model of the pipeline welded joint.

[0016] Based on the above-mentioned method for constructing a prediction model of a heat-affected zone, the present invention also provides a device for constructing a prediction model of a heat-affected zone.

[0017] A device for constructing a prediction model of a heat affected zone includes a processor, a memory, and a computer program stored in the memory. When the computer program is executed by the processor, the method for constructing a prediction model of a heat affected zone as described above is implemented.

[0018] Based on the above-mentioned heat affected zone prediction model construction method, the present invention also provides a heat affected zone prediction method.

[0019] Heat affected zone prediction methods include:

[0020] Obtain pipeline welding process parameters to be predicted;

[0021] Inputting the pipeline welding process parameters to be predicted into the heat-affected zone strength prediction model of the pipeline welding joint constructed by using the heat-affected zone prediction model construction method described above to obtain the heat-affected zone analysis results;

[0022] According to the analysis results, the strength performance of the heat affected zone of the pipeline weld joint is determined.

[0023] The beneficial effects of the present invention are as follows: the prediction model construction method, system and equipment of the heat affected zone of the present invention and the prediction method correct the phase change parameters based on the thermal simulation test to obtain the finite element model of the metal microstructure evolution; then the expansion database corresponding to the finite element prediction correction model of the heat affected zone microstructure is calculated and the corresponding expansion database is determined in combination with the ring weld simulation and the method of correcting the associated parameters based on the small punch test, and finally the heat affected zone strength prediction model of the pipeline welding joint is obtained through machine learning training; the present invention is based on the material performance measurement test, the finite element model construction method and the finite element model expansion database method, which greatly improves the accuracy of the heat affected zone strength prediction model, reduces the research cost, realizes the prediction from microstructure to macro performance, and makes up for the deficiency that the traditional macro mechanical performance test analysis cannot explain the softening mechanism of the hot zone material. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is an overall flow chart of the method for constructing a prediction model of a heat affected zone according to the present invention;

[0025] Figure 2 To obtain the flow chart of the finite element model of metal microstructure evolution;

[0026] Figure 3 Flow chart for obtaining phase transformation parameters of metal microstructure phase transformation model;

[0027] Figure 4 Flow chart for obtaining the finite element prediction and correction model of the heat affected zone microstructure;

[0028] Figure 5 A flow chart for obtaining a first associated parameter;

[0029] Figure 6 A flow chart for obtaining a second associated parameter;

[0030] Figure 7 A structural block diagram of a prediction model building system for heat affected zones of the present invention;

[0031] Figure 8 The present invention is a structural block diagram of the heat affected zone prediction model building system. DETAILED DESCRIPTION

[0032] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0033] Embodiment 1:

[0034] like Figure 1 As shown, the method for constructing a prediction model of a heat affected zone of the present invention comprises:

[0035] S1, based on the thermal simulation test, the phase change parameters of the metal microstructure phase change model are corrected to obtain the finite element model of the metal microstructure evolution;

[0036] S2, constructing a finite element prediction model of the heat affected zone microstructure according to the finite element model of the girth weld simulation calculation and the finite element model of the metal microstructure evolution;

[0037] S3, based on the method of correcting the associated parameters of the heat affected zone hardness test and the small punch test, the heat affected zone microstructure finite element prediction model is corrected to obtain the heat affected zone microstructure finite element prediction correction model;

[0038] S4, determining an expanded database corresponding to the heat-affected zone microstructure finite element prediction and correction model, and using the expanded database to train the heat-affected zone strength prediction model to obtain the heat-affected zone strength prediction model of the pipeline welded joint.

[0039] In the prediction model construction method of the heat affected zone of the present invention, the thermal simulation test, the heat affected zone hardness test and the small punch test are specifically as follows:

[0040] Thermal simulation test: Thermal simulation test is an advanced material processing and performance testing method, which is mainly used to simulate the physical processes such as heat and force of the thermal simulation specimen during preparation and thermal processing. The working principle of the thermal simulation test machine is to heat the specimen to a certain temperature through a heating device, and then apply a certain stress to the specimen through a loading device to make the specimen deform at high temperature. By measuring parameters such as force and displacement during the deformation process, the thermal deformation constitutive equation of the metal material can be obtained, thereby analyzing the thermal deformation behavior of the metal material.

[0041] Heat affected zone hardness test: The heat affected zone hardness test is a method used to evaluate the quality of welded joints. During the welding process, a heat affected zone will be generated under high temperature, and the structure and properties in this area may change. The heat affected zone hardness test can measure the hardness value in the heat affected zone of the welded joint to judge the quality of the welded joint. This test method usually uses an indentation hardness tester for measurement. By performing hardness tests at different positions in the heat affected zone, the distribution of hardness values ​​in the area can be obtained. The hardness test results can be used to evaluate the strength, toughness and durability of the welded joint, thereby ensuring that the quality of the welded joint meets the requirements.

[0042] Small Punch Test: Small Punch Test (SPT) is a new testing method for evaluating material properties. It is particularly important in evaluating the performance and remaining life of in-service equipment such as steel pipes. It aims to obtain various performance parameters of materials by obtaining micro-sized samples from in-service equipment. During the test, a certain steel ball or spherical punch is used to apply pressure to a thin sample clamped in a specified die, causing the sample to bend and deform until it breaks. By measuring and recording the load-deformation data of the sample during the test, it is converted into a curve graph to obtain the basic performance parameters of the material.

[0043] The prediction model construction method of the heat affected zone of the present invention greatly improves the accuracy of the strength prediction model of the heat affected zone of the pipeline welding joint by correcting the model and combining the expansion of the database. The research cost is reduced by applying the expansion of the database, and the prediction from microstructure to macroscopic performance is realized, which makes up for the deficiency that the traditional macroscopic mechanical property test analysis cannot explain the softening mechanism of the hot zone material.

[0044] In S1 of this embodiment, Figure 2 As shown, the S1 is specifically:

[0045] S11, obtaining thermal simulation test data, metallographic structure ratio data and phase change parameters of the metal microstructure phase change model based on the thermal simulation test.

[0046] The thermal simulation test used in this embodiment may be a Gleebl e thermal simulation test; the thermal simulation test data is obtained according to the thermal simulation test parameters, the metallographic structure ratio data may be obtained through the thermal simulation test results, and the thermal simulation test data and the metallographic structure ratio data may be test data of the isothermal zone corresponding to the test thermal simulation sample. The Gleebl e thermal simulation test includes the following operations:

[0047] The thermal simulation test material (here X80 pipeline steel is taken as an example) is processed into two shapes of specimens: a rectangular specimen: the specific data of the specimen include a length of 76 mm, a width of 10.5 mm, and a thickness of 10.5 mm; a sheet specimen: the specific data of the specimen include an overall length of 76.7 mm, a width of 10.0 mm, and a thickness of 2.65 mm. The specific data of the middle area of ​​the specimen can be: a length of 11.35 mm and a width of 4.13 mm. Because the shape design can achieve a higher cooling rate, the surface of the specimen needs to be polished, free of oxide film, oil stains, rust, etc., to ensure that the weldment is well heated; after completing the above processing, the thermal simulation test parameters are determined according to the metallographic structure continuous cooling transformation diagram corresponding to the thermal simulation test, the metallographic structure ratio data, and the welding process parameters corresponding to the thermal simulation specimen. Based on the thermal simulation test parameters, the thermal simulation specimens are grouped and tested according to the cooling rate threshold to obtain the thermal simulation test results. The cooling rate threshold is used to indicate the cooling rate threshold corresponding to the thermal simulation specimen. The cooling rate threshold can include four cooling rates: 20, 40, 70, and 100 ° C / s.

[0048] Based on the above four cooling rates, the thermal simulation samples are grouped for test processing. For example, the group test is divided into four groups. Each group in the group conducts experiments at a single cooling rate. Each group of experimental samples may have 29 samples, for example. The first three groups of samples in the group use rectangular samples, and the last group uses sheet samples. The cooling rate thresholds used by the first three groups may be, for example, 20, 40, and 70°C / s, respectively, and the cooling rate threshold used by the last group may be, for example, 100°C / s.

[0049] The parameters of the thermal simulation test are shown in Table 1 below:

[0050] Table 1

[0051]

[0052] The phase change parameters of the metal microstructure phase change model can be obtained through the isothermal transformation curve (TTT diagram) of the thermal simulation sample and the supercooled austenite transformation diagram (CCT diagram) during the continuous cooling process. Specifically, before obtaining the thermal simulation test data and metallographic structure ratio data, the temperature history of the thermal simulation sample is firstly subjected to phase change simulation processing, and a geometric model is constructed according to the isothermal zone in the middle of the thermal simulation sample. The phase change model parameters are determined from the CCT diagram or the TTT diagram, so as to determine the phase change parameters of the metal microstructure phase change model after the phase change treatment.

[0053] S12, correcting the phase change parameters of the metal microstructure phase change model according to the thermal simulation test data and the metallographic structure ratio data to obtain phase change correction parameters.

[0054] In this step, the pipe welding joint corresponding to X80 steel is used as an example for correction processing. The thermal simulation test data may include temperature history data of the thermal simulation test, and the phase change parameters of the metal microstructure phase change model may be based on the TTT diagram and CCT diagram of X80 steel, and obtained by correction processing of the temperature history and metallographic structure ratio test data corresponding to the thermal simulation test. The phase change parameters at different temperatures after correction may be as shown in the following Table 2, where K A→P , L A→P is the diffusion-controlled phase transformation parameter from austenite to pearlite, K A→B , L A→B is the diffusion-controlled phase transformation parameter from austenite to bainite, and the temperature intervals selected in the table are selected based on the test conditions and are not specifically limited here.

[0055] Table 2

[0056]

[0057] S13, determining an initial metal microstructure evolution finite element model, and correcting the initial metal microstructure evolution finite element model according to the phase change correction parameter to obtain the metal microstructure evolution finite element model.

[0058] The initial metal microstructure evolution finite element model is an uncorrected model constructed based on thermal simulation tests.

[0059] Preferably, Figure 3 As shown, in S11, the specific steps of obtaining the phase change parameters of the metal microstructure phase change model based on the thermal simulation test are:

[0060] S111, determining the metallographic structure ratio corresponding to the thermal simulation sample in the thermal simulation test, performing phase change simulation processing on the temperature history of the thermal simulation sample according to the metallographic structure ratio, and obtaining a phase change simulation processing result.

[0061] Among them, the metallographic structure ratio corresponding to the thermal simulation sample is a variety of different metallographic structure ratios corresponding to the thermal simulation sample in different states, and the material corresponding to the thermal simulation sample can be an X80 steel material. The different states of the thermal simulation sample can include five metallographic phases: austenite, ferrite, pearlite, bainite, and martensite. The phase change simulation process can include: the three phase changes from austenite to ferrite, austenite to pearlite, and austenite to bainite are diffusion-controlled phase changes, and the phase change from austenite to martensite is a displacement-type (diffusion-free) phase change.

[0062] The diffusion-controlled phase change uses the Leblond-Devaux model to simulate the phase change. The Leblond-Devaux model can be implemented using the following formula, for example:

[0063]

[0064] Specifically, s is the source phase, ξ d is the target phase, A s→d is the generation rate of the target phase; K s→d and L s→d are universal coefficients, and their expressions are:

[0065]

[0066] Specifically, K0 is the phase transition temperature T m The isotherm width at the time of phase transition is L0, which is the zero temperature value of the phase transition interface movement speed. A and B are parameters related to material properties. m is the phase transition temperature, T is the real-time changing temperature; K0, L0, A, and B can all be fitted according to the specific material properties and the experimental data in the form of the supercooled austenite transformation diagram during continuous cooling.

[0067] The Koistinen-Marburger model is used to simulate the displacement (diffusion-free) phase transformation. This model can better characterize the kinetics of martensitic phase transformation. The Koistinen-Marburger model can be implemented using the following formula:

[0068]

[0069] Specifically, s is the source phase, ξ d is the target phase, A s→d is the generation rate of the target phase, M s is the martensite start temperature, β is the Koistinen-Marburger coefficient, and T is the rate of change of real-time temperature.

[0070] Martensite start temperature M s (℃) is a function of alloying elements (weight %), which can be expressed as follows:

[0071] M s =539-423C-30.4Mn-17.7Ni-12.1Cr-7.5Mo+10Co-7.5Si;

[0072] Specifically, C, Mn, Ni, Cr, Mo, Co and Si are the element weights of the corresponding chemical elements carbon, manganese, nickel, chromium, molybdenum, cobalt and silicon, respectively.

[0073] S112, determining phase change parameters of the metal microstructure phase change model according to the phase change simulation processing result.

[0074] In S2 of this embodiment: the finite element model for the girth weld simulation calculation can be constructed based on a pipeline girth weld fully automatic welding process, and the pipeline can be, for example, an X80 pipeline.

[0075] In S3 of this embodiment, the finite element prediction model of the heat affected zone microstructure is corrected in order to obtain a model with more accurate prediction results. Figure 4 As shown, the S3 is specifically:

[0076] S31, obtaining a first correlation parameter based on a thermal simulation test, and obtaining a second correlation parameter based on a heat affected zone hardness test and a small punch test.

[0077] S32, correcting the heat affected zone microstructure finite element prediction model according to the first associated parameter to obtain an initial corrected heat affected zone microstructure finite element prediction model.

[0078] In this step, after the first associated parameter is determined, for example, part of the test data may be used to verify the first associated parameter.

[0079] S33, correcting the initial correction model of the finite element prediction of the microstructure of the heat affected zone according to the second correlation parameter to obtain the correction model of the finite element prediction of the microstructure of the heat affected zone.

[0080] Specifically, the thermal simulation test is a thermal simulation test based on a thermal simulation specimen corresponding to the pipeline welding joint; Figure 5 As shown, in S31, the specific steps of obtaining the first correlation parameter based on the thermal simulation test are:

[0081] S311, dividing the test data of the thermal simulation test according to a preset ratio to obtain first test data and second test data.

[0082] Among them, the first test data and the second test data can be, for example, data obtained by dividing the test data of the thermal simulation test according to a ratio. The proportional division can be, for example,: the first test data accounts for 70% of the test data, and the second test data accounts for 30%, and the proportional division can be flexibly selected according to actual conditions, and no specific limitation is made here.

[0083] S312: Determine, based on the first test data, a correlation parameter between the hardness data of the thermal simulation sample corresponding to the thermal simulation test and the metallographic structure ratio data.

[0084] The thermal simulation test data and the metallographic structure ratio data may be, for example, test data of the isothermal zone corresponding to the test thermal simulation sample.

[0085] This step may, for example, be to use 70% of the test data as the first test data, and to conduct a test based on the first test data to determine the correlation parameters between the hardness data of the thermal simulation sample corresponding to the thermal simulation test and the metallographic structure ratio data.

[0086] S313, using the second test data, verifying the correlation parameters between the hardness data of the thermal simulation sample corresponding to the thermal simulation test and the metallographic structure ratio data to obtain the first correlation parameters.

[0087] This step may, for example, be to use 30% of the test data as the second test data, and verify the correlation parameters between the hardness data of the thermal simulation sample corresponding to the thermal simulation test and the metallographic structure ratio data based on the second test data.

[0088] like Figure 6 As shown, the specific steps for obtaining the second correlation parameter based on the heat affected zone hardness test and the small punch test are:

[0089] S314, dividing the heat affected zone hardness test data and the small punch test data corresponding to the pipeline welding joint according to a preset ratio to obtain third test data and fourth test data.

[0090] Among them, the third test data and the fourth test data can be, for example, data obtained by dividing the test data of the thermal simulation test according to a ratio. The proportional division can be, for example, that the third test data accounts for 70% of the test data and the fourth test data accounts for 30%, and the proportional division can be flexibly selected according to actual conditions and is not specifically limited here.

[0091] S315, determining, based on the third test data, a correlation parameter between the heat affected zone hardness test data and the small punch test data;

[0092] This step may, for example, be to use 70% of the test data as the third test data, and to conduct a test based on the third test data to determine the correlation parameters between the heat affected zone hardness test data and the small punch test data.

[0093] S316, using the fourth test data to verify the correlation parameters between the heat affected zone hardness test data and the small punch test data, to obtain the second correlation parameters; the heat affected zone hardness test data is obtained by the heat affected zone hardness test test, and the small punch test data is obtained by the small punch test.

[0094] This step may, for example, be to use 30% of the test data as fourth test data, and to conduct a test based on the fourth test data to determine the second correlation parameter.

[0095] In addition, the heat affected zone hardness test data can be the full-size joint heat affected zone hardness data, and the small punch test data can be the tensile strength data of the corresponding pipe welding joint. The heat affected zone hardness test data can be, for example, the test data corresponding to the heat affected zone hardness test. Before performing the heat affected zone hardness test process, for example, the following steps can be performed:

[0096] The hardness test surface of the full-size joint of the pipeline is ground, polished, and etched to show the base material area, heat-affected zone, weld area, and each layer of each weld. On the other hand, in order to make the hardness test point positions spread throughout the heat-affected zone of the full-size joint, it is necessary to set a reasonable point spacing, which may include, for example, 0.3mm or 0.5mm, without specific restrictions. The material of the pipeline may be, for example, X80 steel. After this step, the heat-affected zone hardness test can be carried out according to the hardness test point positions corresponding to the point spacing.

[0097] The small punch test data may be test data obtained by conducting a small punch test, for example, and the small punch test step may include: sampling a small punch sample along the heat-affected zone of the full-size joint sample of the pipeline, and conducting a small punch test according to the test standard. The material of the pipeline may be, for example, X80 steel.

[0098] It should be noted that, in the above steps, the order of correcting the parameters of the finite element prediction model of the microstructure of the heat affected zone can be flexibly selected according to the specific circumstances through the correlation parameters between the hardness data of the thermal simulation specimen corresponding to the thermal simulation test and the metallographic structure proportion data, as well as the correlation parameters between the hardness test data of the heat affected zone and the small punch test data. The correlation parameters between the hardness data of the thermal simulation specimen corresponding to the thermal simulation test and the metallographic structure proportion data, as well as the correlation parameters between the hardness test data of the heat affected zone and the small punch test data can also simultaneously correct the parameters of the finite element prediction model of the microstructure of the heat affected zone.

[0099] By classifying the test data of the thermal simulation test and the test data of the heat affected zone hardness test and the small punch test data, the correlation parameters of the heat affected zone microstructure finite element prediction model are obtained, and the finite element prediction model of the heat affected zone microstructure after correction is obtained according to the correlation parameters. This method obtains the corresponding correlation parameters by classifying the test data, and corrects the finite element prediction model of the heat affected zone microstructure based on the correlation parameters, so that the prediction results of the heat affected zone strength prediction model of the pipeline welded joint obtained based on the model training are more accurate.

[0100] In S4 of this embodiment, the expanded database may be: based on the finite element prediction and correction model of the heat affected zone microstructure, batch calculation of the heat affected zone strength performance under different welding process parameters, and a unified expanded database established in combination with performance measurement test data.

[0101] The heat affected zone strength prediction model can be: taking the pipeline full-automatic welding process parameters (current, voltage, speed) as input, and the strength performance of the heat affected zone of the welded joint as output, a suitable machine learning algorithm (multi-layer feedforward neural network based on error back propagation algorithm, multi-layer perceptron, radial basis function neural network and other machine learning algorithms) is selected to build a heat affected zone strength prediction model based on machine learning. The pipeline can be an X80 pipeline, and the current corresponding to the X80 pipeline can be 100-250A, the voltage can be 20-25V, and the speed can be 600-1000cm / min. The machine learning model can be any machine learning model that meets the present conditions in the prior art, and no specific restrictions are made here.

[0102] The training processing result can be the most suitable machine learning model selected from the above-mentioned multiple machine learning models, or the most suitable machine learning model selected after reaching the set number of training times. The termination condition of the training process can be flexibly selected according to actual conditions and is not specifically limited here.

[0103] The method of this embodiment starts from the prediction model of phase change performance of heat-affected zone materials, and corrects the phase change parameters of the metal microstructure phase change model based on experimental data, so as to obtain a more accurate finite element model of metal microstructure evolution, which is helpful to improve the accuracy of the finite element prediction model of the microstructure of the affected zone in the future. The finite element prediction model of the microstructure of the heat-affected zone is corrected according to the test data of the thermal simulation test, the test data of the hardness test of the heat-affected zone, and the small punch test data, so that the prediction results of the heat-affected zone strength prediction model of the pipeline welding joint obtained based on the training of the model are more accurate, and based on the test and simulation data, it breaks through the difficulty that the traditional material strengthening theory is difficult to accurately predict the performance of the hot zone materials, and couples the three physical fields of solid mechanics, solid heat transfer, and metal microphase change to achieve the prediction from microstructure to macro performance, making up for the deficiency that the traditional macroscopic mechanical performance test analysis cannot explain the softening mechanism of the hot zone material.

[0104] Embodiment 2:

[0105] Based on the above-mentioned heat-affected zone prediction model construction method, the present invention also provides a heat-affected zone prediction model construction system. Figure 7 As shown, the prediction model construction system of the heat affected zone is applied to the prediction model construction method of the heat affected zone described above, and includes:

[0106] A phase change parameter correction module, which is used to correct the phase change parameters of the metal microstructure phase change model based on thermal simulation tests to obtain a finite element model of the metal microstructure evolution;

[0107] A finite element modeling module, which is used to construct a finite element prediction model of the heat-affected zone microstructure based on the girth weld simulation calculation finite element model and the metal microstructure evolution finite element model;

[0108] An associated parameter correction module, which is used to correct the heat-affected zone microstructure finite element prediction model based on a heat-affected zone hardness test and a small punch test to obtain a heat-affected zone microstructure finite element prediction correction model;

[0109] A database expansion module is used to determine an expanded database corresponding to the heat-affected zone microstructure finite element prediction and correction model, and use the expanded database to train the heat-affected zone strength prediction model to obtain the heat-affected zone strength prediction model of the pipeline welded joint.

[0110] The specific functions of each module in the prediction model construction system of the heat affected zone of the present invention refer to the specific steps in the prediction model construction method of the heat affected zone of the present invention, which will not be repeated here.

[0111] Embodiment three:

[0112] Based on the above-mentioned prediction model construction method for heat-affected zones, the present invention also provides a prediction model construction device for heat-affected zones. The prediction model construction device for heat-affected zones includes a processor, a memory, and a computer program stored in the memory, and when the computer program is executed by the processor, the prediction model construction method for heat-affected zones as described above is implemented.

[0113] Embodiment 4:

[0114] Based on the above-mentioned heat-affected zone prediction model construction method, the present invention also provides a heat-affected zone prediction method. Figure 8 As shown in the figure, the prediction method of heat affected zone includes:

[0115] Obtain pipeline welding process parameters to be predicted;

[0116] Inputting the pipeline welding process parameters to be predicted into the heat-affected zone strength prediction model of the pipeline welding joint constructed by using the heat-affected zone prediction model construction method described above to obtain the heat-affected zone analysis results;

[0117] According to the analysis results, the strength performance of the heat affected zone of the pipeline weld joint is determined.

[0118] Specifically, the pipeline material corresponding to the pipeline welding process parameters to be predicted may be, for example, X80 steel, which is not specifically limited herein. The pipeline welding process parameters corresponding to the X80 pipeline may include, for example: 100-250A current, 20-25V voltage, 600-1000cm / min speed.

[0119] The heat affected zone prediction method provided in this embodiment realizes accurate prediction of the strength of the heat affected zone of the pipeline welded joint by inputting specific data of the pipeline welding process parameters (current, voltage, speed) into the heat affected zone strength prediction model of the pipeline welded joint. At the same time, it is convenient to guide the optimization of the welding process parameters in engineering practice.

[0120] The heat affected zone prediction method of the present invention realizes accurate prediction of the strength of the heat affected zone of the pipeline welded joint by inputting specific data of the pipeline welded joint into the heat affected zone strength prediction model of the pipeline welded joint. The prediction is performed based on the prediction model, which greatly improves the prediction accuracy and reduces the research cost.

[0121] The prediction model construction method, system and equipment of the heat affected zone of the present invention and the prediction method correct the phase change parameters based on the thermal simulation test to obtain the finite element model of the metal microstructure evolution; then combine the ring weld simulation and the method of correcting the associated parameters based on the small punch test to calculate the expanded database corresponding to the finite element prediction correction model of the microstructure of the heat affected zone and determine the corresponding expanded database, and finally obtain the heat affected zone strength prediction model of the pipeline welding joint through machine learning training; the present invention is based on the material performance measurement test, the finite element model construction method and the finite element model expansion database method, which greatly improves the accuracy of the heat affected zone strength prediction model, reduces the research cost, realizes the prediction from microstructure to macro performance, and makes up for the deficiency that the traditional macro mechanical performance test and analysis cannot explain the softening mechanism of the hot zone material.

[0122] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a prediction model of a heat affected zone, characterized in that: include: S1, based on the thermal simulation test, the phase change parameters of the metal microstructure phase change model are corrected to obtain the finite element model of the metal microstructure evolution; S2, constructing a finite element prediction model of the heat affected zone microstructure according to the finite element model of the girth weld simulation calculation and the finite element model of the metal microstructure evolution; S3, based on the method of correcting the associated parameters of the heat affected zone hardness test and the small punch test, the heat affected zone microstructure finite element prediction model is corrected to obtain the heat affected zone microstructure finite element prediction correction model; S4, determining an expanded database corresponding to the heat-affected zone microstructure finite element prediction and correction model, and using the expanded database to train the heat-affected zone strength prediction model to obtain the heat-affected zone strength prediction model of the pipeline welded joint.

2. The method for constructing a prediction model of a heat affected zone according to claim 1, characterized in that: The S1 is specifically: S11, obtaining thermal simulation test data, metallographic structure ratio data and phase change parameters of the metal microstructure phase change model based on the thermal simulation test; S12, correcting the phase change parameters of the metal microstructure phase change model according to the thermal simulation test data and the metallographic structure ratio data to obtain phase change correction parameters; S13, determining an initial metal microstructure evolution finite element model, and correcting the initial metal microstructure evolution finite element model according to the phase change correction parameter to obtain the metal microstructure evolution finite element model.

3. The method for constructing a prediction model of a heat affected zone according to claim 2, characterized in that: In S11, the specific steps of obtaining the phase change parameters of the metal microstructure phase change model based on the thermal simulation test are: S111, determining the metallographic structure ratio corresponding to the thermal simulation sample in the thermal simulation test, and performing phase change simulation processing on the temperature history of the thermal simulation sample according to the metallographic structure ratio to obtain a phase change simulation processing result; S112, determining phase change parameters of the metal microstructure phase change model according to the phase change simulation processing result.

4. The method for constructing a prediction model of a heat affected zone according to claim 3, characterized in that: In S111, the phase change simulation includes diffusion-controlled phase change simulation and displacement-type phase change simulation; wherein, the diffusion-controlled phase change simulation adopts the Leblond-Devaux model for phase change simulation, and the displacement-type phase change simulation adopts the Koistinen-Marburger model for phase change simulation.

5. The method for constructing a prediction model of a heat affected zone according to claim 1, characterized in that: The S3 is specifically: S31, obtaining a first correlation parameter based on a thermal simulation test, and obtaining a second correlation parameter based on a heat affected zone hardness test and a small punch test; S32, correcting the heat affected zone microstructure finite element prediction model according to the first correlation parameter to obtain an initial corrected heat affected zone microstructure finite element prediction model; S33, correcting the initial correction model of the finite element prediction of the microstructure of the heat affected zone according to the second correlation parameter to obtain the correction model of the finite element prediction of the microstructure of the heat affected zone.

6. The method for constructing a prediction model of a heat affected zone according to claim 5, characterized in that: The thermal simulation test is a thermal simulation test based on a thermal simulation specimen corresponding to the pipeline welding joint; in S31, the specific steps of obtaining the first associated parameter based on the thermal simulation test are: S311, dividing the test data of the thermal simulation test according to a preset ratio to obtain first test data and second test data; S312, determining, according to the first test data, a correlation parameter between the hardness data of the thermal simulation sample corresponding to the thermal simulation test and the metallographic structure ratio data; S313, using the second test data, verifying the correlation parameters between the hardness data of the thermal simulation sample corresponding to the thermal simulation test and the metallographic structure ratio data to obtain the first correlation parameters.

7. The method for constructing a prediction model of a heat affected zone according to claim 5, characterized in that: The thermal simulation test is a thermal simulation test based on a thermal simulation specimen corresponding to the pipeline welding joint; in S31, the specific steps of obtaining the second associated parameter based on the heat affected zone hardness test and the small punch test are: S314, dividing the heat-affected zone hardness test data and the small punch test data corresponding to the pipeline welding joint according to a preset ratio to obtain third test data and fourth test data; S315, determining, based on the third test data, a correlation parameter between the heat affected zone hardness test data and the small punch test data; S316, using the fourth test data to verify the correlation parameters between the heat affected zone hardness test data and the small punch test data, to obtain the second correlation parameters; The heat-affected zone hardness test data is obtained from a heat-affected zone hardness test, and the small punch test data is obtained from a small punch test.

8. A prediction model building system for heat affected zones, characterized in that: A method for constructing a prediction model for a heat affected zone as claimed in any one of claims 1 to 7, comprising: A phase change parameter correction module, which is used to correct the phase change parameters of the metal microstructure phase change model based on thermal simulation tests to obtain a finite element model of the metal microstructure evolution; A finite element modeling module, which is used to construct a finite element prediction model of the heat-affected zone microstructure based on the girth weld simulation calculation finite element model and the metal microstructure evolution finite element model; An associated parameter correction module, which is used to correct the heat-affected zone microstructure finite element prediction model based on a heat-affected zone hardness test and a small punch test to obtain a heat-affected zone microstructure finite element prediction correction model; A database expansion module is used to determine an expanded database corresponding to the heat-affected zone microstructure finite element prediction and correction model, and use the expanded database to train the heat-affected zone strength prediction model to obtain the heat-affected zone strength prediction model of the pipeline welded joint.

9. A device for constructing a prediction model of a heat-affected zone, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory, wherein when the computer program is executed by the processor, the method for constructing a prediction model of a heat affected zone according to any one of claims 1 to 7 is implemented.

10. A method for predicting a heat affected zone, characterized in that: include: Obtain pipeline welding process parameters to be predicted; Inputting the pipeline welding process parameters to be predicted into the heat affected zone strength prediction model of the pipeline welded joint constructed by the heat affected zone prediction model construction method according to any one of claims 1 to 7 to obtain the heat affected zone analysis result; According to the analysis results, the strength performance of the heat affected zone of the pipeline weld joint is determined.

Images