JP4189908B2 - Weld material prediction method - Google Patents

Description

【００５３】
［実施例２］
成分（mass％、残部実質的にＦｅ）がＣ：０．１５％、Ｓｉ：０．３７％、Ｍｎ：１．４３％の母材（板厚２０mmの厚鋼板）と、Ｃ：０．０６％、Ｓｉ：０．８０％、Ｍｎ：１．６０％、Ｔｉ：０．２６％の溶接ワイヤを用いて、ＣＯ₂ ガスシールドアーク溶接を行うこととし、前記晶出計算、オーステナイト組織計算、析出計算および変態組織情報としてフェライト粒情報およびアシキュラフェライト粒情報を考慮した変態組織計算により予測した組織情報を基に溶接金属部の材質（引張強さＴＳ、０℃でのシャルピー衝撃値vＥ₀）を予測するとともに、実際に溶接を行って同部の材質を実測した。
【００５４】
溶接条件は、実施例１と同様の母材開先条件の下で、溶接パス数５、溶接速度２５ｃｍ／分とし、溶接後の冷却速度を種々設定するため、入熱量およびパス間温度を種々変えて冷却速度を調整した。
また、フェライト粒情報の計算に際して、オーステナイト粒界面積を実施例１の手法Ｂにより、またフェライトの核生成速度、成長定数を実施例１の手法Ｂにより求めた。また、アシキュラフェライト粒情報の計算に際して、アシキュラフェライトモデルのｋ（定数）、析出速度Ｉaf(T)、成長定数αaf(T)はデータベースを回帰分析して得られた下記値、式を用いて計算した。

【００５５】
本実施例による材質予測計算による計算結果の一例を表２、図６および図７に示す。また、表２には、図２の座標系でｘ＝０mm、ｙ＝１８mmの位置におけるメッシュ領域（溶接金属中央上部）におけるフェライト、アシキュラフェライトの量（体積％）についての計算値および実測値を示す。また、同部における、オーステナイト変態開始温度Ｔas以上に加熱される最後の加熱冷却過程における最高到達温度からフェライト変態終了温度に至る平均冷却速度を同表に併せて示す。フェライト量などの実測値は以下の要領にて測定した。溶接部断面を観察することができるように組織観察用試験片を切り出し、研磨後、ナイタールエッチングで金属組織を現出させ、４００倍の光学顕微鏡写真を撮り、画像解析装置（Image Pro PLUS（プラネトロン社製））にて組織量を面積率で測定し、その値を体積率に等しいものと見なした。
【００５６】
図６および図７より、本実施例による材質予測結果と実測結果とは誤差が小さく、良好な一致が認められる。また、フェライト量、アシキュラフェライト量についても、予測精度に優れることが看取される。
【００５７】
【表２】

【００５８】
【発明の効果】
本発明の材質予測方法によれば、晶出物の生成を伴う溶接金属部について、晶出物を考慮した組織を予測計算し、予測計算された組織に基づいて溶接部内の被予測部の材質特性を予測するので、被予測部の材質特性を精度良く予測することができる。また、溶接条件に応じて被予測部の材質を精度良く予測することができるので、目標とする特性を得るための最適な溶接条件を予測結果を基に決定することができる。
【図面の簡単な説明】
【図１】実施形態の材質予測方法を示す主フローチャートである。
【図２】溶接部におけるメッシュ領域を示す断面説明図である。
【図３】あるメッシュ領域の温度履歴を示す時間−温度線図である。
【図４】実施例１における引張強さの予測計算値と実測値との関係を示すグラフである。
【図５】実施例１におけるシャルピー衝撃値の予測計算値と実測値との関係を示すグラフである。
【図６】実施例２における引張強さの予測計算値と実測値との関係を示すグラフである。
【図７】実施例２におけるシャルピー衝撃値の予測計算値と実測値との関係を示すグラフである。[0001]
[Technical field to which the invention belongs]
The present invention relates to a method for predicting the material (mechanical characteristics) of a predicted part in a welded part including a weld metal part and a heat-affected part after welding in general welding such as arc welding, laser welding, and electron beam welding. .
[0002]
[Prior art]
In recent years, various welding methods have been applied to welding constructions of ships, steel frames, bridges, and other building structures. In carrying out the welding work, the mechanical characteristics of the predicted part in the welded part after the work (the predicted part means the whole welded part or a part thereof, hereinafter may be simply referred to as the predicted part). Preliminary prediction is important in obtaining a weld having a target characteristic.
[0003]
As a method for predicting the mechanical properties of the weld metal, for example, Japanese Patent No. 2850773 predicts the weld metal composition from the base material component, welding current, welding voltage, welding speed, wire component, etc., for submerged arc welding. A technique for predicting mechanical characteristics based on the above has been proposed.
[0004]
[Problems to be solved by the invention]
  However, since this technique predicts the material based on the weld metal composition, there is a problem that the prediction accuracy of the characteristics of the predicted portion is inferior. This is because the welded part is repeatedly heated and cooled, and a complicated transformation phenomenon occurs inside it.thingIs not taken into account, and crystallized products mainly composed of oxides are formed in the weld metal part and the transformation structure may change greatly depending on the welding conditions, but these effects are ignored. It is.
[0005]
Note that Japanese Patent Laid-Open No. 5-93720 describes a method for predicting the structure of a steel material after hot rolling, but remarkable crystallization of a crystallized product is observed in a weld metal part depending on welding conditions. In addition, since it is generally formed by repeated heating and cooling by welding with multiple welding passes, it is subject to cooling conditions in which it is continuously cooled from high temperature to low temperature as in the conventional steel material prediction. The structure of the portion to be predicted cannot be predicted by predicting the material.
[0006]
The present invention has been made in view of such a problem, and a material prediction method capable of improving the prediction accuracy of a material to be predicted by predicting a structure of a predicted portion and predicting a material based on the predicted structure. The purpose is to provide.
[0007]
[Means for Solving the Problems]
  The method for predicting the material quality of a welded portion according to the present invention includes a welded structure in which a base material is welded through one or a plurality of welding passes through a welded metal portion in a welded portion including the welded metal portion and the heat affected zone. A method for predicting the material of the prediction unit,
  Assuming a plurality of mesh regions by dividing the weld, for each mesh region,(1) , (2) , (3) ,and (Four)Perform each calculation of
  (1)  For each unit time until the end of the final welding pass, the first welding is performed by sequentially obtaining the mesh region temperature after the unit time elapses based on the base material component, the wire component or the weld metal composition and the welding conditions before the unit time elapses. Mesh area temperature history calculation to calculate the temperature history from pass to final welding pass,
  (2)  After the unit time elapses based on the base metal component, wire component, shield gas component or weld metal composition before the unit time elapses and mesh region temperature after the unit time elapses until the end of the final welding pass Weld metal composition calculation to obtain the weld metal composition of
  (3) After the mesh region temperature history calculation and the weld metal composition calculation, the crystallized substance information including the composition of the crystallized material, the average particle size and the crystallized amount based on the weld metal composition and the temperature history, and the crystallized material is crystallized. Crystallization calculation to determine the weld metal composition after crystallization after crystallization,
  (Four) The weld metal composition after crystallizationTransformation structure calculation to obtain transformation structure information including the fraction of transformation structure in the mesh region based on temperature history
  Thereafter, after the final welding pass in each mesh region in the predicted portionThe crystallized information andBased on the transformation structure information, a material prediction calculation is performed to predict the material of the predicted portion from the relationship between the structure and the material obtained in advance by actual measurement. As the material prediction calculation, predicted part structure calculation for obtaining structure information in the predicted part from the transformed structure information after the end of the final welding pass in each mesh region in the predicted part is performed, and obtained by the predicted part structure calculation. Based on the predicted portion tissue information obtained, it is possible to perform the tissue / material prediction calculation for predicting the material of the predicted portion from the relationship between the tissue and the material obtained in advance by actual measurement. Alternatively, based on the transformation structure information in each mesh region in the predicted portion, perform a mesh region material prediction calculation to predict the material in each mesh region from the relationship between the welded portion structure and the material obtained in advance by actual measurement, The material / material prediction calculation for predicting the material of the predicted portion can be performed based on the material in each mesh region in the predicted portion determined by the mesh region material prediction calculation.
[0008]
  According to the material prediction method of the present invention, the mesh region temperature history for a plurality of mesh regions partitioning the welded portion.CalculationTo obtain the temperature history from the first welding pass to the final welding pass.Based on the weld metal composition calculation, the weld metal composition is calculated. Based on the weld metal composition and the temperature history, the crystallization information and the post-crystallization weld metal composition are obtained by the crystallization calculation. Based on composition and temperature historySince the transformation structure information after the end of the final welding pass in each mesh region in the predicted portion is obtained and the material of the predicted portion is predicted based on this, the material of the predicted portion can be predicted with high accuracy. . In particular, weld metalIn the departmentUnlike ordinary steel,Since the shield gas component contains active elements such as oxygen, it is accompanied by the formation of crystallized products mainly composed of oxides.Changes in weld metal composition due to remelting of the next weld pass affect the transformation structure. In the present invention, suchCrystallized products andConsider remelting informationAfter crystallizationSince the composition of the weld metal part is determined and the structure information of the predicted part in the weld part is predicted, the prediction accuracy is improved. In the weld metal composition calculation in the present invention, it was considered that the composition of the weld metal part differs from the wire component due to dilution from the base metal, and that the alloy component is discharged as slag by the slag / metal reaction. Calculation is performed. Specifically, the dilution calculation from the base material isAbove melting pointBy averaging the composition of all mesh regions, the slag / metal reaction can be calculated, for example, from a database. As a result, the composition of the weld metal part is roughly predicted.
[0009]
  As a preferred aspect of the material prediction method, before performing the transformation structure calculation,Weld metal composition after crystallizationThe austenite grain calculation for obtaining austenite grain information is performed based on the temperature history and the austenite grain information is also used as a basis for the transformation structure calculation.
  Prediction accuracy can be further improved by using the austenite grain information in the transformation structure calculation. At this time, if the form of dendritic austenite grains peculiar to the weld metal is taken into consideration, the prediction accuracy can be further improved.
[0010]
  Further, as a preferred embodiment, before performing the transformation structure calculation,Weld metal composition after crystallizationPrecipitation calculation to obtain precipitate information including the composition, average particle size and precipitation amount of precipitates precipitated in the structure based on the temperature history and the precipitate after the final welding pass as the basis of the material prediction calculation Information is also used.
  When it is necessary to consider the influence of precipitates in the material to be predicted, the prediction accuracy can be further improved by using the precipitate information together with other structure information in the material prediction calculation.
[0013]
Further, as a preferred embodiment of the material prediction method, a final heating / cooling process that is heated above the solidus temperature is obtained from the temperature history, a crystallization calculation is performed for the heating / cooling process, and from the temperature history The last heating / cooling process heated above the temperature Taf at which the austenite transformation ends in the heating process is obtained, the austenite structure calculation is performed for this last heating / cooling process, and the austenite transformation starts in the heating process from the temperature history. The last heating / cooling process heated above the temperature Tas to be obtained is obtained, and the transformation structure is calculated for this last heating / cooling process.
According to these aspects, by performing a predetermined calculation only for the heating and cooling process that most affects the final structure, the crystallized substance information and transformation structure information after the final welding pass and the transformation after the final welding pass are completed. The austenite grain information that determines the structure can be easily calculated. For this reason, a calculation load becomes light and a material prediction can be performed promptly.
[0014]
The transformation structure includes various structures such as ferrite, acicular ferrite, pearlite, martensite, and bainite. By predicting more transformation structure information, the prediction accuracy of mechanical properties is improved. However, in order to predict all of these transformed tissue information, the calculation load increases. Therefore, it is desirable to select and compute the transformed tissue information that has the greatest influence on the desired mechanical properties. Currently, there are 490 MPa class steels among the steel materials that are generally used for welded structures, but the welds of such general-purpose steels often have ferrite or even acicular ferrite as the main structure. The impact on mechanical properties is also significant. For this reason, when predicting the material quality of welds of such general-purpose steel materials, ferrite grain information including ferrite average grain diameter and ferrite content (ferrite transformation fraction) as transformation structure information, or further acicular ferrite average grain It is preferable to adopt the acicular ferrite grain information including the diameter and the amount of acicular ferrite (acicular ferrite transformation fraction). The ferrite grain information is preferably calculated by the following calculation method. A method for calculating acicular ferrite grain information will be described later.
[0015]
As the transformation structure calculation, calculating the ferrite transformation fraction using the austenite grain information, the weld metal composition and the austenite grain interface area determined by the temperature history, the nucleation rate of ferrite and the growth constant in the transformation structure calculation. Is preferred.
As a result, ferrite grain information (ferrite transformation fraction) can be calculated from two aspects of ferrite nucleation and grain growth, and the prediction accuracy of ferrite grain information is improved.
[0016]
Further, as the austenite grain interfacial area, it is preferable to use the grain interfacial area for the elongated grains based on the major axis and minor axis of the austenite grain information.
By this, in the transformation structure calculation, the shape of the austenite grains can be approximated to elongated grains such as a rectangular parallelepiped in conformity with the dendrites (columnar crystals) remarkable in the weld metal, and the grain boundary area corresponding to this can be used. The ferrite grain information can be predicted and calculated more accurately than the grain interface area based on the spherical grains in the case of steel materials.
[0017]
Further, as the nucleation rate of the ferrite, the grain boundary nucleation rate obtained based on the austenite grain interfacial area generated with the austenite grain boundary as the nucleus, and the crystallized product nucleation rate with the crystallized product as the nucleus, It is preferable to use the complex nucleation rate obtained based on the above.
As a result, not only nucleation from austenite grain boundaries but also nucleation from crystallized materials can be taken in as nucleation rate used in transformation structure calculation, and ferrite grain information can be predicted and calculated more accurately. .
[0018]
Furthermore, as the growth constant, it is preferable to use a corrected parabolic rate constant obtained by correcting the influence of pinning due to crystallized substances on the parabolic rate constant.
As a result, in the transformation structure calculation, the pinning effect due to the remarkable crystallized material in the weld metal (the effect that the crystallized material pins the growth and suppresses the growth) can be reflected in the calculation of the ferrite content, It is possible to predict and calculate ferrite grain information more accurately.
[0019]
On the other hand, when not only ferrite grain information but also acicular ferrite grain information is used as transformation structure information, the acicular ferrite transformation fraction is preferably calculated using the nucleation rate and growth constant of the acicular ferrite.
As a result, acicular ferrite grain information (acicular ferrite transformation fraction) can be calculated from the two aspects of acicular ferrite nucleation and growth, and the accuracy of acicular ferrite grain information prediction can be improved. it can.
[0020]
Further, as a preferred embodiment of the present invention, for each mesh region, a final heating / cooling process that is heated to a temperature Tas or higher at which austenite transformation starts in the heating process is obtained from the temperature history, and after this last heating / cooling process In the heating and cooling process of the above, the material of the predicted portion with respect to the material change caused by the recovery and tempering of each mesh region during the temperature range from the temperature Tff at which the ferrite transformation ends in the cooling process to the tempering temperature Perform tempering correction calculation to correct As the tempering correction calculation, in each heating and cooling process after the last heating and cooling process, a holding time held in a temperature region from a temperature Tff at which the ferrite transformation is completed in the cooling process to a tempering temperature is obtained, and each mesh region is obtained. An average holding time in the predicted portion is obtained from the holding time, and a method of correcting the material of the predicted portion predicted by the average holding time can be taken.
According to this aspect, it is possible to correct material changes due to structural changes due to recovery and tempering, and it is possible to further improve the material prediction accuracy of the welded portion or a predicted portion in the welded portion.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
The material prediction method of the present invention will be described for a case where base metal thick steel plates (hypereutectoid composition) are butted and arc welding is performed by a plurality of welding passes.
The material prediction method according to the embodiment is executed by a material prediction computer in which a program for executing various calculations shown in a flowchart described later is stored in a storage device. Information necessary for calculation, for example, a base material component, a welding wire component, a shielding gas component, and welding conditions are input and stored in the storage device by an operator or from a central management computer. Welding conditions include welding current, welding voltage, welding speed, interpass temperature (weld zone allowable temperature when welding of the next pass is started), base metal shape (plate thickness), groove shape, weld length, There are number of passes. Secondary data such as pass welding time (welding time required per pass = weld length / welding speed) is calculated from these data, and these are also stored in a storage device.
[0022]
The storage device stores, for example, various databases shown in the following items. The data in these databases has been investigated in advance through experiments. Data in the database is discrete, but is used after linear interpolation if necessary. It is also possible to obtain a relational expression by regression analysis of discrete data and use it in a formulated form.
I. Composition (base material, weld metal) and liquidus temperature, solidus temperature, temperature at which austenite transformation ends in the heating process (temperature at which all the structures become austenite) Taf, temperature at which austenite transformation starts in the heating process ( Temperature at which austenite begins to be generated) Tas, temperature Tfs at which ferrite transformation starts in the cooling process, temperature Tff at which ferrite transformation ends in the cooling process, tempering temperature (lower limit temperature at which material changes substantially due to recovery and tempering, (Normally about 300 ° C), and a phase diagram database showing the relationship with other transformation temperatures (for example, martensite transformation start temperature, pearlite transformation start temperature) as necessary
B. A crystal showing the weld metal composition, the type of crystallized material to be crystallized, the melting heating temperature and the solidification time from the same temperature to the solidus temperature, the average particle size of the crystallized material, and the amount of crystallized (number per unit area) Birth database
C. Austenite structure database showing the relationship between the austenite heating temperature (above Taf and below the solidus temperature), the cooling rate from the heating temperature to Tfs, and the major axis and minor axis of the austenite grains
D. Precipitate database showing weld metal composition and types of precipitates to be precipitated, austenite heating temperature, precipitation time from the same heating temperature to tempering temperature, average particle size of precipitates and amount (number per unit area)
E. Material prediction database showing the material (mechanical properties) from the types and amounts of crystallized materials and precipitates, the ferrite grain size and the amount (fraction)
F. A material prediction database that shows the material (mechanical properties) from the types and amounts of crystallized materials and precipitates, and the particle sizes and amounts (fractions) of ferrite and acicular ferrite. Tempering correction database showing the relationship between tempering time from Tff to tempering temperature and material correction factor
[0023]
Hereinafter, the material prediction method according to this embodiment will be described in detail with reference to the main flowchart shown in FIG.
[0024]
First, the base material component, the wire component, the shield gas component, and the welding conditions, which are the basis of the prediction calculation, are input to the material prediction computer (S1) and stored in the storage device.
[0025]
Next, as shown in FIG. 2, the welded portion including the weld metal portion 2 and the heat affected zone assumed from the groove shape of the base materials 1 and 1 is partitioned into a large number of mesh regions 3 (S2). Although it is preferable in terms of prediction accuracy that the welded part is divided more than the number of passes in the width and height directions of the weld metal part, the calculation time increases if it is divided more finely than necessary. Usually, it may be divided into about 2 to 8 times the number of passes. The position of each mesh region is determined by setting an appropriate coordinate axis. In the illustrated example, the X axis is set to the right, the Y axis is set upward, and the Z axis is set in a direction perpendicular to the paper surface.
[0026]
Then, initial values are set for the welding pass i and the elapsed time t per unit time (S3). Since the welding pass starts from the first pass, i = 1. The unit time may be a time at which the metallurgical change can be observed. In this embodiment, the unit time is 1 second, so the elapsed time t is set to t = 1.
[0027]
  Next, the following calculation is performed for each mesh region.
(1)  Mesh region temperature history calculation (S4)
  This calculation is based on the base metal component, the wire component and the welding conditions, or the weld metal composition before the unit time has elapsed (when the base metal and the wire have already melted to form a weld metal) and the welding conditions. The temperature of each mesh region is obtained every unit time, and this is sequentially repeated up to the final welding pass N to obtain the temperature history from the first pass to the final pass. The temperature of the mesh region can be calculated using a three-dimensional heat conduction equation. It should be noted that data such as the thermal conductivity and density of the heat conduction medium (base material, wire or weld metal) necessary for this calculation is appropriately read from the database. Moreover, since the software used for this calculation is supplied to the market as thermal analysis software, it can be used.
[0028]
(2)  Weld metal composition calculation (S5)
  This calculation is based on the base metal component, wire component, shield gas component, and mesh region temperature after the unit time obtained by the mesh region temperature history calculation, or the weld metal composition before the unit time (already with the base material). Wires melt together to form weld metaldo itAnd the weld metal composition of the mesh region after the unit time has elapsed is determined based on the mesh region temperature after the unit time has elapsed. More specifically, the weld metal composition can be calculated by averaging the compositions of all mesh regions that are equal to or higher than the liquidus temperature after the unit time has elapsed.
[0029]
  the above(1) , (2)Is calculated for each unit time, whether or not the elapsed time t has passed the pass welding time tp is determined (S6), and the state after the unit time is sequentially calculated until the time elapses (S7). After the elapse of time, it is determined whether or not the temperature T of the mesh region is equal to or lower than the interpass temperature Tpass (S8), and the state after the unit time is sequentially calculated until the temperature becomes lower than the interpass temperature (S7). Further, when the temperature is lower than the interpass temperature, it is determined whether or not the current pass is the final pass (S9), and the state after the unit time elapses in the next pass is sequentially calculated until the final pass (S10).
[0030]
The temperature history of a certain mesh area calculated as described above is illustrated in FIG. The figure is an example of the temperature history of the mesh region in the weld metal part formed in the first pass when welding in five passes, and for each pass from the first pass to the final pass (fifth pass). The maximum temperature is recorded, and this maximum temperature is gradually decreased as the process proceeds to the subsequent stage. “Pn” in the figure indicates the peak temperature of the n-th pass. The figure also shows the liquidus temperature, the solidus temperature, Tas, Taf, Tfs, Tff, and the tempering temperature Tp.
[0031]
Next, crystallization calculation (S11) is performed. This calculation is based on the weld metal composition and the temperature history to obtain crystallized information such as the composition of the crystallized material, the average particle diameter, and the amount of crystallized (number per unit area). The crystallization component calculated based on the composition and the amount of crystallization is subtracted to obtain the weld metal composition after crystallization. More specifically, the composition of the crystallized product is obtained from the database based on the weld metal composition, and the final heating / cooling process heated above the liquidus temperature is obtained from the temperature history, and the maximum temperature in this last heating / cooling process is obtained. The elapsed time from the maximum temperature to the solidus temperature is obtained, and the particle size and the amount of crystallized matter are obtained from the database based on the maximum temperature and the elapsed time.
Referring to FIG. 3 as an example, the last cooling process that is heated above the liquidus temperature is a heating and cooling process that peaks at P2, and the elapsed time from P2 to the passage of the solidus temperature from P2. Ask for. In the heating / cooling process having a peak at P1, the maximum temperature is P1, and crystallized substances are crystallized during cooling, but most of the crystallized substances are remelted in the next heating / cooling process. For this reason, in this embodiment, the crystallization calculation is performed for the heating and cooling process in which the crystallized material remaining after solidification is mainly crystallized and P2 is the peak. This makes it easy to calculate. The crystallized material remaining after the end of the final welding pass is strictly (crystallized product generated during cooling in the heating / cooling process with P1 as the peak)-(heating during the heating / cooling process with P2 as the peak). Recrystallized crystallized product) + (crystallized product generated during cooling in the heating and cooling process with P2 as the peak).
[0032]
Next, austenite structure calculation (S12) is performed. This calculation is to obtain austenite grain information such as the major axis, minor axis, and average grain size of austenite grains based on the post-crystallization weld metal composition and temperature history obtained by crystallization calculation. More specifically, the last heating / cooling process over Taf is obtained from the temperature history, and the maximum temperature and the cooling rate from the highest temperature to Tfs are obtained for this last heating / cooling process. Based on the above, the austenite grain information is obtained from the database. There are various ways of expressing the average particle diameter, but it is usually derived from the major axis and minor axis. For example, it can be represented by a circular diameter having an area equivalent to an elliptical area having a major axis and a minor axis.
Referring to FIG. 3 as an example, the last heating / cooling process heated to Taf or higher is a heating / cooling process having a peak at P3, and the maximum temperature P3 and the cooling rate from P3 to Tfs are obtained. By obtaining austenite grain information about the last heating and cooling process exceeding Taf, information on austenite grains before the ferrite transformation can be obtained, and useless calculation can be omitted.
[0033]
Next, precipitation calculation (S13) is performed. This calculation is to obtain precipitate information such as the composition of the precipitate, the average particle size, and the amount of precipitation (number per unit area) based on the post-crystallization weld metal composition and temperature history obtained by the crystallization calculation. It is. More specifically, the composition of the precipitate is obtained from the database based on the weld metal composition after crystallization, and the final heating / cooling process heated to the solidus temperature or higher is obtained from the temperature history. The elapsed time from reaching the solidus temperature to the tempering temperature is determined, and the particle size and crystallization amount of the precipitate are determined from the database based on the solidus temperature and the elapsed time.
Referring to FIG. 3 as an example, the last heating / cooling process that is heated above the solidus temperature is a heating / cooling process that peaks at P2, and the elapsed time t1 from the time when the solidus temperature is reached to the tempering temperature Tp. Ask for. Since the precipitate is deposited up to about Tp, the elapsed time during this period is determined.
[0034]
Next, transformation structure calculation (S14) is performed. For example, when ferrite transformation is considered, this calculation includes ferrite grains containing the average grain size and amount (transformation fraction) of ferrite precipitated from austenite grain boundaries based on the weld metal composition after crystallization, austenite grain information, and temperature history. Information is obtained from the ferrite transformation model.
As the ferrite transformation model, a model described by the following formula (No. 131, 132, Nishiyama Memorial Technology Lecture, “Present and Future of Material Prediction / Control Technology of Steel Materials”, edited by Japan Iron and Steel Institute, 1989 Published in 2009, see the paper “Quantification of Phase Transformation Behavior from Unprocessed and Work-Hardened Austenite”).
Ferrite grain size Dα = f (α, Is, t, S)
Ferrite fraction X = g (α, Is, t, S)
Here, t is the holding time, S is the grain interface area of austenite in the unit volume, α is the growth coefficient (called parabolic constant), and Is is the nucleation rate per unit area. S is calculated from austenite grain information, Is is calculated from the austenite composition and holding temperature, and α is calculated theoretically or experimentally from the austenite composition (C concentration in the weld metal composition after crystallization) and holding temperature. Desired.
[0035]
In the present embodiment, the last heating / cooling process (heating / cooling process having a peak at P4 in FIG. 3) that is heated to Tas or higher is obtained, and the highest temperature (P4 in FIG. 3) (maximum temperature) in this last heating / cooling process. When Tf exceeds Taf, Tfs) is set as the holding temperature, and the elapsed time from the maximum temperature to Tff is used as the holding time t. Accordingly, it is possible to omit the transformation structure calculation not related to the final ferrite structure in the heating and cooling process before the last heating and cooling process.
[0036]
In the calculation of the ferrite transformation fraction X and the ferrite grain size Dα, the austenite grain interface area S is used, and this S is calculated from the austenite grain information as described above. As the austenite grain interfacial area, when the austenite grains can be assumed to be almost spherical as in the case of a steel material, the grain interfacial area obtained when the austenite grains having an average grain size are densely packed in a unit volume can be used. However, since there is a region in which the austenite grains are not spherical but columnar crystals in the weld metal, accuracy is reduced if the austenite grains are approximated to a spherical shape. Therefore, in the present embodiment, the austenite grain information includes the major axis and minor axis (or one of these dimensions and the aspect ratio (= major axis / minor axis), or the average grain size and aspect ratio), and the ellipsoid or A rectangular parallelepiped-shaped particle shape is assumed, and the grain boundary area when the unit volume is packed densely is used. Thereby, prediction accuracy can be improved.
[0037]
The nucleation rate Is means the nucleation rate from the austenite grain boundary, and this Is can also be used in the present invention. However, since many crystallization products exist in the weld metal, nucleation occurs not only from the austenite grain boundaries but also from the crystallization products. For this reason, as the nucleation rate Ia, it is preferable to use (Is + Ic) (referred to as a composite nucleation rate) obtained by adding the nucleation rate Ic from the crystallized product to Is, rather than using Is alone. Ic is obtained from the database based on crystallized information and temperature.
[0038]
Further, as described above, the growth coefficient α is determined theoretically or experimentally by equilibrium calculation, and this α can also be used in the present invention. However, as described above, a large number of crystallized substances exist in the weld metal. This crystallized substance exerts a pinning effect so as to hinder the growth when the ferrite grows, and delays the growth. Therefore, it is preferable to correct this effect as the growth coefficient and use aα + b (a (0 <a ≦ 1), where b is a constant) (referred to as a corrected parabolic constant) as the growth coefficient αa. The constants a and b are obtained from the database based on the crystallized substance information.
[0039]
  When calculating acicular ferrite grain information as transformation structure information in transformation structure calculation, average particle size and amount of acicular ferrite precipitated from austenite based on weld metal composition and temperature history after crystallization (transformation fraction) ) Is obtained from the acicular ferrite transformation model.
  As the acicular ferrite (hereinafter, acicular ferrite may be abbreviated as AF) transformation model, a model described by the following equation can be used.
AF fraction = 1−expΣ (−k · Iaf (T) · αaf (T) · Δt^1/2)
AF average particle size = {ΣIaf (T) · αaf (T) · Δt}/ ΣIaf (T)
  Here, k is a constant, Iaf (T) is an AF nucleation rate at a certain temperature T, and αaf (T) is an AF growth constant at a certain temperature T. Δt is a minute time, and the accumulated time is from the starting temperature of the acicular ferrite transformation described later to the martensitic transformation starting temperature or the pearlite transformation starting temperature.
[0040]
Since the values of k, Iaf (T), and αaf (T) vary depending on the composition and temperature of the weld metal, they are calculated from a database obtained by experiment in advance or from a relational expression obtained by regression analysis of this database. Is done. When stricter calculation is performed, k, Iaf (T), and αaf (T) vary depending on the type of crystallized substance and its amount (number), and also the type of precipitate and its amount (number). Based on the above factors, the nucleation rate and growth constant calculation values are corrected by correction factors determined in advance by experiments. In addition to the composition and temperature of the weld metal, details including the type and amount of crystallized materials and precipitates are included. K, Iaf (T), and αaf (T) may be determined from the database.
[0041]
The transformation start temperature of the acicular ferrite (the end temperature of the ferrite transformation) pushes the transformation structure calculation for the ferrite grain information according to cooling, and the free energy of austenite regardless of whether the ferrite transformation is in progress or stopped (G (γ)) and the free energy (G (α)) of ferrite (ΔG = G (γ) −G (α)) is the temperature at the time when it reaches 600 J / mol (“Materials and Process ”, VOL.3, No. 3, p871, 1990). The free energy difference (ΔG) is calculated using a thermodynamic function before each transformation structure calculation. According to the above-mentioned document, bainite transformation starts at 600 J / mol or more. However, since acicular ferrite can be regarded as bainite generated from within austenite grains, it becomes higher than this value in the present invention. It was assumed that the transformation of acicular ferrite started.
The difference in free energy (ΔG) between austenite and ferrite is obtained by expressing G (γ) and G (α) as thermodynamic functions and taking the difference in the case of the same composition. Therefore, the value of ΔG is Ae_ThreeThe point is zero, and takes a larger value as the temperature decreases. G (γ) and G (α) are each expressed as a sum of thermodynamic data of each component, and in the present invention, Urenius: Hardenability Concepts with Applications to Steels (1978) ed. The calculation was performed using thermodynamic data by Doane and Kirkaldy, P28-81).
[0042]
For each mesh region, the crystallized information obtained by the crystallization calculation, the precipitate information obtained by the precipitation calculation, and the ferrite obtained by the transformation structure calculation, or further the particle size and amount of the acicular ferrite (transformation fraction) ) Is averaged over the entire region of the predicted portion by the predicted portion structure calculation (S15), and is calculated from the database based on the predicted portion tissue information or based on the database by the structure / material prediction calculation (S16). The material (mechanical characteristics) is predicted by the structure-material relational expression obtained by the regression analysis.
[0043]
The flow chart of FIG. 1 shows a case where the predicted part structure calculation and the structure / material prediction calculation are used as the material prediction calculation. However, as another method, the transformed tissue information calculated for each mesh region is used. Based on the above, the mesh region material prediction calculation may be performed for each mesh region, and then the material / material prediction calculation for obtaining the material of the entire predicted portion from the calculated material in each mesh region may be performed.
[0044]
Although the material of the welded portion is predicted as described above, the prediction accuracy can be further improved by correcting the influence of recovery and tempering by tempering correction calculation (S17).
This calculation assumes that recovery and tempering occur in a temperature range below Tff and above the tempering temperature, and in the temperature region from Tff to the tempering temperature, the recovery and tempering of each mesh region are taken into account and the prediction is made. The material of the part is corrected. As the correction method, for example, the time held in the temperature region is obtained for each mesh region, the average retention time in the entire weld is obtained based on this, the correction coefficient is obtained from the database based on this average retention time, and the material The material obtained by the prediction calculation can be corrected.
The calculation method of the holding time will be described in more detail. The last heating / cooling process (heating / cooling process having a peak at P4 in FIG. 3) in which the maximum temperature is equal to or higher than Tas is obtained from the temperature history. The total time (t2 in FIG. 3) that is equal to or lower than Tff and equal to or higher than the tempering temperature Tp is obtained. In FIG. 3, when there is a portion where the mesh region temperature once falls below Tp in the cooling process having a peak at P4, the time of the temperature interval above Tp is integrated.
As another correction method, when predicting the material from the tissue by the mesh region material prediction calculation for each mesh region and predicting the material of the predicted part by the material / material prediction calculation, the same method as in the above example is used. Perform a tempering correction calculation that corrects the material of the mesh region based on the holding time, and predicts the material of the predicted portion by the material / material prediction calculation based on the corrected material of each mesh region corrected by this calculation. You can also.
[0045]
By the way, as a method for predicting the material of the predicted portion, it is conceivable to directly predict the material of the predicted portion from the weld metal composition and the temperature history without obtaining the structure after welding as described above. However, steel materials including weld metals often undergo transformations and the like, and characteristics frequently change abruptly at certain chemical components and temperatures. For this reason, a chemical component, temperature history, and a material do not necessarily have a linear relationship. Therefore, when a material is predicted using a database composed of discrete data based on the calculated chemical composition and temperature history, even if the desired data point is interpolated by linear interpolation, the prediction is performed. The accuracy tends to vary, and the accuracy generally decreases. On the other hand, since the mechanical characteristics change continuously and more linearly with respect to the tissue, according to the method of obtaining the tissue based on the chemical composition and temperature history and predicting the material from the tissue as in the present invention. The prediction accuracy can be improved.
[0046]
【Example】
[Example 1]
Ingredients (mass%, the balance being substantially Fe) are C: 0.15%, Si: 0.37%, Mn: 1.43% of the base material (thick steel plate with a plate thickness of 20 mm), C: 0.06 %, Si: 0.80%, Mn: 1.60%, Ti: 0.26% CO with the following welding conditions₂Gas shielded arc welding is performed, and the material of the weld metal part (tensile) based on the structure information predicted by the crystallization calculation, austenite structure calculation, precipitation calculation, and transformation structure calculation considering only ferrite grain information as transformation structure information. Strength TS, Charpy impact value at 0 ° C vE₀) And actual welding was performed to actually measure the material of the same part.
・ Welding conditions
Under the following conditions, a groove V shape (opening angle 45 °), a gap of 6 mm, a welding speed of 25 cm / min, and a welding length of 0.35 m (pass welding time 84 seconds) were used.
Condition A: Number of passes 5, heat input 30 kJ / cm, temperature between passes 150 ° C.
Condition B: Number of passes 5, heat input 30 kJ / cm, interpass temperature 250 ° C.
Condition C: Number of passes 5, heat input 30 kJ / cm, interpass temperature 350 ° C.
Condition D: Number of passes 4, heat input 40 kJ / cm, interpass temperature 250 ° C.
Condition E: Number of passes 4, heat input 40 kJ / cm, interpass temperature 350 ° C
[0047]
As shown in FIG. 2, the size of the mesh area is determined from the height h of the groove shape, the width W, the number of passes N, and the weld length L, x-direction size = (h / N) / 4, y-direction size = (H / N) / 4, size in the welding length direction = L / 10.
[0048]
In the transformation structure calculation, the austenite grain interfacial area obtained by the following two methods A and B was used.
・ Method A
This is a conventional method using a steel material, and the austenite grain interfacial area S was determined by the following formula, where the average austenite grain size was Dγ.
S = 4 / (√π · Dγ)
・ Method B
Considering the crystal form (dendritic form) in the weld metal, when the major axis of the austenite grain is a and the minor axis is b, the austenite grain is assumed to be a rectangular parallelepiped of length a, width a, and height b. The grain interface area was determined.
[0049]
In the transformation structure calculation, ferrite nucleation rate and growth constant obtained by the following two methods A and B were used.
・ Method A
Theoretical calculations with conventional steel materials determined the nucleation rate Is from the austenite grain boundary as the nucleation rate Ia and the parabolic rate constant α as the growth coefficient αa (Ia = Is, αa = α).
・ Method B
Considering the crystallization product in the weld metal material, the compound nucleation rate (Ia) is obtained by adding Ic (nucleation rate from the crystallized product) to Is (nucleation rate from the austenite grain boundary) as the nucleation rate Ia. = Is + Ic). Further, a corrected parabolic rate constant was obtained as the growth coefficient αa by the following formula.
αa = aα + b
However, a = 0.86, b = −n / D × 0.6 × 10^{-twenty three}N is the number of crystallized substances (pieces / m²), D is the average particle size (m) of the crystallized product.
[0050]
An example of the calculation result by the material prediction calculation according to this embodiment is shown below. The welding condition is the condition A, and the mesh region illustrating the calculation result is a region at the position of x = 0 mm and y = 18 mm in the coordinate system of FIG. 2 as the final raw material portion, and x = 0 mm, y as the reheating portion. = Indicates a region at a position of 12 mm.
(1) Result of calculation of weld metal composition after the final heating and cooling process (mass%, the balance being substantially Fe)
C: 0.087%, Si: 0.52%, Mn: 1.16%, Ti: 0.05%, O: 0.39%
・ Reheating unit
C: 0.083%, Si: 0.51%, Mn: 1.08%, Ti: 0.05%, O: 0.033%
(2) Crystallization calculation results after the final heating and cooling process
・ Final quality part
Precipitate composition: TiO₂Average particle size: 660 nm, number: 9600 / mm²
・ Reheating unit
Precipitate composition: TiO₂, Average particle size: 480 nm, Number: 11000 / mm²
(3) Austenite structure calculation results after the final heating and cooling process
・ Final quality part
Austenite average particle diameter: 187 μm, aspect ratio: 2.68
・ Reheating unit
Austenite average particle size: 89 μm, aspect ratio: 1.18
(4) Transformation structure calculation
(1) Austenite grain interface area per unit volume after final pass S (m²/ M^Three)
・ Final quality part
Method A: 1.2 × 10^Four, Method B: 2.3 × 10^Four
・ Reheating unit
Method A: 2.5 × 10^FourMethod B: 3.4 × 10^Four
(2) Ferrite nucleation rate Ia and growth constant αa at 750 °
Method A: Ia = Is = 2.6 × 10⁷, Αa = α = 1.6 × 10^-6
Method B: Ia = Is + Ic = 1.1 × 10⁸, Αa = 1.3 × 10^-6
[0051]
Table 1, FIG. 4, and FIG. 5 show the material prediction calculation results and the actual measurement results. From the figure, there is little error between the material prediction result and the actual measurement result according to the present embodiment, and good agreement is recognized.
[0052]
[Table 1]

[0053]
[Example 2]
Ingredients (mass%, the balance being substantially Fe) are C: 0.15%, Si: 0.37%, Mn: 1.43% of the base material (thick steel plate with a plate thickness of 20 mm), C: 0.06 %, Si: 0.80%, Mn: 1.60%, Ti: 0.26%₂Welded metal based on microstructure information predicted by transformation structure calculation considering ferrite grain information and acicular ferrite grain information as crystallization calculation, austenite structure calculation, precipitation calculation and transformation structure information. Material (tensile strength TS, Charpy impact value at 0 ° C vE₀) And actual welding was performed to actually measure the material of the same part.
[0054]
The welding conditions are the same as the base material groove conditions as in Example 1, the number of welding passes is 5, the welding speed is 25 cm / min, and the cooling rate after welding is variously set. Changed and adjusted the cooling rate.
In calculating the ferrite grain information, the interfacial area of austenite grains was determined by Method B of Example 1, and the nucleation rate and growth constant of ferrite were determined by Method B of Example 1. In calculating the acicular ferrite grain information, k (constant), precipitation rate Iaf (T), and growth constant αaf (T) of the acicular ferrite model are the following values and formulas obtained by regression analysis of the database. Calculated.

[0055]
An example of the calculation result by the material prediction calculation by a present Example is shown in Table 2, FIG. 6, and FIG. Table 2 also shows calculated values and actual measurement values for the amount of ferrite and acicular ferrite (volume%) in the mesh region (upper center of the weld metal) at the position of x = 0 mm and y = 18 mm in the coordinate system of FIG. Indicates. In addition, the average cooling rate from the highest reached temperature to the ferrite transformation end temperature in the final heating and cooling process in which heating is performed at an austenite transformation start temperature Tas or higher is also shown in the same table. Measured values such as the amount of ferrite were measured as follows. A specimen for structure observation was cut out so that the cross section of the welded portion could be observed, and after polishing, the metal structure was revealed by nital etching, a 400 × optical microscope photograph was taken, and an image analyzer (Image Pro PLUS ( The amount of tissue was measured in terms of the area ratio at Planetron, Inc.)), and the value was considered to be equal to the volume ratio.
[0056]
From FIG. 6 and FIG. 7, there is a small error between the material prediction result and the actual measurement result according to the present embodiment, and good agreement is recognized. In addition, it can be seen that the amount of ferrite and the amount of acicular ferrite are also excellent in prediction accuracy.
[0057]
[Table 2]

[0058]
【The invention's effect】
  According to the material prediction method of the present invention,For the weld metal part with the formation of crystallized material, predict and calculate the structure considering the crystallized material,Since the material property of the predicted portion in the welded portion is predicted based on the predicted and calculated structure, the material property of the predicted portion can be predicted with high accuracy. Moreover, since the material of the to-be-predicted part can be accurately predicted according to the welding conditions, the optimum welding conditions for obtaining target characteristics can be determined based on the prediction results.
[Brief description of the drawings]
FIG. 1 is a main flowchart showing a material prediction method according to an embodiment.
FIG. 2 is a cross-sectional explanatory view showing a mesh region in a welded portion.
FIG. 3 is a time-temperature diagram showing a temperature history of a mesh region.
4 is a graph showing a relationship between a predicted calculated value of tensile strength and an actually measured value in Example 1. FIG.
FIG. 5 is a graph showing a relationship between a predicted calculation value of a Charpy impact value and an actual measurement value in Example 1.
6 is a graph showing a relationship between a predicted calculated value of tensile strength and an actually measured value in Example 2. FIG.
7 is a graph showing the relationship between a predicted calculation value of Charpy impact value and an actual measurement value in Example 2. FIG.

Claims (17)

A method for predicting a material of a predicted part in a welded part including a welded metal part and a heat-affected part of a welded structure in which base materials are welded via a welded metal part by one or a plurality of welding passes. ,
A plurality of mesh regions are assumed by dividing the welded portion, and for each mesh region, the following calculations (1) , (2) , (3) , and (4) are performed,
(1) For each unit time until the end of the final welding pass, the mesh region temperature after the unit time has been sequentially determined based on the base metal component, the wire component or the weld metal composition and the welding conditions before the unit time has elapsed. Mesh region temperature history calculation for calculating the temperature history from the first welding pass to the final welding pass,
(2) Every unit time until the end of the final welding pass, the unit is based on the base metal component, wire component, shield gas component or weld metal composition before the unit time elapses and the mesh region temperature after the unit time elapses. Weld metal composition calculation to determine the weld metal composition after the passage of time,
(3) After the mesh region temperature history calculation and weld metal composition calculation, based on the weld metal composition and temperature history, crystallized information including crystallized composition, average grain size and crystallized substance amount, and crystallized Crystallization calculation for obtaining the weld metal composition after crystallization after the object has crystallized,
(4) transformation structure calculation to obtain transformation structure information including the fraction of transformation structure in the mesh region based on the post-crystallization weld metal composition and temperature history,
Thereafter, based on the crystallized material information and the transformation structure information after the end of the final welding pass in each mesh region in the predicted portion, the material of the predicted portion is determined from the relationship between the structure and the material obtained by actual measurement in advance. A method for predicting the material quality of a welded part, which performs a material prediction calculation.   The material prediction calculation was obtained by the predicted portion structure calculation for obtaining the structure information in the predicted portion from the transformation structure information after the final welding pass in each mesh region in the predicted portion, and the predicted portion structure calculation. The material prediction method according to claim 1, further comprising: a structure / material prediction calculation for predicting a material of the predicted portion based on a relationship between the structure and the material obtained in advance by actual measurement based on the predicted portion structure information.   The material prediction calculation is a mesh region material prediction calculation that predicts the material in each mesh region from the relationship between the welded tissue and the material obtained in advance by actual measurement based on the transformation structure information in each mesh region in the predicted portion. And a material / material prediction calculation for predicting a material of the predicted portion based on a material in each mesh region in the predicted portion determined by the mesh region material prediction calculation. Prediction method. The austenite grain information is obtained to obtain austenite grain information based on the post-crystallization weld metal composition and temperature history before the transformation structure calculation, and the austenite grain information is also used as a basis for the transformation structure calculation. The material prediction method described in any one of 1 to 3.   For each mesh region, find the last heating and cooling process heated above the temperature Taf at which the austenite transformation ends in the heating process from the temperature history, and perform the austenite structure calculation for this last heating and cooling process, The material prediction method according to claim 4. Before performing the transformation structure calculation, perform precipitation calculation to obtain precipitate information including the composition of the precipitates precipitated in the structure based on the weld metal composition and temperature history after crystallization , the average particle size, and the precipitation amount, The material prediction method according to any one of claims 1 to 5, wherein precipitate information after the final welding pass is also used as a basis for the material prediction calculation. Wherein for each mesh region, determined the final heating and cooling step, which is heated above the solidus temperature from the temperature history, performing crystallization calculated for this final heating and cooling step, of claims 1-6 The material prediction method described in any one of the items . For each of the mesh regions, a final heating / cooling process that is heated to a temperature Tas or higher at which austenite transformation starts in the heating process is determined from the temperature history, and a transformation structure calculation is performed for the last heating / cooling process. The material prediction method according to any one of claims 1 to 7 . The material prediction method according to any one of claims 1 to 8 , wherein the transformation structure information is ferrite grain information including an average ferrite grain diameter and an amount of ferrite. In the transformation structure calculations, to calculate the ferrite transformation fraction using austenite grain boundary area is determined, the nucleation rate and growth constants of the ferrite the austenite grain information and weld metal composition and temperature history, according to claim 9 Material prediction method. The material prediction method according to claim 10 , wherein a grain interface area for elongated grains based on a major axis and a minor axis of austenite grain information is used as the austenite grain interface area. As the nucleation rate of the ferrite, based on the grain boundary nucleation rate determined based on the austenite grain interfacial area calculated from the austenite grain information, and the crystallized product nucleation rate that is generated using the crystallized product as a nucleus. The material prediction method according to claim 10 or 11 , wherein the obtained complex nucleation rate is used. The material prediction method according to any one of claims 10 to 12 , wherein a corrected parabolic rate constant obtained by correcting an influence of pinning due to a crystallized substance is used as the growth constant. The material prediction method according to any one of claims 9 to 13 , wherein the transformation structure information is acicular ferrite grain information including an acicular ferrite amount and an acicular ferrite average particle diameter. The material prediction method according to claim 14 , wherein the acicular ferrite transformation fraction is calculated using the nucleation rate and growth constant of acicular ferrite obtained from the weld metal composition and temperature history after crystallization . For each mesh region, a final heating / cooling process in which heating is performed at a temperature Tas or higher at which austenite transformation starts in the heating process is obtained from the temperature history, and in the heating / cooling process after the last heating / cooling process, ferrite is Tempering correction calculation is performed to correct the material of the predicted portion with respect to the recovery and tempering of the material in each mesh region during the temperature range from the temperature Tff at which transformation ends to the tempering temperature. The material prediction method according to any one of claims 1 to 15 . The tempering correction calculation calculates a holding time held in a temperature region from a temperature Tff at which the ferrite transformation is completed in the cooling process to a tempering temperature in the heating and cooling process after the last heating and cooling process, The material prediction method according to claim 16 , wherein an average holding time in the predicted part is obtained from a holding time, and the material of the predicted part predicted by the average holding time is corrected.

Images