Fatigue intensity estimation method and fatigue intensity estimation device

Abstract

This method allows for easy estimation of the fatigue strength of welded joints subjected to additional loads. [Solution] In numerical analysis using an analysis model of a welded joint that has been subjected to impact treatment on the welded portion, the combination of stress change that occurs at stress concentration points when a repeated load is applied to the analysis model and the fatigue strength that is expected when the repeated load is actually applied to the welded joint is prepared as stress-fatigue strength data, and reference information is prepared showing multiple stress-fatigue strength data obtained by changing the repeated load in multiple ways (Step S11). In numerical analysis using the analysis model, after applying an additional load different from a predetermined repeated load to the analysis model, the stress change that occurs at stress concentration points when the predetermined repeated load is applied to the analysis model is obtained (Step S12), and the fatigue strength of the welded joint subjected to the additional load is estimated by referring to the reference information using this stress change (Step S13).

Description

[Technical Field]

[0001] This invention relates to a technique for estimating the fatigue strength of welded joints. [Background technology]

[0002] Conventionally, the fatigue strength of welded joints has been improved by applying compressive residual stress to the weld. Impact treatments such as needle peening are frequently used to apply compressive residual stress.

[0003] Furthermore, Patent Document 1 discloses a method for estimating the fatigue strength of a component without conducting fatigue tests. In this method, a stress distribution is generated at stress concentration points, with the set nominal stress causing the stress to flow from the surface into the base material. The stress intensity factor range is calculated assuming that the tip of the natural crack experiences stress corresponding to the distance from the surface of the stress concentration point. If this stress intensity factor range matches the lower limit stress intensity factor range of the material obtained from the set stress ratio, the set nominal stress at that time is determined as the estimated fatigue strength.

[0004] Patent Document 2 discloses a method for predicting the fatigue life of micro-notched materials having minute notches. In this method, fatigue tests are performed using multiple samples with different notch depths and notch tip radii, and an S / N curve is created for each sample. In addition, the stress distribution at the notch cross section of each sample is estimated for the nominal stress applied in the fatigue test, and a characteristic distance defined by a predetermined formula is determined for each sample. Based on the stress distribution and characteristic distance, the characteristic distance mean stress, which is the average stress from the bottom of the notch to the characteristic distance, is determined, and the relationship between the characteristic distance mean stress and the fatigue crack initiation life is determined in advance from this characteristic distance mean stress and the S / N curve. This relationship is used when predicting the fatigue life of micro-notched materials.

[0005] Non-Patent Document 1 discloses the so-called Siebel diagram. Non-Patent Document 2 describes setting the value of m in the slope of the SN curve, represented by (-1 / m), to 5 for welded joints that have undergone peening treatment with a needle or the like. Furthermore, Table 4 of Non-Patent Document 2 describes how the FAT class can be reduced by the stress ratio. Figure 3.11 of Non-Patent Document 3 shows the effect of mean stress on fatigue strength. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2018-141703 [Patent Document 2] Patent No. 5212146 [Non-patent literature]

[0007] [Non-Patent Document 1] E. Siebel, M. Stieler, "Ungleichformige Spannungsverteilung bei schwingender Beanspruchung", VDI-Z, 1955, vol. 97, no. 5, pp. 121-126. [Non-Patent Document 2] Gary B. Marquis, Zuheir Barsoum, “IIW Recommendations for the HFMI Treatment : For Improving the Fatigue Strength of Welded Joints (IIW Collection),” Springer, 2016. [Non-Patent Document 3] Japan Society of Steel Construction, "Guidelines for Fatigue Design of Steel Structures and Commentary - with Design Examples," Gihodo Publishing, 2012. [Overview of the Initiative] [Problems that the invention aims to solve]

[0008] Incidentally, fatigue strength of welded joints is measured by fatigue tests in which repeated loads are applied to the welded joint. On the other hand, in actual structures, relatively large loads (excessive loads) may be applied to welded joints due to typhoons, earthquakes, etc. Also, for example, compressive loads may be applied to welds that are expected to be subjected to tensile loads. It is thought that the fatigue strength of a welded joint subjected to a load different from the repeated load (hereinafter referred to as "additional load") will change compared to when no additional load is applied. Therefore, there is a need for a method to easily estimate the fatigue strength of a welded joint subjected to an additional load.

[0009] This invention has been made in view of the above problems, and aims to easily estimate the fatigue strength of a welded joint subjected to an additional load. [Means for solving the problem]

[0010] One aspect of the present invention is a fatigue strength estimation method for estimating the fatigue strength of a welded joint, comprising: a) a step of preparing reference information that shows a combination of stress change occurring at stress concentration points when a repeated load is applied to the analysis model in a numerical analysis using an analysis model of a welded joint that has been subjected to impact treatment on the welded portion, and the fatigue strength expected when the repeated load is actually applied to the welded joint, as stress-fatigue strength data obtained by changing the repeated load in multiple ways; b) a step of obtaining the stress change occurring at the stress concentration points when the predetermined repeated load is applied to the analysis model after applying an additional load different from a predetermined repeated load in a numerical analysis using the analysis model; and c) a step of estimating the fatigue strength of the welded joint subjected to the additional load by referring to the reference information using the stress change obtained in step b).

[0011] A second aspect of the present invention is the fatigue strength estimation method of the first aspect, wherein the stress concentration area is the surface of the indentation caused by the impact treatment.

[0012] Aspect 3 of the present invention is a fatigue strength estimation method according to Aspect 1 (which may be Aspect 1 or 2), wherein the stress-fatigue strength data represents a combination of a value derived from the maximum and minimum values of stress in the stress change and the fatigue strength.

[0013] Aspect 4 of the present invention is a fatigue strength estimation method according to Aspect 1 (which may be any one of Aspects 1 to 3), wherein the additional load is a load that generates a compressive stress at the stress concentration portion.

[0014] Aspect 5 of the present invention is a fatigue strength estimation method according to any one of Aspects 1 to 4. In the step a), by numerical analysis using the analysis model, the stress gradient directed inward from the stress concentration portion and the stress concentration coefficient at the stress concentration portion are obtained. The fatigue strength in the stress-fatigue strength data is obtained from the fatigue strength reduction coefficient obtained using the stress gradient and the stress concentration coefficient and the fatigue strength of the smoothing material.

[0015] Aspect 6 of the present invention is a fatigue strength estimation device for estimating the fatigue strength of a welded joint. In numerical analysis using an analysis model of a welded joint subjected to impact treatment at the welded portion, when a repeated load is applied to the analysis model, a combination of the stress change generated at the stress concentration portion and the fatigue strength assumed when the repeated load is actually applied to the welded joint is used as stress-fatigue strength data. A reference information storage unit stores reference information indicating a plurality of stress-fatigue strength data obtained by changing the repeated load in a plurality of ways. An input reception unit receives an input of the stress change generated at the stress concentration portion when the predetermined repeated load is applied to the analysis model after applying an additional load different from the predetermined repeated load to the analysis model in the numerical analysis using the analysis model. A fatigue strength estimation unit estimates the fatigue strength of the welded joint that has received the additional load by referring to the reference information using the stress change received by the input reception unit.

Advantages of the Invention

[0016] According to the present invention, the fatigue strength of a welded joint subjected to an additional load can be easily estimated.

Brief Description of the Drawings

[0017] [Figure 1] It is a perspective view showing a part of a welded structure. [Figure 2] It is a longitudinal sectional view showing a part of the welded joint in an enlarged manner. [Figure 3] It is a view showing a part of an analysis model of the welded joint. [Figure 4A] It is a view showing the change in the nominal stress of the base material when only a repeated load is applied. [Figure 4B] It is a view showing the change in the nominal stress of the base material when an additional load and a repeated load are applied. [Figure 5] It is a view showing the change in the stress at the stress concentration part. [Figure 6] It is a view showing the SN curve. [Figure 7A] It is a view showing the change in the nominal stress of the base material when only a repeated load is applied. [Figure 7B] It is a view showing the change in the stress at the stress concentration part when only a repeated load is applied. [Figure 8A] It is a view showing the change in the nominal stress of the base material when an additional load and a repeated load are applied. [Figure 8B] It is a view showing the change in the stress at the stress concentration part when an additional load and a repeated load are applied. [Figure 9] It is a view showing the change in the stress at the stress concentration part. [Figure 10] It is a view showing the configuration of the fatigue strength estimation system. [Figure 11] It is a view showing the flow of the process for estimating the fatigue strength of the welded joint. [Figure 12] It is a view showing the flow of the reference information preparation process. [Figure 13] It is a view showing the stress at each position from the stress concentration part toward the inside. <000This is a diagram of Siebel's lines. [Figure 15] This figure shows an example of an S / N curve for a welded joint. [Modes for carrying out the invention]

[0018] Figure 1 is a perspective view showing a part of welded structure 1. Welded structure 1 is a structure in which multiple base materials are joined by welding. Welded structure 1 is a relatively large structure such as the floating body of a semi-submersible offshore wind power generation device or a transfer press. In Figure 1, the mutually orthogonal X, Y, and Z directions are indicated by arrows.

[0019] The welded structure 1 in Figure 1 includes a flat plate-shaped base material 11 that is substantially perpendicular to the Z direction and a flat plate-shaped base material 12 that is substantially perpendicular to the Y direction. The upper surface of the base material 11 facing the (+Z) direction is joined to the lower end surface of the base material 12 facing the (-Z) direction by welding, forming a welded joint 3. The welded joint 3 is a T-joint in which the base material 11 and the base material 12 are welded to each other substantially perpendicularly. At the connection between the two sides of the base material 12 facing the Y direction and the upper surface of the base material 11, weld metal 31 (i.e., a weld line) extending substantially in a straight line is formed. The weld metal 31 is metal that has melted due to the heat during welding and then solidified, and extends in the X direction. The weld metal 31 is also called a weld bead. In the welded joint 3, the base material 11 and the base material 12 are joined by the weld metal 31.

[0020] Figure 2 is a longitudinal cross-sectional view showing an enlarged portion of the welded joint 3 shown in Figure 1. In Figure 2, a cross-section perpendicular to the X direction is shown in the portion of the welded joint 3 near the weld metal 31. Of the base materials 11 and 12, the portion 32 enclosed by the dashed line around the weld metal 31 is the heat-affected zone 32, which was not melted by the heat during welding, but underwent changes in its structure, metallurgical properties, mechanical properties, etc., due to the heat. The weld metal 31 and the heat-affected zone 32 constitute the welded portion 33. A recess 311 is provided on the surface of the weld metal 31. The recess 311 is not shown in Figure 1. The recess 311 is an indentation formed by impact treatment of the welded portion 33. Impact treatment is a process in which the welded portion 33 is struck to cause plastic deformation to the surface layer of the welded portion 33. By applying impact treatment to the welded portion 33, the fatigue strength of the welded joint 3 is increased. The fatigue strength of the welded joint 3 is, for example, the stress range in the SN curve described later (see Figures 6 and 15). The fatigue strength may also be expressed as stress amplitude, etc.

[0021] The impact treatment of the weld 33 is performed, for example, by continuously striking the surface of the weld 33 with a metal or non-metallic needle having a roughly spherical, rounded tip. In other words, the impact treatment in this example is needle peening. As a result, the surface of the weld metal 31 is indented, and the aforementioned recess 311 is formed. The pin impact on the weld 33 is performed, for example, on the toe end of the weld metal 31 on the base metal 11 side (i.e., the portion of the weld metal 31 near the boundary between the surface of the base metal 11 and the surface of the weld metal 31). The recess 311 may be formed by an impact treatment other than needle peening (for example, shot peening, roll peening, laser peening, water jet peening, etc.).

[0022] The impact treatment on the weld 33 may be performed on parts of the weld metal 31 other than the toe as described above. Alternatively, the impact treatment on the weld 33 may be performed on the boundary between the weld metal 31 and the heat-affected zone 32 of the base metal 11, so that the recess 311 is formed across both the weld metal 31 and the heat-affected zone 32. Or, it may be performed on the heat-affected zone 32 of the base metal 11 instead of the weld metal 31. Furthermore, the impact treatment on the weld 33 may be performed on the toe of the weld metal 31 on the base metal 12 side, on the boundary between the weld metal 31 and the heat-affected zone 32 of the base metal 12, or on the heat-affected zone 32 of the base metal 12.

[0023] Next, the fatigue evaluation of the welded joint 3 will be explained. Generally, in the welded portion 33 of the welded joint 3, the fatigue strength in the direction perpendicular to the weld line (in the example of Figure 2, the Y direction or Z direction) is smaller than the fatigue strength in the direction parallel to the weld line. In the following explanation, we will focus on the stress in the Y direction when a load in a predetermined direction (for example, a load in the Y direction or Z direction) is applied to the base material 11, and will simply refer to the stress in the Y direction as "stress". Also, in the welded joint 3 in which a recess 311 is formed as shown in Figure 2, stress concentrates at the bottom of the surface of the recess 311 (labeled A in Figure 2), making it prone to fatigue cracks, so this bottom is referred to as "stress concentration area A". In this processing example, stress concentration area A is the surface of the indentation caused by the impact treatment, and is typically the part of the welded portion 33 where the stress is maximum.

[0024] Figure 3 shows a part of the analysis model of the welded joint 3 including the recess 311. The analysis model is used for numerical analysis. In the numerical analysis described herein, the finite element method (FEM) is used, but other analysis methods such as the finite difference method and boundary element method may also be used. Figure 4A shows the change in the nominal stress (nominal stress in the Y direction) of the base material 11 when only a cyclic load is applied to the analysis model. Figure 4B shows the change in the nominal stress of the base material 11 when an additional load different from the cyclic load is applied to the analysis model, and then the cyclic load is applied. In Figure 4B, the period during which the additional load was applied is indicated by arrow P1, and the period during which the cyclic load was applied is indicated by arrow P2 (the same applies to Figures 5, 8A, 8B, and 9 described later). The average stress during the cyclic load period P2 is shown by a dashed line. In Figures 4A and 4B, positive values ​​indicate tensile stress, and negative values ​​indicate compressive stress. The same applies to other figures showing changes in stress. In the following explanation, the expression "large (or small) stress value" means a large (or small) signed stress value, where a positive value indicating tensile stress is greater than a negative value indicating compressive stress.

[0025] Figure 5 shows the change in stress at stress concentration point A. In Figure 5, line L11 shows the change in stress at stress concentration point A when only repeated loads are applied to the analysis model, and line L12 shows the change in stress at stress concentration point A when additional loads are applied to the analysis model before the repeated loads are applied. Due to the impact treatment on the welded joint 33, a compressive residual stress U1 (hereinafter also referred to as "initial residual stress U1") is applied to stress concentration point A. Similar to Figure 4A, when only repeated loads are applied to the analysis model, the change in stress at stress concentration point A changes periodically in accordance with the repeated load, as shown by line L11. At this time, the maximum value, which is the value at the peak of each peak, is approximately constant, and the minimum value, which is the value at the bottom of each valley, is also approximately constant. In this way, the stress at stress concentration point A changes periodically in accordance with the period of the repeated load, between approximately constant maximum and minimum values. Note that the minimum value on line L11 is approximately equal to the initial residual stress U1.

[0026] On the other hand, similar to Figure 4B, when additional loads and repeated loads are applied, as shown by line L12, the stress value at stress concentration A becomes smaller than the initial residual stress U1 during the first half of the additional load period P1 (until the additional load reaches its maximum). Subsequently, as the magnitude of the additional load decreases, the stress value at stress concentration A increases. As a result, at the end of period P1, when no load is applied to the analysis model (no-load state), the stress value at stress concentration A becomes significantly larger than the initial residual stress U1. In the example in Figure 5, the stress at stress concentration A in the no-load state is tensile stress. In other words, the additional load applies a compressive stress exceeding the yield point to stress concentration A, and a portion of the compressive residual stress U1 at stress concentration A is released.

[0027] During the period P2 of repeated loading, the stress at stress concentration point A periodically changes between approximately constant maximum and minimum values ​​in accordance with the period of repeated loading. When additional load and repeated loading are applied, the maximum value during the period P2 of repeated loading is greater than the maximum value when no additional load is applied (maximum value on line L11). The same applies to the minimum value. As a result, when additional load and repeated loading are applied, the average stress during the period P2 of repeated loading (shown by dashed line H12 in Figure 5) is greater than the average stress when no additional load is applied (shown by dashed line H11 in Figure 5). In the example in Figure 5, the average stress shown by dashed line H12 is tensile stress, and the average stress shown by dashed line H11 is compressive stress. In the fatigue evaluation of the welded joint 3, repeated loading is applied to the welded joint 3 many times (for example, several million times). As will be discussed later, due to the above-mentioned differences in mean stress, the fatigue strength differs between cases where only repeated loads are applied and cases where additional loads and repeated loads are applied.

[0028] Figure 6 shows the SN curve and is presented as a log-log graph. The vertical axis of Figure 6 represents the stress range, and the horizontal axis represents the number of load applications until fracture. In Figure 6, the solid line L21 shows the SN curve when only repeated load is applied to a welded joint 3 that has been treated with impact, and the dashed line L22 shows the SN curve when additional load and repeated load are applied to a welded joint 3 that has been treated with impact. The solid line L23 shows the SN curve when only repeated load is applied to a welded joint 3 that has not been treated with impact.

[0029] As is clear from the solid lines L21 and L23, applying impact treatment to the welded joint 3 improves the stress range at each number of cycles, i.e., the fatigue strength of the welded joint 3. On the other hand, the stress range of the dashed line L22 when an additional load and a cyclic load are applied to the impacted welded joint 3 is smaller than the stress range of the solid line L21 when only a cyclic load is applied. Thus, the fatigue strength of the welded joint 3 subjected to an additional load decreases. For the SN curve of the welded joint 3 when only a cyclic load is applied, publicly available data provided by the Japan Society of Steel Construction (JSSC) and the International Welding Society (IIW), etc., is available. However, there is no available data for the SN curve of the welded joint 3 subjected to an additional load.

[0030] Next, an outline of the method for estimating the fatigue strength of the welded joint 3 subjected to an additional load will be described. Figure 7A shows the change in nominal stress of the base material 11 when only a cyclic load is applied to the analysis model. The stress ratio of the nominal stress of the base material 11 caused by the cyclic load is 0.45, and the stress range is 0.5σy (where σy is the yield stress). Figure 7B shows the change in stress at stress concentration point A when only the cyclic load is applied to the analysis model. Figure 8A shows the change in nominal stress of the base material 11 when an additional load and a cyclic load are applied to the analysis model. The nominal stress of the base material 11 caused by the additional load (the nominal stress when the absolute value is maximum) is -0.5σy. The stress ratio of the nominal stress of the base material 11 caused by the cyclic load is 0, and the stress range is 0.5σy. Figure 8B shows the change in stress at stress concentration point A when the additional load and the cyclic load are applied to the analysis model.

[0031] Figure 9 is a diagram that shows the stress changes at stress concentration point A superimposed on Figures 7B and 8B. In Figure 9, the stress change when only repeated load is applied is shown by the solid line L31, and the stress change when additional load and repeated load are applied is shown by the dashed line L32. As shown in Figure 9, the stress change at stress concentration point A during the period P2 of repeated loading is approximated by the solid line L31 and the dashed line L32. Therefore, from the viewpoint of causing such stress changes at stress concentration point A, the conditions for additional load and repeated load that produce the stress change shown by the dashed line L32 can be considered equivalent to the conditions for repeated load that produce the stress change shown by the solid line L31. In the example above, the load conditions in which the nominal stress due to the additional load is -0.5σy, and the stress ratio and stress range of the nominal stress due to the repeated load are 0 and 0.5σy, respectively, are equivalent to the load conditions in which the nominal stress due to the additional load is 0 MPa (i.e., no additional load), and the stress ratio and stress range of the nominal stress due to the repeated load are 0.45 and 0.5σy, respectively.

[0032] In the fatigue evaluation of the welded joint 3, a cyclic load is applied to the welded joint 3 multiple times. Typically, fracture of the welded joint 3 is caused by fatigue cracks that occur at the stress concentration point A. Therefore, the SN curve of the welded joint 3 under load conditions in which arbitrary additional loads and cyclic loads are applied is considered to approximate the SN curve under load conditions in which only cyclic loads are applied, as the stress change is equivalent to the stress change at the stress concentration point A during the cyclic loading period P2 under those load conditions. As mentioned above, publicly available data for the SN curve of the welded joint 3 under conditions in which only cyclic loads are applied can be used from organizations such as the Japan Society of Steel Construction (JSSC) and the International Welding Society (IIW). Below, a method for estimating the fatigue strength of the welded joint 3 subjected to an additional load will be described using the above relationship.

[0033] Figure 10 shows the configuration of the fatigue strength estimation system 5. The fatigue strength estimation system 5 comprises a fatigue strength estimation device 6 and a numerical analysis unit 7. The numerical analysis unit 7 performs the aforementioned numerical analysis (in this case, finite element analysis) on the analysis model of the welded joint 3 to obtain stress changes, etc., that occur at the stress concentration point A. The numerical analysis unit 7 is typically realized by a computer having a CPU and memory executing a predetermined program.

[0034] The fatigue strength estimation device 6 is a device for estimating the fatigue strength of a welded joint 3, and comprises an input receiving unit 61, a fatigue strength estimation unit 62, and a reference information storage unit 63. The input receiving unit 61 receives input such as stress changes of stress concentration points A obtained by the numerical analysis unit 7. The reference information storage unit 63 stores the reference information 631 described later. The fatigue strength estimation unit 62 estimates the fatigue strength of the welded joint 3 subjected to an additional load by referring to the reference information 631 using the stress changes received by the input receiving unit 61. The fatigue strength estimation device 6 in this embodiment includes a computer having a CPU and memory, and the functions of the input receiving unit 61, the fatigue strength estimation unit 62, and the reference information storage unit 63 are realized by the computer executing a predetermined program. Some or all of these components may be realized by a dedicated electrical circuit. Furthermore, the computer may be the same as or different from the computer that realizes the numerical analysis unit 7.

[0035] Next, the process for estimating the fatigue strength of the welded joint 3 subjected to an additional load will be explained with reference to Figure 11. In the fatigue strength estimation process, first, an analysis model of the welded joint 3 (see Figure 3) is created, and numerical analysis is performed by the numerical analysis unit 7 using the analysis model, and the reference information 631 shown in Table 1 is prepared (step S11). The reference information 631 is stored in the reference information storage unit 63.

[0036] [Table 1]

[0037] Here, in the numerical analysis described above, the combination of the stress change that occurs at stress concentration point A when only the repeated load is applied to the analysis model, and the fatigue strength that is expected when only the repeated load is actually applied to the welded joint 3, is called the stress-fatigue strength data. Reference information 631 shows multiple stress-fatigue strength data obtained by changing the repeated load in multiple ways.

[0038] In reference information 631, the stress change at stress concentration point A in the stress-fatigue strength data is represented by a group of index values ​​including the stress ratio and mean stress (see "Stress Ratio" and "Mean Stress" in "Group of Index Values" in Table 1). The group of index values ​​may also include other values ​​derived from the maximum and minimum stress values ​​in the stress change, such as the stress range and stress amplitude, or it may include the maximum and / or minimum values. In this case, the stress ratio and / or mean stress may be omitted. Preferably, the group of index values ​​includes two or more values ​​that represent the stress change. In Table 1, the maximum, minimum, and stress range of the stress change at stress concentration point A are shown as "σmax", "σmin", and "Δσ" in "Stress at Stress Concentration Point". The fatigue strength in the stress-fatigue strength data is the stress range at 2 million cycles in the SN curve, and is written as "2 Million Cycle Fatigue Strength" in Table 1. The stress-fatigue strength data in Table 1 is a combination of the group of index values ​​and the 2 Million Cycle Fatigue Strength.

[0039] By the way, in the numerical analysis performed by the numerical analysis unit 7 in this processing example, the load applied to the analysis model is set as the nominal stress generated in the analysis model. That is, multiple types of repeated loads in multiple stress-fatigue strength data are represented by changes in the nominal stress. In Table 1, "σmax", "σmin", "Δσ", "stress ratio", and "mean stress" in "Nominal Stress" show the maximum value, minimum value, stress range, stress ratio, and mean stress of the nominal stress generated by each repeated load. Each row in Table 1 shows the load conditions in which the nominal stress due to the added load is 0 MPa (i.e., no added load), and the stress ratio and stress range of the nominal stress due to the repeated load are the values ​​shown in Table 1. The multiple load conditions in Table 1 include conditions in which the stress ratio of the nominal stress is changed in various ways. The stress range and other parameters may also be changed in the load conditions. Details of the reference information preparation process will be described later.

[0040] Once reference information 631 is prepared, in the numerical analysis performed by the numerical analysis unit 7 using the above analysis model, after applying an additional load to the analysis model, a predetermined repeated load (hereinafter referred to as the "set repeated load") is applied to the analysis model, and the stress change that occurs at the stress concentration point A during the period of the set repeated load is obtained (step S12). The stress change at the stress concentration point A during the period of the set repeated load is represented by a group of index values ​​(hereinafter referred to as the "acquired index value group") in the same way as reference information 631.

[0041] Table 2 shows the stress ratio and mean stress of the acquired index values ​​for the duration of the set cyclic load when multiple different additional loads are applied to the analysis model. These values ​​are shown in the "Stress Ratio" and "Mean Stress" columns of the "Acquired Index Values" column under "Duration of Set Cyclic Load." The maximum, minimum, and stress range of the stress change at stress concentration point A during this period are shown in "σmax," "σmin," and "Δσ" under "Stress at Stress Concentration Point" in the "Duration of Set Cyclic Load." The period of the set cyclic load is the same as the period of the cyclic load under the multiple load conditions in Table 1.

[0042] [Table 2]

[0043] As described above, in this embodiment, the load applied to the analysis model is set in terms of nominal stress, and the additional load is also expressed as nominal stress (the nominal stress at which the absolute value is maximum). In Table 2, the nominal stress corresponding to each additional load is shown in "Nominal Stress Before Repeated Loading" and "When Additional Load is Applied". Similarly, in "Nominal Stress" for the "Period of Set Repeated Loading", the maximum value, minimum value, stress range, stress ratio, and mean stress of the nominal stress generated by the set repeated load are shown in "σmax", "σmin", "Δσ", "Stress Ratio", and "Mean Stress". In Table 2, only the additional load is changed and the set repeated load remains the same, so the maximum value, minimum value, stress range, stress ratio, and mean stress of the nominal stress generated by the set repeated load are the same for all load conditions. Furthermore, the stress at stress concentration point A before the application of the additional load (residual stress due to impact treatment), the stress at stress concentration point A when the additional load is applied, and the stress at stress concentration point A under no load after the application of the additional load are shown in "Stress at stress concentration point before repeated loading" under "After impact treatment," "During application of additional load," and "After application of additional load," respectively.

[0044] The acquired index values ​​under predetermined load conditions are input from the numerical analysis unit 7 to the input receiving unit 61 of the fatigue strength estimation device 6 and accepted. As described above, the acquired index values ​​indicate the stress changes that occur at the stress concentration point A during the period of the set repeated load. The fatigue strength estimation unit 62 estimates the fatigue strength of the welded joint 3 subjected to the additional load by referring to the reference information 631 using the acquired index values ​​(step S13).

[0045] For example, as shown in Table 2, under load conditions where the nominal stress due to the added load is -150 MPa, and the stress ratio and stress range of the nominal stress due to the set repeated load are 0 and 150 MPa, respectively, the stress ratio of the acquired index value group becomes -0.54, and the average stress becomes 66.9 MPa. The fatigue strength estimation unit 62 determines that the stress-fatigue strength data (stress-fatigue strength data in the fourth row from the bottom) in the reference information 631 of Table 1, where the stress ratio of the index value group is -0.55 and the average stress is 65.7 MPa, is the closest approximation to the acquired index value group. The fatigue strength of 90 MPa indicated by this stress-fatigue strength data is then obtained as the estimated fatigue strength of the welded joint 3 subjected to the added load. In Table 2, the estimated fatigue strength obtained by the fatigue strength estimation unit 62 is shown in "Estimated fatigue strength for 2 million cycles".

[0046] In the fatigue strength estimation process shown in Figure 11, in essence, step S12 acquires the stress change occurring at stress concentration point A during the set repeated load period, and step S13 identifies stress-fatigue strength data that best approximates this stress change. The fatigue strength indicated by this stress-fatigue strength data is then obtained as the estimated fatigue strength of the welded joint 3 subjected to the additional load.

[0047] Next, the reference information preparation process in step S11 described above will be explained with reference to Figure 12. In the reference information preparation process, first, the shape of the weld 33 in the weld joint 3 is measured by the operator. In the example of Figure 2, where the recess 311 is formed at the toe of the weld metal 31, the shape of the part including the toe is obtained. A contact-type shape measuring machine or a non-contact-type shape measuring machine may be used to measure the shape of the weld 33. After that, an analysis model of the weld joint 3 (see Figure 3) is created by the operator in accordance with the measurement results of the shape of the weld 33 (step S20).

[0048] Next, the numerical analysis unit 7 performs numerical analysis for each of the multiple load conditions using the analysis model, thereby determining the stress ratio and mean stress (i.e., the stress change at stress concentration point A) of the index value group (step S21). In each load condition in the reference information preparation process, the nominal stress due to the additional load is 0 MPa (i.e., no additional load), and the stress ratio and stress range of the nominal stress due to the repeated load are the values ​​shown in Table 1. The period of the repeated load is the same as the set repeated load.

[0049] Furthermore, the numerical analysis unit 7 obtains the stress at each location extending inward from the stress concentration point A under a single load condition (for example, the stress at which the absolute value of the nominal stress is maximum). Figure 13 shows the stress at each location extending inward from the stress concentration point A. The vertical axis of Figure 13 shows the normalized stress, which is the value obtained by dividing the stress at each location by the nominal stress (the nominal stress at which the absolute value is maximum), and the horizontal axis shows the depth from the stress concentration point A (the surface of the recess 311 in Figure 3).

[0050] Once the stress at each location is obtained, the stress gradient between the stress concentration point A and a predetermined reference position can be determined. If the depth at stress concentration point A is L1 mm (0 mm in this case) and the normalized stress is M1, and the depth at the reference position is L2 mm and the normalized stress is M2, then the stress gradient χ can be calculated using (χ = (M1 - M2) / (L2 - L1)). In this example, the depth at the reference position is 1 mm, and the magnitude of the slope of the dashed line L4 in Figure 13 represents the stress gradient χ.

[0051] Furthermore, the stress concentration factor α can be obtained by dividing the stress at stress concentration point A (which is the maximum stress at the welded joint 33) by the nominal stress. In this example, the stress concentration factor α is the normalized stress at stress concentration point A. In this way, the stress gradient χ extending inward from stress concentration point A and the stress concentration factor α at stress concentration point A are obtained (step S22). The stress gradient χ and stress concentration factor α may be determined by an operator, or they may be determined by the fatigue strength estimation device 6 or the computer implementing the numerical analysis unit 7 or by the functions of another computer. The same applies to the processing in steps S23 to S25 described later.

[0052] Next, the fatigue strength reduction coefficient β is obtained using the stress gradient χ and the stress concentration factor α. In obtaining the fatigue strength reduction coefficient β, the Siebel diagram disclosed in "Ungleichformige Spannungsverteilung bei schwingender Beanspruchung" (VDI-Z, 1955, Vol. 97, No. 5, pp. 121-126) by E. Siebel and M. Stieler (Non-Patent Literature 1 above) is used. Figure 14 shows the Siebel diagram. The vertical axis of Figure 14 shows the ratio value α / β of the stress concentration factor to the fatigue strength reduction coefficient, and the horizontal axis shows the stress gradient χ. The value of α / β is obtained by referring to the Siebel diagram using the stress gradient χ obtained in step S22, and the fatigue strength reduction coefficient β is obtained by dividing the stress concentration factor α by this value (step S23).

[0053] On the other hand, for the smooth material, which is a flat plate made of the same material as the base materials 11 and 12, the fatigue strength σ after 2 million cycles w0 This has been obtained in advance. The fatigue strength of the smooth material after 2 million cycles σ w0 Publicly available data provided by organizations such as the Japan Society of Steel Construction (JSSC) and the International Welding Society (IIW) may be used. Fatigue strength of smooth material after 2 million cycles σ w0 By dividing by the fatigue strength reduction coefficient β (i.e., σ w0 (Due to / β), the fatigue strength of welded joint 3 after 2 million cycles σ wis required (step S24). As will be described later, since the fatigue strength σ for 2 million cycles w is corrected according to the stress ratio and the like, hereinafter, it is also referred to as "reference fatigue strength σ w ".

[0054] Here, in a double logarithmic graph where the vertical axis represents the stress range and the horizontal axis represents the number of repetitions, the SN curve is known to be a substantially straight line with a negative slope. Let the slope of the SN curve be (-1 / m), the stress range be Δσ, the number of repetitions be N, and a predetermined constant be C0. The SN curve of the welded joint 3 subjected to the impact treatment is represented by Equation (1).

[0055] (Equation (1)) (Δσ) m ×N = C0

[0056] In "IIW Recommendations for the HFMI Treatment: For Improving the Fatigue Strength of Welded Joints (IIW Collection)" (Springer, 2016) by Gary B. Marquis and Zuheir Barsoum (Non-Patent Document 2 above), it is described that for welded joints subjected to peening treatment with needles or the like, the above m is set to 5. In this treatment example as well, the above m is set to 5. Also, since the stress range Δσ when N is 2 million is the above reference fatigue strength σ w , the constant C0 is obtained.

[0057] Table 3 shows the stress concentration coefficient α, stress gradient χ, α / β, fatigue strength reduction coefficient β, the fatigue strength σ for 2 million cycles of the smoothing material, w0 and the fatigue strength σ for 2 million cycles of the welded joint 3 w (reference fatigue strength σ w ) in Cases 1 to 3 where the radius of curvature of the concave portion 311 (root radius ρ) and the thickness of the base material 11 (plate thickness t) are changed. FIG. 15 shows an example of the SN curve of the welded joint 3 in Case 1.

[0058]

Table 3

[0059] Subsequently, the reference fatigue strength σ of the welded joint 3 w is corrected (step S25). Here, the International Institute of Welding (IIW) has proposed classifying fatigue strength into a plurality of FAT classes, and in Table 4 of the above-mentioned Non-Patent Document 2, it is described that the FAT class is lowered according to the stress ratio. Specifically, with the stress ratio as R, when R ≤ 0.15, it is not necessary to lower the FAT class due to the stress ratio. When 0.15 < R ≤ 0.28, the FAT class is lowered by one class. When 0.28 < R ≤ 0.4, the FAT class is lowered by two classes. When 0.4 < R ≤ 0.52, the FAT class is lowered by three classes. When R > 0.52, confirmation by testing is required.

[0060] In this processing example, after the reference fatigue strength σ w is classified into one FAT class, the FAT class of the load condition is lowered or maintained according to the range including the stress ratio of the nominal stress under each load condition. In the example of Table 1, the FAT class of the load condition where the stress ratio of the nominal stress is 0 to 0.15 remains the same as the FAT class of the reference fatigue strength σ w . The FAT class of the load condition where the stress ratio of the nominal stress is 0.2 to 0.28 is lowered by one class from the FAT class of the reference fatigue strength σ w . The FAT class of the load condition where the stress ratio of the nominal stress is 0.3 to 0.4 is lowered by two classes from the FAT class of the reference fatigue strength σ w . The FAT class of the load condition where the stress ratio of the nominal stress is 0.45 to 0.52 is lowered by three classes from the FAT class of the reference fatigue strength σ w . The FAT class of the load condition where the stress ratio of the nominal stress is 0.6 is not available.

[0061] In this way, the final FAT class is determined for each load condition, and the 2 million cycle fatigue strength of that FAT class becomes the value of "2 million cycle fatigue strength" in Table 1. Through the above process, reference information 631 showing multiple stress-fatigue strength data is created and prepared. Standard fatigue strength σ w This may be modified to take into account the effects of mean stress, etc. In this case, for example, Figure 3.11 of "Fatigue Design Guidelines for Steel Structures and Commentary - with Design Examples -" by the Japan Society of Steel Construction (Gihodo Publishing, 2012) (Non-Patent Literature 3 above) or Goodman diagrams may be used. The fatigue strength for each load condition may be obtained or corrected by actually performing fatigue tests under the load condition.

[0062] As explained above, in the fatigue strength estimation method for estimating the fatigue strength of a welded joint 3, in numerical analysis using an analysis model of a welded joint 3 in which the welded part 33 has been subjected to impact treatment, a combination of the stress change that occurs at the stress concentration point A when a repeated load is applied to the analysis model and the fatigue strength that is expected when the repeated load is actually applied to the welded joint 3 is prepared as stress-fatigue strength data, and reference information 631 is prepared showing multiple stress-fatigue strength data obtained by changing the repeated load in multiple ways (step S11). Subsequently, in numerical analysis using the analysis model, after applying an additional load different from a predetermined repeated load to the analysis model, the stress change that occurs at the stress concentration point A when the predetermined repeated load is applied to the analysis model is obtained (step S12). Then, by referring to the reference information 631 using the stress change obtained in step S12, the fatigue strength of the welded joint 3 subjected to the additional load is estimated (step S13). With the above fatigue strength estimation method, the fatigue strength of the welded joint 3 subjected to the additional load can be easily estimated.

[0063] Preferably, the stress-fatigue strength data shows a combination of the maximum and minimum stress values ​​in the stress change at stress concentration point A and the fatigue strength. This makes it easy to identify stress-fatigue strength data that includes a stress change that most closely approximates the stress change obtained in step S12, and further facilitates the estimation of the fatigue strength of the welded joint 3. On the other hand, in the fatigue strength estimation method, the stress-fatigue strength data may include a curve showing the stress change occurring at stress concentration point A (hereinafter referred to as the "stress change curve"). In this case, the fatigue strength of the welded joint 3 is estimated by identifying stress-fatigue strength data that includes a stress change curve that most closely approximates the stress change curve of stress concentration point A obtained in step S12 (for example, a stress change curve in which the sum of the absolute values ​​of the differences with the stress change curve obtained in step S12 is minimized).

[0064] Preferably, the additional load is a load that generates compressive stress in the stress concentration area A. Even if a portion of the compressive residual stress in the stress concentration area A is released by the additional load, the fatigue strength estimation method described above can appropriately estimate the fatigue strength of the welded joint 3. Of course, the additional load may also be a load that generates tensile stress in the stress concentration area A. Furthermore, the magnitude of the additional load may be greater than or less than the magnitude of the set repeated load.

[0065] Preferably, in step S11, the stress gradient from stress concentration A to the interior and the stress concentration coefficient at stress concentration A are determined by numerical analysis using an analytical model. The fatigue strength in the stress-fatigue strength data is then determined from the fatigue strength reduction coefficient obtained using the stress gradient and stress concentration coefficient, and the fatigue strength in the smooth material. This makes it possible to prepare stress-fatigue strength data easily and in a short time compared to actually performing fatigue tests.

[0066] The fatigue strength estimation device 6 for estimating the fatigue strength of a welded joint 3 comprises a reference information storage unit 63, an input receiving unit 61, and a fatigue strength estimation unit 62. The reference information storage unit 63 stores the reference information 631. The input receiving unit 61 receives input of stress changes that occur at stress concentration points A when an additional load different from a predetermined repeated load is applied to the analysis model in a numerical analysis using an analysis model, and then the predetermined repeated load is applied to the analysis model. The fatigue strength estimation unit 62 estimates the fatigue strength of the welded joint 3 subjected to the additional load by referring to the reference information 631 using the stress changes received by the input receiving unit 61. The fatigue strength estimation device 6 can easily estimate the fatigue strength of the welded joint 3 subjected to the additional load.

[0067] The fatigue strength estimation method and fatigue strength estimation device 6 described above can be modified in various ways.

[0068] In the example shown in Figure 1, the welded joint 3 is a T-joint, but the welded joint 3 whose fatigue strength is estimated by the above method may be other types of joints such as butt joints, lap joints, or cross joints.

[0069] The configurations in the above embodiments and each modified example may be combined as appropriate, as long as they do not contradict each other. [Explanation of Symbols]

[0070] 3. Welded joints 6. Fatigue intensity estimation device 33 Welded section 61 Input Reception Section 62 Fatigue Strength Estimation Unit 63 Reference information storage unit 631 Reference Information A. Stress concentration area S11~S13, S20~S25 Step

Claims

1. A fatigue strength estimation method for estimating the fatigue strength of a welded joint, a) In numerical analysis using an analytical model of a welded joint subjected to impact treatment, the steps include: preparing reference information that shows multiple stress-fatigue strength data obtained by changing the repeated load in multiple ways, where the combination of the stress change that occurs at stress concentration points when repeated loads are applied to the analytical model and the fatigue strength expected when the repeated loads are actually applied to the welded joint is used as stress-fatigue strength data; b) In numerical analysis using the analysis model, the steps include: applying an additional load different from a predetermined repeated load to the analysis model, and then obtaining the stress change that occurs at the stress concentration point when the predetermined repeated load is applied to the model; c) A step of estimating the fatigue strength of the welded joint subjected to the additional load by referring to the reference information using the stress change obtained in step b), A fatigue intensity estimation method comprising the following features.

2. A fatigue strength estimation method according to claim 1, A fatigue strength estimation method wherein the stress concentration area is the surface of the indentation caused by the impact treatment.

3. A fatigue strength estimation method according to claim 1, A fatigue strength estimation method in which the stress-fatigue strength data is a combination of the maximum and minimum stress values ​​in the stress change and the fatigue strength.

4. A fatigue strength estimation method according to claim 1, A fatigue strength estimation method in which the additional load is a load that generates compressive stress at the stress concentration point.

5. A method for estimating fatigue strength according to any one of claims 1 to 4, A fatigue strength estimation method comprising the above-mentioned step a), wherein, by numerical analysis using the analytical model, the stress gradient from the stress concentration area toward the interior and the stress concentration coefficient at the stress concentration area are determined, and the fatigue strength in the stress-fatigue strength data is determined from the fatigue strength reduction coefficient obtained using the stress gradient and the stress concentration coefficient and the fatigue strength in the smooth material.

6. A fatigue strength estimation device for estimating the fatigue strength of a welded joint, In numerical analysis using an analysis model of a welded joint subjected to impact treatment, a reference information storage unit stores reference information indicating multiple stress-fatigue strength data obtained by changing the repeated load in multiple ways, which is a combination of stress change that occurs at stress concentration points when repeated loads are applied to the analysis model and fatigue strength that is expected when the repeated loads are actually applied to the welded joint, as stress-fatigue strength data. In numerical analysis using the aforementioned analysis model, an input receiving unit receives input of stress changes that occur at the stress concentration point when an additional load different from a predetermined repeated load is applied to the analysis model, and then the predetermined repeated load is applied to the analysis model. A fatigue strength estimation unit estimates the fatigue strength of the welded joint subjected to the additional load by referring to the reference information using the stress change received by the input receiving unit, A fatigue intensity estimation device equipped with the following features.

Images