1. Introduction
Orthotropic steel bridge decks have been widely used worldwide due to their light weight, high bearing capacity, and fast construction speed. The fatigue crack problem of the bridge deck is more prominent due to factors such as its structural system, stress characteristics, processing and manufacturing, and service environment. The most vulnerable area for fatigue cracking is the welded joint, where different profiles are connected through welding. This greatly impacts the operational quality, durability, and safety of the structures [
1]. Therefore, it is very necessary to optimize the treatment of the welded joints for the fatigue cracking problem of the bridge decks. Currently, the main methods for optimizing the welded joints treatment include aging treatment, heat treatment, explosive treatment, shot peening, and ultrasonic impact treatment (UIT). Among these methods, UIT is a more ideal post-weld treatment method because it has the advantages of lightweight, flexibility, high efficiency, and low cost. It has been widely applied in the regulation of residual stresses in welded joints and in the improvement of the fatigue life of bridge decks [
2,
3]. UIT strengthens the surface of the welded joints, thereby improving its fatigue life. This type of surface enhancement technology can significantly improve the surface morphology of the welded joints, reduce the geometric discontinuity of the structure, and introduce beneficial compressive stress, thereby reducing residual tensile stress [
4,
5].
UIT is an effective method for improving the fatigue performance of welded joints. This method has gained significant research attention in recent years. Numerous researchers have analyzed the mechanical properties, microstructure, and residual stress of welded structures to investigate the optimization effects of UIT.
In terms of mechanical properties, UIT has proven to be highly effective in enhancing the fatigue life of welded joints [
6,
7,
8,
9,
10,
11,
12,
13,
14]. Studies have shown that UIT can optimize the performance of welded components composed of various materials and structures, such as high-speed steel [
15] and S700 steel [
16], resulting in a fatigue strength increase of over 50%. It is noted that the effectiveness of UIT is influenced by several factors, including duration, amplitude, and temperature. For instance, Yu et al. [
17] conducted UIT on the surface of 16MnR steel butt joint welding parts, finding that longer impact times led to greater increases in fatigue life. The growth rates at treatment durations of 30 and 60 min reached 375.22% and 521.24%, respectively. Chen et al. [
18] carried out UIT on the welded joints of 7A52 (Al-Zn-Mg-Cu) alloy and analyzed the influence of time parameters on the UIT effect.
In addition, UIT not only improves the geometric profiles of the welding seams and reduces the stress concentration factor [
19], but it also optimizes the microstructure of the welded joints, and then improves the overall mechanical properties. Researchers have determined that UIT can generate grain refinement, such as forming a modified layer on the surface of the 2219 aluminum alloy friction stir welding (FSW) joint with plastic deformation morphology and small grains [
20]. This process dissolves larger precipitates and improves the joint’s stress corrosion cracking resistance. Using UIT to treat the circumferential welding seam of austenitic stainless steel can form a grain refinement layer and a transition layer, and significantly reduce the number of precipitates in the transition layer. This improves the microhardness and maximum shear strength by approximately 18.9% and 9.6%, respectively [
21]. Wang et al. [
22] conducted a microstructural study on 7075 aluminum alloy and its friction-welded joints. After UIT, the specimens developed a gradient refinement structure known as the ultra-fine grained (UFG) layer, which is capable of enhancing the material’s hardness and strength. Currently, researchers have confirmed that UIT achieves its optimization effects through mechanisms such as grain refinement and a reduction in the number of precipitates. Further, artificial intelligence is greatly helpful in detecting changes in the strength and toughness of microstructure and chemical composition changes [
23]. These findings provide new ideas for deepening our understanding of UIT’s effects.
Finally, UIT has the ability to improve the residual stress state on the surface of welded joints. Residual stress refers to the stress within a material caused by uneven deformation and alterations to its microstructure during processes such as material processing, heat treatment, and welding. These residual stresses can greatly affect the performance and fatigue life of the material. UIT eliminate the residual stress on material surfaces by applying compressive stress and thermal stress, thereby forming a uniform, stable, and resilient surface. By utilizing high-frequency energy, UIT can alter the grain dislocations of the materials, thereby increasing their density and reducing the overall energy. This process leads to the release of residual stress, providing a more stable stress state for the material surface [
24,
25]. For example, Hu et al. [
26] successfully eliminated residual stress in the weld and heat-affected zones of 316L weld joints using UIT, enhancing the quality and strength of the welding. This also occurred when stainless steel plates were treated. Liu et al. [
27] conducted UIT on a localized area of the circumferential weld of stainless steel pipes. The influence of localized UIT on the stress distribution of the weld was studied: UIT introduces compressive stresses on the impacted surface; it also affects the stress distribution within the circumferential weld of the stainless steel pipes up to a depth of 8 mm, forming a compressive stress layer with a depth of 4 mm in the treated area. This technology is not limited to butt joints; it is also suitable for fillet welding in orthotropic steel bridge decks. Yuan et al. [
28] developed a 3D numerical simulation method to analyze the effect of UIT on the fatigue strength of the welded joints of orthotropic bridge decks.
This study was conducted using a combined approach of experiment and finite element analysis. The study focuses on analyzing the effect of UIT on the fatigue performance of T-joints, as well as studying the changes in the angle misalignment, microstructure grain, and surface micro-hardness of T-joints before and after undergoing UIT. Taking into account the notch effect, the influence of the transition angle change of the T-joint after UIT on its stress concentration is discussed using the traction structural stress method.
4. Discussion
4.1. The Effect of the Notch on the Traction Structural Stress of the Weld Toe
According to the microstructure of the test specimens in
Section 3.1, it can be clearly seen that the weld toe of the specimen forms an obvious arc-shaped notch after UIT. The conventional calculation method of the traction structural stress at the weld toe does not take into account the impact of the notch. Therefore, it is necessary to consider the notch effect on the traction structural stress at the weld toe of the T-joints, so as to meet the traction structural stress obtained when there is a notch on the actual weld toe.
Next, an analysis of the structural stress at the weld toe of the specimen with a plate thickness of
and a crack depth of
caused by the notch was conducted, as shown in
Figure 17. The overall structure is in a self-balancing state at the cross-section A-A, and equivalent stresses
and
are defined on the crack surface. Assuming that the self-balancing stress caused by the gap can be calculated based on any crack depth
l using the principle of equilibrium, the equivalent stresses represented by
and
can be redistributed and calculated using
in another cross-sectional area (at a given gap depth
), thus obtaining the equivalent structural stress at the crack notch. The calculation formula is as follows:
where:
—far field membrane stress,
—far field bending stress,
—membrane stress at a given crack depth
,
—bending stress at a given crack depth
,
—membrane stress on the crack surface,
—bending stress on the crack surface,
—structural stress on the crack surface.
The above only provides the calculation formula for
,
. However, the calculation of
,
requires introducing a bilinear distribution with characteristic depth to estimate the equivalent stress of the self-balancing part (dashed line) under the stress state.
Figure 18a, and
Figure 18b shows the stress distribution of the self-balancing part, which can be considered as two linear distributions in regions 1 and 2, respectively. By conducting equilibrium conditions and traction continuity at point 2, the following formula can be obtained [
35,
36,
37,
38]:
In regions 1 and 2, the equivalent membrane stress and bending stress are:
The dimension of the weld zone without UIT is shown in
Figure 19a; the establishment of the finite element model and the settings of the conditions are shown as
Figure 19b. Next, the traction structural stress at the interior and exterior weld toes are extracted and calculated according to the structural stress formula with the notch effect.
Figure 20 portrays the relationship between the traction structural stresses at the interior and exterior welding seams and the ratio of the notch depth (
) to the overall specimen thickness (
). It can be observed that as the ratio
increases, the structural stress at both the interior and exterior weld toes decreases, indicating the effect of the notch depth. When
is less than 15%, the structural stress at the interior weld toe exceeds that of the exterior weld toe. However, when
exceeds 15%, the structural stresses at both the interior and exterior weld toes become increasingly consistent and display a stable trend. When
= 1, which implies that the given crack depth matches the specimen’s thickness, the structural stress at the interior and exterior weld toes reached their minimum value, equal to 1.029 MPa.
After considering the influence of the notch effect depth, it can be concluded that the structural stress variation trend of the interior and exterior weld toes is essentially consistent. It can be found that the structural stress of the exterior weld toe is slightly smaller than that of the interior weld toe.
.
4.2. Traction Structural Stress Calculation after the UIT
Under UIT, the weld toe morphology of the T-joint specimens changes, forming an arc-shaped notch, and the geometric transition is smoother. We overlooked the change in thickness and focused on studying the impact of the rounding radius of the face transition into the native material (transition angle). Therefore, by changing the transition angle, the effect of UIT on the structural stress at the weld toe of the joint is studied. In this section, three FE models with different weld profiles are used to study. As shown in
Figure 21, the transition angles between the weld zone and the base metal are 100°, 140°, and 180°, respectively. The transition angle of 180° is similar to the morphology of the weld toe after UIT.
According to FE-safe 2019 software, the nodal stress at the weld toe of the three models was extracted, and then the structural stress was calculated according to the notch effect formula in
Section 3.1. The ratio of the structural stress at the weld toe and the given notch effect depth (
) to the overall thickness (
) of the specimen under different transition angles
was obtained, as shown in
Figure 22.
As the ratio increases, the structural stress at the weld toe at different transition angles decreases and finally tends to be consistent. When is less than 15%, the structural stress at the weld toe decreases with the increase in the transition angle, and the trend is obvious. In addition, when the transition angle is 180°, the structural stress of the weld toe is the minimum, indicating that the UIT can effectively reduce the traction structural stress at the welded toe; when is 15%, the structural stress is 1.8 MPa when the transition angle is 100°, 1.6 MPa when the transition angle is 140°, and 1.35 MPa when the transition angle is 180°. If is greater than 15%, with the increase in the ratio, the traction structural stress of the weld toe under different transition angles gradually stabilizes and, finally, tends to be consistent. When =1—that is, when the given crack depth reaches the specimen thickness—the structural stress of different transition fillet toes reaches the minimum value. The structural stress is 1.029 MPa when the transition angle is 100°, 1.027 MPa when the transition angle is 140°, and 1.024 MPa when the transition angle is 180°. With the increase in the rounding radius of the face transition into the native material, the fatigue performance becomes superior due to the decrease in stress concentration factor.
4.3. Fatigue Performance Analysis of T-Joints after UIT
Referring to the traction structure stress curve in
Section 4.1 and
Section 4.2, the structural stress value of the notch effect depth/thickness (
) of 15% at 180° is evaluated for a chosen equivalent structural stress coefficient of 1.35.
The fatigue test results are shown in
Table 5: Compared with the original welded T-joint specimens, the angular misalignment of the test specimens after the UIT is significantly reduced by about 50%; when the stress ratio is 0.1 and the nominal stress is 180 MPa, the fatigue life is up to 1,064,588. When the nominal stress is 220 MPa, the shortest fatigue life is 523,606. When the stress ratio is 0.3 and the nominal stress is 169.6 MPa, the fatigue life is up to 716,606. When the nominal stress is 220 MPa, the fatigue life is 262,582. However, for the specimens without ultrasonic impact treatment, when the stress ratio is 0.1 and the nominal stress is 169.6 MPa, the longest fatigue life is 360,021, and when the nominal stress is 233.2 MPa, the shortest fatigue life is 117,413.
Figure 23 depicts the equivalent structural stress S-N curve of the test specimens before and after UIT. The curve indicates that UIT significantly enhanced the fatigue performance of the specimen by approximately 350% according to fatigue results with a stress ratio of 0.1. The fatigue life of test specimens with a stress ratio of 0.1 is 150% higher than that of test specimens with a stress ratio of 0.3. Additionally, the fatigue life of the specimens with a stress ratio of 0.3 after UIT surpassed that of the original welded specimens with a stress ratio of 0.1, showing an increase of approximately 200%.
The curve indicates that UIT can effectively improve the fatigue life of T-joints. According to the fatigue results, UIT significantly improved the fatigue performance of the sample by about 350% at a stress ratio of 0.1. At a stress ratio of 0.3, UIT increased its fatigue life by 150%. However, as the stress ratio increases, the fatigue life of T-joint specimens gradually decreases.