Introduction

Aluminum alloys have been extensively utilized in shipbuilding, automotive manufacturing, and other industrial applications owing to their exceptional specific strength, low density, and superior corrosion resistance1. The primary welding technologies for aluminum alloys encompass arc welding processes (Tungsten Inert Gas welding, Gas Metal Arc Welding, etc.), friction stir welding2, laser welding3, and hybrid welding techniques4. Among these, arc welding has gained widespread adoption in aluminum alloy joining due to its high efficiency, precise controllability, and excellent mechanical properties of resultant welded joints5.

Heat input represents a critical parameter influencing weld quality in arc welding processes, particularly for thin-gauge aluminum alloy welding. The inherent high thermal conductivity of aluminum alloys necessitates precise heat input control, as excessive heat input may induce weld distortion, porosity formation, and hot cracking in the joint. Conversely, insufficient heat input can lead to stress concentration, incomplete fusion defects, and compromised joint strength. Consequently, optimization of process parameters to regulate heat input is imperative for achieving high-quality welds in aluminum alloy thin plates.

Systematic experimental parameter screening constitutes an effective approach to ensure welding quality. Chen et al.6 investigated 4047 aluminum alloy using pulsed gas arc welding, analyzing parameter effects under constant heat input conditions. Their findings demonstrated significant influences of welding current and voltage on weld quality, with notable improvements in average hardness corresponding to increased current and voltage. Wang et al.7 conducted dissimilar aluminum alloy TIG welding, observing that elevated heat input promoted dissolution of strengthening phases and consequent hardness reduction in the weld center. Hu et al.8 examined 6061 aluminum alloy TIG welding, revealing that high heat input resulted in microstructure coarsening and hardness reduction, while low heat input produced welds with inferior tensile strength due to porosity and fusion defects. Bekir et al.9 evaluated 7075 aluminum alloy GTAW joints, reporting that increased welding current led to grain coarsening in the weld center and thermal cracking at the root, ultimately reducing impact strength.

The advancement of computational methods has facilitated numerical simulation as a powerful tool for welding parameter optimization10. Xia et al.11 implemented laser brazing for dissimilar aluminum alloy joining, developing a thermal cycle model that demonstrated uniform heat distribution and minimal temperature gradients under cross-laser conditions, yielding joints with enhanced mechanical properties. Cui et al.12 employed a dual ellipsoidal heat source model to simulate 6082 aluminum alloy MIG welding, establishing that excessive heat input degraded tensile strength and toughness, with mechanical properties exhibiting a non-monotonic relationship with heat input. Yang et al.13 simulated GMAW molten pool behavior, identifying greater isotherm density and enhanced fluid velocity in the welding direction.

These studies collectively demonstrate that welding current serves as the primary parameter for heat input regulation in practical welding operations. The present study employs integrated numerical simulation and experimental approaches to systematically investigate heat input effects on weld morphology and microstructural evolution, with the ultimate objective of determining optimal heat input parameters for 5083 aluminum alloy thin plate TIG welding.

Experimental materials and methods

Experimental parameters

This experiment used TIG butt welding, with the welding equipment being the Panasonic YC-350WX5 welder. The diameter of the welding wire was 1.2 mm, and the shielding gas was argon with a purity of 99.99%. The welding gap was 0.5 mm, and the distance between the welding torch and the base material was 2 mm. Figure 1 shows the composition and working principle of the experimental platform. During the experiment, the welding parameters were adjusted to change the heat input, with the specific welding parameters shown in Table 1. The formula for calculating the heat input is:

$$E\:=\: \eta UI/V$$
(1)

Where: E represents the heat input; U represents the welding voltage; I represents the welding current; V represents the welding speed; η represents the welding thermal efficiency of 0.5. The welding speed was regulated by controlling the speed of the traveling mechanism during the experiment.

Fig. 1
Fig. 1
Full size image

Schematic diagram of the experimental platform.

Table 1 Welding process parameters.
Full size table

Experimental materials

The base material used in this experiment was 5083 aluminum alloy, with a melting point of approximately 610 °C. The dimensions of the base material were chosen to be 100 mm × 100 mm × 3 mm. The type of welding wire used was ER5356. The chemical compositions of the 5083 aluminum alloy base material and the ER5356 welding wire are shown in Table 2. To achieve numerical simulation of the welding process, it is necessary to define the physical properties of the material, including thermal conductivity, viscosity, etc. The thermal physical properties of 5083 aluminum alloy at different temperatures were calculated using Jmat Pro.

Table 2 Main chemical composition of base metal and welding wire (wt%).
Full size table

Experimental procedures and test methods

After the welding process was completed, a wire-cutting machine was used to cut metallographic samples at the stable position of the weld. Tension specimens and metallographic specimens were cut perpendicular to the weld on each test plate, as shown in Fig. 2. The metallographic specimens were sequentially ground with sandpapers of grit numbers 400#, 600#, 800#, 1000#, 1200#, 1500#, and 2000#, and then polished. Keller’s reagent was applied to etch the polished surface. Optical microscopy was used to examine the weld cross-section and measure the weld reinforcement height, weld width, and weld angle. High-speed industrial cameras captured the arc morphology during the initial stage and the welding process under different parameters. The PHENOM XL scanning electron microscope (SEM) and energy dispersive spectrometer (EDS) were used to analyze the microstructure, elemental content, and distribution near the weld center and fusion line. Tensile tests were conducted using the WDW-500 M tensile testing machine, and the tensile fracture morphology was observed under the Zeiss EV015 electron scanning microscope. Hardness measurements were performed along the weld cross-section using the digital microhardness tester HVS-1000, with a test point taken every 0.5 mm, applying a load of 0.98 N for 15 s.

Fig. 2
Fig. 2
Full size image

Sample cutting position and size.

Numerical simulations

The welding thermal process was numerically simulated using the computational fluid dynamics (CFD) software FLUENT, as illustrated in Fig. 3. To address the inherent non-uniformity of temperature and fluid flow fields during welding while maintaining computational efficiency, an adaptive meshing strategy was implemented. The model features refined mesh elements (minimum size = 0.1 mm) in critical regions including the fusion zone and heat-affected zone (HAZ), transitioning to progressively coarser meshes in peripheral areas. This meshing methodology effectively minimizes numerical artifacts while optimizing computational resource utilization. The simulation domain was simplified by exploiting geometric symmetry about the weld centerline, with half of the base material geometry being modeled to reduce computational overhead14.

Fig. 3
Fig. 3
Full size image

Numerical simulation model.

Boundary conditions

The boundary conditions are mainly composed of heat boundary conditions and momentum boundary conditions, in which the heat boundary conditions mainly consider the heat flow density problem on the surface of the weldment.

The expression of the double ellipsoid heat source model is:

$$\:\:q\left(1\right)=\frac{6\sqrt{3}{f}_{1}Q}{{\pi\:}^{\frac{3}{2}}{a}_{1}bc}\text{e}\text{x}\text{p}\{-3[{\left(\frac{x}{{a}_{1}}\right)}^{2}+{\left(\frac{y}{b}\right)}^{2}+{\left(\frac{z}{c}\right)}^{2}\left]\right\}$$
(2)
$$\:q\left(2\right)=\frac{6\sqrt{3}{f}_{2}Q}{{\pi\:}^{\frac{3}{2}}{a}_{2}bc}\text{e}\text{x}\text{p}\{-3[{\left(\frac{x}{{a}_{2}}\right)}^{2}+{\left(\frac{y}{b}\right)}^{2}+{\left(\frac{z}{c}\right)}^{2}\left]\right\}$$
(3)

where \(\:Q=UI\eta\:\), a, b, c are ellipsoid shape parameters; \(\:{f}_{1}\), \(\:{f}_{2}\) are before and after ellipsoid heat distribution functions.

The welded surface is dominated by convective heat transfer losses \(\:{q}_{c}\) and thermal radiation losses \(\:{q}_{r}\), with the expressions, respectively:

$$\:{q}_{c}={h}_{c}(T-{T}_{a})$$
(4)
$$\:{q}_{r}={\sigma\:}_{SB}\epsilon\:({T}^{4}-{T}_{a}^{4})$$
(5)

The heat flow density at the surface of the weldment is expressed as:

$$\:q=-{q}_{c}-{q}_{r}$$
(6)

\(\:{h}_{c}\) is the surface heat transfer coefficient, \(\:{T}_{a}\) is the ambient temperature, \(\:{\sigma\:}_{SB}\) is the Boltzmann constant, and \(\:\epsilon\:\) is the emissivity.

The momentum boundary condition is mainly driven by the surface tension of the melt pool on the fluid, and the surface tension gradient should be balanced with the shear stress of the fluid14–15:

$$\:-\mu\:\frac{\partial\:u}{\partial\:z}=\frac{\partial\:\gamma\:}{\partial\:T}\times\:\frac{\partial\:T}{\partial\:x}$$
(7)
$$\:-\mu\:\frac{\partial\:u}{\partial\:z}=\frac{\partial\:\gamma\:}{\partial\:T}\times\:\frac{\partial\:T}{\partial\:y}$$
(8)

Melt pool driving force

In the welding process, the driving force of the melt pool is mainly electromagnetic force, surface tension and buoyancy, the change of the temperature of the melt pool changes the fluid density, resulting in buoyancy at the surface, the expression is:

$$\:{F}_{b}=\rho\:\text{g}\beta\:(T-{T}_{m})$$
(9)

\(\:\rho\:\) is the density of the liquid metal, \(\:\beta\:\) is the thermal expansion coefficient of the material, and \(\:{T}_{m}\)is the melting point of the metal.

The above boundary conditions and heat source equations are implemented through FLUENT’s custom program User-defined.

Results and analysis

Effect of heat input on weld morphology

Figure 4 presents the surface morphology and cross-sectional macrostructure of TIG welds fabricated under varying process parameters. With increasing heat input from 276 J/mm to 347 J/mm, the weld surfaces exhibit well-defined ripple patterns and maintain good formation integrity without observable welding defects.

The cross-sectional analysis reveals that at a heat input of 276 J/mm, the weld exhibits porosity defects along the fusion boundary and lack-of-fusion defects at the weld root. These imperfections are attributed to insufficient heat input, which results in a reduced molten pool volume and accelerated solidification. The shortened solidification time restricts gas escape, promoting pore formation upon solidification. Additionally, incomplete penetration into the aluminum alloy base metal, combined with inadequate backside shielding gas coverage, permits atmospheric contamination of the molten pool, further exacerbating internal porosity formation.Under low heat input conditions, the occurrence of lack-of-fusion defects can be attributed to three interrelated factors: (1) insufficient thermal energy input to maintain adequate molten pool temperature, (2) diminished fluidity of the molten metal, and (3) non-uniform distribution of arc energy density. These conditions collectively result in suboptimal thermophysical characteristics of the weld pool, including depressed peak temperatures and restricted lateral spreading capability. Consequently, the molten pool fails to achieve complete melting of the base material or sufficient penetration to adequately fill the joint interface, ultimately manifesting as lack-of-fusion discontinuities at the weld root region.

With the elevation of heat input, a notable reduction in pore formation is observed, along with the complete elimination of lack-of-fusion defects and significant improvement in joint fusion quality. This phenomenon can be attributed to the enhanced thermal dynamics within the molten pool, where increased heat input elevates both the temperature and thermal conduction efficiency. The augmented thermal energy facilitates deeper penetration into the base material, resulting in a substantial increase in weld penetration depth. Concurrently, the expanded heat input promotes greater lateral diffusion of the molten pool, which manifests as an increased weld width in the cross-sectional profile. These combined effects demonstrate the critical role of thermal input in optimizing weld morphology and defect suppression.

When the heat input reaches 347 J/mm, the number of pores in the weld increases, and they are mostly located near the fusion line. This is due to the high thermal conductivity of aluminum, which results in rapid cooling and creates a temperature gradient near the fusion line. The high thermal conductivity of aluminum causes the edges of the molten pool to cool quickly, leading to uneven solidification and restricted flow. As a result, the fluidity of the molten metal at the edges of the pool is poor, and rapid solidification traps gas, preventing it from escaping. Additionally, under high heat input, magnesium evaporates due to the excessive temperature of the molten pool, forming bubbles. These bubbles migrate with the flow of the molten pool and accumulate at the edges or solidification front, ultimately being trapped as the pool solidifies rapidly. This leads to porosity predominantly distributed near the weld center or fusion line. It can be seen that the size of the heat input affects the porosity rate of the weld. Combining the cross-sectional morphology of the weld, weld A with low heat input and weld D with high heat input have a larger number of porosity defects. As the heat input increases, the porosity defects are mostly concentrated at the edge of the weld. Figure 5 shows the arc morphology at the initial stage and during the welding process under different parameters captured by an industrial camera. As the current increases, the arc size gradually increases, leading to an increase in the heat input to the base material, which in turn increases the molten pool volume and reduces the volume of the wire on the surface of the base material in a molten state. This change enlarges the size of the molten pool, enhances its fluidity, and causes its shape to become wider and shallower, while also prolonging the solidification time. As a result, it improves weld formation and reduces defects such as lack of fusion.

Fig. 4
Fig. 4
Full size image

Macroscopic morphology of welded seam.

Fig. 5
Fig. 5
Full size image

Arc morphology under different parameters.

The ratio of the back width to the front width of the weld, denoted as RW. Figure 6 illustrates the effect of heat input on the weld width and the width ratio. When the heat input is below a certain value, the welded joint exhibits incomplete penetration or semi-penetration, resulting in a discontinuous back weld width with numerous dispersed porosity defects that affect the welding quality. Increasing the heat input raises the temperature and fluidity of the molten pool, resulting in an increase in weld width.As the heat input increases, both the front and back weld widths show a growth trend, and the difference between them gradually decreases. Compared to the front weld width, the back weld width grows at a faster rate. This is because an increase in welding heat input enhances the heat transfer in the molten pool, accelerating the growth of the back weld width.

Fig. 6
Fig. 6
Full size image

Effect of heat input on weld width.

Figure 7 illustrates the correlation between heat input and both weld reinforcement height and contact angle. The data demonstrate a general positive correlation between heat input and these dimensional parameters, with observed inter-dataset variations diminishing progressively. At a heat input of 276 J/mm, the limited arc energy results in inadequate molten wire spreading, producing a constricted and shallow weld pool with reduced volume. Under these conditions, the disproportionate wire melting rate leads to material accumulation at the weld surface, consequently increasing both reinforcement height and contact angle. With escalating heat input, the weld exhibits enhanced width and depth dimensions, accompanied by greater overall weld volume. This progression correlates with a systematic reduction in both reinforcement height and contact angle. The underlying mechanisms can be attributed to two primary factors: Firstly, elevated heat input promotes molten pool expansion in both lateral and vertical dimensions. The consequent increase in molten wire distribution within the enlarged pool volume results in diminished surface accumulation, thereby reducing reinforcement height. Secondly, under constant wire feed rate conditions, the wire melting rate reaches a thermodynamic equilibrium that limits the maximum achievable filling capacity within the weld zone. This constraint explains the pronounced disparities in reinforcement height observed across different heat input conditions.

Fig. 7
Fig. 7
Full size image

Effect of heat input on weld margin height and contact angle.

Analysis of simulation results

Experimental validation serves as the most direct and reliable approach to evaluate the accuracy of numerical simulations. By examining the morphology and dimensions of the molten pool via metallographic microscopy and comparing them with simulation predictions, this study primarily assesses the consistency in molten pool width, depth, and geometry to verify simulation fidelity. Figure 8 presents a comparative analysis of the simulated and experimentally obtained temperature distributions across the weld cross-section under varying parameters. The results demonstrate strong agreement between simulation and experimental data. As heat input increases, a significant enhancement in weld penetration depth is observed. When other parameters remain constant, higher welding current elevates arc power density, leading to greater overall heat input and an expanded weld volume along the vertical axis. Consequently, the weld transitions from partial to full penetration. Furthermore, Fig. 9 illustrates the thermal histories at the weld centerline under different parameters at a fixed time interval. The data reveal a progressive rise in peak temperature at the weld center with increasing heat input. Elevated peak temperatures promote deeper weld penetration and contribute to defect suppression by improving molten pool stability and solidification behavior.

Fig. 8
Fig. 8
Full size image

Comparison between simulation and experiment with different parameters: (a) 276 J/mm (b) 296 J/mm (c) 321 J/mm (d) 347 J/mm.

Fig. 9
Fig. 9
Full size image

Temperature profile of the weld with different parameters.

Microstructure of welded joints

The thermal cycling inherent to the welding process generates three distinct microstructural zones in the joint: base metal (BM), heat-affected zone (HAZ), and weld zone (WZ). Figure 10 presents the microstructural transition from BM to WZ under a heat input of 296 J/mm. The HAZ predominantly consists of coarse lamellar structures interspersed with black granular phases.A notable microstructural transition occurs with decreasing distance from the weld center. Thermal gradient effects induce the formation of a narrow fine-grained zone (~ 100 μm wide) adjacent to the fusion boundary. Beyond this transition region, columnar grains nucleate at the fusion line and exhibit preferential growth oriented nearly perpendicular to the interface (Fig. 10a). This crystallographic orientation results from competitive growth along the maximum temperature gradient direction. Progressing toward the weld center (Fig. 10b), the microstructure transitions to a mixed morphology of equiaxed and columnar grains. This evolution reflects decreasing cooling rates and thermal gradients during solidification. The high thermal gradient near the fusion boundary favors directional growth of columnar grains, while the reduced gradient in the weld center promotes equiaxed grain formation through enhanced nucleation and isotropic growth. Consequently, the solidification sequence demonstrates a progressive transformation from lamellar → columnar → equiaxed grain structures. The weld center exhibits equiaxed grains with relatively uniform size distribution and random crystallographic orientation. This isotropic morphology arises from minimal thermal gradient variation in the central region, permitting unconstrained grain growth16. The observed microstructural gradation directly correlates with the spatial variation in thermal conditions during solidification.

Fig. 10
Fig. 10
Full size image

Microstructure and morphology of welded joints.

Figure 11 presents the microstructural characteristics of distinct regions within the welded joint under varying heat input conditions. At a heat input of 347 J/mm, the elevated peak temperature of the molten pool induces a substantial expansion of the heat-affected zone (HAZ). In proximity to the fusion line, the microstructure exhibits coarse columnar grains (Fig. 11d), which can be attributed to the prolonged high-temperature exposure of the weld metal under higher heat input conditions. This thermal regime facilitates enhanced energy transfer from the weld to the base material, thereby promoting grain boundary migration through increased thermodynamic driving forces17. Within the weld center, both equiaxed and columnar grains undergo significant coarsening, with a marked predominance of equiaxed grains. This microstructural evolution stems from the reduced cooling rate and extended solidification duration associated with elevated heat input, which provides favorable conditions for equiaxed grain growth and consequent dendrite arm spacing enlargement. As the heat input diminishes, a progressive refinement of the microstructure becomes evident in both the weld center and fusion line regions, accompanied by a concomitant reduction in HAZ width. Notably, the HAZ exhibits distinct microstructural features compared to the fusion line vicinity, characterized by coarser grains and expanded intergranular regions. The reduced grain boundary density in the HAZ renders this region particularly susceptible to mechanical degradation, potentially establishing it as the critical weak point in the welded joint3.

Fig. 11
Fig. 11
Full size image

Microstructure of welded joints at different heat inputs(A) 276 J/mm (B) 296 J/mm (C) 321 J/mm (D) 347 J/mm.

Distribution of elements in different regions of the joint

Figure 12 shows the scanning electron microscopy (SEM) microstructure at the fusion line of the 5083 aluminum alloy welded joint with a heat input of 347 J/mm, as well as the distribution of the main alloy elements. From the SEM image, it can be observed that the heat-affected zone mainly consists of white flake-like and needle-like microstructures, with larger dimensions and a tendency to grow towards the center of the weld. This is consistent with the phenomenon observed in the metallographic images mentioned earlier. The elemental mapping results indicate that Fe and Mn elements are more concentrated in the heat-affected zone. Quantitative point analysis of different microstructures, as shown in Table 3, suggests that region A corresponds to the Al6(Fe, Mn) phase, and region B corresponds to the Al3(Fe, Mn) phase18. In the heat-affected zone, the Al6(Fe, Mn) phase is more abundant, with a flake-like morphology and larger size, while the Al3(Fe, Mn) phase is less abundant, with a needle-like morphology and smaller size. Region C exhibits a black granular microstructure. The elemental analysis in Table 3 indicates that this region is a concentration of Mg and Si elements. According to reference19, region C corresponds to the Mg2Si phase. The coarse Al6(Fe, Mn) phase in the heat-affected zone can easily cause stress concentration during mechanical testing, weakening the mechanical properties of the welded joint.

Fig. 12
Fig. 12
Full size image

Distribution of alloying elements near the fusion line.

Table 3 Elemental point analysis results (atomic fraction, %).
Full size table

Figure 13 presents the scanning electron microscopic (SEM) characterization of the microstructural morphology and elemental distribution in the weld center region. In contrast to the heat-affected zone (HAZ), the weld center exhibits an isotropic microstructure predominantly composed of randomly oriented needle-like and granular phases. Energy-dispersive X-ray spectroscopy (EDS) mapping analysis indicates a significant depletion of Fe and Mn elements in the weld region compared to the HAZ, leading to the formation of finer Al-Fe-Mn intermetallic phases with predominantly acicular morphology. The observed reduction in Mg content within the weld zone provides direct evidence of “Mg” vaporization during the welding process. Concurrently, the increased Si concentration correlates with the homogeneous dispersion of black granular Mg2Si precipitates. The formation of these finely distributed strengthening phases contributes to the enhancement of mechanical properties in the welded joint.

Fig. 13
Fig. 13
Full size image

Distribution of alloying elements in the center of the weld.

Mechanical properties of welded joints

Effect of heat input on tensile properties of joints

Tensile tests were conducted on welded joints prepared under varying heat inputs (after removing the weld reinforcement), with the results presented in Fig. 14. Both the tensile strength and fracture elongation of the joints initially increased and subsequently decreased with rising heat input. The optimal mechanical properties were achieved at heat inputs of 296 J/mm and 321 J/mm, exhibiting comparable tensile strength and ductility. However, when the heat input was further increased to 347 J/mm, a noticeable deterioration in both tensile strength and elongation was observed. The macroscopic fracture morphology of the welded joints under different heat inputs is displayed in the inset of Fig. 14. At a heat input of 276 J/mm, the fracture surface exhibited incomplete fusion zones and relatively large porosity. With increasing heat input, the weld microstructure became more homogeneous, and no significant cracks or voids were detected. However, at the highest heat input (347 J/mm), porosity defects reappeared on the fracture surface. Correlating these findings with prior metallographic and cross-sectional analyses, it is evident that insufficient heat input leads to incomplete fusion defects. Although low heat input promotes finer grain structures—enhancing tensile strength in accordance with the Hall-Petch relationship—the presence of fusion defects and porosity significantly compromises joint integrity. Conversely, increasing the heat input elevates the molten pool’s peak temperature and extends solidification time, facilitating bubble escape and reducing defect formation, thereby improving joint strength. Thus, variations in heat input critically influence defect formation, ultimately determining the tensile performance of the welded joints.

Figure 15 presents the microfracture morphology of welded joints under varying heat input conditions. At a heat input of 276 J/mm, the fracture surface exhibits numerous quasi-cleavage and cleavage facets, accompanied by limited dimple formation and tear ridges in the surrounding regions, indicative of a mixed ductile-brittle fracture mechanism. As heat input increases, a notable microstructural evolution is observed: the areal fraction of cleavage planes progressively diminishes while dimple density increases significantly. This transition is characterized by the formation of extensive tear ridges, ultimately resulting in a predominantly ductile fracture mode. Correspondingly, the welded joints demonstrate enhanced toughness properties, as evidenced by their elevated elongation at break values. However, at the maximum heat input of 347 J/mm, the fracture morphology reveals a distinct deterioration in joint integrity. The fracture surface is populated with randomly distributed pores of varying dimensions, accompanied by a microstructure comprising multiple quasi-cleavage planes interconnected by tear ridges. This microstructural configuration suggests a reversion to a mixed ductile-brittle fracture mode, with quasi-cleavage becoming the dominant failure mechanism. The observed fracture mode transitions demonstrate a clear correlation between heat input parameters and joint mechanical performance, with intermediate heat inputs yielding optimal toughness characteristics through the promotion of ductile fracture mechanisms.

Fig. 14
Fig. 14
Full size image

Tensile strength of welded joints at different heat inputs.

Fig. 15
Fig. 15
Full size image

Microscopic fracture morphology of welded joints with different heat inputs(A) 276 J/mm (B) 296 J/mm (C) 321 J/mm (D) 347 J/mm.

Effect of heat input on the microhardness of joints

Microhardness measurements were systematically performed along the horizontal axis of the weld cross-section, with Fig. 16(a) presenting the comparative hardness profiles of joints fabricated under different welding parameters. Significant variations in microhardness were observed across the four parameter sets, particularly within the weld zone. Notably, Sample D exhibited both a reduced average hardness and greater hardness fluctuation range compared to other specimens. Microscopic examination of the indentation morphology revealed that regions displaying anomalously low hardness values corresponded to enlarged indentation areas. This phenomenon can be attributed to the presence of porosity defects within the weld structure, which compromise local mechanical integrity and facilitate greater indenter penetration. The inverse correlation between indentation size and measured hardness confirms the detrimental effect of such imperfections on joint performance.

Figure 16(b) illustrates that the hardness profile of the joint exhibits an approximately symmetrical distribution about the weld centerline. At a heat input of 276 J/mm, the average microhardness in the weld zone measures 70.7 HV0.1. With increasing heat input, a marginal enhancement in weld hardness is observed. For heat inputs of 296 and 321 J/mm, the weld hardness stabilizes at a consistent level. However, at the maximum heat input of 347 J/mm, the average microhardness decreases slightly to 69.94 HV0.1. The microhardness distribution across the welded joint displays significant regional variations, exhibiting a characteristic trend of initial reduction followed by subsequent elevation when traversing from the base material to the weld center. This behavior correlates directly with the microstructural evolution observed in Figs. 10 and 11. The weld center and fusion line adjacent regions demonstrate refined crystalline structures, while the heat-affected zone (HAZ) is dominated by coarse, lamellar microstructures. Consistent with the Hall-Petch relationship, which establishes an inverse proportionality between grain size and material hardness, the HAZ exhibits diminished hardness values due to its coarser microstructure20. Furthermore, comparative analysis reveals that the elevated heat input in Sample D induces substantial grain coarsening in the weld region relative to Sample A, resulting in an overall reduction in microhardness. This microstructural coarsening effect accounts for the observed hardness variations across different welding parameters.

Fig. 16
Fig. 16
Full size image

Microhardness of welded joints at different heat inputs.

Conclusions

(1) Numerical simulation of the welding process was conducted using Fluent computational fluid dynamics software, which enabled precise characterization of the thermal distribution within the molten pool during the welding process. This simulation established the optimal parameter window for thin-plate welding applications. Through maintaining a constant wire feed rate while systematically varying the current parameters, distinct heat input conditions were experimentally realized.

(2) Metallurgical analysis revealed that insufficient heat input conditions predominantly resulted in lack-of-fusion defects at the joint interface. Progressive increase in heat input promoted weld pool expansion and complete elimination of fusion-related defects. It was observed that under constant wire feed conditions, the weld reinforcement height exhibited significant variation with increasing heat input, accompanied by progressive improvement in fusion zone characteristics. However, excessive heat input conditions were found to promote porosity agglomeration and consequent degradation of mechanical properties in the weld region.

(3) Mechanical testing demonstrated a non-monotonic relationship between heat input and joint performance, with tensile strength initially increasing before reaching a maximum and subsequently decreasing. Notably, at a critical heat input of 347 J/mm, the weld surface hardness exhibited marked deterioration. Microstructural analysis correlated this behavior with grain coarsening phenomena and the formation of porosity clusters under high heat input conditions.

(4) The combined numerical and experimental investigation established that optimal welded joint performance for 5083 aluminum alloy thin-plate welding was achieved within a precisely controlled heat input range of 296–321 J/mm. Excellent correlation was observed between simulation predictions and experimental measurements, validating the computational model’s predictive capability for process parameter optimization.