Optimisation of friction stir welding parameters using combined Taguchi L9 and genetic algorithm

This paper comprehensively reviews friction stir welding (FSW) as applied to copper and its alloys. FSW is a solid-state joining process that offers significant advantages over traditional fusion welding methods, particularly for materials like copper that are difficult to weld conventionally due to their high thermal conductivity and oxidation issues. Over time, the FSW process has been developed for different industries. Copper structures joined through FSW are utilized for nuclear waste storage, electrical connectors, chemical and petrochemical storage, refrigeration systems, heat exchangers, and the aerospace industry. This covers recent advancements in FSW technology, the geometry of the tools used, the process parameters, and the microstructural characteristics and mechanical properties of the joints. It examines the shapes, sizes, and materials of the tools used for welding copper and its alloys, along with process parameters such as rotational speed and traverse speed, and their influence on the quality of the joints. Additionally, the paper presents syntheses of previously published results, highlighting the values of parameters that indicate the quality of the welds, including grain size, microhardness, mechanical strength, and elongation. The challenges and potential solutions in applying FSW to copper are also discussed, providing a starting point for future research and industrial applications.

Friction stir welding process has been studied extensively in the last decades since its early stage. Most of the research done so far is related to the process development including tool design, material weldability, post-weld mechanical behavior, and microstructural properties. More recently, in-line process monitoring and artificial intelligence algorithms are introduced into this process, but mainly to specific material configuration and joint thicknesses. This study will focus on the evaluation of different machine learning approaches including principle component analysis, K-nearest neighbor, multilayer perceptron, single vector machine, and random forest methods on a friction stir welding cell environment. The input variables provided from this cell environment are namely divided into two groups: one group refers to the application variables and the other group is related to the friction stir welding process variables. The application variables target the aluminum alloys, joint configuration, sheet thicknesses, initial mechanical properties, and their chemical composition. The friction stir welding process variables dictate the rotational speed, travel speed, forging force, longitudinal and transverse forces, torque, and specific energy. The output response to model from these machine learning algorithms is the defect index, which has been quantified using high-resolution immersed bath ultrasounds. This nondestructive evaluation technique has been described previously, which can detect defects ≥150 µm in thin sheets. The defect index has been classified into five classes, which is distinguished by the nature of defect, cold weld, or hot weld, as well as the width of the internal volumetric defect upon ultrasound C-scan result. The dataset, which is composed of around 500 various process conditions, has been generated over the last few years and the variables were taken exclusively in constant weld regime and in the force control mode using the output average values. This paper compares the best resulting machine learning methods applied on a friction stir welding cell basis, which is the K-nearest neighbor and multilayer perceptron algorithms. The K-nearest neighbor model reaches a deviation of 0.55 on the defect index in comparison with the experimental values, which is slightly better than the multilayer perceptron model, which obtains a score of 0.69. Over the initial 59 available model parameters, 10 and 15 of them were retained in the final algorithm using these techniques. The main predictors include the material thickness, base material ultimate tensile stress, rotational speed, travel speed, weld forces, and specific energy. The K-nearest neighbor model was able to provide a map of defect indices with regard to rotational speed and travel speed but was only possible when a higher density of data was found within the prediction area. A data density score was also included within the model to inform the end-user about the prediction reliability. The machine learning models are mainly about differentiating various cases rather than representing the physical phenomena as determined using the finite element analysis. That being said, in order to improve the prediction reliability as well as the machine learning models, the data twinning concept, which consists of generating simulated friction stir welding process conditions by finite element analysis, is briefly discussed.

In this study, Al-5083 and St-12 alloy sheets were friction stir lap welded at different travel (7–23 cm min−1) and rotation (750–1125 rev min−1) speeds of the welding tool. The welded joints were characterised by various methods including shear tensile and Vickers microhardness tests, optical and scanning electron microscopies and X-ray diffraction analysis. The results showed that the weld zone defects decreased, and the joint strength also improved significantly with the reducing tool travel speed from 23 to 7 cm min−1. The travel speed of 11 cm min−1 was an optimum speed within the travel speeds investigated in this work. Additionally, raising the tool rotation speed enhanced the joint strength slightly. The microhardness results confirmed the formation of intermetallic phase layer with the hardness of ∼335 HV at the swirl layered structure.

Friction stir welding is an alternative to traditional welding processes, that can be considered both environmentally friendly and a tool to face weldability problems, specially in aluminium alloys. This process uses a particularly design tool that rotates and travels along the join, using equipment similar to a milling machine. Control in friction stir welding helps to assure weld soundness if done correctly. Different control approaches have been developed using force, torque and temperature sensors as input variables. Depending on the application this control strategies are successful while using dedicated FSW equipment, looking into local implementation it is more viable and cost effective to employ an adapted milling machine for the job. This type of mechanism only allows setting parameters such as travel speed, rotational speed and depth of penetration, which can be insufficient to obtain sound welds. Based on the set outcome of local implementation model-based control is an alternative to classic control and it uses a set of equations that describe the process as the input. The proposed model consists of a set of algebraic balance equations describing how mass and energy, in terms of power consumption, transform along the process predicting the soundness of the joint in every step. This model serves as the base for a control system that can predict the soundness in the next step and react accordingly, modifying basic process parameters such as plunge depth as well as travel and rotational speed minimizing mass loss. Experimental data has been compared to the model output in order to validate it.

Friction stir welding (FSW) is gaining traction as a preferred technique due to its potential to reduce heat input and enhance the mechanical properties of welded joints. However, the path to commercializing FSW for flange joints is not without challenges. Two primary obstacles are the complexity of the welding path and the intricate design requirements for the fixtures. These factors contribute to the difficulty in determining the ideal weld settings and process parameters, which are critical for achieving optimal results. The current study addresses these challenges by applying FSW to flange joints using custom-engineered fixtures. These fixtures are meticulously designed to hold the pipes and plates securely during the welding process. The focus of the research is on optimizing the multi-performance characteristics of FSW for Al 6063 flange joints through the hybrid Grey-based Taguchi method. The integrity of the weld joint is assessed by examining various mechanical properties within the weld zone, including rotation speed, travel speed, tool profile, and shoulder diameter. The study identifies the optimal parameter settings for the FSW process: a rotation speed of 3000 rpm, a travel speed of 3 mm/min2, a shoulder diameter of 20 mm, and a conical tool profile. Under these ideal conditions, the welded material exhibited a tensile strength of 170.169 MPa, a hardness of 63.7709 HV, and a corrosion rate of 0.022 mm/year. These findings underscore the effectiveness of the optimized FSW process in producing robust and durable flange joints.

Friction stir welding (FSW) is a promising technique for joining Aluminium Metal Matrix Composites (AMCs), overcoming defects associated with traditional welding methods. This study investigates the FSW of Aluminium 7075 (AA7075) composites reinforced with varying weight percentages of Carbon Nanotubes (CNTs) and Hexagonal Boron Nitride (HBN) prepared using the Rotational Ultrasonication with Mechanical Stirrer (RUMS) casting technique. FSW was conducted at 900 rpm rotational speed, 25 mm/min tangential velocity, and 6 kN axial force, resulting in distinct zones: Stir Zone (SZ), Thermo-Mechanically Affected Zone (TMAZ), Heat-Affected Zone (HAZ), and Base Material (BM). The SZ exhibited a uniform nanoparticle dispersion, grain size refinement (from 39 µm in BM to 5 µm), and enhanced mechanical properties due to the Zener-pinning effect and Orowan strengthening mechanism. Reinforced samples showed higher microhardness and improved tensile properties. However, the HAZ experienced softening and increased corrosion susceptibility due to coarsening precipitates and precipitate-free zones. Corrosion behaviour was assessed using immersion tests, salt fog spray, and potentiodynamic polarisation (PDP). These results demonstrate the effectiveness of FSW in fabricating high-performance metal matrix composites, offering potential applications in harsh environments such as sewage in-pipe robots.

Ensuring consistent surface quality in friction stir welding (FSW) remains a challenge due to the complex interaction between process parameters such as axial force, rotational speed, and travel speed that govern heat generation and material flow. This study presents a data-driven and physically interpretable machine learning framework to predict weld surface quality directly from these parameters. Thirty robot-assisted FSW trials were performed on sand cast aluminium plates, varying axial force (4–7 kN), rotational speed (700–1700 rpm), and travel speed (180–300 mm min −1 ) at a constant tool tilt of 1°. The resulting welds were classified into four surface-quality categories: good weld, flash/burr, surface roughness, and surface void. Four models logistic regression, support vector machine, random forest, and artificial neural network were trained and validated using stratified five-fold cross validation. The support vector machine achieved the highest performance, with 81.8 % accuracy and a balanced precision recall score of approximately 0.77. Decision-boundary maps and calibrated probability predictions identified an optimal defect-free process window (≈ 4–6 kN, 1200–1500 rpm, ≥ 300 mm min −1 ), consistent with known thermo-mechanical behaviour in FSW. The proposed framework integrates predictive analytics with process physics, offering a deployable foundation for real-time, data-driven quality assurance in automated FSW.

Friction stir welding (FSW) is increasingly used in joining the high strength aluminum alloys. Compared to traditional fusion welding methods, one of advantages of FSW is to obtain defect– free joints more easily. Besides the forms of FSW tools and welding parameters, certain characteristics of the base material can also affect the quality of welded joints. There is usually a protective layer on aluminum alloy surface to prevent corrosion, called Al clad. Generally the mechanical properties of Al clad are significantly lower than those of the base metal. In this paper, tensile and three–point bending tests were applied to investigate the influence of accumulated Al clad on mechanical properties of joints. The mechanism of reducing the mechanical properties of joints, caused by the accumulated Al clad was analyzed by OM, SEM and EDS. The results show that the mechanical properties of the welded joints are lower than those of the joint with which the bottom Al clad on the base material is removed before or after welding. The Al clad tiled on bottom of the base material is squeezed to both sides of the onion ring in the role of the FSW tool and gathered in the bottom after welding. Due to the poorer mechanical properties of Al clad than the base metal, the Al clad aggregation positions become the weak areas in the joints, leading to the formation of the micro cracks in the joints under tensile stress where the micro cracks continue extending to generate macro cracks under the continuous * Q Ik^6~6F? 6 50875146 $xU7rP : 2010–06–10, $x+27rP : 2010–08–17 T 4 : X 0, 9, 1985 A , 5 DOI: 10.3724/SP.J.1037.2010.00276

Numerical simulations of friction stir welding of dual phase titanium alloy for aerospace applications

This paper is to study of dissimilar welding AA6061 aluminium alloy and AZ31B magnesium alloy with ER5356 filler using friction stir welding. 2 mm thick plates of aluminium and magnesium were used. Friction stir welding operations were performed at different rotation and travel speeds and used the fixed tilt angle which is 3°. The rotation speeds varied from 800 to 1100 rpm, and the travel speed varied from 80 to 100 mm/min. In the range rotation speed of 800 to 1000 rpm and welding speed of 80 to 100 mm/min there are no defect at the weld. Tensile test show the higher tensile strength is 198 MPa and the welding efficiency is about 76%.

Friction- stir- welding (FSW) is comparatively a new welding process invented by The Welding Institute in 1991. Since no melting or fusion occurs during the welding process, FSW is free of high heat input and solidification defects. Two main objectives have been set forth in this work. The first object of this thesis is to develop thermomechanical models based on experimental knowledge and understanding the FSW process at a fundamental level. The quality of friction-stir-welding (FSW) joints depends on many critical weld process parameters. The main challenge for the FSW is in the selection of these critical process parameters that would produce a defect-free weld joint. For a particular pin tool, spindle rotational speed, welding speed, plunge rate and vertical plunge force are considered as key factors to generate heat during welding process. In this work, a temperature dependent friction coefficient is employed that takes into account both sticking and sliding friction conditions. Furthermore, both strain rate independent and strain rate dependent plasticity model was applied to develop FSW. Moreover, heat generation as a process in itself is modeled by accounting for friction heat and plastic deformation between tool/boundary conditions. To demonstrate the validity of the model, the model is applied to different weld schedules of Aluminum AA2219 alloys. Finally, the developed model is used to carry out parametric studies on the effect of process parameters such as rotational speed, welding speed, plunge rate and plunge force on heat generation during FSW. This parametric study helps to give an insight into creating defect free weld joints. Also the effect of process parameters on the quality of predictions using Coulomb and modified Coulomb models of FSW has been analyzed in this current work. A second objective of the research is to improve fatigue life of defect free AA2219 Friction welded joints by Post Weld Heat Treatment (PWHT). Later fatigue life was experimentally compared with base material, as welded specimen and post weld heat treated specimen. Post weld heat treated specimens have higher fatigue strength compare to as welded specimen.

Investigating the robust bonding between dissimilar materials such as aluminium alloys and polymers is vital in welding processes. This research delves into the structural interface and mechanical attributes achieved through aluminium alloy 6061 (AA6061) and Polypropylene (PP) friction stir welding by varying traverse and rotational speeds. A travel speed was varied with a range of 50–80 mm/min, and the tool rotational speed between 1300 and 1900 rpm was investigated to analyse their impact on mechanical properties. The lap joint configuration in which the PP is at the top and AA6061 is at the bottom was chosen from trial experiments. A scanning electron microscope observed that many locations had a tight bond in most interface areas. The average hardness values at the stir zone in the joints at different rotational speeds initially increased up to 1700 rpm due to fine fragmentation and proper distribution of these aluminium chips, and then reduced after rising to 1900 rpm. Evaluation of weld quality encompassed assessments of weld nugget hardness and lap tensile shear strength, elucidating the mechanical interlocking mechanism between AA6061 particles and PP as a critical factor contributing to joint stability. The intersection of aluminium chips and solidified polymer contributes to enhancement in the shear strength due to the phenomenon known as mechanical interlocking. The highest tensile shear strength of 5.4 MPa was observed at a rotational speed of 1700 rpm and a traverse speed of 60 mm/min, signifying optimal process conditions for enhancing mechanical integrity.

Friction stir welding (FSW) is one of the primary fabrication techniques for joining different components, and it has become popular, especially in aluminum alloy structures for marine applications. The welded joint with the friction stir process greatly depends on the process parameters, i.e., feed rate, rotational speed, and pin profile of the tool. In the current study, plates of aluminum 5451 alloy were joined by the FSW technique, and the Taguchi method was used to find the process parameters at an optimal level. The maximum value of tensile strength, i.e., 160.6907 MPa, was achieved using optimum welding conditions of a tool rotation speed of 1400, a feed rate of 18 mm/min, and the tool pin with threads. The maximum value of hardness, i.e., 81.056 HV, was achieved using optimum conditions of 1200 tool rotational speed and a feed rate of 18 mm/min with a tool pin profile having threads. In addition, the contribution in terms of the percentage of each input parameter was found by the analysis of variance (ANOVA). The ANOVA results revealed that the pin profile of the tool has the maximum contribution of 67.77% and 62.42% in achieving the optimum value of tensile strength and hardness, respectively. The study also investigated the joint efficiency of the friction stir welded joint, hardness at the weld zone, and metallography on FSW samples at the optimized level. The effectiveness and reliability of FSW joints for shipping industry applications can be observed by joint efficiency. That was investigated at optimum conditions, and it comes out to be 80.5%.

The development of laboratory work on friction stir welding (FSW) and friction stir spot welding (FSSW) FSW of dissimilar materials should provide a good insight into their possible industrial applications; and therefore, enhance their industrial development. Many applications in various industries, especially in the manufacturing sector, have led to the development of processes, such as FSW and FSSW for similar and dissimilar materials. Aluminium and copper have different properties including melting temperatures, which make the two materials difficult to join. Choosing suitable parameters, such as rotation speed, welding speed, and tool plunge depth and dwell time is important when fabricating sound FSWelds and FSSWelds. This chapter presents the current state of FSW and FSSW of aluminium and copper. An overview of the research conducted in the field of FSW and FSSW between aluminium and copper is summarized in terms of the microstructural evolution and the mechanical properties. The quality of the fabricated welds, spot welds and explanations of various properties of the welds by various researchers is presented and summarized. This could provide an insight into the current state of the two processes; and it could also lead to the optimization of the techniques by conducting more research in the field of FSW and FSSW Al/Cu. Furthermore, future scope on the usage of the two techniques is addressed.

Applying a nonlinear, pitch-catch, ultrasonic technique for the detection of kissing bonds in friction stir welds