The current research work investigates the possibility of using machine learning models to deduce the relationship between WAAM (wire arc additive manufacturing) sensor responses and defect presence in the printed part. The work specifically focuses on three materials from the nickel alloy family – Inconel 718, Invar 36 and Inconel 625, and uses three sensor responses (welding voltage, welding current and welding audio) for predictions. A variety of types of prints, including ramp tests, single bead depositions, and walls were explored. Three different machine learning models are used – artificial neural networks (ANNs), K-Means clustering and random forests (RF), and the performances are compared. In addition to separate material analysis, cross-material predictions are conducted using two supervised models to investigate the prediction capabilities of such an approach. The results indicate that models are indeed capable of finding connections between welding parameters and defect formation, and the accuracies range from 60% to 90% and the correlation coefficient is less than 0.5 (indicating weak positive correlation) depending on the model and material. The cross-material predictions are significantly worse, with accuracies ranging from 20% to 27% and very weak correlation coefficients (less than 0.1). Analysis of the results indicates that the importance of audio sensor response depends on the nature of defect, and that additional sensors like spectrometers could give a wider range of information to cover more types of defects, potentially raising the performance of cross-material predictions. Between the models, random forest is found to perform the best overall, with ANNs coming in a close second. The versatility of ANNs indicates that increasing the dataset size and resolving the class imbalance could potentially tip the scales in the favor of ANNs.

In recent years, Wire and Arc Additive Manufacturing (WAAM), is gaining attractions both from academia and industry, because it has the potential to substitute or complement the traditional manufacturing methods. Traditional materials and methods are reaching their limitation when demanding and special applications are considered. The current growth in WAAM applications will boost the WAAM consumable material development in the coming future. WAAM applications mostly refer to small batch production or prototypes, which often requires special wire compositions or can benefit from tailoring the consumable composition for the desired components. On one hand, solid wire production is only economically viable when large volumes are involved. On the other hand, metal-cored wires are particularly suitable to produce tailored or small-batch consumable compositions, which is very attractive for WAAM. However, only limited research has been carried out in the use of metal-cored wires in additive manufacturing. In this research, the focus is to explore the use of metal-cored wire in WAAM applications. Two chemical compositions, one ferrous, medium carbon low steel alloy (AM-XC-45), and one non-ferrous, Stellite 6 (cobalt-based superalloy), metal-cored wires were investigated based on industrial interests at RAMLAB. The microstructure of the WAAM AM-XC-45 thin wall was characterized using optical microscopy and scanning electron microscopy. Pearlite, ferrite, bainite, and martensite are present in the deposited wall. Columnar grains are found near the fusion line. The repeated thermal cycles cause the grains to become finer from the top to bottom layers. The mechanical properties including microhardness and tensile strength were tested and compared with the traditional processing methods like casting, milling, and forging. It showed a comparable or superior microhardness and tensile strength to the traditional process whereas the relative lower elongation of the deposited AM-XC-45 thin wall indicates that further post heat treatment is needed to improve the ductility of the part. Stellite 6 is a cobalt-based superalloy, which has good wear and corrosion resistance and retains these properties at high temperatures. In this study, Stellite 6 metal-cored wire (WEARTECH® WT-6 GMAW-C, AWS A5.21 ERCCoCr-A) is selected and optimized to obtain results comparable to the commonly employed and more costly laser deposition, in terms of microstructure and mechanical properties. The optimal deposition parameters were developed using the S355 steel substrate, and then adopted on depositing on the AISI 420 stainless steel substrate. Characterization showed that Stellite 6 layer mainly contains Co-Cr-Fe solid solution with FCC crystal structure as the matrix and the Cr7C3 and Cr3C2 were identified in the microstructure through SEM, EDS, and XRD. These carbides can contribute to the strength. In addition, the identified Co4W2C carbides can contribute to the wear properties of the Stellite 6 WAAM deposits. The dilution and hardness of the WAAM deposit can reach the same level as laser deposition. Additionally, a 3D temperature distribution model is developed by adopting the moving mesh technique to numerically study the WAAM process. The model helps to gain a better understanding of the physical phenomena (heat and mass transfer) during the WAAM deposition. The AM-XC-45 was used to validate the developed model. The simulated and experimental results were compared. The dilution and HAZ of WAAM deposited single bead was used as validation criteria. The A1 line (1100K) simulation is in a good agreement with the experimental result. The fusion line (1800K) shows some deviation. Compared with laser deposition, the main differences between the two processes lie on the heat source, the boundary conditions, and the material responses to the process.

Wire Arc Additive Manufacturing (WAAM), one of the Additive Manufacturing (AM) deposition methods which employs the basic principles of Gas Metal Arc Welding (GMAW) welding technique was used to show that it is possible to build moulds for ceramic products using AISI 420 stainless steel due to its characteristics such as corrosion resistance, machinability, hardness and dimensional stability. The WAAM process also is efficient in terms of its deposition rate, reduced material wastage and high surface quality. To investigate the feasibility of AM of AISI 420 stainless steel using WAAM, Response Surface Methodology (RSM - a predictive technique) was used to navigate within the input parameter range for process optimization. Bead-on-plate welding experiments were performed with a MIG welding robot on a structural steel (S355J2) as the substrate. In the tested range according to RSM analyses, the optimum weld condition was 261 A (Current), 29 V (Volts) and 0.59 m/min S (Travel Speed) with preheating at 200°C. However, this condition was found to be unsuitable for AM due to its low deposition speed, non - uniform building surface morphology and inter-run porosities when overlapping welds were deposited. Further analyses on the metallurgy of the WAAM AISI 420 stainless steel through Scanning Electron Microscope (SEM) revealed that the weld metal consisted of a martensite matrix and delta-ferrite at the grain boundaries. The Vickers Hardness of the weld metal was 514HV. To achieve WAAM feasibility using AISI 420 stainless steel, modified RSM method was performed by expanding the input parameters and visually inspecting the bead for its shape, size and quality. The experiment adopted was called the Ramping Procedure wherein a single resulting weld bead can represent many input parameter combinations. The optimum condition for the input parameters were identified to be 200 A (Current), 18.5 V (Volts) and 1.00 m/min S (Travel Speed). With the optimal processing conditions, rectangular blocks or walls were modelled and designed in the Autodesk Powermill software and built to evaluate the feasibility of WAAM AISI 420 stainless steel. Samples were built without preheating and no surface defects and cracking were observed. Microstructural and hardness studies were then performed. Results show that the as-welded weld metal consisted of delta-ferrite present in a martensite matrix. The hardness of the weld metal was 623 HV. The WAAM optimization procedure for AISI 420 stainless steel that has not been explored for AM processes due to its high sensitivity to welding thermal cycles that can lead to cracking. In this study, it has been successfully demonstrated that crack free AISI 420 stainless steel can be deposited with WAAM.