A review on wire arc additive manufacturing: Monitoring, control and a

framework of automated system

Chunyang Xiaa,b , Zengxi Pana, *, Joseph Poldena , Huijun Lia ,

Yanling Xub , Shanben Chenb , Yuming Zhangc
ABSTRACT
INTRODUCTION
• Wire arc additive manufacturing (WAAM) is an emerging technology in advanced fabrication.
• In contrast to other additive manufacturing (AM) technologies, WAAM makes use of an electric arc as heat-source to deposit metal material layer-by-
layer, (as illustrated in Fig. 1) which makes up the final part.
• WAAM commonly has three types: Gas Tungsten Arc Welding (GTAW)-based, Gas Metal Arc Welding (GMAW)-based and Plasma Arc Welding
(PAW)-based.
• This method of depositing an entire component using weld metal has been in practice since 1925
• When compared to laser powder AM methods, WAAM also features better material utilization ratio and more environmental friendly production process
• As WAAM technology has a wide range of potential applications across a number of manufacturing sectors, research and development of its general
principles has been increasing at rapid rate[
• . This can be largely attributed to the lack of process monitoring and closed loop control of the overall WAAM process
• process monitoring and feedback control for AM process were also identified as key advancements critical to the overall methods’ success. A
• hose issues also exist in WAAM and the main challenges for application of WAAM can be summarized as manufacturing accuracy, quality assurance and
automation level. In traditional welding and AM area, different monitoring method has been tried to improve those issues
. The sensing methods mainly include
WAAM MONITORING
CONT …
•the deposition process of Wire Arc Additive Manufacturing (WAAM) technologies. Several factors contribute
to inaccuracies, including limitations in wire diameter, fluctuations in arc ignition and extinguishment, selection
of improper welding parameters, distortion from heat accumulation and cooling, and the cumulative error
known as the 'staircase effect' as layers are deposited.
•However, existing approaches often operate offline, relying on static training data. This becomes problematic
as the WAAM process is dynamic, with factors such as heat accumulation varying over time. As more layers are
deposited, the prediction model can deviate from the actual process due to the 'staircase effect.'
DEFECT IDENTIFY
WAAM DEFECTS
Vision sensing
• Visual sensing plays a crucial role in various additive manufacturing (AM) processes, including Wire Arc Additive Manufacturing (WAAM).
By employing vision sensors, researchers can analyze melt pool dynamics, layer/surface morphology, and other process signatures like
spatter. This real-time information can be utilized for feedback control, optimization of process parameters, and ensuring the final quality of
AM components.
• In laser AM, visual sensing has been extensively utilized. For instance, Davis and Shin used CCD cameras and line lasers to monitor bead
profiles during laser cladding processes, enabling optimization of process parameters for subsequent layers. However, the processing time
involved in this method may cause a lag effect, limiting its suitability for feedback control. Clijsters et al. developed an optical system using
high-speed near-infrared (NIR) thermal CMOS cameras and photodiodes to monitor the melt pool in real-time, providing insights into the
final product's quality.
• Moreover, visual sensing can aid in monitoring spatter levels during laser AM processes, reflecting overall process stability. Repossini et al.
utilized high-speed cameras to capture images of spatter in laser powder bed fusion (LPBF) processes, enabling diagnosis of various melting
conditions. Machine learning techniques have been applied to detect unstable melting conditions based on infrared images of the process
plume.
• Furthermore, visual monitoring of the deposited surface can help identify discontinuities or defects in the final component. Researchers have
employed high-resolution digital cameras to collect surface images during AM processes, extracting multidimensional visual features for
defect detection through binary classification techniques.
• In WAAM applications, visual sensing can provide valuable insights into process dynamics, morphology, and defects, enabling real-time
monitoring and control to improve the overall quality of the deposited components.
Vision sensing
Spectral sensor
Spectral sensors, also known as spectrometers or spectroradiometers, are devices used to measure the intensity of light at different wavelengths within the
electromagnetic spectrum. These sensors can be utilized across various fields, including environmental monitoring, agriculture, astronomy, and industrial
applications.
In additive manufacturing (AM), spectral sensors can offer several potential applications:
1. Material Analysis: Spectral sensors can be used to analyze the composition of materials used in additive manufacturing processes. By measuring the
spectral signature of materials, manufacturers can ensure the quality and consistency of feedstock materials, which is crucial for achieving desired
mechanical properties in the final printed parts.
2. Process Monitoring: Spectral sensors can monitor the emission spectra of the arc or laser during the additive manufacturing process. Changes in the
emission spectra can indicate variations in process parameters such as temperature, composition, or gas atmosphere. Real-time monitoring using spectral
sensors can enable prompt adjustments to maintain process stability and quality.
3. Quality Control: Spectral sensors can aid in quality control by detecting defects or inconsistencies in printed parts. For example, variations in spectral
signatures across a printed layer may indicate insufficient melting or bonding, leading to porosity or weak spots in the final part. By detecting these issues
early, manufacturers can take corrective actions to improve part quality.
4. Surface Characterization: Spectral sensors can also be used to characterize the surface properties of printed parts. By analyzing the reflectance or
absorbance spectra of the surface, manufacturers can assess surface roughness, texture, or chemical composition, which is important for applications
requiring specific surface properties or finishing processes.
Overall, spectral sensors offer a versatile and powerful tool for monitoring and controlling additive manufacturing processes, enhancing process efficiency,
quality, and reliability. Integrating spectral sensing technology into AM systems can contribute to the advancement and widespread adoption of additive
manufacturing across various industries.
Spectral sensor
Thermal sensing
. Thermal sensing involves the use of sensors to measure temperature variations in a given environment or system. In additive manufacturing (AM),
thermal sensing plays a crucial role in monitoring and controlling the printing process. Here are several ways in which thermal sensing is utilized in AM:
1. Process Monitoring: Thermal sensors can be integrated into additive manufacturing systems to monitor the temperature of various components,
including the build platform, print head, and surrounding environment. By continuously monitoring temperature variations during the printing process,
manufacturers can detect anomalies or deviations from the desired parameters, allowing for timely adjustments to maintain process stability and quality.
2. Quality Assurance: Thermal sensing can help ensure the quality of printed parts by monitoring temperature gradients and thermal profiles during the
printing process. Discrepancies in temperature distribution across the print bed or within the printed layers can indicate potential defects such as
warping, delamination, or inadequate material fusion. By detecting these issues early, manufacturers can take corrective actions to improve part quality
and integrity.
3. Material Characterization: Thermal sensing can also be used for material characterization in additive manufacturing. By measuring the thermal
properties of feedstock materials, such as their melting temperature, thermal conductivity, or heat capacity, manufacturers can optimize printing
parameters and select suitable materials for specific applications. Additionally, thermal sensing can help identify material inconsistencies or impurities
that may affect print quality.
4. Process Optimization: Thermal sensing data can be used to optimize printing parameters and improve overall process efficiency. By analyzing
temperature trends and thermal responses during the printing process, manufacturers can identify optimal printing conditions, such as nozzle
temperature, print speed, or layer thickness, to achieve desired part quality while minimizing energy consumption and production time.
5. Safety Monitoring: Thermal sensing is also important for ensuring the safety of additive manufacturing systems and operators. By monitoring
temperature levels in critical components and detecting potential overheating or thermal runaway events, thermal sensors can help prevent equipment
damage, mitigate fire hazards, and ensure operator safety during printing operations.
Overall, thermal sensing plays a crucial role in additive manufacturing by providing valuable insights into temperature dynamics, thermal behavior of
materials, and process performance. By integrating thermal sensing technology into AM systems, manufacturers can enhance process control, optimize
part quality, and ensure the reliability and safety of additive manufacturing operations
Thermal sensor
Sensor uses and advantages
Thermal control methodology
In laser additive manufacturing (AM), there has been a significant amount of research focused on process control, particularly in the areas of geometric and
thermal control methodologies. Here are some examples from the literature:
1. Coordination-Control SchemeHan and Jafari developed a coordination-control scheme to ensure uniform bead geometry and deposition accuracy in laser
AM. This scheme regulated material flow in real-time to match the relative changes in speed between the deposition head and the workpiece, aiming for
consistent deposition.
2. High Frame-Rate Camera for Infrared Imaging: Hu and Kovacevic installed a high frame-rate camera above the laser-nozzle setup to capture infrared
images of the molten pool in real-time during the laser AM process. These images were used to measure the dimensions of the melt pool, enabling feedback
control to adjust process parameters as needed.
3. CMOS Camera for Melt Pool Width Measurement: Hofman et al. utilized a CMOS camera to obtain the width of the melt pool during laser AM. They
then employed a proportional-integral (PI) controller to regulate the width of the melt pool during deposition, ensuring consistent deposition quality.
4. Control-Oriented MIMO Model: Cao and Ayalew proposed a control-oriented multiple input multiple output (MIMO) model for laser aided powder
deposition (LAPD) processes. This model, derived in Hammerstein form, combined linearized dynamics with nonlinear relationships based on mass and
heat balance considerations. Inputs to the model included laser power and scanning speed, while outputs included layer height and molten pool temperature.
5. CCD Camera for Molten Pool Height Measurement: Zeinali and Khajepour utilized a CCD camera to measure the height of the molten pool in laser
cladding processes. They developed an adaptive sliding mode control algorithm with an uncertainty estimator to regulate the clad height, validating the
controller's performance through experiments.
These examples demonstrate the diverse approaches researchers have taken to implement process control in laser AM, with a focus on achieving uniform
deposition, controlling melt pool dimensions, and optimizing process parameters for improved part quality and consistency.
cont
• work relating to feedback control systems specifically for WAAM is
beginning to emerge.
• A passive visual sensing system was developed to measure the nozzle
to top surface distance (NTSD) as a feedback signal.
• an adaptive controller was proposed to keep the NTSD constant
through adjust wire-feed speed online
Cont..
the system consists of two subsystems: multiple-sensor monitoring
module and control module. The monitoring system is responsible for
capturing process signals and proving warning of defects. The function
of the control system is to regulate the bead geometry and heat input.
software
Nc program
• The 6 axis
• Using cad software
• They build prototype
Conclusion

• The lack of robustness and repeatability of metal AM processes has been widely pointed out as one major issue that deserves considerable research
efforts and technological advances. Therefore development and implementation of monitoring and control strategies possess a priority to push forward
the industrial breakthrough of metal AM technology. In this paper, a comprehensive review of research efforts in area of process monitoring and control
for AM was provided. Defects of AM, and its related signal and sensing method were categorized and discussed. Different control methods were
reviewed and categorized according to its process type, control object and algorithm type. Furthermore, a framework of automatic WAAM system was
proposed based on literature review. The proposed automatic WAAM system consists of two modules: multiple-sensor monitoring module and multiple-
controller system. The proposed framework can help push forward the development of WAAM in both research and industry.
• WAAMSim
• Simulation software that allows you to visualize RoboWAAM's
deposition environment, run through your RoboWAAM program, and
check for collision
• MetalXL
• Software with built-in checks that enables industrial 3D metal printing
from design to print in one go using robotic WAAM technology
• WAAM3D
• End-to-end software workflow for large-scale metal 3D printing,
including tool path planning and
Cost estimation

Xir 1800 Visual sensor Spectral sensor

3344 8000 6000

Cont …

Acoutic light sensor Thermal sensor Screen monitor

300 142 30000

Total cost

= 78000

Hexagon ab software