Search

Wire arc additive manufacturing technology (WAAM) has become a very promising alternative to high-value large metal components in many manufacturing industries. Due to its long process cycle time and arc-based deposition, defect monitoring, process stability and control are critical for the WAAM system to be used in the industry. Although major progress has been made in process development, path slicing and programming, and material analysis, a comprehensive process monitoring, and control system are yet to be developed. This paper aims to provide an in-depth review of sensing and control design suitable for a WAAM system, including technologies developed for the generic Arc Welding process, the Wire Arc Additive Manufacturing process and laser Additive Manufacturing. Particular focus is given to the integration of sensor-based feedback control, and how they could be implemented into the WAAM process to improve its accuracy, reliability, and efficiency. The paper concludes by proposing a framework for sensor-based monitoring and control system for the GMAW based WAAM process. This framework provides a blueprint for the monitoring and control strategies during the WAAM process and aims to identify and reduce defects using information fusion techniques.

Geometric properties of material deposited by the wire arc additive manufacturing (WAAM) process often deviates from desired setpoints. To improve the accuracy and repeatability of the WAAM process, an effective control strategy to maintain desired deposition geometry that operates robustly under various welding conditions is required. In this work, a control strategy utilizing multi-input multi-output (MIMO) model-predictive control (MPC) is presented. This approach, based on linear autoregressive (ARX) modelling, aims to improve the accuracy and flexibility of deposited bead geometry in the WAAM process. The MPC controller updates welding parameters between successive layers by minimizing a cost function based on sequences of input variables. Measurements of deposited bead geometry are made by laser scanner and input to the linear ARX model, which then makes future bead geometry predictions. Weighting coefficients of the ARX model are trained iteratively throughout the manufacturing process. Experimental results show that the derived control strategy can reduce fluctuations in a part's height by 400% and maintain the fluctuation within an acceptable range. In addition, the fluctuations in bead width along a single weld seam was also improved by more than 50%.

Wire Arc Additive Manufacturing (WAAM) is well suited for the manufacture of sizeable metallic workpieces featuring medium-to-high geometrical complexity due to its high deposition rate, low processing conditions limit, and environmental friendliness. To enhance the current capability of the WAAM process for fabricating structures with complex geometry, this paper proposes a robot-based WAAM strategy adapted specifically for fabricating free-form parts with wire structures composed of multiple struts. Contributions in this work include: (i) The study of bead modelling, which establishes optimal welding parameter selection for the process; (ii) The novel manufacturing strategy, including the adaptive slicing methodology and height control system for accurately depositing every single strut; (iii) Detailed manufacturing procedures for multi-strut branch intersections as well as the collision-free path planning to control the overall fabrication process. To verify the effectiveness of this proposed WAAM approach, two complex wire structures were fabricated successfully, indicating the feasibility of the proposed fabrication strategy.

In more recent times, research on various aspects of the Wire Arc Additive Manufacturing (WAAM) process has been conducted, and efforts into monitoring the WAAM process for defect identification have increased. Rapid and reliable monitoring of the WAAM process is a key development for the technology as a whole, as it will enable components produced by the process to be qualified to relevant standards and hence be deemed fit for use in applications such as those found in the aerospace or naval sectors. Intelligent algorithms provide inbuilt advantages in processing and analysing data, especially for the large data sets generated during the long manufacturing cycles. Interdisciplinary engineering (IDE) furnishes a concept integrating computer science and industrial system manufacturing engineering together to treat large amounts of process monitoring data. In this work, a WAAM process monitoring and defect detection system integrating intelligent algorithms is presented. The system monitors welding arc current and voltage signals produced by the WAAM process and makes use of a support vector machine (SVM) learning method to identify disturbances to the welding signal which indicate the presence of potential defects. The incremental machine learning models developed in this work are trained via statistical feature analysis of the welding signals and a novel quality metric that improves detection rates is also presented. The incremental learning approach provides an efficient means of detecting welding-based defects, as it does not require large quantities of data to be trained to an operational level (addressing a major drawback of other machine learning methods). A case study is presented to validate the developed system, results show that it was able to detect a set of defects with a success rate greater than 90% F1-score.

With the increasing demand for high productivity and low cost, the advanced manufacturing system has become more complex. It is challenging to develop an advanced manufacturing system by engineers or researchers from manufacturing engineering disciplines. To combine knowledge and skills from two or more disciplines, interdisciplinary engineering (IDE), from industrial information integration, was carried out as a promising subject for further developing advanced manufacturing systems. Recently, Wire arc additive manufacturing (WAAM) technology has attracted attention from industrial sectors due to its capability to fabricate medium-to-large scale components with low capital investment and high productivity. In addition, a variant of WAAM, called multi-directional WAAM techniques, has been developed for the direct fabrication of parts with overhanging features. The multi-directional approach can reduce the need for additional supporting structures, reducing (amongst other things) material costs, manufacturing time, and post-process machining requirements. Although the multi-directional WAAM has a great potential for practical industrial use, the process planning also becomes more complex for developing an automated system for industrial use. This paper presents a novel automated processing planning algorithm and integrates with the automated robot offline programming (AOLP) engine and WAAM hardware. The proposed system aims at planning the multi-directional WAAM process automatically. The process planning algorithm consists of four key modules relating to (a) robot motion planning, (b) initial collision processing, (c) layer sequence optimisation, and (d) weld torch pose adjustment. In the first stage, the required robot motions to deposit each layer are obtained through an AOLP engine. Then a collision matrix is generated to guide later steps of the planning process from these robot motions. After this, layer sequence optimisation is performed to eliminate collision between the robotic system and the part. If collision still exists, adjustments to the torch pose are made to avoid the remaining collision. The final chapter of this paper demonstrates the effectiveness of the proposed system via a real-world case study, where a workpiece with overhanging features was fabricated.

In the offshore oil and gas industry, the repair of damaged drilling riser buoyancy modules (DRBMs) currently involves the removal of defective regions using hand-held abrasive tools. This operation is not only time consuming, but also exposes the worker to hazardous materials generated by the grinding process. In order to increase productivity and minimize occupational health hazards, an automated robotic repair system is developed. The DRBM has a large and almost featureless surface making it particularly challenging to accurately and efficiently detect and localize all the defective regions on the surface. To address these challenges, an eye-in-hand monocular vision system is used to generate the DRBM's global surface map. The defective regions can then be identified from the map using ellipse detection algorithms, and subsequently removed with a robotic manipulator. Experimental results have shown that the proposed solution is able to accurately and efficiently remove defective regions on the surface of DRBMs.

As a new emerging technology, wire arc additive manufacturing (WAAM) has attracted extensive interests from both academia and industry during recent years. WAAM uses welding arc as an energy source to fuse metal wire and deposit layer by layer, which provides the advantages of freeform deposition. In order to improve its manufacture precision, stability and repeatability, it is necessary to develop sensing and control strategy for WAAM process. This research develops a passive visual sensing system for a robotic WAAM system. A new deep learning technique (Mask R-CNN) is proposed to detect and segment the melt pool area, and the width of melt pool can be measured based on the coordinate of the bounding rectangle. The pseudo-random ternary (PRT) signals were used to stimulate the WAAM process, and the corresponding width can be measured by the Mask R-CNN. Based on the width data and corresponding PRT input, a dynamic model of adaptive neuro-fuzzy inference system was built for the WAAM process.

In this paper, an automatic rebar counting system based on image processing and machine learning techniques is proposed. The system makes use of several image processing techniques including Canny edge detection, Circle Hough Transform (CHT) calculation and a machine learning system to accurately identify the number of individual rebar in a given bundle under various lighting conditions. This work includes a study of a number of different machine learning algorithms including decision tree, K-Nearest Neighbor (KNN), Support Vector Machine (SVM), traditional neural network and Convolutional Neural Network (CNN). The proposed system is able to transfer the original object detection problem into a more easily solvable image classification problem and is hence achieve an overall accuracy of 95.99% in the presence of reasonable lighting conditions.

Wire arc additive manufacturing (WAAM) is an emerging process for the fabrication of complex metallic components due to its high deposition rate and low equipment costs. This paper presents a WAAM strategy adapted especially for producing freeform wire-structured parts. Contributions in this work include a bead modelling study to establish optimal welding parameters and a novel deposition strategy, which includes an adaptive slicing methodology and height control system to control the overall process. The effectiveness of this approach, termed skeleton arc additive manufacturing (SAAM), was demonstrated with the manufacture of a complex wire-structured part.

The quality of a path generated from an automated motion planning algorithm is of considerable importance, particularly when used in a real world robotic application. In this work a new path optimization algorithm, called the Adaptive Partial Shortcut algorithm, is presented. This algorithm optimizes paths as a post process to motion planning, and is designed specifically for use on industrial manipulators. The algorithm optimizes a robot's degrees of freedom independently allowing it to produce manipulator paths of particularly high quality. This new algorithm utilizes an adaptive method of selecting the degree of freedom to optimize at each iteration, giving it a high level of efficiency. Tests conducted show the effectiveness of the algorithm; over a range of different test paths, the adaptive algorithm was able to generate solutions with a 60 % reduction in collision checks compared to the original partial shortcut approach.

This paper presents a novel sampling-based motion planning method for shape inspection applications with a redundant robotic system. In this paper, a 7-Degree-of-Freedom (DOF) redundant robotic system consisting of a 6-DOF manipulator and a 1-DOF turntable is used for the industrial inspection problem. A Set Covering Problem (SCP) is formulated to select suitable viewpoints that satisfy the inspection requirements, and a Generalized Travelling Salesman Problem (GTSP) is formulated to determine both the robot poses and the visiting sequences. While previous studies solve the two problems separately, we formulate the SCP and GTSP problems as a combined sequencing SC-GTSP problem. A Random-Key Genetic Algorithm (RKGA) is then used to solve the combined SC-GTSP problem in a one-step optimization process. To validate the effectiveness of our method, we applied the proposed method to several motion planning cases. The results show that the proposed method outperforms the previous approaches by requiring up to 28.1% less total inspection time.

In this paper, we propose a novel planning method for the shape inspection applications with a sensor-equipped industrial manipulator. In the proposed approach, the view planning problem is formulated as a Set Covering Problem (SCP) and the path planning problem is formulated as a Travelling Salesman Problem (TSP). The combined SCP and TSP problems are then formulated as a sequencing problem, and solved simultaneously via Random-Key Genetic Algorithm (RKGA) in a one-step optimization process. With the proposed method, time-efficient and collision-free motion plan that satisfies the requirement of the robotic inspection tasks can be generated automatically. Computational studies were performed to verify the effectiveness of the proposed planning method. The method was found to outperform other approaches by generating robot paths that complete the inspection task with shorter execution time. The Overall time reductions were found up to 28.4% compared to the previous approaches. To further demonstrate the effectiveness of the proposed method in a real-world inspection task, an experimental test was also performed using an ABB industrial robotic system equipped with an on-board 3D scanner.

This paper presents a new sampling based path planning algorithm, called the Lazy Significant Edge Algorithm (LSEA). LSEA utilises roadmap connectivity information to bias its sampling strategy towards objects in a robots workspace that have not yet been navigated by the robot. This allows LSEA to avoid redundant sampling of configuration space. The robotic system used in this paper to test LSEA consists of an articulated industrial manipulator mounted on a linear rail. LSEA was tested on this system with a series of different path planning problems in order to judge its overall effectiveness. When compared to a number of other popular sampling based path planning algorithms, it was concluded that LSEA had the best overall performance. It was observed to solve the various path planning problems more quickly than its counterparts, utilising fewer clash checks in order to reach the various solutions.

Probabilistic methods have been proven to be effective for robotic path planning in a geometrically complex environment. In this paper, we propose a novel approach, which utilizes a specialized roadmap expansion phase, to improve lazy probabilistic path planning. This expansion phase analyses roadmap connectivity information to bias sampling towards objects in the workspace that have not yet been navigated by the robot. A new method to reduce the number of samples required to navigate narrow passages is also proposed and tested. Experimental results show that the new algorithm is more efficient than the traditional path planning methodologies. It was able to generate solutions for a variety of path planning problems faster, using fewer samples to arrive at a valid solution.