Manufacture Parts Research Articles

Light Helicopter Manufacturing Returns to the UK

Hill Helicopters is manufacturing a high-tech luxury helicopter from scratch and has secured pre-orders totalling 943. A key part of manufacture is its CAD/CAM platform

Tomographic eigenstrain reconstruction for full-field residual stress analysis in large scale additive manufacturing parts

Current experimental and numerical quantification methods are limited in their ability to full-field mapping of the unpredictable distribution of all residual stress and permanent plastic strain components in additive manufacturing parts with discontinuous processing properties. To address this limitation, a tomographic eigenstrain (inherent strain) reconstruction method, that merges eigenstrain reconstruction with diffraction strain tomography for mapping volumetric distribution of all components of eigenstrains and corresponding elastic deformations like residual stresses non-destructively using minimum amount of tomographic scans is presented through numerical experiments, and then applied to the analysis of a CM 247 LC superalloy additive manufacturing part using diffraction strain tomography data. The method reconstructs all eigenstrain and corresponding residual stress components, parallel to the build direction, aligned with the experimental data component accurately, demonstrating its potential in optimizing the performance and reliability of parts designed for high-tech industries such as aerospace. Subsequent validations using the X-ray diffraction sin2ψ and neutron diffraction strain scanning techniques confirm the method's reliability in reconstructing residual stress components parallel to the plane of powder bed that are different from the experimental data component. Furthermore, the novel findings of this study reveal a characteristic residual stress distribution pattern within additive manufacturing parts particularly those featuring rectangular shapes. Microstructural analysis also validates eigenstrain distribution in accordance with the findings on the characteristic distribution of residual stresses, highlighting the significance of this method in advancing materials research and development.

Direct voxel classification from x-ray projections for 3D pore detection applied to laser sintered parts

X-ray computed tomography (XCT) is a validated and frequently used tool to verify part geometry and to perform a non-destructive inspection of additive manufacturing parts. However, the acquisition of a large number of x-ray projections generates long inspection times. This conflicts with a high throughput of the production process and hinders the integration of XCT as an in-line quality control procedure for low-end parts. In this paper, we propose a method to obtain three-dimensional (3D) information of the internal pores from a limited view or limited angle scan. The method combines a forward projection model of a cone-beam x-ray system and a deep learning neural network to directly classify each individual voxel, based on the x-ray projections in order to avoid the reconstruction and segmentation step. Accompanying reconstruction artifacts for limited view and limited angle XCT scans are thereby reduced, while preserving 3D information of the pores, defects or inclusions inside the material. The method is validated on real x-ray projections of polymer laser sintered industrial parts and shows a significant reduction in the required x-ray projections, hence acquisition time.

Optimising digital signal processor‐based defect detection in smart manufacturing with lightweight convolutional neural networks

AbstractIndustrial defect detection is an important part of intelligent manufacturing, and Internet of things (IoT)‐based defect detection is receiving more and more attention. Although deep learning (DL) can help defect detection reduce the cost and improve the accuracy of traditional manual quality inspection, DL requires huge computational resources and is difficult to be simply deployed on IoT devices with limited computational power and memory resources. Digital signal processor (DSP) is an important IoT device with small size, high performance and low energy consumption, which has been widely used in intelligent manufacturing. In order to perform accurate defect detection on DSP, the authors proposed various optimisation strategies and then used a parallel scheme to scale the model to execute on multiple cores. The authors’ method evaluated it on Northeastern University Surface Defect Dataset, Magnetic Tile Defect Dataset, Rail Surface Defect Dataset and Silk Cylinder Defect Dataset, and the experimental results showed that the authors’ method obtains faster speeds without accuracy loss compared to running the same Convolutional Neural Networks model on a mainstream desktop CPU. This means that the authors’ method can realise efficient and accurate defect detection on IoT devices with limited computational power and memory resources, which opens up new possibilities for future development in the field of smart manufacturing.

Research on an Online Monitoring Device for the Powder Laying Process of Laser Powder Bed Fusion.

Improving the quality of metal additive manufacturing parts requires online monitoring of the powder bed laying procedure during laser powder bed fusion production. In this article, a visual online monitoring tool for flaws in the powder laying process is examined, and machine vision technology is applied to LPBF manufacture. A multiscale improvement and model channel pruning optimization method based on convolutional neural networks is proposed, which makes up for the deficiencies of the defect recognition method of small-scale powder laying, reduces the redundant parameters of the model, and enhances the processing speed of the model under the premise of guaranteeing the accuracy of the model. Finally, we developed an LPBF manufacturing process laying powder defect recognition algorithm. Test experiments show the performance of the method: the minimum size of the detected defects is 0.54 mm, the accuracy rate of the feedback results is 98.63%, and the single-layer laying powder detection time is 3.516 s, which can realize the effective detection and control of common laying powder defects in the additive manufacturing process, avoids the breakage of the scraper, and ensures the safe operation of the LPBF equipment.

Assessing the potential for additive manufacturable spare parts in the railway industry by a data-driven framework

Due to the long-life span of railway vehicles, spare part availability is a major challenge for railway companies. To avoid downtime, high procurement costs, and shortages, railway companies often stock large and costly inventories. However, additive manufacturing (AM) makes it possible to cost-effectively produce spare parts in small quantities to reduce inventory. To assess the suitability of a spare part and choose the optimal supply strategy, economic and technical data must be available in good quality. The lack of centralised information and poor data quality is one of the biggest challenges for AM potential assessment.A workflow was designed, in which system data from various databases was mapped and enhanced with technical information. In addition, the potential for additive manufacturability was evaluated based on costs and savings affected by transportation, design, manufacturing, storage compared to other supply strategies. The resulting database was then used to classify the spare parts based on specific characteristics. The defined classes provide information about the timeline for implementing the part and the type of application. A similarity check based on identified AM spare parts is used to search the entire database to identify additional potential AM parts. The result is a data-driven framework to conduct a holistic potential assessment for additive manufacturable spare parts.

Deep Learning Approach for Enhanced Transferability and Learning Capacity in Tool Wear Estimation

As an integral part of contemporary manufacturing, monitoring systems obtain valuable information during machining to oversee the condition of both the process and the machine. Recently, diverse algorithms have been employed to detect tool wear using single or multiple sources of measurements. In this study, a deep learning approach is proposed for estimating tool wear, considering cutting parameters. The model's accuracy and transferability in tool wear estimation were assessed with milling experiments conducted under varying cutting parameters. The results indicate that the proposed method outperforms conventional methods in terms of both transferability and rapid learning capabilities.

Investigating the influence of inter-printhead bond strategy on tensile strength in multi-printhead PLA parts for collective additive manufacturing

Multiple printheads working in parallel on the same part have been proposed to increase the speed of FDM additive manufacturing systems. However, it is still unclear how this implementation affects part performance, particularly whether optimal inter-printhead bonds can be realised and how these might impact proposed speed gains. Therefore, this study investigates the trade-offs in tensile performance of parts printed using multiple printheads and inter-printhead interface strategies compared to a single printhead. By evaluating various inter-printhead bond orientations and chunking angles at the inter-printhead interface, the research assesses whether introducing a seam, which may weaken the part, can be justified by the potential benefit of workload distribution to enable parallelism in FDM additive manufacturing systems, hence faster print times. Findings show that the inline inter-printhead bond orientation and 30◦ chunking angle specimens exhibit a 9% and < 1% reduction in ultimate tensile strength, respectively. Therefore, with optimal inter-printhead interface settings, multiple printheads can speed up FDM additive manufacturing systems with minimal impact on tensile performance.

Preliminarna studija geometrije navara kod navarivanja na nerđajući čelik 316l korištenjem mig postupka zavarivanja

The application of the gas metal arc welding (GMAW) process can produce metal parts in additive manufacturing (AM) and has the advantages of fast production speed and material saving. There are some different requirements for welding beads between the AM process and the usually welded joints, so preliminary research on the 316L GMAW is conducted to find the optimal voltage, wire feed speed (WFS), and travel speed (TS). Taguchi algorithm was used firstly to obtain parameters to achieve the desired higher reinforcement, lower width, and higher aspect ratio (R/W). The voltage of 19 V, WFS of 500 cm/min, and TS of 15 cm/min were obtained as the optimal one among the samples. To further investigate the influence of the parameters and verify the result, a two-factor-three-level full factorial design was carried out, with the consideration of the interaction between factors. At last, the studied parameters were used in the AM process using a GMAW welding robot to verify the reliability.

METAL ADDITIVE MANUFACTURING IN SPACE AND AEROSPACE EXPLORATION: CURRENT PROCESSES, MATERIAL SELECTION AND CHALLENGES

Space and aerospace exploration have the highest regulations regarding design and manufacturing processing. As a result, an immense multi-tier manufacturing system has been developed because of the tooling and machinery associated with the production of high-grade spacecraft and aircraft. Furthermore, the nature of the supply chain and hefty capital investments further complicate spacecraft and aircraft manufacturing. Metal additive manufacturing (AM) offers the best solution to produce complex parts from digital information without the need for large amounts of tooling and machinery. MAM represents a promising aspect of innovations to revolutionize the way manufacturing is designed and constructed for many applications. Moreover, various research activities are being carried out in the aerospace industry for MAM process, which produces parts that would be impossible to create using conventional manufacturing (CM) methods. MAM has gained popularity in recent years from its initial stages of development in the research phase to providing a wide range of functional applications in the aerospace industry to manufacture and repair a wide range of components used in spacecraft and aircraft. The significant growth in MAM process can be attributed to its commercial and performance advantages within the aerospace industry. In this paper, a detailed assessment of existing MEM processes is presented, emphasizing the ability to manufacture and repair metal parts for the aerospace industry, providing information about the materials currently in use, as well as identifying possible future applications and challenges. There is no doubt that the information provided in this study will support exploring MAM’s new opportunities for future needs in space and aerospace exploration.

Overheating detection in the powder bed fusion of metals using a laser beam through local thermal resistance analysis

The powder bed fusion of metals using a laser beam is a production technique to additively manufacture parts by melting metallic powder layers with a laser. The geometry of a part affects how the heat is dissipated from the process zone. Geometry-induced heat accumulation results in undesirable processing conditions and part properties. Current approaches to identify locations of overheating are mostly based on thermal finite element simulations. This study presents an analytical approach to detect these locations based on a voxel-wise calculation of the local thermal resistance. With a thermal simulation, it was verified that the presented approach efficiently detects areas of overheating. This allows the automated evaluation of the manufacturability of existing geometries.

A model for the rapid prediction of small feature width producible by Metal AM

In this paper, an analytical parametric model is presented for the swift prediction of small features producible in the X-Y plane through the Metal AM Laser Powder Bed Fusion (LPBF) process. This model can be employed to design and create surface textures on Additive Manufactured parts without the need for costly post-processing steps. The Rosenthal equation is the basis for the model, which considers both the build parameters of the LPBF process and the thermo-physical properties of the materials. The initial model was constructed and assessed using one LPBF machine followed by the implementation of a tuning method utilizing the Limited Memory Algorithm for Bound Constrained Optimization to enhance the model’s accuracy. Overall, the findings suggest that with a simple optimization step based on a single printed tuning sample, precise analytical models can be established for specific LPBF machines and materials combinations.

Influence of absorptivity variation on Laser Powder Bed Fusion simulation and impact of process parameters on residual stress formation

Laser powder bed fusion stands out for its precision and capability to manufacture parts with complex geometry, yet it poses challenges due to the significant thermal gradient, causing elevated residual stress and potential damage. Ensuring precise manufacturing of Ti6Al4V, demands control over input parameters. This study employs a simulation model using COMSOL Multiphysics to explore the influence of thermal absorptivity on reliability of predicted results and role of key process parameters including laser velocity and power on temperature distribution and thermal stress. Results indicate that the volumetric heat source with temperature-dependent absorptivity, ended in more accurate predictions compared to constant absorptivity. Moreover, considering absorptivity as a function of laser power and velocity enhanced results particularly in situations with higher energy densities. However, while reducing the laser velocity and increasing laser power typically led to higher residual stress as the area exposed to high temperature was wider, there was a critical point for the impact of laser velocity, where the role of thermal gradient overcomes the role of maximum temperature.

Statistical model for setting up a machining center to achieve geometric tolerances at maximum material requirement

The production adjustment on a machining center aims to manufacture parts that conform to their manufacturing specifications. To achieve this, the adjustment operation consists of determining a set of setting variables associated with a geometric model of the machine. The maximum material requirement proves very useful in guaranteeing the conformity of parts to intended functional requirements. This is the case, for example, for ensuring part assembly by linking the tolerances of dimensional and geometric specifications. However, setting up a machining center to achieve geometric tolerance with the maximum material requirement needs decoupling geometric tolerance from the associated dimensional tolerance(s). In particular, the dimension to be targeted by the feature of linear size must be determined with the performance of the production means from a dimensional and geometric point of view. Therefore, the article aims to develop a computation model for setting the target dimension to be machined, ensuring maximum production of parts conforming to a geometric specification at maximum material requirement. This model is based on the statistical distributions of the history of dimensional and geometric measurements taken on different parts produced by the production means under consideration. A case study is carried out to validate the proposed model.

A prediction method of surface geometric deviation for additive manufacturing parts based on knowledge-integrated deep learning algorithm

Compared with traditional machining processes, additive manufacturing (AM) has received widespread attention in recent years because of its high degree of modeling freedom. However, due to the multiple manufacturing errors and complex physical state changes involved in the process, the geometric deviation on the AM part surface is a challenge for controlling product geometrical quality. To address this problem, data-driven machine learning (ML) techniques have been widely studied in product quality controlling. However, traditional ML greatly depends on the training sample data, and suffers the risk of violating physical mechanisms due to the lack of domain knowledge. In order to take the best advantage of domain knowledge, prior information and deep learning algorithm, this paper proposes a knowledge-integrated deep learning algorithm and constructs the geometric deviation prediction model of the AM part surface. After that, the method was verified with design of experiments. The results show that compared with the data-driven neural network (DDNN), the knowledge-integrated neural network (KINN) has fewer iterations during the training process, less sample data requirement and more accurate prediction results.

Collaborative AI & immersive VR simulation for workplace layout optimization in manufacturing applications

Workplace layouts significantly affect operator well-being and productivity in manufacturing. Traditional methods often neglect to incorporate operator feedback until expensive prototypes are built, leading to costly redesigns. This paper presents an approach that leverages collaborative AI and immersive virtual reality (VR) simulation for early engagement of relevant stakeholders. The proposed approach employs Artificial Intelligence algorithms, particularly genetic algorithms, to recommend layouts to minimize travel distance. The approach considers spatial constraints, workstation characteristics, and task sequences using an A-based pathfinding algorithm. This algorithm estimates the fitness score of each investigated layout plan while ensuring that collision-free operator paths, incorporating travelling to all the required workstations, are evaluated each time. These layouts are evaluated by humans through VR task simulations. In the VR environment, factors such as accessibility, reachability, posture, and occlusions are assessed alongside travel distance. Intuitive and informed layout updates are enabled within VR. This approach was tested in two use cases inspired by in-situ repairing of composite materials and post-processing of additive manufacturing parts. Results demonstrated increased design efficiency, allowing for a comprehensive exploration of alternatives while incorporating end-user feedback. Finally, informed decision-making improved user satisfaction and reduced stress during design.

Generative Design of Progressive Die

The evolution of products, dictated by the necessity to survive in the market, requires multi manufacturing processes to be performed simultaneously. Progressive die is one type of tool in which several individual stations are arranged in sequence. Each station will perform one or more various operations on the part. It is widely used to manufacture parts for automobile and other industries. This process with the help of progressive die is a very inexpensive method for forming moderate to large volume parts while keeping labor and die costs down. This study is dedicated to providing a 3D Computer-Aided Design (CAD) model for simple and economical progressive die construction which is suitable for the mass production of an industrial bracket. Different forces and stress convoluted in performing metal operations were also calculated with an appropriate formula. Generative designing technologies are also applied to yield the most efficient design economically as well as productively without compromising the optimum features of the tool. Various contours representation of data obtained is also presented in this work.

Robots and export quality

Robots are becoming a key part of manufacturing in developed and emerging economies. This paper examines how automation can affect international trade through quality upgrading. The effects of robot use on export quality are estimated, by combining cross-country and cross-industry data on industrial robots with detailed Harmonized System 10-digit trade data. Robot diffusion in (preexisting) foreign customers is used as an instrumental variable to predict robot adoption in the home country-industry. The findings show that robot diffusion increases the quality of exported products. Quality improvements are predominantly driven by upgrading of (initially) lower-quality exports of developed and developing countries. Since developing countries tend to export lower quality goods, we find stronger effects overall for developing economies. The paper also finds some differences in the type of robots – sophisticated or more basic – are associated with quality gains in developing and developed economies.

Metal Surface Defect Detection Based on Metal-YOLOX

Article Metal Surface Defect Detection Based on Metal-YOLOX Xiaoli Yue , Jiandong Chen , and Guoqiang Zhong * 1 College of Computer Science and Technology, Ocean University of China, Qingdao 266000, China * Correspondence: gqzhong@ouc.edu.cn Received: 14 June 2023 Accepted: 13 September 2023 Published: 21 December 2023 Abstract: Due to the limitations of manufacture technologies, working environments and other conditions, metals (such as steel and aluminum) are susceptible to surface defects during the production process. Therefore, defect detection is an indispensable part of metal manufacturing. This paper innovatively proposes a one-stage defect detection model named Metal-YOLOX. Metal-YOLOX addresses the limitations in existing models posed by large variances in defect features and inadequate balance between detection accuracy and efficiency. Firstly, the composite convolution module of Metal-YOLOX integrates texture, dilated and deformable convolutions to filter out irrelevant features and extract effective feature information. Secondly, the feature cross-fusion module (HCNet) alleviates the problem of large dimensional differences in defects. HCNet uses skip connections to establish the connection between the original multi-scale features and the output nodes, and reduces the addition of redundant information. Thirdly, Metal-YOLOX adopts the deep separable convolution and global channel reduction. This lightweight design helps reduce computational complexity. Finally, detailed experiments demonstrate that, in terms of mean average precision, Metal-YOLOX achieves 79.83, 69.14, and 81.22 on the NEU-DET, GC-10 and Aluminum datasets, respectively. Furthermore, Metal-YOLOX dramatically reduces parameter number and computational complexity. The experiments validate that the Metal-YOLOX model improves the detection performance, maintains the detection speed, and meets the real-time requirements.

3D Printer Selection for the Sustainable Manufacturing Industry Using an Integrated Decision-Making Model Based on Dombi Operators in the Fermatean Fuzzy Environment

Three-dimensional printers (3DPs), as critical parts of additive manufacturing (AM), are state-of-the-art technologies that can help practitioners with digital transformation in production processes. Three-dimensional printer performance mostly depends on good integration with artificial intelligence (AI) to outperform humans in overcoming complex tasks using 3DPs equipped with AI technology, particularly in producing an object with no smooth surface and a standard geometric shape. Hence, 3DPs also provide an opportunity to improve engineering applications in manufacturing processes. As a result, AM can create more sustainable production systems, protect the environment, and reduce external costs arising from industries’ production activities. Nonetheless, practitioners do not have sufficient willingness since this kind of transformation in production processes is a crucial and irrevocable decision requiring vast knowledge and experience. Thus, presenting a methodological frame and a roadmap may help decision-makers take more responsibility for accelerating the digital transformation of production processes. The current study aims to fill the literature’s critical theoretical and managerial gaps. Therefore, it suggests a powerful and efficient decision model for solving 3DP selection problems for industries. The suggested hybrid FF model combines the Fermatean Fuzzy Stepwise Weight Assessment Ratio Analysis (FF–SWARA) and the Fermatean Ranking of Alternatives through Functional mapping of criterion sub-intervals into a Single Interval (FF–RAFSI) approaches. The novel FF framework is employed to solve a critical problem encountered in the automobile manufacturing industry with the help of two related case studies. In addition, the criteria are identified and categorized regarding their influence degrees using a group decision approach based on an extended form of the Delphi with the aid of the Fermatean fuzzy sets. According to the conclusions of the analysis, the criteria “Accuracy” and “Quality” are the most effective measures. Also, the suggested hybrid model and its outcomes were tested by executing robustness and validation checks. The results of the analyses prove that the suggested integrated framework is a robust and practical decision-making tool.