Search Results for deep learning

This article presents the analytics challenges in additive manufacturing. It discusses the types and applications of data analytics. Data analytics can be classified into four types: descriptive, diagnostic, predictive, and prescriptive. The diverse applications of data analytics and machine learning include design, process-structure-properties (PSP) relationships, and process monitoring and quality control. The article also presents tools used for data analytics.

Additive manufacturing (AM) is a revolutionary technology that fabricates parts layerwise and provides many advantages. This article discusses polymer AM processes such as material extrusion, vat photopolymerization (VPP), powder-bed fusion (PBF), binder jetting (BJ), material jetting (MJ), and sheet lamination (SL). It presents the benefits of online monitoring and process control for polymer AM. It also introduces the respective monitoring devices used, including the models and algorithms designed for polymer AM online monitoring and control.

Machine vision, also referred to as computer vision or intelligent vision, is a means of simulating the image recognition and analysis capabilities of the human eye and brain system with digital techniques. The machine vision functionality is extremely useful in inspection, supervision, and quality control applications. This article presents a variety of machine vision functions for different purposes and provides a comparison of machine and human vision capabilities in a table. It discusses the processes of a machine vision system: image acquisition, image preprocessing, image analysis, and image interpretation. The article provides information on the uses of machine vision systems in three categories of manufacturing applications: visual inspection, identification of parts, and guidance and control applications.

The use of additive manufacturing (AM) is increasing for high-value, critical applications across a range of disparate industries. This article presents a discussion of high-valued engineering components predominantly used in the aerospace and medical industries. Applications involving metal AM, including methods to identify pores and voids in AM materials, are the focus. The article reviews flaw formation in laser-based powder-bed fusion, summarizes sensors used for in situ process monitoring, and outlines advances made with in situ process-monitoring data to detect AM process flaws. It reviews investigations of ML-based strategies, identifies challenges and research opportunities, and presents strategies for assessing anomaly detection performance.

This article focuses on the technologies and applications of additive manufacturing (AM) in the oil and gas industry. It then presents the challenges of AM and the oil and gas industry. The article provides a detailed description of the critical steps in the AM process chain, including part selection, design optimization, and process planning, control, and inspection. Qualification and certification standardization is discussed, as is a commitment to reduce the carbon footprint of the manufacturing sector through AM. It ends with the future outlook of AM in the oil and gas industry.

This article focuses specifically on material modeling applied to structure-property predictions. It provides general guidelines and considerations in terms of modeling the salient material features that ultimately impact the mechanical performance of parts produced by additive manufacturing (AM). Two of the primary ingredients needed to predict structure-property relationships via material modeling include a geometrical representation of the microstructural features of interest (e.g., grain structure and void defects) and a suitable constitutive model describing the material behavior, both of which can be scale and resource dependent. The article also presents modeling challenges to predict various aspects of (process-) structure-property relationships in AM.

This article details findability, accessibility, interoperability, and reusability (FAIR) additive manufacturing data management principles and examines related motivations, benefits, and challenges. It explains opportunities to advance the state of the AM community efforts in fostering FAIR data management practices/principles and outlines the consequence of such efforts on technology maturation and industrialization for AM technologies.

Additive manufacturing (AM) provides exceptional design flexibility, enabling the manufacture of parts with shapes and functions not viable with traditional manufacturing processes. The two paradigms aiming to leverage computational methods to design AM parts imbuing the design-for-additive-manufacturing (DFAM) principles are design optimization (DO) and simulation-driven design (SDD). In line with the adoption of AM processes by industry and extensive research efforts in the research community, this article focuses on powder-bed fusion for metal AM and material extrusion for polymer AM. It includes detailed sections on SDD and DO as well as three case studies on the adoption of SDD, DO, and artificial-intelligence-based DFAM in real-life engineering applications, highlighting the benefits of these methods for the wider adoption of AM in the manufacturing industry.

Process optimization is the discipline of adjusting a process to optimize a specified set of parameters without violating engineering constraints. This article reviews data-driven optimization methods based on genetic algorithms and stochastic models and demonstrates their use in powder-bed fusion and directed energy deposition processes. In the latter case, closed-loop feedback is used to control melt pool temperature and cooling rate in order to achieve desired microstructure.

The use of an effective control design, along with high-performance hardware and software for controller implementation, allows the use of feedback process control for manufacturing processes to improve part quality and consistency. This article provides an overview of control system design and its application to various manufacturing processes. It presents various examples of control system applications to show that appropriate control strategies increase the robustness of the processes by eliminating process sensitivity to system variations and external disturbances.

This article describes the test methods for evaluating the durability of a metal in soil. It provides useful information on soil characteristics such as soil electrical resistivity, pH value, and soil texture. Specimen design, preparation, burial, and retrieval techniques are discussed. The type of information sought during soil-induced corrosion evaluation controls the design configuration and the nature of the corrosion measurements. Consideration of these factors during the planning stage helps the corrosion engineer to obtain the maximum amount of information with the minimum number of problems.

This article provides an introduction to architected cellular materials, their design, fabrication, and application domain. It discusses design decisions involving the selection, sizing, and spatial distribution of the unit cell, property-scaling relationships, and the integration of cells within an external boundary. It describes how manufacturing constraints influence achievable feature resolution, dimensional accuracy, properties, and defects. It also discusses the mechanical behavior of architected cellular materials and the role of additive manufacturing in their fabrication.

Of the many different nondestructive evaluation (NDE) techniques, ultrasonic inspection continues to be the leading nondestructive method for inspecting composite materials, because measurements can be quantitative and the typical defect geometries and orientations lend themselves to detection and characterization. This article focuses on the three common methods for ultrasonic nondestructive inspection of plastics, namely pitch-catch, through-transmission, and pulse-echo, as well as the three basic types of ultrasonic NDE scans: the A-scan, B-scan, and C-scan. The discussion includes the linear and phased array systems that are sometimes used for large-scale inspection tasks to reduce scan times, the various gating and image processing techniques, and how ultrasonic data are interpreted and presented. A brief section on future trends in ultrasonic inspection is presented at the end of the article.

In all grinding operations, care must be used in the selection of wheels and abrasive belts to meet finish and tolerance requirements without damaging the workpiece. This article discusses the major aspects of the grinding wheel, including production methods, selection considerations, standard marking systems, abrasives, and bonding types. It compares bonded wheel grinding with abrasive belt grinding. The article reviews the types of grinding fluids and discusses their importance in grinding operations. It describes the specific grinding processes and provides recommendations for grinding and grinding wheels.

This article presents the damage tolerance criteria for military composite aircraft structures to safely operate the structures with initial defects or in-service damage. It describes the effects of defects, such as wrinkles in aircraft structures, and the reduction in compressive strength and tensile strength. The article reviews low velocity impacts in aircraft structures in terms of resin toughness, laminate thickness, specimen size and impactor mass, and post-impact fatigue. It explains the tension strength analysis, such as linear elastic fracture mechanics and R-curve methods, to predict the residual strength of the structures.

Corrosion modeling is an essential benchmarking element for the selection and life prediction associated with the introduction of new materials or processes. These models are most naturally expressed in terms of differential equations or in other nonexplicit forms of mathematics. This article discusses the principles and applications of various models developed for understanding the corrosion mechanism. These models include mechanistic models, including Pourbaix model, thermophysical module, electrochemical module, and ion association model; risk-based models; and knowledge models. The risk-based model and knowledge models are illustrated with examples for better understanding. The article also describes boundary-element modeling and pitting corrosion fatigue models.

This article presents a brief history of additive manufacturing (AM). It begins by describing additive manufacturing prehistory, dating back to 1860, which is characterized by additive part creation without the use of a computer. The article then discusses the development of additive manufacturing processes occurring in the period from 1968 to 1984 and is followed by a section on modern additive manufacturing (1981 to the late 2000s). The article concludes by providing information on the growth of additive manufacturing since 2010 and the development of standards.

Hardening and depth of hardening of steel is a critically important material and process design parameter. This article presents a selective overview of experimental and predictive procedures to determine steel hardenability. It also covers the breadth of steel hardenability, ranging from shallow, to very difficult to harden, to air-hardening steels.