A5037531533- Additive Manufacturing and 3D Printing Technologies
- Manufacturing Process and Optimization
- Industrial Vision Systems and Defect Detection
- Advanced Multi-Objective Optimization Algorithms
- Spacecraft Design and Technology
- Big Data and Business Intelligence
- Digital Transformation in Industry
- Simulation Techniques and Applications
- Injection Molding Process and Properties
- Ultrasonics and Acoustic Wave Propagation
- Modular Robots and Swarm Intelligence
- Advanced Machining and Optimization Techniques
- Optical measurement and interference techniques
- Additive Manufacturing Materials and Processes
Stevens Institute of Technology
2019-2024
Interface (United Kingdom)
2013
Bringing a new AI system into production environment involves multiple different stakeholders such as business owners, risk officer, ethics officers approving the System for specific usage. Governance frameworks typically include manual steps, including curating information needed to assess risks and reviewing outcomes identify appropriate actions governance strategies. We demo human-in-the-loop automation that takes natural language description of an intended use case in order create...
Abstract Part defects in additive manufacturing are more frequent compared to machining or molding. Failures can go unnoticed for hours, wasting resources and extending process cycle times. This paper describes a Machine Learning based method automated sensing of onset failure machinery. Investigations conducted on Fused Filament Fabrication (FFF) 3D printer, the same methods then applied digital light processing printer. The investigation focuses signal-based analysis, specifically passive...
Evaluating the safety of AI Systems is a pressing concern for organizations deploying them. In addition to societal damage done by lack fairness those systems, deployers are concerned about legal repercussions and reputational incurred use models that unsafe. Safety covers both what model does; e.g., can it be used reveal personal information from its training set, how was built; only trained on licensed data sets. Determining an system requires gathering wide set heterogeneous sources...
Abstract The open-loop process by which 3D printers operate often leads to significant time and material losses due print failures. A challenge in Additive Manufacturing (AM) is assessing 3D-printed parts mid-print for the dimensional stability of internal structures their microstructure properties. These features are typically opaque visual inspections after can only be accessed property characterization if part cut open. An assessment methods evaluating situ using air-coupled ultrasonic...
Heterogeneous materials exhibit considerable spatial variations in properties, impacting structural performance and local stress strain fields. Recent research has focused on considering material behaviour uncertainties quantifying the impact of response, requiring definition random fields for describing variability. In this paper, a stochastic modeling methodology additively manufactured structures is implemented (forward model). The parameters are determined from experimental Digital Image...
Simulations quantifying the uncertainty in structural response and damage evolution require accurate representation of randomness underlying material stiffness strength behaviors. In this paper, mean variance descriptions variability additively manufactured composite specimens are augmented with random field correlation descriptors that represent process dependence on property heterogeneity through microstructure variations. Two lengths a rotation parameter introduced into randomized...
Abstract In the field of additive manufacturing (AM), mid-print failure is exceedingly common due to user error, bad design, or environmental factors that cannot be readily prepared for. This holds for most if not all types AM, but perhaps none more so than popular Filament Deposition Modeling (FDM) method machines. Absent total power failure, bulk modes in FDM can expressed as having an immediate impact on mechanical system, whether a head collision warping, increased pressure stepper it...
The goal of DARPA's Symbiotic Design Cyber Physical Systems (SDCPS) program is to develop tools for "correct-by-synthesis" design cyber physical systems (CPS) and reduce the time from concept deployment years months. Achieving this poses several hard challenges. spaces are high-dimensional cross-products discrete continuous spaces. It can take minutes hours evaluate performance a design. human designer's intent often not concretely articulated. Sometimes designs created scratch but rather by...
Simulations used for navigating design spaces or finding optimal points and Pareto fronts require accuracy resolution to guide the designers towards effective decisions. High-fidelity high-resolution simulators are computationally expensive. Due inconsistent parameter settings physically invalid outcomes, such can also fail return a solution during automation optimization loops. In CPS systems, failures be expected when there is hard execution-time deadline on simulator producing solution....