26800 research outputs found
Sort by
Multi-Modal Explainable Artificial Intelligence for neural network-based tool wear detection in machining
No full textAs data in manufacturing becomes more accessible and larger in scope, solutions are required to explain the underlying reasoning behind predictive models that use data from various modalities. This research aims to develop multi-modal Explainable Artificial Intelligence approaches to facilitate human-interpretable solutions for tool wear prediction in machining. Initially, a multi-modal neural network was used as the supervised Machine Learning classifier for training and binary classification by applying a set of features comprised of different modalities. The data consisted of two sets of image data representing the Flank and the Rake views of the tools accompanied by time series data including acceleration, acoustics, temperature, spindle speed, and feed rate during the orthogonal tube turning process. The classifier was used to predict the condition of the tool after the cutting process. After the training process and the network performance evaluation were completed, two multi-modal neural network explainability approaches were investigated. The Full Multi-Modal Explainable Artificial Intelligence approach provided a general explainability of the feature importance based on latent space representations of the input data while the Decomposed Multi-Modal Explainable Artificial Intelligence approach produced explainability results of input feature importance by decomposing the multi-modal model into its constituent single modal sub-models. The developed explainability methods were demonstrated to provide sufficient information regarding explaining the multi-modal network performance and the decision-making processes. Hence, the presented methods will enable end-users to understand the underlying logic for advanced neural network-based models that incorporate data from different modalities through the selection of multiple algorithms
The impact of by-product production on the availability of critical metals for the transition to renewable energy
No full textThe availability of critical metals for the transition to renewable energy is of growing importance as the scale of the materials needed becomes apparent. There is a concern some production could be constrained as by-products of other metals. This paper provides a conceptional model about the behavior of by-product metals. From this model, scenarios are developed about the types of price-production interaction patterns between by-product and main product metals that might be observed given different situations of by-product metal availability. The paper\u27s aim is to determine if there is evidence of by-product metal production being constrained by the output of the main product metal as the transition to renewable energy continues to grow. For the rare earth metals, the four metals used for renewable energy are shown to move together significantly in price (0.672–0.956) and not to move significantly with the other rare earths prices (0.028–0.351). These results suggest the possibility of a mismatch between overall joint rare earth production and the availability of the separated individual metals. For the base metals, significant statistical correlations (0.787–0.962) between by-product metal and main product metal production with weaker significance for prices (0.093–0.750) was found. These results indicate a possible pattern of production constraint for critical by-product metals as their demand grows. Policy recommendations in the case of constrained by-product metal production are to enhance recycling programs to increase the metals\u27 supply, find new sources for the metals, and develop technology that reduces the metals\u27 need
Advances in pyrazolo[1,5-a]pyrimidines: Synthesis and their role as protein kinase inhibitors in cancer treatment
Get PDFPyrazolo[1,5-a]pyrimidines are a notable class of heterocyclic compounds with potent protein kinase inhibitor (PKI) activity, playing a critical role in targeted cancer therapy. Protein kinases, key regulators in cellular signalling, are frequently disrupted in cancers, making them important targets for small-molecule inhibitors. This review explores recent advances in pyrazolo[1,5-a]pyrimidine synthesis and their application as PKIs, with emphasis on inhibiting kinases such as CK2, EGFR, B-Raf, MEK, PDE4, BCL6, DRAK1, CDK1 and CDK2, Pim-1, among others. Several synthetic strategies have been developed for the efficient synthesis of pyrazolo[1,5-a]pyrimidines, including cyclization, condensation, three-component reactions, microwave-assisted methods, and green chemistry approaches. Palladium-catalyzed crosscoupling and click chemistry have enabled the introduction of diverse functional groups, enhancing the biological activity and structural diversity of these compounds. Structure-activity relationship (SAR) studies highlight the influence of substituent patterns on their pharmacological properties. Pyrazolo[1,5- a]pyrimidines act as ATP-competitive and allosteric inhibitors of protein kinases, with EGFR-targeting derivatives showing promise in non-small cell lung cancer (NSCLC) treatment. Their inhibitory effects on B-Raf and MEK kinases are particularly relevant in melanoma. Biological evaluations, including in vitro and in vivo studies, have demonstrated their cytotoxicity, kinase selectivity, and antiproliferative effects. Despite these advances, challenges such as drug resistance, off-target effects, and toxicity persist. Future research will focus on optimizing synthetic approaches, improving drug selectivity, and enhancing bioavailability to increase clinical efficacy
Development of a HAND-based flood risk assessment tool in Google Earth Engine for a data-scarce region in the US
No full textDespite the decreased disaster resilience of rural communities in the Great Lakes region to flooding, flood mitigation efforts have been impeded by inadequate data and lack of appropriate tools for understanding flood risk. Development of such resources often requires data and computationally intensive approaches, which are challenging in data-scarce conditions. This study presents the development of a web application in Google Earth Engine (GEE) for flood risk assessment. The application utilizes the Height Above the Nearest Drainage (HAND) model and synthetic rating curve (SRC) for fluvial flood inundation modeling, the Simulating WAves Nearshore (SWAN) model for coastal flood inundation modeling, the United States Geological Survey (USGS) regional regression equations for estimating peak discharge, and depth-damage functions of the HAZUS-MH flood model for estimating losses due to building-level impacts. The GEE-based geospatial web application, which is operational across five counties in the Western Upper Peninsula (WUP) of Michigan, fulfills the requirement of the community and decision-makers to assess the risks caused by flooding in the region. We demonstrated the applicability of the tool in the Ontonagon River, Michigan, and the results indicate the suitability of the platform for implementing decisions, long-term planning, and understanding flood risk with a reasonable degree of accuracy
Co-training of multiple neural networks for simultaneous optimization and training of physics-informed neural networks for composite curing
No full textThis paper introduces a Physics-Informed Neural Network (PINN) technique that co-trains neural networks (NNs) that represent each function in a system of equations to simultaneously solve equations representing an out-of-autoclave (OOA) cure process while conducting optimization in adherence to process requirements. Specifically, this co-training approach benefits from using NNs to represent OOA inputs (air temperature profile) and outputs (part and tool temperature profiles and degree of cure). Production requirements can then be levied on the inputs, such as maximum air temperature and minimum cure cycle, and simultaneously on the outputs, such as degree of cure, maximum part temperature, and part temperature rate limits. Co-training the NNs results in an optimized input producing outputs that meet all OOA process requirements. The technique is validated with finite element (FE) simulations and physical experiments for curing a Toray T830H-6 K/3900-2D composite panel. Hence, this novel approach efficiently models and optimizes the OOA cure process
Coupled Lake-Atmosphere-Land Physics Uncertainties in a Great Lakes Regional Climate Model
Get PDFThis study develops a surrogate-based method to assess the uncertainty within a convective permitting integrated modeling system of the Great Lakes region, arising from interacting physics parameterizations across the lake, atmosphere, and land surface. Perturbed physics ensembles of the model during the 2018 summer are used to train a neural network surrogate model to predict lake surface temperature (LST) and near-surface air temperature (T2m). Average physics uncertainties are determined to be 1.5°C for LST and T2m over land, and 1.9°C for T2m over lake, but these have significant spatiotemporal variations. We find that atmospheric physics parameterizations alone are the dominant sources of uncertainty (45%–53%), while lake and land parameterizations account for 33% and 38% of the uncertainty of LST and T2m over land respectively. Interactions of atmosphere physics parameterizations with those of the land and lake contribute to an additional 13%–17% of the total variance. LST and T2m over the lake are more uncertain in the deeper northern lakes, particularly during the rapid warming phase that occurs in late spring/early summer. The LST uncertainty increases with sensitivity to the lake model\u27s surface wind stress scheme. T2m over land is more uncertain over forested areas in the north, where it is most sensitive to the land surface model, than the more agricultural land in the south, where it is most sensitive to the atmospheric planetary boundary and surface layer scheme. Uncertainty also increases in the southwest during multiday temperature declines with higher sensitivity to the land surface model
Higher-order diffusion and Cahn–Hilliard-type models revisited on the half-line
No full textIn this paper, we solve explicitly and analyze rigorously inhomogeneous initial-boundary-value problems (IBVP) for several fourth-order variations of the traditional diffusion equation and the associated linearized Cahn–Hilliard (C-H) model (also Kuramoto–Sivashinsky equation), formulated in the spatiotemporal quarter-plane. Such models are of relevance to heat-mass transfer phenomena, solid-fluid dynamics and the applied sciences. In particular, we derive formally effective solution representations, justifying a posteriori their validity. This includes the reconstruction of the prescribed initial and boundary data, which requires careful analysis of the various integral terms appearing in the formulae, proving that they converge in a strictly defined sense. In each IBVP, the novel formula is utilized to rigorously deduce the solution’s regularity and asymptotic properties near the boundaries of the domain, including uniform convergence, eventual (long-time) periodicity under (eventually) periodic boundary conditions, and null noncontrollability. Importantly, this analysis is indispensable for exploring the (non)uniqueness of the problem’s solution and a new counter-example is constructed. Our work is based on the synergy between: (i) the well-known Fokas unified transform method and (ii) a new approach recently introduced for the rigorous analysis of the Fokas method and for investigating qualitative properties of linear evolution partial differential equations (PDE) on semi-infinite strips. Since only up to third-order evolution PDE have been investigated within this novel framework to date, we present our analysis and results in an illustrative manner and in order of progressively greater complexity, for the convenience of readers. The solution formulae established herein are expected to find utility in well-posedness and asymptotics studies for nonlinear counterparts too
Exploring Italian consumers’ willingness to pay for sustainable fashion: the roles of eco-consciousness and vintage preference
Get PDFThe former study suggests that Italian consumers are willing to pay premiums for new bio-based products but expect discounts on second-hand clothing, revealing a sustainability bias. This study adds more insights into such a bias by examining factors influencing consumers’ willingness to pay a premium for bio-based clothing and their expected discounts for second-hand items. Using Bayesian Mindsponge Framework (BMF) analytics on a dataset of 402 Italian consumers, the findings reveal that environmental concerns are positively associated with premiums for bio-based clothing, while higher income and education levels are also associated with the higher premium that consumers are willing to pay. For second-hand clothing, the preference for vintage appeal is linked to lower expected discounts. Men and younger consumers tend to expect higher discounts for second-hand clothing. By providing insights into Italian consumers’ sustainable fashion choices, this study offers implications for businesses, policymakers, and researchers aiming to promote eco-conscious consumption and sustainability in the fashion industry
Forb diversity globally is harmed by nutrient enrichment but can be rescued by large mammalian herbivory
Get PDFForbs (“wildflowers”) are important contributors to grassland biodiversity but are vulnerable to environmental changes. In a factorial experiment at 94 sites on 6 continents, we test the global generality of several broad predictions: (1) Forb cover and richness decline under nutrient enrichment, particularly nitrogen enrichment. (2) Forb cover and richness increase under herbivory by large mammals. (3) Forb richness and cover are less affected by nutrient enrichment and herbivory in more arid climates, because water limitation reduces the impacts of competition with grasses. (4) Forb families will respond differently to nutrient enrichment and mammalian herbivory due to differences in nutrient requirements. We find strong evidence for the first, partial support for the second, no support for the third, and support for the fourth prediction. Our results underscore that anthropogenic nitrogen addition is a major threat to grassland forbs, but grazing under high herbivore intensity can offset these nutrient effects
