10784 research outputs found
Sort by
Advancing towards Safe Reinforcement Learning over Sparse Environments with Out-of-Distribution Observations: Detection and Adaptation Strategies
No full textPublisher Copyright: © 2024 IEEE.Safety in AI-based systems is among the highest research priorities, particularly when such systems are deployed in real-world scenarios subject to uncertainties and unpredictable inputs. Among them, the presence of long-tailed stimuli (Out-of-Distribution data, OoD) has captured much interest in recent times, giving rise to many proposals over the years to detect unfamiliar inputs to the model and adapt its knowledge accordingly. This work analyzes several OoD detection and adaptation strategies for Reinforcement Learning agents over environments with sparse reward signals. The sparsity of rewards and the impact of OoD objects on the state transition distribution learned by the agent are shown to be crucial when it comes to the design of effective knowledge transfer methods once OoD objects are detected. Furthermore, different approaches to detect OoD elements within the observation of the agent are also outlined, stressing on their benefits and potential downsides. Experiments with procedurally generated environments are performed to assess the performance of the considered OoD detection techniques, and to gauge the impact of the adaptation strategies on the generalization capability of the RL agent. The results pave the way towards further research around the provision of safety guarantees in sparse open-world Reinforcement Learning environments.Peer reviewe
Wear properties of a new Al80Mg10Si5Cu5 multicomponent alloy
Get PDFPublisher Copyright: © 2024 The Author(s)The present study investigates the tribological properties of a newly developed multicomponent aluminium weight-light multicomponent alloy for wear based on the Al80Mg10Si5Cu5 system for lightweight automotive applications, especially back drum discs. The samples were manufactured by High-Pressure Die Casting (HPDC) employing cast alloy returns and secondary aluminium ingots and were tested at room temperature (RT) and 200 °C. It has been observed that the Al80Mg10Si5Cu5 alloy offers a higher hardness and wear resistance at RT and especially at 200 °C compared with the AlSi9Cu3 reference alloy (x10 times reduction in wear rate). The impact of maintaining the external surface layer (skin) of HPDC cast parts has been studied for the ball-on disc test, showing improved tribological properties and the possibility of avoiding the machining of contact surfaces. The as-cast Al80Mg10Si5Cu alloy with the surface layer showed a wear rate coefficient of 5 × 10−4 mm3/N.m2 at RT, a 50 % lower than that of the sample without skin. Solution heat-treated samples (72 h at 440 °C, water quenching at 75 °C, and natural aging) with the surface layer showed a wear rate coefficient of 11 × 10−4 mm3/N.m2, approximately 20 % lower than the sample without a surface layer. The wear rate of AlSi9Cu3 alloy decreased by more than 50 % in the samples without skin at RT. At 200 °C, wear rate coefficients were lower in the samples with the surface layer.Peer reviewe
An Age-Friendly Neighbourhood Index as a Long-Term Urban Planning Decision-Making Tool
Get PDFPublisher Copyright: © 2024 by the authors.People responsible for shaping the future of cities often seek valuable tools to assist in their decision-making processes. Using objective, quantified, and analysed data proves highly beneficial when determining where to focus interventions at the city level. Various urban indexes have been established to measure different aspects of urban life, ranging from sustainability to liveability. These indexes encompass multiple dimensions of a city, including mobility and walkability, among others. The age-friendly cities initiative developed indicators for assessing the age-friendliness of cities. Some researchers further refined these indicators to focus on urban planning competencies. Building on this foundation, this article aims to present an Age-Friendly Neighbourhood Index (AFNI) validated by a panel of experts using the Delphi method. This index can serve as a valuable tool for urban planners when they need to prioritise interventions to enhance age-friendliness at neighbourhood scale. The article also outlines the necessary data and measurement techniques for these indicators. The AFNI has been applied to a real case study in the city of Santander (Spain). This application assesses the age-friendliness of various neighbourhoods in Santander, demonstrating the challenges in acquiring sub-local quality data and emphasising the need for data-driven urban management.Peer reviewe
Enzyme activity monitoring in industrial solid-state fermentation processes based on colorimetric LoC compatible with R2R fabrication
No full textPublisher Copyright: © 2024, Avestia Publishing. All rights reserved.In this work we present a user-friendly, disposable lab-on-a-chip (LoC) for colorimetric enzyme activity monitoring in industrial solid-state fermentation (SSF) processes. The chip, compatible with roll-to-roll (R2R) extrusion coating fabrication, significantly reduces the cost per device improving the efficiency [1]. COC was chosen for the chip fabrication for its flexibility required in R2R manufacturing and its transparency necessary for colorimetric detection [2]. An adhesive tape is used to cover the structured microfluidic foil. The chip works on the principle of capillary-driven flow microfluidics [3]. The user adds a drop in the inlet then the chip is self-filled by capillary action and the sample arrives to the reaction chamber. The flows continue until reaching the capillary pump, which removes the liquid excess and isolates the detection area where the enzymatic reaction takes place. Amylase and cellulase were selected for this work due to its industrial importance in SSF processes. To be suitable on LoC, enzymatic assays should provide a robust signal output and easy handling for the final user. These requirements are fulfilled by one-step colorimetric assays, where a dye is released from a substrate by enzymatic action. Two strategies have been studied for the integration of the enzymatic substrates in the chip. One the one hand, drop-casting of the reagents followed by freeze-drying. On the other hand, spotting of the reagents by piezoelectric deposition, with subsequently air-drying. Different storage conditions were tested to calculate the shelf-life of the LoC and provide optimal reagent stability. To measure changes in enzymatic activity on-site in the SSF plant, a miniaturised colorimetric reader adapted to the LoC geometry and the specific enzymatic reactions has been developed. This device incorporates customized miniaturized electronics design for electro-optical output and optic detection and data acquisition elements. Besides, an Android application to enable communication with the device. First, a functional validation of the LoC was carried out with commercial enzymes. A calibration curve with increasing enzymatic concentrations was build up to calculate the limit of detection (LoD) of the system. Functional validation was then performed using enzymes extracted from real samples from a SSF plant. A simple and fast procedure was set up for the extraction of the enzymes from solid samples. Finally, the enzymatic activity of the samples obtained at different time-points of the fermentation process was measured in 5-10 minutes. The samples show increasing enzymatic activity indicating that as the fermentation process progresses, the amount of generated enzymes is higher. In conclusion a LoC for one-step detection of amylase and cellulase was developed using R2R extrusion coating. Enzymatic activity was measured using a customized colorimetric reader, and the performance of the chip was validated and tested with commercial enzymes and real samples from a SSF plant. This technology enables quick analysis of enzymatic activity in enzyme-containing ferments.Peer reviewe
Enabling the injection of hydrogen in high-pressure gas grids: Investigation of the impact on materials and equipment
Get PDFPublisher Copyright: © 2023 Hydrogen Energy Publications LLCGreen hydrogen is a renewable gas that can help to reach the goal of decarbonizing the energy sector. The use of the natural gas grid for seasonal storage and transport of hydrogen needs previous assessment of its tolerance, ensuring safe and viable operation. In this work, the tolerance of the most typical material pipelines and key elements of high-pressure gas grids to the transport of 20 mol% hydrogen blends at 80 barg has been investigated at pilot scale. For this, the experimental campaign carried out at a testing platform replicating a high-pressure gas grid lasted 3000 h. The tightness of different kind of valves led to hydrogen losses below 1 Nml·h−1. No embrittlement or other kind of damage was found on these valves or the different parts of the equipment (regulator, flowmeter) tested. C-ring, 4pb and CT-WOL specimens have been prepared from carbon steel pipes (API5L Gr X42 to X70), showing no damage after exposure to hydrogen. H2/CH4 deblending has also been successfully achieved using Pd-based membrane technology, obtaining a high-purity separated streams.This project has received funding from the Fuel Cells and Hydrogen 2 Joint Undertaking (now Clean Hydrogen Partnership) under Grant Agreement No. 875091 ‘HIGGS’. This Joint Undertaking receives support from the European Union's Horizon 2020 Research and Innovation program, Hydrogen Europe and Hydrogen Europe Research . Financial support from the T13-20D (the Aragón Government and the ESF) is also gratefully acknowledged.Peer reviewe
Time of the week AutoRegressive eXogenous (TOW-ARX) model to predict thermal consumption in a large commercial mall
Get PDFPublisher Copyright: © 2024This paper proposes a procedure to build a Time of the Week AutoRegressive eXogenous (TOW-ARX) model, indexed with respect to time and day of the week, to characterize heat consumption in tertiary buildings. Models for building heat load characterization and prediction are crucial to enhance energy efficiency. The proposed model can be used for different purposes, e.g., control of indoor climate, or characterization of the thermal response of the building. A case study is described where the TOW-ARX model is used to characterize the energy consumption of a large retail building in Madrid. In order to discard the risk of model overfitting, cross validation is applied using the k-fold technique. The performance of the TOW-ARX model is compared with a set of different models: a reduced version of the model where similar segments are clustered using the k-means method (R-TOW-ARX), a general ARX model, a linear regression steady-state TOW model (TOW-LR), a version of the latter reduced through clustering (R-TOW-LR), and a general multiple linear regression model (LR). The results reveal that ARX-based models notably outperforms the rest. The TOW-ARX model shows the best metrics, but also outnumbers the number of coefficients of the other models by far. The selection of the most suitable model is not straightforward and should depend on the purpose of such model: the TOW-ARX model would arguably be the best for control purposes due to its low mean absolute error, but the ARX model would be preferable for an efficient characterization of the thermal response of a building due to its reduced number of parameters.Peer reviewe
Towards the Design, Quality Assessment and Explainability of Synthetic Tabular Data Generation Techniques for Metabolic Syndrome Diagnosis
No full textPublisher Copyright: © 2024 IEEE.In last years decision-making Machine Leaning (ML) approaches have evolved from traditional methods to evidence-based approaches, particularly in healthcare sector. However, sharing data with third parties raises significant security and privacy concerns. To address these issues, researchers have explored data anonymization, distributed privacypreserving data mining, and synthetic data generation (SDG). SDG, in particular, shows promise in enabling secure data sharing while preserving privacy, crucial for developing advanced AI models. This paper focuses on Metabolic Syndrome (MetS) data, a condition affecting a significant portion of the population, and investigates various synthetic tabular data generation (STDG)techniques. It evaluates the performance of an AutoML approach for predicting MetS using different percentages of synthetic data assessed through a specific evaluation framework. Moreover, presents an explainability and feature relevance analysis of the proposed STDG methods.Peer reviewe
Compressive Behaviour of Concrete Mixtures with Simultaneous Additions of Waste Wind-Turbine Blade and Coarse Recycled Concrete Aggregates
No full textPublisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.The development of more sustainable materials for the construction sector begins with optimal use of natural resources and appropriately recycling the residues produced at the end of the lifecycle of infrastructures. In the coming years, many wind-farms will be dismantled, providing big amounts of waste wind-turbine blade (WWTB). This waste can be an opportunity to decarbonize the construction industry, which causes huge green-house-gas emissions each year endangering the environment. Waste generated during the demolition of concrete elements has been used in the past few years for structural concrete manufacturing in the form of Recycled Concrete Aggregate (RCA). Feasibility of combination of RCA with WWTB has yet to be study in concrete mixtures. This study aims to analyze the simultaneous use of RCA and WWTB to produce a greener concrete mixture as an effort to transform an industry hitherto considered polluting into a more sustainable one. For this purpose, three concrete mixtures were produced with different volume combinations of each waste: 0% WWTB and 0% RCA; 5% WWTB and 50% RCA; and 10% WWTB and 100% RCA. Compressive strength, modulus of elasticity and Poisson´s coefficient have all been evaluated at 28 days. Results show that an adequate dosage of both types of residues can result in a structural concrete mixture that preserves basic compressive strength properties while at the same time providing an outlet for a large amount of waste generated in both construction and wind energy sector.Peer reviewe
Electrodeposition of Zn and Cu Nanoparticles into TiO2 Nanotubes on Ti6Al4V: Antimicrobial Effect against S. Epidermidis and Cytotoxicity Assessment
No full textPublisher Copyright: © 2024 by the authors.Surface modification of the Ti6Al4V alloy (ASTM grade 5), with the fabrication of vertically oriented TiO2 nanotubes, has been receiving increasing attention both as a way to provide advanced bioactive features and the ability to act as reservoirs for a localized, controlled drug release. In this work, TiO2 nanotubes were grown on the surface of a Ti6Al4V alloy through electrochemical anodization. An ethylene glycol-based electrolyte containing 0.5 wt.% NH4F and 2.5% (v/v) H2O was used. Post-anodizing heat treatments at 500 °C in air atmosphere were performed to achieve a crystalline oxide layer with a higher mechanical stability. Following these treatments, Zn or Cu nanoparticles were incorporated into the nanotubular structures through electrodeposition processes. Then, the antimicrobial performance of the obtained surfaces was assessed against Staphylococcus epidermidis, a Gram-positive bacterium common in implant-related infections. Lastly, the cytotoxicity of the produced surface was evaluated against MC3T3-E1 mouse pre-osteoblast cells. In general, Cu-doped TiO2 nanotubes presented an almost total antimicrobial action, while Zn doped samples had a lower, but still significant antibacterial effect. However, a highly cytotoxic effect against MC3T3-E1 cells was observed on all anodized samples due to the release of vanadium from the alloy. In spite of this, the surface modification reported in this work can be a valid solution for existing commercially available orthopedic implants, considering that similar solutions were already studied in in vivo assays.Peer reviewe