UPCommons (Universitat Politècnica de Catalunya)
Not a member yet
281550 research outputs found
Sort by
Predicting EU orchard farmers' adoption intentions for pesticide-reducing innovations using an extended theory of planned behavior
No full text© 2025 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ).Reducing synthetic pesticide use is central to the EU Green Deal. However, with legislative progress stalled, farmers increasingly bear the responsibility for implementing pesticide reduction. This highlights the need to understand how decisions are made and what shapes farmers' intentions to adopt technologies that reduced pesticide use, particularly in orchard systems where pesticide dependence remains high despite the availability of promising alternatives. This study examines stakeholder roles in orchard pest management, identifies key drivers of farmers’ adoption intentions, and develops a tool to predict adoption likelihood. A mixed-methods approach was applied, combining expert interviews (n = 16) and a structured survey of orchard farmers (n = 354). We extended the Theory of Planned Behavior (TPB) by incorporating three additional constructs: perceived cost, relative advantage of current practices, and perceived ease of use, all based on established behavioral theories. Results show that decision-making involves multiple stakeholders and is influenced by factors such as farm size, ownership structure, and external pressures. Meanwhile, structural equation modeling shows that attitudes and perceived ease of use significantly boost adoption intentions. In contrast, perceived cost and the relative advantage of current practices serve as barriers. Subjective norms and perceived behavioral control were not significant predictors. Based on these findings, a web-based adoption intention predictive tool was developed to support stakeholders in assessing adoption likelihood. The study offers practical insights for reducing adoption barriers, strengthening positive attitudes toward innovation, and supporting sustainable pest management in orchard systems.This study was supported by the NOVATERRA project under the research grant agreement number: 101000554. The NOVATERRA project has received funding from the European Union’s Horizon 2020. We sincerely appreciate the valuable administrative and technical support provided by Cristina Poyato Santiago and Filippo Alfonso Baldaro to the NOVATERRA project.Peer ReviewedPostprint (published version
Coupled r-h mesh adaptation in embedded finite element frameworks
No full textAmong mesh-based methods, embedded methods are widely used in the literature for solving partial differential equations (PDEs) in complex domains without the challenge of creating bodyfitted (or, more generally, domain-fitted) meshes. In these methods, the physical domain is embedded in a simple background domain that can be easily meshed. However, the uniform accuracy of the background mesh in these methods is often inadequate in critical regions, such as boundary layers and vortex zones. While h-refinement methods for mesh adaptation are commonly employed to enhance resolution in regions with complex local phenomena like vortices, they have limitations in areas with steep gradients in one direction but little to no gradients in the other(s), such as boundary layers. These limitations include increased computational costs due to the lack of mesh alignment strategies. Mesh alignment can be achieved through ¿-methods, which deform the mesh without adding new elements, avoiding overhead computational costs and mesh repartitioning. However, while ¿-methods improve mesh alignment in boundary layers, they can lead to insufficient element density in other areas where local phenomena, such as vortices, occur. In this study, we introduce a variational moving adaptive mesh method (VMAM), a subset of ¿-methods, combined with h-refinement (h-method), designed to operate on distributed memory parallel systems. This ¿ - h method, coupled with embedded finite element method (EFEM), addresses the challenge of inadequate resolution in key regions. The ¿-method aligns the mesh in boundary layers, while the h-method refines it in local critical areas, such as regions where vortices occur. The alignment in VMAM is achieved through a mapping strategy that transfers the simple computational domain to the physical domain. This mapping, which is challenging in domain-fitted approaches, is applied easily in the embedded method due to the simplicity of the background domain. In the present method, a level-set approach is used to define the physical domain. For simple cases, this is determined using analytical expressions, whereas in more complex cases, explicit boundary representations from computer aided design (CAD) are employed. The method is examined on several two-dimensional (2D) and three-dimensional (3D) cases, demonstrating its effectiveness and improvements in the simulation results compared to those obtained by non-adaptive EFEM on Cartesian grids.The authors acknowledge grant PID2021-123611OB-I00, funded by MCIN/AEI/10.13039/501100011033 and “ERDF: A Way of Making Europe” as well as grant CEX2018-000797-S, funded by the Ministerio de Ciencia e Innovación, MCIN/AEI/10.13039/501100011033, for supporting this research.Peer ReviewedPostprint (published version
Identification of abnormal conditions in induction motors from current spectrum images using a two-stage approach with progressive learning
No full textBackground and objectives: This study presents a fault diagnosis system for induction machines based on a two-stage architecture using Convolutional Neural Networks (CNN). The aim is to improve fault identification by simplifying the classification process through sequential modeling. Methods: A dataset of 5404 images was generated from the Fourier spectra of current signals acquired from a test bench under five conditions: healthy machine, rotor asymmetry fault, broken rotor bar, race bearing fault, and ball bearing fault. A CNN based on the Visual Geometry Group (VGG) architecture was trained from scratch and then adapted using transfer learning. The classification strategy follows two steps: first, distinguishing healthy from faulty machines; then, identifying the specific fault type. Results: The system reached 96.5% accuracy in the first stage and 98.5% in the second. All main performance metrics (sensitivity, specificity, precision, F1-score) remained above 95%. The behavior of the models was examined using Uniform Manifold Approximation and Projection (UMAP), which showed clearer separation between conditions in the latent space when using the sequential approach. In addition, Gradient-Weighted Class Activation Mapping (Grad-CAM) visualizations provided insights into the image regions influencing each decision, showing consistent focus on spectral areas related to each condition. Conclusions: The combination of image-based preprocessing, sequential classification, and model interpretation techniques leads to accurate predictions and helps to understand how the models behave. These features support its use in predictive maintenance tasks for industrial applications.The article is funded by the R+D+i project (PID2021-128013OB-I00) through the MCIN/AEI/10.13039/-501100011033 and ERDF A way of making Europe, as well as by Generalitat Valenciana, Spain (CIAICO/2022/042). Additionally, I personally acknowledge funding from the Secretary of Higher Education, Science, Technology and Innovation of Ecuador as an individual grant holder.Peer ReviewedPostprint (published version
Self-reference UAV motion elimination and structural modal parameter restoration method
No full textTo ensure the accuracy of vision-based vibration testing, close-range measurement using unmanned aerial vehicles (UAV) demonstrates significant advantages. However, the required fixed reference point in each measurement view to eliminate the UAV ego motion is hardly satisfied. To address this limitation, this paper proposes a self-reference UAV motion elimination and structural modal parameter restoration method. First, different from the traditional fixed reference point required in each segment vibration test, a vibration measurement point within the field of view is selected as self-reference point to eliminate the UAV motion. Second, the modal parameters are identified using self-reference relative pixel displacement responses. The modal frequencies and damping ratios can be obtained directly, while the mode shape segments are distorted except the segment that include pier reference point. Therefore, the mode shape segments are restored sequentially through the relationship between the false mode shape and the true mode shape. Third, a numerically simulated example was generated and used to verify the effectiveness of the derived restoration method. Finally, a set of laboratory vibration tests on a clamped-supported beam, monitored with UAV vision, fixed camera, and accelerometers were conducted to verify the method. The results showed that the proposed method can identify modal parameters with high measurement point density and high accuracy.The authors are grateful for the financial support from the Key Scientific and Technological Research Projects of Henan Province, China (Grant No. 222102320006) and the Open Fund Project of Zhejiang Engineering Center of Road and Bridge Intelligent Operation and Maintenance Technology (Grant No. 202501G).Peer ReviewedPostprint (published version
Effect of pressure and supply frequency on the mean free path and mean energy of corona-generated electrons in air
No full textCorona discharges have many industrial applications, but in high-voltage systems, they often create unwanted effects that need to be addressed. Therefore, design engineers need simple tools to control their effects and, when they do occur, to determine the potential impact. Since the mean free path and mean energy of corona-generated electrons are key indicators of plasma characteristics and the efficiency of various processes occurring within the discharge, these parameters provide important information about the intensity and performance of the discharge. In this work, an easy-to-apply method is developed to determine the mean free path and mean energy of coronagenerated electrons once the local electric field is known, which can be determined from finite element method simulations. The mean electron free path and mean electron energy can then be determined from published experimental data relating the total electron collision cross sections with air molecules to the mean electron energy and the mean electron energy to the reduced electric field. The method presented here is applied to experimental data obtained at different supply frequencies and different pressures typical of aircraft systems, which, due to the imperative need for electrification, are exposed to a higher risk of discharges due to the higher operating voltages, higher supply frequencies, and low pressure environment.This project received funding from grant PID2023-147016OB-I00, by MICIU/AEI/10.13039/501100011033/ and by ERDF “A way of making Europe,” by the European Union and from the Agència de Gestió d’Ajuts Universitaris i de Recerca-AGAUR (2021 SGR 00392).Peer ReviewedPostprint (published version
An a priori error analysis of a thermoelastic problem with history dependence on the mechanical and thermal components
No full textHere, we provide an a priori error analysis of a thermoelastic problem, with the Moore–Gibson–Thompson (MGT) equation, where the history dependence is assumed on both the mechanical and thermal parts. An existence and uniqueness result, and the exponential stability, are recalled. Then, a fully discrete approximation is introduced by using the finite element method and the implicit Euler scheme, to approximate the spatial variable and the time derivatives, respectively. A discrete stability property is proved and a main a priori error estimates result is obtained. The linear convergence of the approximations is deduced under suitable regularity conditions. Finally, we perform some one- and two-dimensional simulations to show the accuracy of the algorithm, the exponential decay of the discrete energy and the behavior of the solution.This paper is part of the project “Qualitative and numerical analyses of some thermomechanical problems (ACUANUTER)” (Ref. PID2024-156827NB-I00), which is currently under evaluation by the Spanish Ministry of Science, Innovation and UniversitiesPeer ReviewedPostprint (published version
A semigroup approach to a linear elastostatic problem in a semi-infinite strip
No full textWe describe via semigroup techniques the space decaying solutions to the system modeling an isotropic and homogeneous elastostatic semi-infinite band, both in the isothermal case and when thermal effects are present. The semigroup approach allows us to transfer some properties that typically occur in evolution problems to our model, such as analyticity and exponential decay at infinity. These results are closely related to the Saint-Venant principle. We conclude the article by recalling some consequences of the analyticity of the semigroup. The resulting properties are rather innovative compared to the usual results in the literature concerning the spatial decay of solutions.Peer ReviewedPostprint (published version
Finite strain thermoelasticity and the Third Law of thermodynamics
No full textThis paper shows that commonly used large strain thermoelastic models in which the specific heat coefficient is constant or, at most, changes with temperature, are incompatible with the Third Law of thermodynamics, namely, that “entropy should be zero at the Kelvin state, that is, absolute zero temperature”. In particular, it will be shown that the Third Law implies that the specific heat coefficient must vary with deformation for the coupling between mechanical and thermal effects to take place. In line with this result, a simple analytical constitutive model consistent with the Third Law will be proposed. The model will be based on a multiplicative decomposition of the specific heat into a deformation dependent part and a temperature dependent component. The resulting thermoelastic model complies with the Third Law and, in addition, the necessary convexity conditions that ensure the existence of real wave speeds. It can replicate existing entropic elasticity models for rubber, describe melting and softening behaviour, and converge to the classical relationships for linear thermoelasticity in the small strain regime.The authors acknowledge funding received from grants PID2022-141957OB-C21 and PID2022-141957OA-C22 financed by MCIN/AEI /10.13039/501100011033/ FEDER, UE. A. J. Gil wishes to acknowledge the support provided by the Defence, Science and Technology Laboratory (Dstl) and The Leverhulme Trust Foundation (UK) through a Leverhulme Fellowship.Peer ReviewedPostprint (published version
DASPack: controlled data compression for distributed acoustic sensing
No full textWe present DASPack, a high-performance, open-source compression tool specifically de- signed for distributed acoustic sensing (DAS) data. As DAS becomes a key technology for real-time, high-density and long-range monitoring in fields such as geophysics, infrastructure surveillance and environmental sensing, the volume of collected data is rapidly increasing. Large-scale DAS deployments already generate hundreds of terabytes and are expected to increase in the coming years, making long-term storage a major challenge. Despite this urgent need, few compression methods have proven to be both practical and scalable in real-world scenarios. DASPack is a fully operational solution that consistently outperforms existing tech- niques for DAS data. It enables both controlled lossy and lossless compression by allowing users to choose the maximum absolute difference per datum between the original and com- pressed data. The compression pipeline combines wavelet transforms, linear predictive coding, and entropy coding to optimise efficiency. Our method achieves up to 3 × file size reductions for strain and strain rate data in lossless mode across diverse data sets. In lossy mode, compres- sion improves to 6 × with near-perfect signal fidelity, and up to 10 × is reached with acceptable signal degradation. It delivers fast throughput (100–200 MB s −1 using a single-thread and up to 750 MB s −1 using 8-threads), enabling real-time deployment even under high data rates. We validated its performance on 15 data sets from a variety of acquisition environments, demon- strating its speed, robustness and broad applicability. DASPack provides a practical foundation for long-term, sustainable DAS data management in large-scale monitoring networks.This work was supported in part by the ‘Severo Ochoa Centre of Excellence’ program (CEX2019-000928-S-21-2); the European Union NextGenerationEU/PRTR Program under projects PSI (PLEC2021-007875), TREMORS (CPP2021-008869), the CSIC contract SAFE with Telxius Cable España, S. L. (ref. 20233069) and the Spanish Ministry of Science, Innovation and Universities under DeeLight project (PID2020-117142GB-I00).Peer ReviewedPostprint (published version
A machine learning approach for EVCS integration in distribution network based on optimal investment actions
No full textAchieving the European climate targets for 2050 requires the large-scale integration of electric vehicles and public Electric Vehicle Charging Stations (EVCS). This energy transition requires the reinforcement planning of the electrical infrastructure to ensure the quality of supply for the next years. Distribution operators need efficient planning tools able to evaluate reinforcement actions for each new connection of EVCS in MV distribution networks. This paper presents a Machine Learning (ML) approach for optimal EVCS allocation based on the Mixed Integer Second Order Cone Programming (MISOCP) formulation. First, a formulation for distribution network operational planning based on an MISOCP model is presented. This formulation aims to minimize total investments while considering operational constraints, the branch flow model, passive and active decision variables. Secondly, a large set of expansion scenarios are solved based on a sensitivity analysis and then used as training data for supervised learning models, allowing them to learn the mapping between scenario inputs and optimal investment outcomes. The strength of the trained model is the fast and accurate predictions of optimal reinforcement actions and investments from unseen new EVCS expansion scenarios, opening new opportunities for efficient large-scale EVCS assessments. Results on a 124-bus MV Network show that decision tree-based models can predict reinforcement actions and investments with a deviation of less than 1%.This publication is part of the I+D+i project ATLAS with reference PID2021-128101OB-I00 funded by MCIN/AEI/10.1 3039/501100011033.Peer ReviewedPostprint (published version