Scipedia
Not a member yet
33380 research outputs found
Sort by
FEM modelling of the mechanical phenomena at the mesoscopic scale within the cross section of a synthetic subrope for the mooring lines of floating offshore wind turbines.
No full text
Entropy-Guided k-Core Pruning Balancing Redundancy Reduction and Information Preservation for Efficient CNN Compression
No full textConvolutional Neural Networks (CNNs) are widely used in computer vision, but their massive computational cost and parameter redundancy hinder deployment on resource-constrained devices (e.g., edge terminals). Existing filter pruning methods often struggle to balance two critical goals: aggressive redundancy reduction and effective preservation of taskcritical information—either leading to excessive accuracy loss or insufficient compression. To address this challenge, we are the first to jointly exploit k-core decomposition and information entropy in a unified pruning criterion, and we instantiate this idea in a novel graph–entropy collaborative framework that achieves Pareto-optimal compression-accuracy trade-offs. The key steps are as follows: First, we use perceptual hashing (pHash) to calculate the similarity of output feature maps between filters, then model each filter as a node in an undirected graph—edges are established only when filter similarity exceeds a predefined threshold, forming a “redundancy graph” that quantifies inter-filter redundancy. Second, kcore decomposition is applied to this graph to identify high-order redundant substructures, which helps locate redundant filters at the structural level. Finally, information entropy is introduced to evaluate the “informational value” of each node (filter) in the k-core: only filters with low redundancy and high information content are retained, ensuring minimal loss of critical features. Extensive experiments are conducted on CIFAR10 and CIFAR-100 datasets, using representative CNN architectures (VGGNet-16, ResNet-56/110, DenseNet-40). Specifically, VGGNet-16 achieves a 65.8% reduction in floating point operations (FLOPs) and an 88.8% reduction in parameters while experiencing only a 1.24% decrease in Top-1 accuracy. ResNet-56 attains a 50.1% reduction in FLOPs with a nearly imperceptible accuracy loss of 0.03%, markedly surpassing the Fire together wire together (FTWT) method which reduces FLOPs by 54% at the cost of a 1.38% accuracy decline. DenseNet-40 accomplishes a 76.5% FLOPs reduction with a 1.55% accuracy decrease, demonstrating the method’s strong applicability for high-intensity compression of densely connected networks. Furthermore, the method’s scalability is validated on the large-scale ImageNet dataset with ResNet-50, where it achieves a 73.65% FLOPs reduction with competitive accuracy, underscoring its practicality for real-world applications. These outcomes collectively affirm the effectiveness and broad applicability of the proposed graphentropy collaborative pruning framework
Cuantificación de incertidumbre sobre parámetros en modelos no lineales
No full textEn este trabajo se estudia la cuantificación de incertidumbre en parámetros de modelos no lineales mediante el enfoque bayesiano. Se parte del planteamiento clásico de problemas inversos, en los cuales los parámetros del modelo deben inferirse a partir de observaciones ruidosas y de un modelo directo formulado como un sistema de ecuaciones diferenciales. Dado que estos problemas suelen estar mal planteados, se introduce la inferencia bayesiana como estrategia de regularización, permitiendo incorporar información a priori y actualizarla con datos mediante la distribución a posteriori. Se presentan los fundamentos teóricos del enfoque bayesiano, así como su aplicación al caso particular del modelo de crecimiento logístico, destacando el uso de métodos computacionales para aproximar las distribuciones resultantes de los parámetros del modelo
Serbian energy sector in the global political landscape amid the Russia‑Ukraine war: a focus on perspectives of integration into the European Union
Get PDFThis article consists of two main parts, both in relation to Serbia’s accession to the European Union EU in relation to its energy sector: (1) Political and policy issues, and (2) Energy production, consumption and pricing. Each is heavily influenced by the Russia-Ukraine War. Political issues are primarily related to Kosovo, which unilaterally declared independence from Serbia. Regarding supply, Serbia’s energy sector is affected unevenly: while electricity production remains self-sufficient, shortages in oil and gas necessitate imports. Natural gas imports, primarily from Russia, now bypass Ukraine, aided by the new Turk Stream pipeline via Black Sea and also new EU-funded interconnector with Bulgaria. Newly introduced EU sanctions restrict Russian crude oil imports especially via maritime routes, but Russian influence still dominates domestic refineries and hydrocarbon extraction. Serbia’s heavy reliance on environmentally unfriendly lignite conflicts with EU renewable goals, but on the other hand protests against small hydropower projects for capturing rivers in pipes are frequent nowadays. Recent issues in the largest thermal power plant led to costly temporary imports. Despite challenges, Serbia navigates a complex energy landscape, balancing geopolitical realities with domestic and EU objectives while addressing environmental concerns, energy security and its national interest
CHAPTER 16: BIO-MONOMERS AND THERMOSTABLE BIO-RESINS
No full textThermosetting resins, especially epoxy resins, have historically been based on petroleum-derived monomers, posing problems of toxicity, high costs, and dependence on non-renewable resources. This chapter presents the use of natural plant sources to develop more sustainable resin precursors.
The synthesis of bio-epoxy monomers from vegetable oils, polysaccharides, lignin, polyphenols, and natural resins is currently the subject of interest in several research projects and scientific papers, in some cases reaching the level of product commercialization. Vegetable oils, such as linseed and soybean, and the transformation of saccharides into epoxy monomers are examples of explored options. Naturally occurring epoxy monomers derived from polyphenols from various plant sources can also be found, although their epoxidation requires the use of toxic compounds such as epichlorohydrin.
Several natural sources, such as natural rubber, resin acids, and lignin are examined as alternatives to synthesize epoxy resins of natural origin. Leutelin, recently identified in fruits and medicinal herbs, is highlighted as a promising compound to produce bioepoxy monomers.
Although initially the research focused mostly on the development of monomers of natural origin, research also extends to hardeners of natural origin, highlighting the synthesis of amines from vanillin and curing agents based on phenalkamines. The development of hardeners with reversible bonds, such as lignin imines, is also being explored, and the catalytic effect of hemp fibers in the curing of epoxy resins is highlighted.
The combination with traditional monomers or the development of recyclable resins and vitrimers with reversible bonds are current fields of interest for the development of these resins, especially in the context of their competitors of petrochemical origin
CHAPTER 14: REUSE OF POLYMERIZED COMPOSITE MATERIALS
No full textSpain, holding a prominent position in Europe's composites industry and notably in aviation composites, faces significant waste management challenges within its 500-strong composite company sector. As it shifts towards the Circular Economy to enhance competitiveness and resource efficiency, Spain adheres to the EU's waste hierarchy, emphasizing the critical need for waste minimization and the reuse of materials. This transition is particularly vital given the environmental impact of disposing of polymer-matrix composites, especially cured composites, with Europe generating roughly 400,000 tonnes of thermoset composite waste annually, a substantial portion of which is from Spain.
This study delves into the reuse and repurposing of polymer composites, promoting their integration within the Circular Economy to preserve material integrity and value. It showcases innovative repurposing projects in Spain and across Europe, such as transforming wind turbine blades into materials for construction, which demonstrates the feasibility of extending these materials' lifecycles. These efforts align with sustainability goals aimed at waste reduction and resource conservation. However, challenges persist, including matching waste volume and condition with market demands and scaling these practices effectively. The concept of structural re-use, turning cured composite waste into high-value, reusable products, highlights the potential of merging reuse and recycling strategies. Innovative approaches to reuse not only mitigate sustainability challenges but also foster economically viable solutions, marking a significant stride towards sustainable and efficient resource utilization in the composites sector.
 
Gestió del Canvi, Lideratge Transformacional i Intel·ligència Emocional en l'Era de les Tecnologies Emergents
No full textAquesta recerca explora com el lideratge transformacional i la intel·ligència emocional impacten en la gestió del canvi organitzatiu en el context de la Quarta Revolució Industrial. Aquesta revolució, marcada per la convergència de tecnologies digitals, físiques i biològiques, està transformant profundament les empreses, que han d'adaptar-se constantment per mantenir la competitivitat. La recerca analitza models de canvi i factors disruptius, destacant el lideratge transformacional com una eina clau per promoure el canvi cultural a les organitzacions. Aquest estil de lideratge, centrat en les persones, s’associa a la intel·ligència emocional, una habilitat essencial per gestionar resistències i fomentar un clima de confiança, innovació i creativitat. Amb una gestió proactiva del canvi i el desenvolupament d’aquestes competències, les empreses poden prosperar en un entorn incert i altament tecnològic, posicionant-se per a l’èxit en una era de transformació constant