Search
Filters
total: 2753
filtered: 2435
-
Catalog
Chosen catalog filters
displaying 1000 best results Help
Search results for: SPARSE AUTOREGRESSIVE MODELS
-
Multi-Fidelity Local Surrogate Model for Computationally Efficient Microwave Component Design Optimization
PublicationIn order to minimize the number of evaluations of high-fidelity (“fine”) model in the optimization process, to increase the optimization speed, and to improve optimal solution accuracy, a robust and computational-efficient multi-fidelity local surrogate-model optimization method is proposed. Based on the principle of response surface approximation, the proposed method exploits the multi-fidelity coarse models and polynomial interpolation...
-
Heat transfer intensification by jet impingement – numerical analysis using RANS approach
PublicationJet impingement is a method of the heat transfer enhancement applied in the engineering systems. The idea is to generate fast-flow fluid jet which impinge on the heated (or cooled) surface, causing significantly higher heat transfer rate. Although some flat surface jet impingement cases are described in the literature, the validated data is still limited. The reason is coming from the fact, that these flows are hydrodynamically...
-
Triple correlation states between in-situ tested soil parameters
PublicationWhen testing soil parameters, the measured parameter values are only approximations of the true values. The measurand is determined based on metrological uncertainties or using statistical models for analysing data. Some parameters of the soil state present strong correlations, but others do not always provide simple correspondences. Multiple correlations between geotechnical parameters can provide a new perspective regarding the...
-
MagMax: Leveraging Model Merging for Seamless Continual Learning
PublicationThis paper introduces a continual learning approach named MagMax, which utilizes model merging to enable large pre-trained models to continuously learn from new data without forgetting previously acquired knowledge. Distinct from traditional continual learning methods that aim to reduce forgetting during task training, MagMax combines sequential fine-tuning with a maximum magnitude weight selection for effective knowledge integration...
-
Sign Language Recognition Using Convolution Neural Networks
PublicationThe objective of this work was to provide an app that can automatically recognize hand gestures from the American Sign Language (ASL) on mobile devices. The app employs a model based on Convolutional Neural Network (CNN) for gesture classification. Various CNN architectures and optimization strategies suitable for devices with limited resources were examined. InceptionV3 and VGG-19 models exhibited negligibly higher accuracy than...
-
Performance‐driven modeling of compact couplers in restricted domains
PublicationFast surrogate models can play an important role in reducing the cost of EM-driven design closure of miniaturized microwave components. Unfortunately, construction of such models is challenging due to curse of dimensionality and wide range of geometry parameters that need to be included in order to make it practically useful. In this letter, a novel approach to design-oriented modeling of compact couplers is presented. Our method...
-
Thermal decomposition kinetics of dynamically vulcanized polyamide 6-acrylonitrile butadiene rubber-halloysite nanotube nanocomposites
PublicationThermally stable thermoplastic elastomer nanocomposites based on polyamide 6 (PA6), acrylonitrile butadiene rubber (NBR), and halloysite nanotubes (HNTs) were dynamically vulcanized, and their nonisothermal decomposition kinetics were examined. The Friedman, Kissinger–Akahira–Sunose (KAS), Ozawa–Wall–Flynn (FWO), and modified Coats–Redfern (m‐CR) isoconversional models were used to obtain information about the kinetics of the thermal...
-
Studying road restraint systems to develop new guidelines
PublicationThe paper will present a proposed approach to such work along with preliminary results of numerical studies for selected problems using road safety devices such as safety barriers on horizontal curves, the effect of kerbs on bridges or the location of obstacles within the barrier’s working width. The paper will discuss assumptions to a methodology of numerical models, calculations and automated processing of data to help with assessing...
-
Mechanical behaviour of the implant used in human hernia repair under physiological loads
PublicationIn laparoscopic operations of abdominal hernias some recurrences still take place, even when applying a surgical mesh. This is usually caused by a failure of the connection between the tissue and the implant. The study deals with the influence of an implant’s orientation on forces in joints, which connect the mesh to human tissues. In the paper, the implant is modelled as a membrane structure within framework of the Finite Element...
-
SAWING PROCESS AS A NEW ALTERNATIVE WAY OF DETERMINING SOME WOOD PROPERTIES
PublicationCutting forces (power) could be considered from a point of view of modern fracture mechanics. The developed cutting model, derived from fracture mechanics, includes work of separation (fracture toughness) in addition to plasticity and friction, and also dullness of the cutting edge described by the cutting edge radius. Moreover, forecasting of the shear plane angle for the cutting models, broaden possibilities of energetic effects...
-
Efficient Multi-Fidelity Design Optimization of Microwave Filters Using Adjoint Sensitivity
PublicationA simple and robust algorithm for computationally efficient design optimiza-tion of microwave filters is presented. Our approach exploits a trust-region (TR)-based algorithm that utilizes linear approximation of the filter response obtained using adjoint sensitivity. The algorithm is sequentially executed on a family of electromagnetic (EM)-simulated models of different fidelities, starting from a coarse-discretization one, and...
-
Abstract class method use on common characteristics of UE regions for better road safety management in these regions.
PublicationRoad Safety is the significant problem requiring still a lot of attention. Despite lots of actions for road fatalities reduction, in many countries or regions situation doesn't improve as good as in the other cases. Even if these actions are very similar. It is so difficult issue, that the answer why is that happens hasn't been found yet. In this paper authors focused on the road network of European regions, classified as NUTS...
-
Low-Cost Multi-Objective Optimization Yagi-Uda Antenna in Multi-Dimensional Parameter Space
PublicationA surrogate-based technique for fast multi-objective optimization of a multi-parameter planar Yagi-Uda antenna structure is presented. The proposed method utilizes response surface approximation (RSA) models constructed using training samples obtained from evaluation of the low-fidelity antenna model. Utilization of the RSA models allowsfor fast determination of the best possible trade-offs between conflicting objectives in multi-objective...
-
Reduced-cost surrogate modelling of compact microwave components by two-level kriging interpolation
PublicationFull-wave electromagnetic (EM) analysis is a versatile tool for evaluating the performance of high-frequency components. Its potential drawback is its high computational cost, inhibiting the execution of EM-driven tasks requiring massive simulations. The applicability of equivalent network models is limited owing to the topological complexity of compact microstrip components because of EM cross-coupling effects. Development of...
-
How to Sort Them? A Network for LEGO Bricks Classification
PublicationLEGO bricks are highly popular due to the ability to build almost any type of creation. This is possible thanks to availability of multiple shapes and colors of the bricks. For the smooth build process the bricks need to properly sorted and arranged. In our work we aim at creating an automated LEGO bricks sorter. With over 3700 different LEGO parts bricks classification has to be done with deep neural networks. The question arises...
-
Wind Conditions at Pedestrian Level in Different Types of Residential Urban Development for a High Degree of Land Use Efficiency
PublicationThe paper concerns wind conditions around urban building development at the pedestrian level. The article aims to identify aerodynamic phenomena around three types of multi-family housing developmentswith different forms and the same urban parameters of building development intensity (high intensity was taken into account). The aim of the research was mainly to achieve qualitative results that would lead to understanding fundamental...
-
Non-volatile molecular composition and discrimination of single grape white of chardonnay, riesling, sauvignon blanc and silvaner using untargeted GC–MS analysis
PublicationThis study developed and applied a GC–MS method aiming at molecular fingerprinting of 120 commercial single grape white wines (Chardonnay, Riesling, Sauvignon Blanc and Silvaner) for possible authentication according to grape variety. The method allowed detection of 372 peaks and tentative identification of 146 metabolites including alcohols, organic acids, esters, amino acids and sugars. The grape variety effect explained 8.3%...
-
Brownian Motion in Optical Tweezers, a Comparison between MD Simulations and Experimental Data in the Ballistic Regime
PublicationThe four most popular water models in molecular dynamics were studied in large-scale simulations of Brownian motion of colloidal particles in optical tweezers and then compared with experimental measurements in the same time scale. We present the most direct comparison of colloidal polystyrene particle diffusion in molecular dynamics simulations and experimental data on the same time scales in the ballistic regime. The four most...
-
MARSTRUCT benchmark study on nonlinear FE simulation of an experiment of an indenter impact with a ship side-shell structure
PublicationThis paper presents a benchmark study on collision simulations that was initiated by the MARSTRUCT Virtual Institute. The objective was to compare assumptions, finite element models, modelling techniques and experiences between established researchers within the field. Fifteen research groups world-wide participated in the study. An experiment involving a rigid indenter penetrating a ship-like side structure was used as the case...
-
Food Classification from Images Using a Neural Network Based Approach with NVIDIA Volta and Pascal GPUs
PublicationIn the paper we investigate the problem of food classification from images, for the Food-101 dataset extended with 31 additional food classes from Polish cuisine. We adopted transfer learning and firstly measured training times for models such as MobileNet, MobileNetV2, ResNet50, ResNet50V2, ResNet101, ResNet101V2, InceptionV3, InceptionResNetV2, Xception, NasNetMobile and DenseNet, for systems with NVIDIA Tesla V100 (Volta) and...
-
Economics of credit scoring management
PublicationCredit scoring models constitute an inevitable element of modern risk and profitability management in retail financial lending institutions. Quality,or separation power of a credit scoring model is usually assessed with the Gini coefficient. Generally, the higher Gini coefficient the better, as in this way a bank can increase number of good customers and/or reject more bad applicants. In...
-
Multi-factor fuzzy sets decision system forecasting consumer insolvency risk
PublicationThe objective of this study is to develop a multi-factor decision system predicting insolvency risk for natural persons with the use of fuzzy sets. Considering that the financial situation of households is affected by various endogenous and exogenous factors, the main assumption of this study is that the system for predicting financial difficulties should not be limited to the use of only a few financial variables concerning consumers,...
-
High-speed multi-stage gas-steam turbine with flow bleeding in a novel thermodynamic cycle for decarbonizing power generation
PublicationIn the global pursuit of sustainable energy and reduced carbon footprints, advances in power generation techniques play a crucial role, not only in meeting the ever-increasing energy demands but also in ensuring that environmental standards are maintained and that the health of our planet is prioritized for future generations. In the ongoing quest for sustainable energy solutions, novel high-speed multi-stage gas-steam turbine...
-
Novel 2-(2-alkylthiobenzenesulfonyl)-3-(phenylprop-2-ynylideneamino)guanidine derivatives as potent anticancer agents – Synthesis, molecular structure, QSAR studies and metabolic stability
PublicationA series of new 2-(2-alkylthiobenzenesulfonyl)-3-(phenylprop-2-ynylideneamino)guanidine derivatives have been synthesized and evaluated in vitro by MTT assays for their antiproliferative activity against cell lines of colon cancer HCT-116, cervical cancer HeLa and breast cancer MCF-7. The obtained results indicated that these compounds display prominent cytotoxic effect. The best anticancer properties have been observed for derivatives...
-
Settlement Networks in Polish Spatial Development Regional Plans
PublicationIn 1999, ten years after the great political changes in Poland, 16 self-governed regions (in Polish: voivodeship) were created. According to Polish law, voivodeship spatial development plans, or regional plans in short, determine basic elements of the settlement network. No detailed regulations indicate the specific elements of the settlement network or what features of these elements should be determined. For this reason, centres...
-
Towards an experience based collective computational intelligence for manufacturing
PublicationKnowledge based support can play a vital role not only in the new fast emerging information and communication technology based industry, but also in traditional manufacturing. In this regard, several domain specific research endeavors have taken place in the past with limited success. Thus, there is a need to develop a flexible domain independent mechanism to capture, store, reuse, and share manufacturing knowledge. Consequently,...
-
In silico modelling for predicting the cationic hydrophobicity and cytotoxicity of ionic liquids towards the Leukemia rat cell line, Vibrio fischeri and Scenedesmus vacuolatus based on molecular interaction potentials of ions
PublicationIn this study we present prediction models for estimating in silico the cationic hydrophobicity and the cytotoxicity (log [1/EC50]) of ionic liquids (ILs) towards the Leukemia rat cell line (IPC-81), the marine bacterium Vibrio fischeri and the limnic green algae Scenedesmus vacuolatus using linear free energy relationship (LFER) descriptors computed by COSMO calculations. The LFER descriptors used for the prediction model (i.e....
-
Model-based identification of the dominant N2O emission pathway in a full-scale activated sludge system
PublicationActivated sludge models (ASMs), extended with an N2O emission module, are powerful tools to describe the operation of full-scale wastewater treatment plants (WWTPs). Specifically, such models can investigate the most contributive N2O production pathways and guide towards N2O and carbon footprint (CF) mitigation measures. A common practice is to develop and validate models using data from a single WWTP. In this study, a successfully...
-
Closed Form Constraint Equations Used to Express Frictionless Slip of Multibody Systems Attached to Finite Elements—Application to a Contact between a Double Pendulum and a Beam
PublicationThis paper focuses on the numerical modeling of the dynamics of mechanical systems. Robots that can inspect high-voltage lines inspired this research. Their control systems must anticipate potential grab positions appropriately. We intend to formulate equations dedicated to the numerical description of the robot/cable contact. The investigated problem is not straightforward, since parts of the modeled systems are numerically inhomogeneous....
-
An algorithm for selecting a machine learning method for predicting nitrous oxide emissions in municipal wastewater treatment plants
PublicationThis study presents an advanced algorithm for selecting machine learning (ML) models for nitrous oxide (N2O) emission prediction in wastewater treatment plants (WWTPs) employing the activated sludge process. The examined ML models comprised multivariate adaptive regression spline (MARS), support vector machines (SVM), and extreme gradient boosting (XGboost). The study explores the concept that involves new criteria to select the...
-
Experimental and computational study on mechanical behaviour of carpentry corner log joints
PublicationThis work concerns experimental and numerical research on carpentry joints used in historic wooden buildings in southeastern Poland and western Ukraine. These structures are mainly sacred buildings, and the types of corner log joints characteristic of this region are primarily saddle-notch and dovetail joints; thus, these two types of joints were analysed in this study. The modelling of historic timber structures is a complex...
-
Implementing decisional trust: a first approach for smart reliable systems
PublicationIn this article, we introduce the necessary elements that must be integrated in order to achieve a decisional technology that is trustworthy. Thus, we refer to such technology as decisional trust. For us, decisional trust can be achieved through the use of elements such as the decisional DNA, reflexive ontologies, and security models; and therefore, we present in this article a framework that was used for the implementation of...
-
Stress analysis of the bridge hangers in terms of the fatigue verification
PublicationThe paper presents the results of numerical stress analysis of the bridge rod hangers. Two types of hangers were taken into account, i.e. the bolted hanger and the welded hanger. A detailed FEM shell models of the connection area between the hanger and the bridge span have been developed. Local stress concentration effects have been investigated in terms of fatigue verification.
-
IT Business Standards as an Ontology Domain
PublicationThe aim of this paper is to report a selection of Semantic Web aspects pertaining to ontology development activities in the domain of the IT business standard (TOGAF 9) such as formulating competency questions, conceptualization of the domain, resolution of the source knowledge deficiency and applying common design patterns in the OWL formalization. Authors also try to determine target groups that may benefit from such ontology...
-
Comparison of buckling resistance of columns modelled by beam and shell elements using non-linear analysis
PublicationThe aim of the paper is to investigate the stability process in the axially compressed columns modelled by beam and shell elements using static and dynamic finite element analysis by taking both the geometric and material non-linearity into account. The perfect columns and columns with geometric imperfections were analysed. The differences between the results of static and dynamic analyses in shell and beam models were discussed.
-
Raz jeszcze o obciążeniu hydrodynamicznym falochronu pionowościennego falą stojącą
PublicationPrzedstawienie istniejących modeli matematycznych opisujących rozkłady ciśnienia hydrodynamicznego na falochron pionowościenny w wyniku oddziaływania fali stojącej. Analiza nieciągłości ciśnienia. Jakościowe porównanie metod obliczeniowych dla wybranych warunków wodno-falowych.
-
Przegląd metod wyznaczania parcia mieszanki betonowej na deskowanie
PublicationW pracy poruszono tematykę wyznaczania parcia mieszanki betonowej na deskowanie. Zebrano i opisano wybrane cztery modele stosowane do wyznaczania rozkładu ciśnienia poziomego świeżej mieszanki betonowej oddziaływującej na deskowanie. Dokonano krótkiego przeglądu literatury z zakresu omawianej tematyki parcia mieszanki betonowej na deskowanie.
-
PROBLEMY RACJONALNEGO ZASTOSOWANIA SYSTEMÓW DIAGNOZUJĄCYCH DO PODEJMOWANIA DECYZJI W FAZIE EKSPLOATACJI URZĄDZEŃ ENERGETYCZNYCH Z UWZGLĘDNIENIEM ZAGADNIEŃ KWANTOWYCH
PublicationW rozdziale przedstawiono najistotniejsze problemy pojawiające się w fazie projektowania, wytwarzania i eksploatacji systemów diagnozujących (SDG) urządzeń energetycznych (będących systemami diagnozowanymi – SDN), w przypadku przysposabiania ich do opracowania finalnej diagnozy eksploatacyjnej. Przestawiono opisową i formalną interpretację pojęcia systemu diagnozującego (SDG). Zaproponowano modele procesu diagnozowania urządzeń...
-
Skuteczne prognozowanie krótkoterminowe mocy farm wiatrowych
PublicationPrognozowanie mocy wytwórczej konkretnej farmy wiatrowej (FW) w horyzoncie 24-godzinnymwymaga zarówno wiarygodnej prognozy wietrzności, jak i narzędzi wspomagających. Narzędzie to jest dedykowanym modelem mocy farmy. Model powinien uwzględniać nie tylko ogólne zasady przetwarzania energii wiatru na energię mechaniczną, ale także cechy szczególnekonkretnej farmy. Liczba czynników wpływających na moc farmy jest duża i dokładna prognozamocy,...
-
Coarse-grained simulation - an efficient approach for studying motions of large proteins
PublicationOne of the most important challenges in performing Molecular Dynamics (MD) simulations of large protein complexes is to accommodate the model accuracy and the simulation timescale. Hitherto, for the most relevant dynamics of protein aggregates in an explicit aqueous environment, the timescale reachable for the all-atoms simulations is of hundreds of nanoseconds. This range is four to six orders of magnitude smaller than processes...
-
Determination of the shape of the CFGFT cylindrical column based on laboratory tests
PublicationAnalyses were carried out on glass-fibre-reinforced polymer tube columns with reference to laboratory tests. The angles of the glass fibre beams were 20◦, 55◦ and 85◦. The study employed non-classical operational calculus. Various modulated harmonic signal shapes were considered for columns and tubes at buckling. The buckling loads were assessed and compared for different models.
-
Blackout initial disturbances in electric Power System searching
PublicationThe paper contains short survey of the blackouts that took place in the last few years. The sources and initial disturbances are discussed. The typical scenario of the disturbance, leading to blackout in the electrical power system are presented.Next, models of a power system eligible to considered topic are discussed. Finally the algorithm of the blackout initial disturbances in power system searching is presented.The consideration...
-
Application of Reverse Engineering Technology in Part Design for Shipbuilding Industry
PublicationIn the shipbuilding industry, it is difficult to create CAD models of existing or prototype parts, especially with many freeform surfaces. The paper presents the creation of the CAD 3D model of a shipbuilding component with the application of the reverse engineering technology. Based on the data obtained from the digitization process, the component is reconstructed in point cloud processing programs and the CAD model is created....
-
Design of precise control systems of industrial plants
PublicationA method for developing precision control systems in industry has been created based on a set of models that take into account the nonlinearity of the reactor model, the nonlinearities of the hydrodynamic regime and the accuracy of the control elements, the nonlinear characteristics and accuracy of the actuators, and the precision of the sensors used. A multidimensional precision control system has been developed for a continuous...
-
Visualization of Spatial Issues in Adobe Acrobat 3D PDF
PublicationThe ability to capture and embed model information in a 3D PDF file and subsequently share it for free is highly needed in the education of descriptive geometry. The described technology and in particular inserting 3D models into the other files can improve the existing teaching materials. The strongest point of 3D PDF technology is that the model can be visualized and manipulated in multiple ways by viewers which enhance the...
-
Robust output prediction of differential – algebraic systems – application to drinking water distribution system
PublicationThe paper presents the recursive robust output variable prediction algorithm, applicable for systems described in the form of nonlinear algebraic-differential equations. The algorithm bases on the uncertainty interval description, the system model, and the measurements. To improve the algorithm efficiency, nonlinear system models are linearised along the nominal trajectory. The effectiveness of the algorithm is demonstrated on...
-
PHYSICAL MODEL OF ENERGY-RELATED PROCESSES IN A ROTATING MECHANICAL SYSTEM FOR THE PURPOSES OF DIAGNOSIS OF TRANSMISSION SHAFTS
PublicationIn the present article, selected problems connected with identification of slowly changing energy-related processes accompanying high cyclic fatigue of construction materials of simple mechanical systems, with special consideration of transmission shafts, are mentioned for diagnostic purposes. The article also presents physical models of object and processes that in a simplified way reflect mechanical energy transmission from transmission...
-
Modelling of Geared Multi-Rotor System
PublicationIn the paper the method of modelling a speed-varying geared rotor system is presented. The proposed approach enables us to obtain an accurate low-order lumped parameter representation of the investigated system. The final model consists of reduced modal models of an undamped beam/torsional shaft system as well as a spatially lumped model of other linear and nonlinear phenomena including gear mesh interaction.
-
Process of Medical Dataset Construction for Machine Learning-Multifield Study and Guidelines
PublicationThe acquisition of high-quality data and annotations is essential for the training of efficient machine learning algorithms, while being an expensive and time-consuming process. Although the process of data processing and training and testing of machine learning models is well studied and considered in the literature, the actual procedures of obtaining data and their annotations in collaboration with physicians are in most cases...
-
Divide and not forget: Ensemble of selectively trained experts in Continual Learning
PublicationClass-incremental learning is becoming more popular as it helps models widen their applicability while not forgetting what they already know. A trend in this area is to use a mixture-of-expert technique, where different models work together to solve the task. However, the experts are usually trained all at once using whole task data, which makes them all prone to forgetting and increasing computational burden. To address this limitation,...