Filters
total: 7822
-
Catalog
- Publications 6375 available results
- Journals 249 available results
- Conferences 107 available results
- People 338 available results
- Projects 18 available results
- Laboratories 3 available results
- e-Learning Courses 306 available results
- Events 11 available results
- Open Research Data 415 available results
displaying 1000 best results Help
Search results for: SURROGATE MODELING , ANTENNA DESIGN , DOMAIN CONFINEMENT , NESTED KRIGING , DEEP NEURAL NETWORKS
-
Applications of neural networks and perceptual masking to audio restoration
PublicationOmówiono zastosowania algorytmów uczących się w dziedzinie rekonstruowania nagrań fonicznych. Szczególną uwagę zwrócono na zastosowanie sztucznych sieci neuronowych do usuwania zakłócających impulsów. Ponadto opisano zastosowanie inteligentnego algorytmu decyzyjnego do sterowania maskowaniem perceptualnym w celu redukowania szumu.
-
Low-Cost and Precise Automated Re-Design of Antenna Structures Using Interleaved Geometry Scaling and Gradient-Based Optimization
PublicationDesign of contemporary antennas is an intricate endeavor involving multiple stages, among others, tuning of geometry parameters. In particular, re-designing antennas to different operating frequencies, makes parametric optimization imperative to ensure the best achievable system performance. If the center frequency at the current design is distant from the target one, local tuning methods generally fail, whereas global algorithms...
-
Resource constrained neural network training
PublicationModern applications of neural-network-based AI solutions tend to move from datacenter backends to low-power edge devices. Environmental, computational, and power constraints are inevitable consequences of such a shift. Limiting the bit count of neural network parameters proved to be a valid technique for speeding up and increasing efficiency of the inference process. Hence, it is understandable that a similar approach is gaining...
-
From Linear Classifier to Convolutional Neural Network for Hand Pose Recognition
PublicationRecently gathered image datasets and the new capabilities of high-performance computing systems have allowed developing new artificial neural network models and training algorithms. Using the new machine learning models, computer vision tasks can be accomplished based on the raw values of image pixels instead of specific features. The principle of operation of deep neural networks resembles more and more what we believe to be happening...
-
Design and Characterization of a Planar Structure Wideband Millimeter-Wave Antenna with Wide Beamwidth for Wearable off-body Communication Applications
PublicationThis letter presents the design of a planar single-layer wideband antenna featuring wide beamwidth has well as high and stable in-band gain. The proposed antenna is a planar monopole fed by a bottom-grounded coplanar waveguide to realize wide beamwidth in both the xz- and yz-planes. Simultaneous optimization of all adjustable antenna parameters, carried out at the full-wave electromagnetic simulation level. The constructive interference...
-
A Conformal Circularly Polarized Series-Fed Microstrip Antenna Array Design
PublicationA conformal circularly polarized series-fed microstrip array design for broadside radiation is presented. The array aperture under design is conformal to a cylindrical surface of a given radius. The approach we present primarily addresses focusing of the circularly polarized major lobe of the conformal array by proper dimensioning of the aperture spacings. The proposed analytical models yield the values of the element spacings...
-
Low-Cost Surrogate Models for Microwave Filters
PublicationA novel low-cost kriging-based multivariable parametric macromodeling technique for microwave filters is presented. Kriging is used to model both the residues and poles of a microwave filter's reflection coefficient, and the zeros of the transmission coefficient. The proposed residue-pole-zero (RPZ) technique is demonstrated to efficiently model a high dimensional (8D) microwave filter with pseudoelliptic characteristics.
-
Modeling the economic dependence between town development policy and increasing energy effectiveness with neural networks. Case study: The town of Zielona Góra
Publication -
Tolerance Optimization of Antenna Structures by Means of Response Feature Surrogates
PublicationFabrication tolerances and other types of uncertainties, e.g., the lack of precise knowledge of material parameters, have detrimental effects on electrical and field performance of antenna systems. In the case of input characteristics these are particularly noticeable for narrow- and multi-band antennas where deviations of geometry parameters from their nominal values lead to frequency shifts of the operating frequency bands. Improving...
-
High-Efficacy Global Optimization of Antenna Structures by Means of Simplex-Based Predictors
PublicationDesign of modern antenna systems has become highly dependent on computational tools, especially full-wave electromagnetic (EM) simulation models. EM analysis is capable of yielding accurate representation of antenna characteristics at the expense of considerable evaluation time. Consequently, execution of simulation-driven design procedures (optimization, statistical analysis, multi-criterial design) is severely hindered by the...
-
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...
-
Improved-Efficacy EM-Driven Optimization of Antenna Structures Using Adaptive Design Specifications and Variable-Resolution Models
PublicationOptimization-driven parameter tuning is an essential step in the design of antenna systems. Although in many cases it is still conducted through parametric studies, rigorous numerical methods become a necessity if truly optimum designs are sought for, and the problem intricacies (number of variables, multiple goals, constraints) make the interactive approaches insufficient. The two practical considerations of electromagnetic (EM)-driven...
-
Classifying Emotions in Film Music - A Deep Learning Approach
PublicationThe paper presents an application for automatically classifying emotions in film music. A model of emotions is proposed, which is also associated with colors. The model created has nine emotional states, to which colors are assigned according to the color theory in film. Subjective tests are carried out to check the correctness of the assumptions behind the adopted emotion model. For that purpose, a statistical analysis of the...
-
Cost-Efficient Multi-Objective Design of Miniaturized Microwave Circuits Using Machine Learning and Artificial Neural Network
PublicationDesigning microwave components involves managing multiple objectives such as center frequencies, impedance matching, and size reduction for miniaturized structures. Traditional multi-objective optimization (MO) approaches heavily rely on computationally expensive population-based methods, especially when exe-cuted with full-wave electromagnetic (EM) analysis to guarantee reliability. This paper introduces a novel and cost-effective...
-
Age Prediction from Low Resolution, Dual-Energy X-ray Images Using Convolutional Neural Networks
PublicationAge prediction from X-rays is an interesting research topic important for clinical applications such as biological maturity assessment. It is also useful in many other practical applications, including sports or forensic investigations for age verification purposes. Research on these issues is usually carried out using high-resolution X-ray scans of parts of the body, such as images of the hands or images of the chest. In this...
-
An interactive system for remote modeling and design validation of hybrid photovoltaic systems
PublicationIn the paper a multi-functional demonstrator of the interactive system designed to modeling, monitoring and validation of hybrid photovoltaic systems assisted by fuel cells and thermoelectric generators is presented. The purpose of this paper is to report the system solution expressed in the form of a block diagram. Technical parameters of demonstrator components such as: silicon photovoltaic modules, fuel cells, thermoelectric...
-
Inverse modeling for fast design optimization of small-size rat-race couplers incorporating compact cells
PublicationIn the paper, a framework for computationally-efficient design optimization of compact rat-race couplers (RRCs) is discussed. A class of hybrid RRCs with variable operating conditions is investigated, whose size reduction is obtained by replacing ordinary transmission lines with compact microstrip resonant cells (CMRCs). Our approach employs a bottom-up design strategy leading to the development of compact RRCs through rapid design...
-
Inverse surrogate models for fast geometry scaling of miniaturized dual-band couplers
PublicationRe-design of microwave structures for various sets of performance specifications is a challenging task, particularly for compact components where considerable electromagnetic (EM) cross-couplings make the relationships between geometry parameters and the structure responses complex. Here, we address geometry scaling of miniaturized dual-band couplers by means of inverse surrogate modeling. Our approach allows for fast estimation...
-
JOURNAL OF MOLECULAR MODELING
Journals -
Mathematical modeling and prediction of pit to crack transition under cyclic thermal load using artificial neural network
PublicationThe formation of pitting is a major problem in most metals, which is caused by extremely localized corrosion that creates small holes in metal and subsequently, it changes into cracks under mechanical load, thermo-mechanical stress, and corrosion process factors. This research aims to study pit to crack transition phenomenon of steel boiler heat tubes under cyclic thermal load, and mathematical modeling...
-
Neural network training with limited precision and asymmetric exponent
PublicationAlong with an extremely increasing number of mobile devices, sensors and other smart utilities, an unprecedented growth of data can be observed in today’s world. In order to address multiple challenges facing the big data domain, machine learning techniques are often leveraged for data analysis, filtering and classification. Wide usage of artificial intelligence with large amounts of data creates growing demand not only for storage...
-
Fast Antenna Optimization Using Gradient Monitoring and Variable-Fidelity EM Models
PublicationAccelerated simulation-driven design optimization of antenna structures is proposed. Variable-fidelity electromagnetic (EM) analysis is used as well as the trust-region framework with limited sensitivity updates. The latter are controlled by monitoring the changes of the antenna response gradients. Our methodology is verified using three compact wideband antennas. Comprehensive benchmarking demonstrates its superiority over both...
-
Piotr Rajchowski dr inż.
PeoplePiotr Rajchowski (Member, IEEE) was born in Poland, in 1989. He received the E.Eng., M.Sc., and Ph.D. degrees in radio communication from the Gdańsk University of Technology (Gdańsk Tech), Poland, in 2012, 2013, and 2017, respectively. Since 2013, he has been working at the Department of Radiocommunication Systems and Networks, Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, as a IT...
-
An interactive system for remote modeling and design validation of hybrid photovoltaic systems
PublicationAbstract: A multi-functional demonstrator of the interactive system is presented. The demostrator enables modeling, monitoring and design validation of hybrid photovoltaic systems assisted by fuel cells and thermoelectric generators, as well as experimentation and scientific research. A block diagram of the system is presented and the selection of its components is discussed. Availability of the system via Ethernet or GSM...
-
The methods of design reliable distributed networks
PublicationOpisano dwie metody projektowania niezawodnych sieci rozległych. W pierwszej metodzie przyjęto, że koszt konstrukcji każdego połączenia jest jednakowy, w drugiej koszty opisuje zadana macierz kosztów. Jako parametr niezawodnościowy przyjęto liczbę niezależnych dróg pomiędzy węzłami sieci. Pierwsza z metod jest metodą dokładną, druga heurystyczną. Zdefiniowano i oceniono błąd metody heurystycznej. Opublikowano wyniki otrzymane...
-
Piotr Szczuko dr hab. inż.
PeoplePiotr Szczuko received his M.Sc. degree in 2002. His thesis was dedicated to examination of correlation phenomena between perception of sound and vision for surround sound and digital image. He finished Ph.D. studies in 2007 and one year later completed a dissertation "Application of Fuzzy Rules in Computer Character Animation" that received award of Prime Minister of Poland. His interests include: processing of audio and video, computer...
-
Towards neural knowledge DNA
PublicationIn this paper, we propose the Neural Knowledge DNA, a framework that tailors the ideas underlying the success of neural networks to the scope of knowledge representation. Knowledge representation is a fundamental field that dedicates to representing information about the world in a form that computer systems can utilize to solve complex tasks. The proposed Neural Knowledge DNA is designed to support discovering, storing, reusing,...
-
Predicting Performance of Lightweight Concrete with Granulated Expanded Glass and Ash Aggregate by Means of Using Artificial Neural Networks
PublicationLightweight concrete (LWC) is a group of cement composites of the defined physical, mechanical, and chemical performance. The methods of designing the composition of LWC with the assumed density and compressive strength are used most commonly. The purpose of using LWC is the reduction of the structure’s weight, as well as the reduction of thermal conductivity index. The highest possible strength, durability and low thermal conductivity...
-
Tolerance-Aware Optimization of Microwave Circuits by Means of Principal Directions and Domain-Restricted Metamodels
PublicationPractical microwave design is most often carried out in the nominal sense. Yet, in some cases, performance degradation due to uncertainties may lead to the system failing to meet the prescribed specifications. Reliable uncertainty quantification (UQ) is therefore important yet intricate from numerical standpoint, especially when the circuit at hand is to be evaluated using electromagnetic (EM) simulation tools. Tolerance-aware...
-
Explicit Size-Reduction-Oriented Design of a Compact Microstrip Rat-Race Coupler Using Surrogate-Based Optimization Methods
PublicationIn this paper, an explicit size reduction of a compact rat-race coupler implemented in a microstrip technology is considered. The coupler circuit features a simple topology with a densely arranged layout that exploits a combination of high- and low-impedance transmission line sections. All relevant dimensions of the structure are simultaneously optimized in order to explicitly reduce the coupler size while maintaining equal power...
-
Comparison of single best artificial neural network and neural network ensemble in modeling of palladium microextraction
Publication -
Neural networks based NARX models in nonlinear adaptive control
Publication -
Prediction of antimicrobial activity of imidazole derivatives by artificial neural networks
Publication -
Application of Artificial Neural Networks in Investigations of Steam Turbine Cascades
PublicationZaprezentowano wyniki badań numerycznych zastosowania sieci neuronowych przy obliczeniach przepływów w palisadach turbin parowych. Na podstawie uzyskanych wyników wykazano, że sieci neuronowe mogą być używane do szacowania przestrzennego rozkładu parametrów przepływu, takich jak entalpia, entropia, ciśnienie czy prędkość czynnika w kanale przepływowym. Omówiono również zastosowania tego typu metod przy projektowaniu palisad, stopni...
-
Forecasting of currency exchange rates using artificial neural networks
PublicationW rozdziale tym autor przedstawił wyniki swoich badań nad wykorzystaniem sztucznych sieci neuronowych do prognozowania kursu walut (na przykładzie pary walutowej PLN-USD).Głównym celem badań było porównanie skuteczności przewidywania kursu złotówki w latach 1997 - 2005 przy pomocy różnych rodzajów sieci neuronowych.
-
Using neural networks to examine trending keywords in Inventory Control
Publication -
Application of neural networks for description of pressure distribution in slide bearing.
PublicationBadano rozkład ciśnienia hydrodynamicznego w łożysku ślizgowym dla wybranych wariantów łożyska. Wykazano, że zastosowanie sieci neuronowych umożliwia opis rozkładu ciśnienia hydrodynamicznego z uwzględnieniem zmian geometrycznych (bezwymiarowa długość - L) i mechanicznych (mimośrodowość względem H) łożyska.
-
Identification of slide bearing main parameters using neural networks.
PublicationWykazano, że sieci neuronowe jak najbardziej nadają się do identyfikacji głównych parametrów geometrycznych i ruchowych hydrodynamicznych łożysk ślizgowych.
-
Estimation the rhythmic salience of sound with association rules and neural networks
PublicationW referacie przedstawiono eksperymenty mające na celu automatyczne wyszukiwanie wartości rytmicznych we frazie muzycznej. W tym celu wykorzystano metody data mining i sztuczne sieci neuronowe.
-
Space and spatial practices in times of confinement. Evidence from three European countries: Austria, France and Poland
PublicationIn the first half of 2020, millions of people were subjected to drastic restrictions aimed at limiting the spread of the Covid-19 disease. Austria, France and Poland have implemented a lockdown to varying degrees and for varying lengths of time. This is an unprecedented situation in Europe: until now, even in times of war, curfew measures have never been applied 24 h a day. The research presented in this article was carried out...
-
Techno‐economic evaluation of a natural deep eutectic solvent‐based biorefinery: Exploring different design scenarios
PublicationThis paper presents a comprehensive techno‐economic evaluation of an integrated natural deep eutectic solvent (NADES)‐based biorefinery – a 1 ton day−1 capacity design plant. The key parameters include payback period, net present value (NPV), and internal rate of return (IRR). These were compared with the parameters of conventional biorefineries. The ‘n th plant’ results clearly revealed that the single product‐based biorefinery...
-
Deep CNN based decision support system for detection and assessing the stage of diabetic retinopathy
PublicationThe diabetic retinopathy is a disease caused by long-standing diabetes. Lack of effective treatment can lead to vision impairment and even irreversible blindness. The disease can be diagnosed by examining digital color fundus photographs of retina. In this paper we propose deep learning approach to automated diabetic retinopathy screening. Deep convolutional neural networks (CNN) - the most popular kind of deep learning algorithms...
-
A Multi-Fidelity Surrogate-Model-Assisted Evolutionary Algorithm for Computationally Expensive Optimization Problems
PublicationIntegrating data-driven surrogate models and simulation models of different accuracies (or fideli-ties) in a single algorithm to address computationally expensive global optimization problems has recently attracted considerable attention. However, handling discrepancies between simulation models with multiple fidelities in global optimization is a major challenge. To address it, the two major contributions of this paper include:...
-
Expedited Metaheuristic-Based Antenna Optimization Using EM Model Resolution Management
PublicationDesign of modern antenna systems heavily relies on numerical opti-mization methods. Their primary purpose is performance improvement by tun-ing of geometry and material parameters of the antenna under study. For relia-bility, the process has to be conducted using full-wave electromagnetic (EM) simulation models, which are associated with sizable computational expendi-tures. The problem is aggravated in the case of global optimization,...
-
Rapid surrogate-assisted statistical analysis of compact microstrip couplers
PublicationIn this paper, a technique for low-cost statistical analysis and yield estimation of compact microwave couplers has been presented. The analysis is executed at the level of a fast surrogate model representing selected characteristic points of the coupler response that are critical to determine satisfaction/violation of the prescribed design specifications. Because of less nonlinear dependence of the characteristic points on geometry...
-
Fast multi-objective optimization of antenna structures by means of data-driven surrogates and dimensionality reduction
PublicationDesign of contemporary antenna structures needs to account for several and often conflicting objectives. These are pertinent to both electrical and field properties of the antenna but also its geometry (e.g., footprint minimization). For practical reasons, especially to facilitate efficient optimization, single-objective formulations are most often employed, through either a priori preference articulation, objective aggregation,...
-
IEEE Transactions on Neural Networks and Learning Systems
Journals -
Optical Memory and Neural Networks (Information Optics)
Journals -
Deep Learning Optimization for Edge Devices: Analysis of Training Quantization Parameters
PublicationThis paper focuses on convolution neural network quantization problem. The quantization has a distinct stage of data conversion from floating-point into integer-point numbers. In general, the process of quantization is associated with the reduction of the matrix dimension via limited precision of the numbers. However, the training and inference stages of deep learning neural network are limited by the space of the memory and a...
-
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...