Filtry
wszystkich: 804
wybranych: 742
-
Katalog
Filtry wybranego katalogu
Wyniki wyszukiwania dla: machine learning algorithmsupervised learningfracture loadfracture toughnessdata-driven techniquesprediction model
-
Automatic labeling of traffic sound recordings using autoencoder-derived features
PublikacjaAn approach to detection of events occurring in road traffic using autoencoders is presented. Extensions of existing algorithms of acoustic road events detection employing Mel Frequency Cepstral Coefficients combined with classifiers based on k nearest neighbors, Support Vector Machines, and random forests are used. In our research, the acoustic signal gathered from the microphone placed near the road is split into frames and converted...
-
Path-based methods on categorical structures for conceptual representation of wikipedia articles
PublikacjaMachine learning algorithms applied to text categorization mostly employ the Bag of Words (BoW) representation to describe the content of the documents. This method has been successfully used in many applications, but it is known to have several limitations. One way of improving text representation is usage of Wikipedia as the lexical knowledge base – an approach that has already shown promising results in many research studies....
-
Will NILM Technology Replace Multi-Meter Telemetry Systems for Monitoring Electricity Consumption?
PublikacjaThe estimation of electric power utilization, its baseload, and its heating, light, ventilation, and air-conditioning (HVAC) power component, which represents a very large portion of electricity usage in commercial facilities, are important for energy consumption controls and planning. Non-intrusive load monitoring (NILM) is the analytical method used to monitor the energy and disaggregate total electrical usage into appliance-related...
-
Design-Oriented Constrained Modeling of Antenna Structures
PublikacjaFast surrogate models are crucially important to reduce the cost of design process of antenna structures. Due to curse of dimensionality, standard (data-driven) modeling methods exhibit serious limitations concerning the number of independent geometry parameters that can be handled but also (and even more importantly) their parameter ranges. In this work, a design-oriented modeling framework is proposed in which the surrogate is...
-
IDENTIFICATION OF THE INTERFECE OF LATERAL SLIDE BEARINGS
PublikacjaIn the paper, the model of slide bearings of motor engine, the metodology of research on this type of bearings and its results are presented measurement standpoint consisting of main elements of Briggs & Stratton 550 serie 10 T 802 model piston motor engine driven by three phase elektric engine is introduced . On the measurement standpoint are measurements were done, needed for specification...
-
Empirical Analysis of Forest Penalizing Attribute and Its Enhanced Variations for Android Malware Detection
PublikacjaAs a result of the rapid advancement of mobile and internet technology, a plethora of new mobile security risks has recently emerged. Many techniques have been developed to address the risks associated with Android malware. The most extensively used method for identifying Android malware is signature-based detection. The drawback of this method, however, is that it is unable to detect unknown malware. As a consequence of this problem,...
-
Design of microstrip antenna subarrays: a simulation-driven surrogate-based approach
PublikacjaA methodology for computationally efficient simulation-driven design of microstrip antenna subarrays is presented. Our approach takes into account the effect of the feed (here, a corporate network) on the subarray side-lobe level and allows adjustment of both radiation and reflection responses of the structure under design within a single automated process. This process is realized as surrogate-based optimization that produces...
-
Simulation-Driven Design of Microstrip Antenna Subarrays
PublikacjaA methodology for computationally efficient simulation-driven design of microstrip antenna subarrays is presented. Our approach takes into account the effect of the feed (e.g., a corporate network) on the subarray side lobe level and allows adjusting both radiation and reflection responses of the structure under design within a single automated process. This process is realized as surrogate-based optimization that produces designs...
-
Sensorless field oriented control for five-phase induction motors with third harmonic injection and fault insensitive feature
PublikacjaThe paper presents a solution for sensorless field oriented control (FOC) system for five-phase induction motors with improved rotor flux pattern. In order to obtain the advantages of a third harmonic injection with a quasi-trapezoidal flux shape, two vector models, α1–β1 and α3–β3, were transformed into d1– q1, d3– q3 rotating frames, which correlate to the 1st and 3rd harmonic plane respectively. A linearization approach of the...
-
Intelligent Decision Forest Models for Customer Churn Prediction
PublikacjaCustomer churn is a critical issue impacting enterprises and organizations, particularly in the emerging and highly competitive telecommunications industry. It is important to researchers and industry analysts interested in projecting customer behavior to separate churn from non‐churn consumers. The fundamental incentive is a firm’s intent desire to keep current consumers, along with the exorbitant expense of gaining new ones....
-
Deep CNN based decision support system for detection and assessing the stage of diabetic retinopathy
PublikacjaThe 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...
-
Half-Order Modeling of Saturated Synchronous Machine
PublikacjaNoninteger order systems are used to model diffusion in conductive parts of electrical machines as they lead to more compact and knowledge models but also to improve their precision. In this paper a linear half-order impedance model of a ferromagnetic sheet deduced from the diffusion of magnetic field is briefly introduced. Then, from physical considerations and finite elements simulation, the nonlinear half-order impedance model...
-
Optimal spindle speed determination for vibration reduction during ball-end milling of flexible details
PublikacjaIn the paper a method of optimal spindle speed determination for vibration reduction during ball-end milling of flexible details is proposed. In order to reduce vibration level, an original procedure of the spindle speed optimisation, based on the Liao–Young criterion, is suggested. As the result, an optimal, constant spindle speed value is determined. For this purpose, on-stationary computational model of machining process is...
-
Recent Advances in Performance-Driven Surrogate Modeling of High-Frequency Structures
PublikacjaDesign of high‐frequency structures, including microwave and antenna components, heavily relies on full‐wave electromagnetic (EM) simulation models. Their reliability comes at a price of a considerable computational cost. This may lead to practical issues whenever numerous EM analyses are to be executed, e.g., in the case of parametric optimization. The difficulties entailed by massive simulations may be mitigated by the use of...
-
WebCT - platforma i tłumaczenie.
PublikacjaW artykule przedstawiono doświadczenia Centrum Edukacji Niestacjonarnej i Międzywydziałowego Koła Naukowego Studentów Politechniki Gdańskiej DEC@TUG w tłumaczeniu i implementacji platform LMS (ang. Learning Management System). Krótko scharakteryzowano funkcjonalność platform BSCW I Moodle, których interfejsy zostały przetłumaczone przez członków Koła DEC@TUG w latach 2000-2003. Opisano główne funkcje platformy WebCT, takie jak...
-
Parametrical 3D model of an asynchronous motor - AutoCAD application.
PublikacjaA computer program for automatic drawing of 3D models of squirrel-cage induction motors is presented in this paper. The created model of a machine is of parametrical nature, which means that the user defines geometry of particular elements (shaft, stator core, bearing, windings etc.) on the basis dimensions and some other parameters e.g. type of bearing, kind of winding, number of slots. This program is useful for 3D modelling...
-
Adversarial attack algorithm for traffic sign recognition
PublikacjaDeep learning suffers from the threat of adversarial attacks, and its defense methods have become a research hotspot. In all applications of deep learning, intelligent driving is an important and promising one, facing serious threat of adversarial attack in the meanwhile. To address the adversarial attack, this paper takes the traffic sign recognition as a typical object, for it is the core function of intelligent driving. Considering...
-
Microgrinding with single-disk lapping kinematics.
PublikacjaGrinding operations are carried out with a variety of tool-workpiece configurations. The selection of a grinding process for a particular application depends on part shape, part size, ease of fixture, requirements concerning the acceptable shape errors. It is evident that lapping is very effective in eliminating the waviness while surface grinding is not. Dual disk machines for the double face grinding with planetary kinematics...
-
Sieciowe systemy przetwarzania rozproszonego typu GRID – rozwiązania systemowe oraz przykłady aplikacyjne
PublikacjaZaprezentowano możliwości wykorzystania oraz integracji rozproszonych mocy obliczeniowych komputerów Internautów w globalnej sieci www. Pokazano paradygmaty internetowego przetwarzania rozproszonego typu grid computing oraz volunteer computing. Zwrócono uwagę na istotność tego typu przetwarzania w rozwiązywaniu zagadnień wymagających bardzo dużych mocy obliczeniowych. Pokazano reprezentatywne przykłady rozwiązań systemowych tego...
-
Internetowe systemy przetwarzania rozproszonego typu grid w zastosowaniach biznesowych
PublikacjaSkoncentrowano się na możliwościach wykorzystania oraz integracji rozproszonych mocy obliczeniowych komputerów Internautów w globalnej sieci www. Zaprezentowano paradygmaty sieciowego przetwarzania typu grid computing oraz volunteer computing. Podkreślono istotność tego typu przetwarzania w zagadnieniach wymagających bardzo dużych mocy obliczeniowych. Zaprezentowano przykłady rozwiązań systemowych tego typu: system BOINC, będący...
-
Power Hardware-In-the-Loop Approach for Autonomous Power Generation System Analysis
PublikacjaThe article presents the Power Hardware-In-the-Loop (PHIL) dynamic model of a synchronous generator of 125 kVA for autonomous power generation system analysis. This type of system is typically composed of electrical energy sources in the form of several diesel generator units with synchronous machines, the main distribution switchboard and different loads. In modern power distribution systems, the proposed power management strategies...
-
Real-time hybrid model of a wind turbine with doubly fed induction generator
PublikacjaIn recent years renewable sources have been dominating power system. The share of wind power in energy production increases year by year, which meets the need to protect the environment. Possibility of conducting, not only computer simulation, but also laboratory studies of wind turbine operation and impact on the power system and other power devices in laboratory conditions would be very useful. This article presents a method...
-
A review on analytical models of brushless permanent magnet machines
PublikacjaThis study provides an in-depth investigation of the use of analytical and numerical methods in analyzing electrical machines. Although numerical models such as the finite-element method (FEM) can handle complex geometries and saturation effects, they have significant computational burdens, are time-consuming, and are inflexible when it comes to changing machine geometries or input values. Analytical models based on magnetic equivalent...
-
Data governance: Organizing data for trustworthy Artificial Intelligence
PublikacjaThe rise of Big, Open and Linked Data (BOLD) enables Big Data Algorithmic Systems (BDAS) which are often based on machine learning, neural networks and other forms of Artificial Intelligence (AI). As such systems are increasingly requested to make decisions that are consequential to individuals, communities and society at large, their failures cannot be tolerated, and they are subject to stringent regulatory and ethical requirements....
-
Metal–Organic Frameworks (MOFs) for Cancer Therapy
PublikacjaMOFs exhibit inherent extraordinary features for diverse applications ranging from catalysis, storage, and optics to chemosensory and biomedical science and technology. Several procedures including solvothermal, hydrothermal, mechanochemical, electrochemical, and ultrasound techniques have been used to synthesize MOFs with tailored features. A continued attempt has also been directed towards functionalizing MOFs via “post-synthetic...
-
Energy-Aware Scheduling for High-Performance Computing Systems: A Survey
PublikacjaHigh-performance computing (HPC), according to its name, is traditionally oriented toward performance, especially the execution time and scalability of the computations. However, due to the high cost and environmental issues, energy consumption has already become a very important factor that needs to be considered. The paper presents a survey of energy-aware scheduling methods used in a modern HPC environment, starting with the...
-
Nonlinear planar modeling of massive taut strings travelled by a force-driven point-mass
PublikacjaThe planar response of horizontal massive taut strings, travelled by a heavy point-mass, either driven by an assigned force, or moving with an assigned law, is studied. A kinematically exact model is derived for the free boundary problem via a variational approach, accounting for the singularity in the slope of the deflected string. Reactive forces exchanged between the point-mass and the string are taken into account via Lagrange...
-
A New Type of Macro-Elements for Efficient Two-Dimensional FEM Analysis
PublikacjaThis letter deals with a model order reduction technique applicable for driven and eigenvalue problems solved using the finite element method (FEM). It allows one to efficiently compute electromagnetic parameters of structures comprising small features that require strong local mesh refinement. The subdomains of very fine mesh are separated from the global domain as so called macro-elements that undergo model reduction. The macro-elements...
-
EM-driven topology evolution for bandwidth enhancement of hybrid quadrature patch couplers
PublikacjaA broad operational bandwidth is one of the key performance figures of hybrid patch couplers. Due to the lack of systematic design procedures, bandwidth enhancement is normally obtained through manual modifications of the structure geometry. In this work, an optimization-based topology evolution for EM-driven design of patch couplers with enhanced bandwidth has been proposed. The method exploits a novel spline-based EM model where...
-
A Multi-Fidelity Surrogate-Model-Assisted Evolutionary Algorithm for Computationally Expensive Optimization Problems
PublikacjaIntegrating 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:...
-
Efficient model order reduction for FEM analysis of waveguide structures and resonators
PublikacjaAn efficient model order reduction method for three-dimensional Finite Element Method (FEM) analysis of waveguide structures is proposed. The method is based on the Efficient Modal Order Reduction (ENOR) algorithm for creating macro-elements in cascaded subdomains. The resulting macro-elements are represented by very compact submatrices, leading to significant reduction of the overall number of unknowns. The efficiency of the model...
-
Simulation-Driven Antenna Modeling by Means of Response Features and Confined Domains of Reduced Dimensionality
PublikacjaIn recent years, the employment of full-wave electromagnetic (EM) simulation tools has become imperative in the antenna design mainly for reliability reasons. While the CPU cost of a single simulation is rarely an issue, the computational overhead associated with EM-driven tasks that require massive EM analyses may become a serious bottleneck. A widely used approach to lessen this cost is the employment of surrogate models, especially...
-
Antenna Modeling Using Variable-Fidelity EM Simulations and Constrained Co-Kriging
PublikacjaUtilization of fast surrogate models has become a viable alternative to direct handling of fullwave electromagnetic (EM) simulations in EM-driven design. Their purpose is to alleviate the difficulties related to high computational cost of multiple simulations required by the common numerical procedures such as parametric optimization or uncertainty quantification. Yet, conventional data-driven (or approximation) modeling techniques...
-
Nested Space Mapping Technique for Design and Optimization of Complex Microwave Structures with Enhanced Functionality
PublikacjaIn this work, we discuss a robust simulation-driven methodology for rapid and reliable design of complex microwave/RF circuits with enhanced functionality. Our approach exploits nested space mapping (NSM) technology, which is dedicated to expedite simulation-driven design optimization of computationally demanding microwave structures with complex topologies. The enhanced func-tionality of the developed circuits is achieved by means...
-
Computationally Efficient Multi-Objective Optimization of and Experimental Validation of Yagi-Uda Antenna
PublikacjaIn this paper, computationally efficient multi-objective optimization of antenna structures is discussed. As a design case, we consider a multi-parameter planar Yagi-Uda antenna structure, featuring a driven element, three directors, and a feeding structure. Direct optimization of the high-fidelity electromagnetic (EM) antenna model is prohibitive in computational terms. Instead, our design methodology exploits response surface...
-
Identyfikacja instrumentu muzycznego z nagrania fonicznego za pomocą sztucznych sieci neuronowych
PublikacjaCelem rozprawy jest zbadanie algorytmów do identyfikacji instrumentów występujących w sygnale polifonicznym z wykorzystaniem sztucznych sieci neuronowych. W części teoretycznej przywołano podstawy przetwarzania sygnałów fonicznych w kontekście ekstrakcji parametrów sygnałów wykorzystywanych w treningu sieci neuronowych. Dodatkowo dokonano analizy rozwoju metod uczenia maszynowego z uwzględnieniem podziału na sieci neuronowe pierwszej,...
-
Classification of objects in the LIDAR point clouds using Deep Neural Networks based on the PointNet model
PublikacjaThis work attempts to meet the challenges associated with the classification of LIDAR point clouds by means of deep learning. In addition to achieving high accuracy, the designed system should allow the classification of point clouds covering an area of several dozen square kilometers within a reasonable time interval. Therefore, it must be characterized by fast processing and efficient use of memory. Thus, the most popular approaches...
-
Accurate Modeling of Antenna Structures by Means of Domain Confinement and Pyramidal Deep Neural Networks
PublikacjaThe importance of surrogate modeling techniques has been gradually increasing in the design of antenna structures over the recent years. Perhaps the most important reason is a high cost of full-wave electromagnetic (EM) analysis of antenna systems. Although imperative in ensuring evaluation reliability, it entails considerable computational expenses. These are especially pronounced when carrying out EM-driven design tasks such...
-
Komputerowo wspomagana klasyfikacja wybranych sygnałów elektromiografii powierzchniowej
PublikacjaWykorzystywanie sygnałów elektromiografii powierzchniowej (ang. Surface Electromyography, SEMG) w procesach sterowania systemami rehabilitacyjnymi stanowi obecnie standardową procedurę. Popularność SEMG wynika z nieinwazyjności metody oraz możliwości szybkiej i precyzyjnej identyfikacji funkcji mięśniowej. W przypadku osób małoletnich proces klasyfikacji sygnałów jest utrudniony ze względu na mniejsze rozmiary i wyższą dynamikę...
-
Five-phase squirrel-cage motor. Construction and drive properties
PublikacjaThis paper presents the simulation and experimental results of a five-phase squirrel-cage induction motor. The new machine has been designed to operate in a drive system with third harmonic rotor flux injection in order to improve the motor torque utilization. The motor structure, the mathematical model as well as the laboratory prototype have been described. The motor speed-torque characteristics and transients are elaborated...
-
THE PHASE SHIFTERS INFLUENCE ON THE POWER SYSTEM STABILITY
PublikacjaThe paper presents the effect of phase shifters as FACTS devices on the possibility of improving the angle stability. Presented results obtained by the dynamic simulation performed on the mathematical model of the three machine system cooperating with the 400 kV network. The generative blocks models include turbine models with their controllers and models of synchronous generators with their excitation systems and voltage regulation....
-
Looking through the past: better knowledge retention for generative replay in continual learning
PublikacjaIn this work, we improve the generative replay in a continual learning setting to perform well on challenging scenarios. Because of the growing complexity of continual learning tasks, it is becoming more popular, to apply the generative replay technique in the feature space instead of image space. Nevertheless, such an approach does not come without limitations. In particular, we notice the degradation of the continually trained...
-
Musical Instrument Tagging Using Data Augmentation and Effective Noisy Data Processing
PublikacjaDeveloping signal processing methods to extract information automatically has potential in several applications, for example searching for multimedia based on its audio content, making context-aware mobile applications (e.g., tuning apps), or pre-processing for an automatic mixing system. However, the last-mentioned application needs a significant amount of research to reliably recognize real musical instruments in recordings....
-
Ionosphere variability II: Advances in theory and modeling
PublikacjaThis paper aims to provide an overview on recent advances in ionospheric modeling capabilities, with the emphasis in the efforts relevant to electron density variability. The discussion spans a wide range of model formulations (e.g., from purely empirical to physics-based ones and data-driven approaches) seeking for advances or gaps with regard to present challenges. This discussion is further supported by consideration of the...
-
Interpretable deep learning approach for classification of breast cancer - a comparative analysis of multiple instance learning models
PublikacjaBreast cancer is the most frequent female cancer. Its early diagnosis increases the chances of a complete cure for the patient. Suitably designed deep learning algorithms can be an excellent tool for quick screening analysis and support radiologists and oncologists in diagnosing breast cancer.The design of a deep learning-based system for automated breast cancer diagnosis is not easy due to the lack of annotated data, especially...
-
Attention-Based Deep Learning System for Classification of Breast Lesions—Multimodal, Weakly Supervised Approach
PublikacjaBreast cancer is the most frequent female cancer, with a considerable disease burden and high mortality. Early diagnosis with screening mammography might be facilitated by automated systems supported by deep learning artificial intelligence. We propose a model based on a weakly supervised Clustering-constrained Attention Multiple Instance Learning (CLAM) classifier able to train under data scarcity effectively. We used a private...
-
Approximation algorithms for job scheduling with block-type conflict graphs
PublikacjaThe problem of scheduling jobs on parallel machines (identical, uniform, or unrelated), under incompatibility relation modeled as a block graph, under the makespan optimality criterion, is considered in this paper. No two jobs that are in the relation (equivalently in the same block) may be scheduled on the same machine in this model. The presented model stems from a well-established line of research combining scheduling theory...
-
Machining process sequencing and machine assignment in generative feature-based CAPP for mill-turn parts
PublikacjaProcess selection and sequencing, as one of the most complex issues when evaluated from a mathematical point of view and crucial in CAPP, still attract research attention. For the current trend of intelligent manufacturing, machining features (MFs) are the information carriers for workpiece geometry and topology representation. They are basically derived from CAD models and are used by downstream engineering applications. A feature-based...
-
No-Wait & No-Idle Open Shop Minimum Makespan Scheduling with Bioperational Jobs
PublikacjaIn the open shop scheduling with bioperational jobs each job consists of two unit operations with a delay between the end of the first operation and the beginning of the second one. No-wait requirement enforces that the delay between operations is equal to 0. No-idle means that there is no idle time on any machine. We model this problem by the interval incidentor (1, 1)-coloring (IIR(1, 1)-coloring) of a graph with the minimum...
-
Halucynacje chatbotów a prawda: główne nurty debaty i ich interpretacje
PublikacjaGeneratywne systemy sztucznej inteligencji (SI) są w stanie tworzyć treści medialne poprzez zastosowanie uczenia maszynowego do dużych ilości danych szkoleniowych. Te nowe dane mogą obejmować tekst (np. Bard firmy Google, LLaMa firmy Meta lub ChatGPT firmy OpenAI) oraz elementy wizualne (np. Stable Diffusion lub DALL-E OpenAI) i dźwięk (np. VALL-E firmy Micro- soft). Stopień zaawansowania tych treści może czynić je nieodróżnialnymi...