Wyniki wyszukiwania - BazTech

1

Development of Ensemble Tree Models for Generalized Blood Glucose Level Prediction

Annals of Computer Science and Information Systems

|

2022

|

Vol. 33

55--61

EN

Type-1 diabetes (T1D) patients must carefully monitor their insulin doses to avoid serious health complications. An effective regimen can be designed by predicting accurate blood glucose levels (BGLs). Several physiological and data-driven models for BGL prediction have been designed. However, less is known on the combination of different traditional machine learning (ML) algorithms for BGL prediction. Furthermore, most of the available models are patient-specific. This research aims to evaluate several traditional ML algorithms and their novel combinations for generalized BGL prediction. The data of forty T1D patients were generated using the Automated Insulin Dosage Advisor (AIDA) simulator. The twenty-four hour time-series contained samples at fifteen-minute intervals. The training data was obtained by joining eighty percent of each patient's time-series, and the remaining twenty percent time-series was joined to obtain the testing data. The models were trained using multiple patients' data so that they could make predictions for multiple patients. The traditional non-ensemble algorithms: linear regression (LR), support vector regression (SVR), k-nearest neighbors (KNN), multi-layer perceptron (MLP), decision tree (DCT), and extra tree (EXT) were evaluated for forecasting BGLs of multiple patients. A new ensemble, called the Tree-SVR model, was developed. The BGL predictions from the DCT and the EXT models were fed as features into the SVR model to obtain the final outcome. The ensemble approach used in this research was based on the stacking technique. The Tree-SVR model outperformed the non-ensemble models (LR, SVR, KNN, MLP, DCT, and EXT) and other novel Tree variants (Tree-LR, Tree-MLP, and Tree-KNN). This research highlights the utility of designing ensembles using traditional ML algorithms for generalized BGL prediction.

2

Identification of stable elementary bilinear time-series model

Malinski L.

Archives of Control Sciences

|

2016

|

Vol. 26, no. 4

577--595

EN

The paper presents new approach to estimation of the coefficients of an elementary bilinear time series model (EB). Until now, a lot of authors have considered different identifiability conditions for EB models which implicated different identifiability ranges for the model coefficient. However, all of these ranges have a common feature namely they are significantly narrower than the stability range of the EB model. This paper proposes a simple but efficient solution which makes an estimation of the EB model coefficient possible within its entire stability range.

3

Chaotic Behavior of Soil Radon Gas and Applications

Kamislioglu M., Kulahci F.

Acta Geophysica

|

2016

|

Vol. 64, no. 5

1563--1592

EN

The soil 222Rn concentration non-linear patterns are investigated by the application of various chaos methodologies based on 70 272 measurement data from the East Anatolian Fault Zone, which is one of the world’s most active faults. Among these methodologies are Lyapunov exponent, surrogate data, rescaled range (R/S) analysis, Fourier spectrum, phase space reconstruction, mutual information, false nearest neighbors, and correlation dimension. The results indicate that the nonlinear dynamical approach is convenient for characterization and prediction of the 222Rn concentration dynamics, which are in turn usually used as an earthquake precursor. Behaviour of 222Rn gas is important in earthquake prediction researches.

4

Wpływ obciążeń eksploatacyjnych na strukturę mobilnej platformy pływającej

Trzaska M., Barnat W., Kiczko A.

Logistyka

|

2015

|

nr 5

1573--1580, CD1

PL

W pracy przedstawiono wyniki badań eksperymentalnych mobilnej platformy pływającej, podczas których badano poziom naprężeń w łożu silnika obciążonego w warunkach eksploatacyjnych. Testy zostały przygotowane i przeprowadzone przez Wojskową Akademię Techniczną i PIMOT. Do pomiaru odkształceń konstrukcji wykorzystano czujniki tensometryczne wraz z mostkiem pomiarowym ESAM CF. W wyniku przeprowadzonych testów z różnym obciążeniem stwierdzono, że poziom odkształceń oraz naprężeń zredukowanych w łożu silnika nie osiąga poziomu naprężeń plastycznych dla stali.

EN

In an article presents results obtained during experimental investigation of mobile floating platform. During tests was measured the level of strains in engine mount in operational conditions. The tests were performed by Military University of Technology and Automotive Industry Institute. To measure strains of construction used strain gauges with ESAM Traveller CF Signal Conditioner Amplifier System with CF-Card data Storage. The results obtained during trials shows the level of strains and stress in engine mount did not exceed the level of yield stress for material.

5

Advanced methods of foundry processes control

Perzyk M., Kozlowski J., Wislocki M.

Archives of Metallurgy and Materials

|

2013

|

Vol. 58, iss. 3

899--902

EN

The paper discusses two main approaches utilized in contemporary industry to control of discrete and continuous manufacturing processes: Statistical Process Control and Engineering Process Control as well as applications of learning systems and time-series analysis in the control systems. The use of time-series techniques for anticipated control of selected foundry processes is presented and positively evaluated using industry data obtained from the green molding sand processing.

PL

W artykule omówiono dwa podejścia stosowane we współczesnym przemyśle do sterowania dyskretnymi i ciągłymi procesami wytwarzania: Statystyczne Sterowanie Procesem oraz sterowanie techniczne (ang. Engineering Process Control), a także zastosowania systemów uczących się i analizy szeregów czasowych w systemach sterowania. Zaprezentowano i poddano pozytywnej ocenie wykorzystanie technik szeregów czasowych w antycypacyjnym sterowaniu wybranymi procesami odlewniczymi, z użyciem danych przemysłowych uzyskanych z procesu przerobu wilgotnych mas formierskich.

6

An identification of indivertible elementary bilinear time-series models

Maliński Ł.

Prace Naukowe Politechniki Śląskiej. Elektryka

|

2011

|

z. 3

7--15

EN

One of the issues of identification of coefficient of elementary bilinear time-series model is indivertibility of the model due to displacement of global minimum of identification algorithm cost function. The paper shows that it is possible to change placement of this global minimum, so it is corresponding to real value of model coefficient.

PL

Jednym z podstawowych problemów identyfikacji współczynnika elementarnego biliniowego modelu ciągu czasowego jest nieodwracalność modelu. Skutkuje ona przemieszczeniem minimum globalnego funkcji kosztu algorytmu identyfikacji, tak że odpowiada ono nieprawidłowej wartości współczynnika. Niniejszy artykuł przedstawia propozycję rozwiązania tego problemu oraz analizę jego powtarzalności.

7

The problem of measurement data complexity for example of the respiratory system and recurrent plots analysis

Jabłoński I., Mroczka J.

Metrology and Measurement Systems

|

2008

|

Vol. 15, nr 4

457-472

EN

The purpose of this modeling paper is to show and, in some extent, to explore the utility of one nonlinear tool, the recurrence plot, in assessing physiological systems and states, for example of the respiratory system. During investigations, the authors first explore the question of interrelations between regular and disordered processes, implying remarkable influences of noise on rhythmicity of the physiological systems, here: the respiratory system. Next, the signals acquired in the Bonhoeffer-van der Pol model of the central respiratory generator are the basis for data analysis; accordingly to Paydarfar et al., computational studies of the equations (1), whose qualitative behaviour is representative of many excitable system, are able to show phase responses adequate to experimental findings in the animal, in the context of phase resetting of the central respiratory oscillator. The applied topological (qualitative) description and its quantitative measures of complexity document the important sensitivity to temporal variations of the data set compositions.