Wyniki wyszukiwania - BazTech

1

A hybrid wavelet–machine learning model for qanat water flow prediction

Samani Saeideh, Vadiati Meysam, Delkash Madjid, Bonakdari Hossein

Acta Geophysica

|

2023

|

Vol. 71, no. 4

1895--1913

EN

In many parts of semiarid and arid regions, qanats are the leading supplier of water demand for agricultural and drinking usage. Qanat is an ancient collecting water system, and qanat water flow (QWF) varies in different seasons and decreases gradually by pumping groundwater wells. The present research utilized a set of supervised machine learning (ML) models to predict the QWF in the Chaghlondi Aquifer in Iran using monthly intervals of 14 years (2007–2021). The wavelet transform (WT) technique was also applied to enhance the QWF prediction quality of ML models for three lead months utilizing QWF, precipitation, evapotranspiration, temperature and GWL signal datasets as input. The five widely used ML models, i.e., artificial neural network (ANN), adaptive neuro-fuzzy inference system, group method of data handling (GMDH), gene expression programming and least square support vector machine, were applied and then compared with their hybrid wavelet models. To assess the performance of the models, the following four evaluation criteria were employed: correlation coefficient (R), Nash–Sutcliffe efficiency (NSE), root means squared error (RMSE) and mean absolute error (MAE). The outcomes showed that the hybrid-wavelet ML considerably improved the standalone model performance. The best QWF predictions for a one-month ahead QWF prediction were acquired from the WT-GMDH model results from input scenario 3 with RMSE, MAE, R and NSE equal to 14.46, 10.78, 0.93 and 0.85, respectively. In addition, the result of this study indicates that ML's performance was improved by using wavelet transform for two and three months ahead of QWF predictions.

2

R peak determination using a WDFR algorithm and Adaptive threshold

Nguyen Thanh-Nghia, Nguyen Thanh-Hai, Ngo Ba-Viet

Applied Computer Science

|

2022

|

Vol. 18, no 3

19--30

EN

The determination of the R peak position in the ECG signal helps physicians not only to know the heart rate per minute, but also to monitor the patient’s health related to heart disease. This paper proposes a system to accurately determine the R peak position in the ECG signal. The system consists of a pre-processing block for filtering out noise using a WDFR algorithm and highlighting the amplitude of the R peak and a threshold value is calculated for determining the R peak. In this research, the MIT-BIH ECG dataset with 48 records are used for evaluation of the system. The results of the SEN, +P, DER and ACC parameters related to the system quality are 99.70%, 99.59%, 0.70% and 99.31%, respectively. The obtained performance of the proposed R peak position determination system is very high and can be applied to determine the R peak of the ECG signal measuring devices in practice.

3

Induction motor fault classification via entropy and column correlation features of 2D represented vibration data

Basaran Murat, Fidan Mehmet

Eksploatacja i Niezawodność

|

2021

|

Vol. 23, no. 1

132--142

EN

Due to long-term use under challenging conditions, the sub-elements of induction motors may suffer certain defects over time. Such defects impair the vibration characteristics of the motors in different ways, depending on the type of defect. Therefore, the change in vibration characteristic provides indicators about the fault type and can be used in preventive maintenance strategies to ensure safe operation of the system. In this work, discrete-time vibration data were transformed into 2-dimensional grey-level images and decomposed into individual components by the Wavelet decomposition method. Features based on entropy and column correlation were extracted from these components and used to classify motor faults by using the Support Vector Machine method implemented by using the Sequential Minimal Optimisation algorithm. When the selected classifier is compared with other popular Machine Learning algorithms, it is observed that motor faults are more successfully classified, and these observations are presented in detail with comparative classification performance results.

4

Wavelet method of edge detection in images with high-noise level

Niedziela Tadeusz

Journal of civil engineering and transport

|

2020

|

Vol. 2, No. 3

119--130

EN

In the paper a mathematical model addressed to non-sharp edges in the images is proposed. This model is based on and integral transform with Haar-Gauss wavelet and matching algorithm of bandwidth, such model is used to detection of the edges in images with high-level noises, both in the x plane and the frequency domains. There is shown that applying the integral Haar-Gaussian transformation the detection of single and double edges is possible. Demonstrated in the paper results confirm that wavelet transform supported by the matching wavelet algorithm of wavelength bandwidth make an important exploration tool of the images with the edges possessing a large depth of sharpness.

PL

W artykule zaproponowano model matematyczny nieostrej krawędzi oraz całkową transformatę z falka Haara-Gaussa wraz z algorytmem dopasowania pasma zarówno w przestrzeni x jak i w dziedzinie częstotliwości. Zilustrowano detekcję pojedynczej i podwójnej krawędzi wykorzystując całkową transformatę Haara-Gaussa. Proponowany model krawędzi wraz z transformatą falkową i algorytmem dopasowania szerokości pasma częstotliwości falki może być ważnym narzędziem w rozpoznawaniu obiektów przez nowoczesne wizyjne systemy transportowe.

5

A Fast Classification Method of Faults in Power Electronic Circuits Based on Support Vector Machines

Cui J., Shi G., Gong C.

Metrology and Measurement Systems

|

2017

|

Vol. 24, nr 4

701--720

EN

Fault detection and location are important and front-end tasks in assuring the reliability of power electronic circuits. In essence, both tasks can be considered as the classification problem. This paper presents a fast fault classification method for power electronic circuits by using the support vector machine (SVM) as a classifier and the wavelet transform as a feature extraction technique. Using one-against-rest SVM and one-against-one SVM are two general approaches to fault classification in power electronic circuits. However, these methods have a high computational complexity, therefore in this design we employ a directed acyclic graph (DAG) SVM to implement the fault classification. The DAG SVM is close to the one-against-one SVM regarding its classification performance, but it is much faster. Moreover, in the presented approach, the DAG SVM is improved by introducing the method of Knearest neighbours to reduce some computations, so that the classification time can be further reduced. A rectifier and an inverter are demonstrated to prove effectiveness of the presented design.

6

A heuristic method for detecting and locating faults employing electromagnetic acoustic transducers

Gómez C. Q. M., García F. P. M., Arcos A. J., Cheng L., Kogia M., Mohimi A., Papaelias M. A.

Eksploatacja i Niezawodność

|

2017

|

Vol. 19, no. 4

493--500

EN

The objective of this paper is to demonstrate a novel signal processing for detection, identification and flaw sizing of structural damage using ultrasonic testing with Electromagnetic Acoustic Transducers (EMATs). Damage detection involves the recognition of a defect that exists within a structure. Damage location is the identification of the geometric position of the defect. Defect classification is the cluster of the damage type into multiple damage scenarios. In the absence of external interferences, a good measure of detectability of a flaw is its signal-to-noise ratio (SNR). Although the SNR depends on various parameters such as electronics used, material properties, e.g. homogeneity and damping, and flaw size, it can be improved using advanced signal processing. The main scientific novelties presented in this paper focus on filtering signal noise through advanced digital signal processing; incorporating wavelet transforms for image and signal representation enhancements; investigating multi-parametric analysis for noise identification and defect classification; studying attenuation curves properties for defect localisation improvement and flaw sizing and location algorithm development.

PL

Celem niniejszego artykułu jest omówienie nowatorskiego sposobu przetwarzania sygnałów w celu wykrywania, identyfikacji i oceny uszkodzeń strukturalnych przy użyciu ultrasonograficznych testów za pomocą elektromagnetycznych przetworników akustycznych (EMAT). Wykrywanie uszkodzeń polega na rozpoznaniu istniejących defektów wewnątrz danej struktury. Lokalizacja uszkodzeń sprowadza się do identyfikacji geometrycznego położenia defektu. Klasyfikacja defektu to klaster typu uszkodzenia w wielu scenariuszach uszkodzeń. W przypadku braku zewnętrznych zakłóceń, dobrym wskaźnikiem wykrywalności błędu jest stosunek sygnału do szumu (SNR). Pomimo tego, że SNR zależy od różnych parametrów, takich jak użyta elektronika, właściwości materiału, np. jednorodność i tłumienie, a także wielkość wady, wskaźnik ten można poprawić przy użyciu zaawansowanego przetwarzania sygnałów. Główne nowe zagadnienia naukowe przedstawione w niniejszym artykule skupiają się na filtrowaniu szumu sygnału za pomocą zaawansowanego przetwarzania sygnału cyfrowego, w tym wykorzystując transformaty falkowe w celu ulepszenia obrazu i sygnału; badanie analizy wieloparametrycznej w celu identyfikacji szumów i klasyfikacji defektów; badanie właściwości krzywych osłabiania w celu sprawniejszego wykrywania i oceny wad oraz rozwoju algorytmu lokalizacji.

7

Development of the automatic control system of pumping station for urban water supply

Pistun E., Yakymchuk N.

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2014

|

nr 2

68--71

EN

The analysis of the automatic control of water intake area of the pumping station, which is based on a controller with fuzzy logic and systematization using seasonally and hourly water consumption by using wavelet transforms.

PL

Analiza automatycznej kontroli poboru wody pompowni, opartej jest na regulatorze rozmytym i systematyzacji na podstawie sezonowego i godzinowego zużycia wody z wykorzystaniem transformaty falkowej.

8

Power system frequency estimation algorithm for electric energy metering of nonlinear loads

Peng Z., Hong-Bin L.

Metrology and Measurement Systems

|

2012

|

Vol. 19, nr 2

307-320

EN

In this paper, a discrete wavelet transform (DWT) based approach is proposed for power system frequency estimation. Unlike the existing frequency estimators mainly used for power system monitoring and control, the proposed approach is developed for fundamental frequency estimation in the field of energy metering of nonlinear loads. The characteristics of a nonlinear load is that the power signal is heavily distorted, composed of harmonics, inter-harmonics and corrupted by noise. The main idea is to predetermine a series of frequency points, and the mean value of two frequency points nearest to the power system frequency is accepted as the approximate solution. Firstly the input signal is modulated with a series of modulating signals, whose frequencies are those frequency points. Then the modulated signals are decomposed into individual frequency bands using DWT, and differences between the maximum and minimum wavelet coefficients in the lowest frequency band are calculated. Similarities among power system frequency and those frequency points are judged by the differences. Simulation results have proven high immunity to noise, harmonic and inter-harmonic interferences. The proposed method is applicable for real-time power system frequency estimation for electric energy measurement of nonlinear loads.

9

Detection of Synthetic and Real Microcalcifications based on Statistical Analysis of Original and Highpass-Filtered Mammograms

Abdel-Qader I., Zyout I., Jacobs C.

Machine Graphics and Vision

|

2009

|

Vol. 18, No. 3

267-288

EN

Detection of clustered microcalcifications in digitized mammograms can be very useful for early detection of breast cancer. Clustered microcalcifications have a distinguished signature in both spatial and frequency domains. In the spatial domain, they appear as white spots which represent local maxima, while in the frequency domain microcalcifications represent local anomalies that can be captured within the high frequency subbands. In this work, we propose an algorithm for detection of clustered microcalcifications by utilizing these signatures, integrating the statistical parameters of both spatial and frequency domains. The results prove the effectiveness of the proposed method, and indicate that the exploitation of both domain signatures of the clustered microcalcifications yields significantly better detection results.

10

Lattice Structures for Synthesis and Implementation of Wavelet Transforms

Yatsymirskyy M.

Journal of Applied Computer Science

|

2009

|

Vol. 17, nr 1

133-141

EN

In this paper the novel lattice structure composed of homogeneous invertible two-point operations which are connected in regular and simple structure is proposed. Further on the pipeline scheme for implementation of such a structure is presented. It is proved with the orthogonal variant of the presented scheme that with respect to the computational complexity it is equivalent to the lifting technique. It means that the proposed scheme belongs to the class of the most effective algorithms for calculation of orthogonal wavelet transforms. The variant of the lattice structure with simplified two-point operations is also proposed. Finally the fundamentals of the synthesis of lattice structure coefficients with the aid of artificial neural networks and some aspects of lattice structures implementation on basic computational architectures are discussed.

11

QRS detection and identification by means of wavelet technique

Duraj A., Krawczyk A.

Przegląd Elektrotechniczny

|

2006

|

R. 82, nr 5

54-56

EN

Automatic ECG processing is recently being developed in the detection and location of the signal characteristic points and it is an important tool in the treatment of cardiac diseases. In this work we discuss the application of Wavelet Transform techniques to the detection QRS complexes. The first aim of this work is to describe tools of wavelet analysis. The second aim is to present some of the wavelet transforms (Haar, Daubiechies, Symplet, Coiflet). The exemplary method is used to establish values for soft and hard thresholding of wavelets coefficients to detection QRS.

PL

W ostatnich latach można zaobserwować bardzo szybki rozwój metod przetwarzania sygnałów biomedycznych oferujących nowe możliwości diagnostyczne. Obok metod wykorzystujących pochodne sygnałów i sieci neuronowe w ostatnim czasie coraz bardziej popularne stają się metody czasowo - częstotliwościowe w tym zwłaszcza transformacje falkowe. Wcześniejsze prace autorów dotyczyły tych metod detekcji zespołu QRS, które wykorzystywały sieci neuronowe i pochodne sygnału. Przedmiotem tej pracy jest problematyka zastosowań przekształceń falkowych do detekcji zespołu QRS. Autorzy opracowania prezentują wstępne wyniki badań dotyczące powyższego zagadnienia.

12

Data compression using Prony's method and wavelet transform in power quality monitoring systems

Zygarlicki J., Mroczka J.

Metrology and Measurement Systems

|

2006

|

Vol. 13, nr 3

325-251

EN

The paper presents a new method for data compression that can be used in power quality monitoring systems. Described algorithm offers high compression ratio and keeps good accuracy of the reconstructed signals. The algorithm uses a modified Prony's method for initial power waveform modeling (basic parameters of estimation) as well as wavelet transform for additional reduction of compression artifacts present in the Prony model of the analysed signal. The combination of the Prony's method and wavelet transform enables obtaining effective compression of real signals observed in monitoring systems. It also gives information about harmonics and transient oscillatory components of power waveforms that can be used for the power quality analysis. Examples of real signals recorded in power monitoring systems are presented in the paper as well as discussion of compression efficiency for different sets of compression artefacts level.

PL

Artykuł opisuje metodę kompresji sygnału elektroenergetycznego rejestrowanego w celu analizy jakości energii. Opisany algorytm umożliwia uzyskanie wysokiego współczynnika kompresji sygnału przy zachowaniu dużej dokładności rekonstrukcji oryginału. Algorytm wykorzystuje zmodyfikowaną metodę Pronego analizy sygnałów w celu wyznaczenia podstawowego modelu sygnału oraz dodatkowo dla zmniejszenia błędów rekonstrukcji stosuje takie dyskretną analizę falkową. Połączenie tych metod umożliwia efektywną kompresję rzeczywistych sygnałów z urządzeń pomiarowych oraz dodatkowo realizuje analizę składowych harmonicznych i zdarzeń o charakterze oscylacyjnym przebiegów, co może ułatwić późniejszą analizę zjawisk zachodzących w monitorowanych obiektach. Artykuł zawiera analizę przykładowych rzeczywistych sygnałów z systemu monitorującego, pokazuje analizę wydajności algorytmu dla różnych założonych błędów kompresji oraz opisuje błędy generowane przez metodę.

13

Zastosowanie przekształcenia falkowego do detekcji zespołu QRS

Duraj A., Krawczyk A., Koźluk E., Kumor M.

Przegląd Elektrotechniczny

|

2005

|

R. 81, nr 1

72-75

PL

W ostatnich latach można zaobserwować bardzo szybki rozwój metod przetwarzania sygnałów biomedycznych oferujących nowe możliwości diagnostyczne. Obok metod wykorzystujących pochodne sygnałów i sieci neuronowe w ostatnim czasie coraz bardziej popularne stają się metody czasowo - częstotliwościowe w tym zwłaszcza transformacje falkowe. Wcześniejsze prace autorów dotyczyły tych metod detekcji zespołu QRS, które wykorzystywały sieci neuronowe i pochodne sygnału. Przedmiotem tej pracy jest problematyka zastosowań przekształceń falkowych do detekcji zespołu QRS. Autorzy opracowania prezentują wstępne wyniki badań dotyczące powyższego zagadnienia.

EN

Automatic ECG processing is recently being developed in the detection and location of the signal characteristic points and it is an important tool in the treatment of cardiac diseases. In this work we discuss the application of Wavelet Transform technigues to the detection QRS complexes. The first aim of this work is to describe tools of wavelet analysis. The second aim is to present some of the wavelet transforms (Haar, Daubiechies, Symplet, Coiflet). The exemplary method is used to establish values for soft and hard thresholding of wavelets coefficients to detection QRS.