scholarly journals Naïve Bayes and linear discriminate analysis based diagnostic analytic of harmonic source identification

2020 ◽  
Vol 20 (3) ◽  
pp. 1626
Author(s):  
M. H Jopri ◽  
MR Ab Ghani ◽  
A.R Abdullah ◽  
Tole Sutikno ◽  
M Manap ◽  
...  
Keyword(s):  
Naive Bayes ◽  
Geometric Mean ◽  
Naïve Bayes ◽  
Linear Discriminate Analysis ◽  
Discriminate Analysis ◽  
Time Frequency ◽  
Frequency Representation ◽  
S Transform ◽  
Harmonic Voltage ◽  
Harmonic Source

<span>The diagnostic analytic type of harmonic source is a vital research due to diagnose and identify type of harmonic source that exist in the power system. This paper presents a comparison of machine learning (ML) algorithm namely as the Naïve Bayes (NB) and linear discriminate analysis (LDA) in identifying and diagnosing the harmonic sources.  The MLs inputs are the voltage and current feature sets that estimated from the time-frequency representation (TFR) of S-transform analysis. Four specific cases of harmonic source location are considered in this research, whereas harmonic voltage (H<sub>V</sub>) and harmonic current (H<sub>C</sub>) source type-load are used in the diagnosing process. The sufficiency of the proposed methodology is tested and verified on the IEEE 4-bust test feeder, and to prevent overfitting, the K-fold cross-validation technique is implemented for performance evaluation. To identify the best ML, the performance measurement consist of the accuracy, precision, geometric mean, F-measure, sensitivity, and specificity are conducted.</span>

scholarly journals Support-vector machine and naïve bayes based diagnostic analytic of harmonic source identification

2020 ◽  
Vol 20 (1) ◽  
pp. 1
Author(s):  
Mohd Hatta Jopri ◽  
Abdul Rahim Abdullah ◽  
Jingwei Too ◽  
Tole Sutikno ◽  
Srete Nikolovski ◽  
...  
Keyword(s):  
Support Vector Machine ◽  
Naive Bayes ◽  
Current Source ◽  
Naïve Bayes ◽  
Support Vector ◽  
Time Frequency ◽  
Frequency Representation ◽  
S Transform ◽  
Harmonic Voltage ◽  
Harmonic Source

<span>A harmonic source diagnostic analytic is a vital to identify the location and type of harmonic source in the power system. This paper introduces a comparison of machine learning (ML) algorithm which are support vector machine (SVM) and Naïve Bayes (NB). Voltage and current features are used as the input for ML are extracted from time-frequency representation (TFR) of S-transform. Several unique cases of harmonic source location are considered, whereas harmonic voltage and harmonic current source type-load are used in the diagnosing process. To identify the best ML, the performance measurement of the propose method including accuracy, specificity, sensitivity, and F-measure are calculated. The adequacy of the proposed methodology is tested and verified on IEEE 4-bust test feeder and each ML algorithm is executed for 10 times due to different partitions and to prevent any overfitting result.</span>

scholarly journals Linear discriminate analysis and k-nearest neighbor based diagnostic analytic of harmonic source identification

2021 ◽  
Vol 10 (1) ◽  
pp. 171-178
Author(s):  
Mohd Hatta Jopri ◽  
Abdul Rahim Abdullah ◽  
Mustafa Manap ◽  
M. Badril Nor Shah ◽  
Tole Sutikno ◽  
...  
Keyword(s):  
Nearest Neighbor ◽  
Geometric Mean ◽  
Performance Criteria ◽  
K Nearest Neighbor ◽  
Linear Discriminate Analysis ◽  
Discriminate Analysis ◽  
Time Frequency ◽  
Frequency Representation ◽  
S Transform ◽  
Harmonic Source

The diagnostic analytic of harmonic source is crucial research due to identify and diagnose the harmonic source in the power system. This paper presents a comparison of machine learning (ML) algorithm known as linear discriminate analysis (LDA) and k-nearest neighbor (KNN) in identifying and diagnosing the harmonic sources. Voltage and current features that estimated from time-frequency representation (TFR) of S-transform analysis are used as the input for ML. Several unique cases of harmonic source location are considered, whereas harmonic voltage (HV) and harmonic current (HC) source type-load are used in the diagnosing process. To identify the best ML, each ML algorithm is executed 10 times due to prevent any overfitting result and the performance criteria are measured consist of the accuracy, precision, geometric mean, specificity, sensitivity, and F measure are calculated.

scholarly journals K-nearest neighbor and naïve Bayes based diagnostic analytic of harmonic source identification

2021 ◽  
Vol 9 (6) ◽  
pp. 2650-2657
Author(s):  
Mohd Hatta Jopri ◽  
Mohd Ruddin Ab Ghani ◽  
Abdul Rahim Abdullah ◽  
Mustafa Manap ◽  
Tole Sutikno ◽  
...  
Keyword(s):  
Nearest Neighbor ◽  
Naive Bayes ◽  
Single Point ◽  
Naïve Bayes ◽  
K Nearest Neighbor ◽  
Time Frequency ◽  
System A ◽  
Frequency Representation ◽  
S Transform ◽  
Harmonic Source

This paper proposes a comparison of machine learning (ML) algorithm known as the k-nearest neighbor (KNN) and naïve Bayes (NB) in identifying and diagnosing the harmonic sources in the power system. A single-point measurement is applied in this proposed method, and using the S-transform the measurement signals are analyzed and extracted into voltage and current parameters. The voltage and current features that estimated from time-frequency representation (TFR) of S-transform analysis are used as the input for MLs. Four significant cases of harmonic source location are considered, whereas harmonic voltage (HV) and harmonic current (HC) source type-load are used in the diagnosing process. To identify the best ML, the performance measurement of the proposed method including the accuracy, precision, specificity, sensitivity, and F-measure are calculated. The sufficiency of the proposed methodology is tested and verified on IEEE 4-bust test feeder and each ML algorithm is executed for 10 times due to prevent any overfitting result.

A Time–Frequency Representation Model for Seismic Ground Motions

2020 ◽  
Author(s):  
Xi Zhong Cui ◽  
Han Ping Hong
Keyword(s):  
Ground Motions ◽  
Seismic Source ◽  
Model Parameters ◽  
Time Frequency ◽  
Motion Models ◽  
Power Spectral ◽  
Frequency Representation ◽  
Site Characteristics ◽  
S Transform ◽  
Proposed Model

ABSTRACT A probabilistic model of the time–frequency power spectral density (TFPSD) is presented. The model is developed, based on the time–frequency representation of records from strike-slip earthquakes, in which the time–frequency representation is obtained by applying the S-transform (ST). The model for the TFPSD implicitly considers the amplitude modulation and frequency modulation for the nonstationary ground motions; this differs from the commonly used evolutionary PSD model. Predicting models for the model parameters, based on seismic source and site characteristics, are developed. The use of the model to simulate ground motions for scenario seismic events is illustrated, in which the simulation is carried out using a recently developed model that is based on the discrete orthonormal ST and ST. The illustrative example highlights the simplicity of using the proposed model and the physical meaning of some of the model parameters. A model validation analysis is carried out by comparing the statistics of the pseudospectral acceleration obtained from the simulated records to those obtained using a few ground-motion models available in the literature and considered actual records. The comparison indicates the adequacy of the proposed model.

Performance Comparison of VSI Switches Faults Analysis Using STFT and S Transform

2015 ◽  
Vol 785 ◽  
pp. 210-214 ◽  
Author(s):  
M. Manap ◽  
A.R. Abdullah ◽  
N.Z. Saharuddin ◽  
N.A. Abidullah ◽  
Nur Sumayyah Ahmad ◽  
...  
Keyword(s):  
Power Converter ◽  
Performance Comparison ◽  
Voltage Source ◽  
Percentage Error ◽  
Identification System ◽  
Time Frequency ◽  
Signal Parameters ◽  
Detection And Identification ◽  
Frequency Representation ◽  
S Transform

Switches fault in power converter has become compelling issues over the years. To reduce cost and maintenance downtime, a good fault detection technique is an essential. In this paper, the performance of STFT and S transform techniques are analysed and compared for voltage source inverter (VSI) switches faults. The signal from phase current is represented in jointly time-frequency representation (TFR) to estimate signal parameters and characteristics. Then, the degree of accuracy for both STFT and S transform are determined by the lowest value of mean absolute percentage error (MAPE). The results demonstrate that S transform gives better accuracy compare to STFT and is suitable for VSI switches faults detection and identification system.

Sentiment Classification Using Text Embedding for Thai Teaching Evaluation

2019 ◽  
Vol 886 ◽  
pp. 221-226 ◽  
Author(s):  
Kesinee Boonchuay
Keyword(s):  
Naive Bayes ◽  
Geometric Mean ◽  
Nearest Neighbors ◽  
Naïve Bayes ◽  
Teaching Evaluation ◽  
Sentiment Classification ◽  
Teaching Skills ◽  
K Nearest Neighbors ◽  
Overall Performance ◽  
F Measure

Sentiment classification gains a lot of attention nowadays. For a university, the knowledge obtained from classifying sentiments of student learning in courses is highly valuable, and can be used to help teachers improve their teaching skills. In this research, sentiment classification based on text embedding is applied to enhance the performance of sentiment classification for Thai teaching evaluation. Text embedding techniques considers both syntactic and semantic elements of sentences that can be used to improve the performance of the classification. This research uses two approaches to apply text embedding for classification. The first approach uses fastText classification. According to the results, fastText provides the best overall performance; its highest F-measure was at 0.8212. The second approach constructs text vectors for classification using traditional classifiers. This approach provides better performance over TF-IDF for k-nearest neighbors and naïve Bayes. For naïve Bayes, the second approach yields the best performance of geometric mean at 0.8961. The performance of TF-IDF is better suited to using decision tree than the second approach. The benefit of this research is that it presents the workflow of using text embedding for Thai teaching evaluation to improve the performance of sentiment classification. By using embedding techniques, similarity and analogy tasks of texts are established along with the classification.

scholarly journals Application of improved time–frequency representation to local noise removal

2021 ◽  
Vol 11 (5) ◽  
pp. 2091-2096
Author(s):  
Baotong Liu ◽  
Qiyuan Liu ◽  
Xuefu Kang
Keyword(s):  
Temporal Resolution ◽  
Real Data ◽  
Noise Removal ◽  
Coherent Noise ◽  
Exponential Factor ◽  
Time Frequency ◽  
Coupling Noise ◽  
Frequency Representation ◽  
S Transform ◽  
Parameter Values

AbstractThe temporal resolution of conventional S transform (ST) is not sufficient for the separation of local coherent noise. We present a revised S transform (RST) which uses an analyzing window function with two control parameters of the scalar σ and the exponential factor γ. Selecting proper parameter values (say σ = 1.1, γ = 1.08), the time–frequency representation (TFR) acquired by our method exhibits a higher temporal resolution. Applying an appropriate filter in the time–frequency domain, we are able to remove specific local noise. Distributed acoustic sensing (DAS) VSP section may suffer from fiber cable coupling noise, hindering the subsequent data processing and geologic interpretation. The real data example shows the coupling noise occurred in the DAS VSP can be removed by the presented RST.

A SIGNAL-TUNED GABOR TRANSFORM WITH APPLICATION TO EEG ANALYSIS

2013 ◽  
Vol 24 (04) ◽  
pp. 1350017 ◽  
Author(s):  
JOSÉ R. A. TORREÃO ◽  
SILVIA M. C. VICTER ◽  
JOÃO L. FERNANDES
Keyword(s):  
Experimental Study ◽  
Fourier Transform ◽  
Input Signal ◽  
Fourier Component ◽  
Gabor Transform ◽  
Eeg Analysis ◽  
Time Frequency ◽  
Frequency Representation ◽  
S Transform ◽  
The Fourier Transform

We introduce a time-frequency transform based on Gabor functions whose parameters are given by the Fourier transform of the analyzed signal. At any given frequency, the width and the phase of the Gabor function are obtained, respectively, from the magnitude and the phase of the signal's corresponding Fourier component, yielding an analyzing kernel which is a representation of the signal's content at that particular frequency. The resulting Gabor transform tunes itself to the input signal, allowing the accurate detection of time and frequency events, even in situations where the traditional Gabor and S-transform approaches tend to fail. This is the case, for instance, when considering the time-frequency representation of electroencephalogram traces (EEG) of epileptic subjects, as illustrated by the experimental study presented here.

Complex matrix interpolation model of the S-transform for electric load movement forecast

2014 ◽  
Vol 05 (03) ◽  
pp. 1450011
Author(s):  
Zong-Chang Yang
Keyword(s):  
Nearest Neighbor ◽  
Interpolation Method ◽  
Electric Load ◽  
Complex Matrix ◽  
Time Frequency ◽  
Frequency Representation ◽  
S Transform ◽  
Load Movement ◽  
Matrix Interpolation ◽  
Symmetric Property

Electric load movement forecast is increasingly importance for the industry. This study addresses the load movement forecast modeling based on complex matrix interpolation of the S-transform (ST). In complex matrix of time-frequency representation of the ST, each row follows conjugate symmetric property and each column appears a certain degree of similarity. Based on these characteristics, a complex matrix interpolation method for the time-frequency representation of the ST is proposed to interpolate each row of the complex matrix based on the conjugate symmetric property, and then to perform nearest-neighbor interpolation on each column. Then with periodic extension for daily and yearly electric load movement, a forecast model employing the complex matrix interpolation of the ST is introduced. The forecast approach is applied to predict daily load movement of the European Network on Intelligent Technologies (EUNITE) load dataset and annual electric load movement of State Gird Corporation of China and its branches in 2005 and 2006. Result analysis indicates workability and effectiveness of the proposed method.

Sign in / Sign up

Export Citation Format

Share Document