scholarly journals Power system fault identification and localization using multiple linear regression of principal component distance indices

2020 ◽  
Vol 9 (2) ◽  
pp. 113
Author(s):  
Alok Mukherjee ◽  
Palash Kr. Kundu ◽  
Arabinda Das
Keyword(s):  
Linear Regression ◽  
Power System ◽  
Transmission Line ◽  
Multiple Linear Regression ◽  
Fault Location ◽  
Principal Component ◽  
Correct Prediction ◽  
Error Matrix ◽  
Ratio Error ◽  
Index Matrix

<span>This paper is focused on the application of principal component analysis (PCA) to classify and localize power system faults in a three phase, radial, long transmission line using receiving end line currents taken almost at the midpoint of the line length. The PCA scores are analyzed to compute principal component distance index (PCDI) which is further analyzed using a ratio based analysis to develop ratio index matrix (R) and ratio error matrix (RE) and ratio error index (REI) which are used to develop a fault classifier, which produces a 100% correct prediction. The later part of the paper deals with the development of a fault localizer using the same PCDI corresponding to six intermediate training locations, which are analyzed with tool like Multiple Linear Regression (MLR) in order to predict the fault location with significantly high accuracy of only 87 m for a 150 km long radial transmission line.</span>

Modeling of TEC over the Iberian Peninsula using PCA decomposition and multiple linear regression on space weather parameters

2021 ◽  
Author(s):  
Anna Morozova ◽  
Tatiana Barlyaeva ◽  
Teresa Barata
Keyword(s):  
Linear Regression ◽  
Multiple Linear Regression ◽  
Iberian Peninsula ◽  
Space Weather ◽  
Multiple Linear Regression Model ◽  
Principal Component ◽  
Electron Content ◽  
Time Lags ◽  
Total Electron ◽  
Weather Parameters

&lt;p&gt;The total electron content (TEC) over the Iberian Peninsula was modeled using a three-step procedure. At the 1&lt;sup&gt;st&lt;/sup&gt; step the TEC series is decomposed using the principal component analysis (PCA) into several daily modes. Then, the amplitudes of those daily modes is fitted by a multiple linear regression model (MRM) using several types of space weather parameters as regressors. Finally, the TEC series is reconstructed using the PCA daily modes and MRM fitted amplitudes.&lt;/p&gt;&lt;p&gt;The advantage of such approach is that seasonal variations of the TEC daily modes are automatically extracted by PCA. As space weather parameters we considered proxies for the solar UV and XR fluxes, number of the solar flares, parameters of the solar wind and the interplanetary magnetic field, and geomagnetic indices. Different time lags and combinations of the regressors are tested.&lt;/p&gt;&lt;p&gt;The possibility to use such TEC models for forecasting was tested. Also, a possibility to use neural networks (NN) instead of MRM is studied.&lt;/p&gt;

scholarly journals Superpixel-Based Mixed Noise Estimation for Hyperspectral Images Using Multiple Linear Regression

2020 ◽  
Vol 12 (8) ◽  
pp. 1324
Author(s):  
Lei Sun ◽  
Bujin Li ◽  
Yongjian Nian
Keyword(s):  
Linear Regression ◽  
Multiple Linear Regression ◽  
Least Squares Method ◽  
Real Life ◽  
Principal Component ◽  
Noise Estimation ◽  
Hyperspectral Images ◽  
Spectral Correlation ◽  
Local Sample ◽  
Mixed Noise

HSIs (hyperspectral images) obtained by new-generation hyperspectral sensors contain both electronic noise and photon noise with comparable power. Therefore, both the SI (signal-independent) component and the SD (signal-dependent) component have to be considered. In this paper, a superpixel-based noise estimation algorithm using MLR (multiple linear regression) is proposed for the above mixed noise to estimate the noise standard deviation of both SI component and SD component. First, superpixel segmentation is performed on the first principal component obtained by MNF (minimum noise fraction)-based dimensionality reduction to generate non-overlapping regions with similar pixels. Then, MLR is performed to remove the spectral correlation, and a system of linear equations with respect to noise variances is established according to the local sample statistics calculated within each superpixel. By solving the equations in terms of the least-squares method, the noise variances are determined. The experimental results show that the proposed algorithm provides more accurate local sample statistics, and yields a more accurate noise estimation than the other state-of-the-art algorithms for simulated HSIs. The results of the real-life data also verify the effectiveness of the proposed algorithm.

The Research of Wine Quality Evaluation Based on Multiple Linear Regression

2013 ◽  
Vol 756-759 ◽  
pp. 2489-2493
Author(s):  
Huai Hui Liu ◽  
Wen Long Ji ◽  
Peng Zhang ◽  
Chuan Wen Yao
Keyword(s):  
Principal Component Analysis ◽  
Linear Regression ◽  
Multiple Linear Regression ◽  
Linear Regression Model ◽  
Quality Evaluation ◽  
Evaluation Model ◽  
Multiple Linear Regression Model ◽  
Principal Component ◽  
Wine Grape ◽  
Wine Quality

Through the establishment of evaluation model based on principal component analysis, select 8 principal components from nearly 30 indexes of wine grape. Then we establish the multiple linear regression model and analyse the association between physicochemical indexes of wine grape and wine, and the influence of physicochemical indexes of wine grape and wine on wine quality. Finally study whether we could use the physicochemical indexes to evaluate the wine quality.

Research on Optimization of "Deoxidation Alloying" of Molten Steel Based on Linear Programming

2020 ◽  
Vol 3 (1) ◽  
pp. 1
Author(s):  
Anna Li ◽  
Dongqing Xu
Keyword(s):  
Linear Programming ◽  
Linear Regression ◽  
Multiple Linear Regression ◽  
Molten Steel ◽  
Programming Model ◽  
Linear Regression Analysis ◽  
Optimal Solution ◽  
Principal Component ◽  
Multiple Linear Regression Analysis ◽  
Main Factors

<p>Aiming at the optimization of the supporting solution for molten steel "deoxidation alloying", the cost of "deoxidation alloying" is minimized from an economic perspective. Using Excel, Eviews and spss software programming, through factor analysis, clustering dimension reduction, principal component analysis Multiple linear regression analysis and linear programming optimization analysis, the author found out the main factors that affected the yield of alloy elements. This paper establishes a multiple linear regression mathematical model that affects the main factors of alloy elements and yield. According to the reference alloy price, the linear programming model is adopted to find the optimal solution of alloy ingredients.</p>

scholarly journals Predicting Students' Academic Performance Using Multiple Linear Regression and Principal Component Analysis

2018 ◽  
Vol 26 (0) ◽  
pp. 170-176 ◽  
Author(s):  
Stephen J.H. Yang ◽  
Owen H.T. Lu ◽  
Anna Y.Q. Huang ◽  
Jeff C.H. Huang ◽  
Hiroaki Ogata ◽  
...  
Keyword(s):  
Principal Component Analysis ◽  
Academic Performance ◽  
Linear Regression ◽  
Multiple Linear Regression ◽  
Principal Component ◽  
Component Analysis

scholarly journals Relationship between the Physical Properties and Hand of Jean Fabric

2016 ◽  
Vol 16 (3) ◽  
pp. 138-145 ◽  
Author(s):  
Atsushi Kawamura ◽  
Chunhong Zhu ◽  
Julie Peiffer ◽  
KyoungOk Kim ◽  
Yi Li ◽  
...  
Keyword(s):  
Principal Component Analysis ◽  
Linear Regression ◽  
Physical Properties ◽  
Multiple Linear Regression ◽  
Principal Components ◽  
Evaluation System ◽  
Principal Component ◽  
Semantic Differential ◽  
Differential Method ◽  
Fabric Hand

Abstract We investigated the distinctive characteristics of jean fabrics (denim fabrics obtained from jeans) and compared the physical properties and the hand. We used 13 kinds of jean fabric from commercial jeans and 26 other fabric types. The physical properties were measured using the Kawabata evaluation system, and the fabric hand was evaluated by 20 subjects using a semantic differential method. To characterise the hand of jean fabrics compared with other fabrics, we used principal component analysis and obtained three principal components. We found that jean fabrics were characterised by the second principal component, which was affected by feelings of thickness and weight. We further characterised the jean fabrics according to ‘softness & smoothness’ and ‘non-fullness’, depending on country of origin and type of manufacturer. The three principal components were analysed using multiple linear regression to characterise the components according to the physical properties. We explained the hand of fabrics including jean fabrics using its association with physical properties.

scholarly journals Community Mobility and COVID-19 Dynamics in Jakarta, Indonesia

2021 ◽  
Author(s):  
Ratih Oktri Nanda ◽  
Aldilas Achmad Nursetyo ◽  
Aditya Lia Ramadona ◽  
Muhammad Ali Imron ◽  
Anis Fuad ◽  
...  
Keyword(s):  
Linear Regression ◽  
Multiple Linear Regression ◽  
Linear Model ◽  
Generalized Linear Model ◽  
Negative Binomial ◽  
Human Mobility ◽  
Management Strategies ◽  
Principal Component ◽  
Outbreak Management ◽  
Community Mobility

Background Human mobility could act as a vector to facilitate the spread of infectious diseases. In response to the COVID-19 pandemic, Google Community Mobility Reports (CMR) provide the necessary data to explore community mobility further. Therefore, we aimed to examine the relationship between community mobility on COVID-19 dynamics in Jakarta, Indonesia. Methods We utilized the mobility data from Google from February 15 to December 31, 2020. We explored several statistical models to estimate the COVID-19 dynamics in Jakarta. Model 1 was a Poisson Regression Generalized Linear Model (GLM), Model 2 was a Negative Binomial Regression Generalized Linear Model (GLM), and Model 3 was a Multiple Linear Regression (MLR). Results We found that Multiple Linear Regression (MLR) with some adjustments using Principal Component Analysis (PCA) was the best fit model. It explained 52% of COVID-19 cases in Jakarta (R-Square: 0.52, p<0.05). All mobility variables were significant predictors of COVID-19 cases (p<0.05). More precisely, about 1% change in grocery and pharmacy would contribute to a 4.12% increase of the COVID-19 cases in Jakarta. Retails and recreations, workplaces, transit stations, and parks would result in 3.11%, 2.56%, 2.26%, and 1.93% of more COVID-19 cases, respectively. Conclusion Our study indicates that increased mobility contributes to increased COVID-19 cases. This finding will be beneficial to assist policymakers to have better outbreak management strategies, to anticipate increased COVID-19 cases in the future at certain public places and during seasonal events such as annual religious holidays or other long holidays in particular.

scholarly journals Um modelo alternativo de risco para companhias não-financeiras aplicado ao setor brasileiro de papel e celulose

2009 ◽  
Vol 7 (3) ◽  
pp. 347
Author(s):  
Hsia Hua Sheng ◽  
Cristiane Karcher ◽  
Paulo Hubert Jr.
Keyword(s):  
At Risk ◽  
Linear Regression ◽  
Multiple Linear Regression ◽  
Financial Risk ◽  
Risk Measure ◽  
Multiple Linear Regression Model ◽  
Principal Component ◽  
Linear Regression Models ◽  
Explanatory Factors ◽  
Paper And Pulp

Earnings at Risk (EaR) is a financial risk measure that can be applied to non-financial companies, similarly to Cash Flow at Risk (CFaR). It is based on a relation that can be quantified using a multiple linear regression model, where the dependent variable is the change on the company's results and the independent variables are changes in distinct risk factors. The presence of correlation between explanatory factors (multicollinearity) in this kind of model may cause problems when calculating EaR and CFaR. In this paper, we indicate some possible consequences of these problems when calculating EaR, and propose a method to solve it based on Principal Component Analysis technique. To test the model, we choose the Brazilian agriculture-business industry, more specifically the paper and pulp sectors. We will show that, on the absence of significant correlation between variables, the proposed model has equivalent performance to usual multiple linear regression models. We find evidence that when correlation appears, the model here proposed yields more accurate and reliable forecasts.

Sign in / Sign up

Export Citation Format

Share Document