Wyniki wyszukiwania - BazTech

1

Optimized Approach of Feature Selection Based on Binary Genetic Algorithm in Classification of Induction Motor Faults

Le Truong-An

Annals of Computer Science and Information Systems

|

2022

|

Vol. 33

145--150

EN

In this paper, an effective model for detection and classification of multiple faults in induction motors is presented. It used S-transform method is used to analyze current signals measured from four different motors including a healthy motor, broken rotor bars, bearing damage, stator winding short-circuits fault. The feature set is extracted based on signal spectrum. With strong exploration capabilities in the search space, binary genetic algorithm (BGA) is proposed to select the optimal feature subset. As the classifier, the backpropagation neural network and support vector machine are used. The simulation results showed that the average accuracy of 100 trails is 98.3\% and the optimal feature subset equal to 36\% of total original features. That means the number of redundant features removed is 64\%. In conclusion, the proposed model combined with BGA reached highly effective in the classification of induction motor.

2

Anomaly detection in a cutting tool by k-means clustering and support vector machines

Lahrache A., Cocconcelli M., Rubini R.

Diagnostyka

|

2017

|

Vol. 18, No. 3

21--29

EN

This paper concerns the analysis of experimental data, verifying the applicability of signal analysis techniques for condition monitoring of a packaging machine. In particular, the activity focuses on the cutting process that divides a continuous flow of packaging paper into single packages. The cutting process is made by a steel knife driven by a hydraulic system. Actually, the knives are frequently substituted, causing frequent stops of the machine and consequent lost production costs. The aim of this paper is to develop a diagnostic procedure to assess the wearing condition of blades, reducing the stops for maintenance. The packaging machine was provided with pressure sensor that monitors the hydraulic system driving the blade. Processing the pressure data comprises three main steps: the selection of scalar quantities that could be indicative of the condition of the knife. A clustering analysis was used to set up a threshold between unfaulted and faulted knives. Finally, a Support Vector Machine (SVM) model was applied to classify the technical condition of knife during its lifetime.

3

Multimodal biometric system for identity verification based on hand geometry and hand palm's veins

Christo Luiz Eduardo de, Zimmer Alessandro

Annals of Computer Science and Information Systems

|

2017

|

Vol. 13

207--212

EN

This project was developed aiming at the implementation of a multibiometric systemcapable to handle hand palm images acquired using a touchless approach. Thisconsiderable increases the difficult of the image processing task due to the fact thatthe images from the same person may vary significantly depeding on the relative position of the hand regarding the sensor. A modular sofware tool was developed, providing the user a method for each of these steps: initialimage preparation, the feature extraction, processing and fusion, ending withthe classification, thus making the researcher'stask a lot easier and faster. The biometric features used for identification includehand geometry features as well palm vein textures. For the hand geometry data, analgorithm for determining finger tips and hand valleys was proposed and from there was possibleto extract a handful of other features related to the geometry of the hand. The handpalm veins' texture features were extracted from a rectangle generated based on thehand's center of mass. The texture descriptor chosen was the Histogram ofGradients. In possession with all the biometric data, the fusion was done on featurelevel. Support Vector Machine technique was used for the classification. Thedatabase chosen for the development of this project was the CASIA Multi- Spectral Palmprint Image Database V1.0. The images used corresponds to the 940nmspectrum due to allowing the visualization of the hand palm's veins. The achievedresult for the hand geometry was an EER of 4,77\%, for the palm veins an EER of3,11\% and changing the threshold value a FAR of 0,50\% and a FRR of 4,82\% wereachieved. For the fusion of both biometrics systems the final result was an EER of 2,33\% witha FAR of 1,30\% and a FRR of 4,27\%.