Mepco Schlenk Engineering College

About: Mepco Schlenk Engineering College is a based out in . It is known for research contribution in the topics: Wavelet & Wavelet transform. The organization has 1307 authors who have published 1665 publications receiving 18690 citations.

On Improving the Performance of Promoter Prediction Classifier for Eukaryotes Using Fuzzy Based Distribution Balanced Stratified Method

Abstract: In the field of molecular biology, identifying eukaryotic promoters computationally is a demanding task. To improve the accuracy of a classifier, an effort is made in this paper to apply fuzzy based distribution balanced stratified `n' fold cross validation technique on a promoter classifier. This technique is applied with both the artificial neural network and support vector machine classifiers. It is evaluated on a data set of human promoters and non-promoters and is found that the accuracy is improved considerably. This proposal also makes it possible to identify the rogue patterns. This greatly enhances the way to trace the specific functional and structural properties of these support vectors which may reveal some strong signals.

Fault Diagnosis and Asset Management of Power Transformer Using Adaptive Boost Machine Learning Algorithm

Abstract: Dissolved Gas Analysis (DGA) data of liquid insulation used to find the incipient faults such as partial discharge, thermal faults of various temperatures, discharge of high and low energy faults, combination of electrical and thermal faults in transformers. The conventional approaches of DGA namely Gas Ratio method, Duval triangle method and the Neural Network seems to be time consuming and sometimes yield erroneous results. In this paper, Adaptive BOOST machine learning algorithm is proposed, which is effective in classifying the transformer incipient faults. The results of proposed algorithm is compared with the results of different other machine learning algorithms such as K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree, Ensembler algorithm for the same set of transformers data. From the comparison, it is evident that ADABOOST machine learning algorithm performs well.

Wavelet transform based land cover classification of hyperspectral images

Abstract: This paper aims at the wavelet transform based algorithm for landcover classification of Hyperspectral remote sensing images using Support Vector Machines (SVM) In this paper Feature Extraction and Hyperspectral pixel classification are done based on Discrete Wavelet Transform (DWT) features which includes the Statistical Features and the Gray Level Co-occurrence Features The experiment is performed on a hyperspectral dataset acquired from ROSIS sensor and the experimental results indicate that it provides an Overall accuracy of about 9828% When compared to the other methods, the wavelet transform based method increases the overall classification accuracy

A review on speech enhancement algorithms and why to combine with environment classification

Abstract: Speech enhancement has been an intensive research for several decades to enhance the noisy speech that is corrupted by additive noise, multiplicative noise or convolutional noise. Even after decades of research it is still the most challenging problem, because most papers rely on estimating the noise during the nonspeech activity assuming that the background noise is uncorrelated (statistically independent of speech signal), nonstationary and slowly varying, so that the noise characteristics estimated in the absence of speech can be used subsequently in the presence of speech, whereas in a real time environment such assumptions do not hold for all the time. In this paper, we discuss the historical development of approaches that starts from the year 1970 to, the recent, 2013 for enhancing the noisy speech corrupted by additive background noise. Seeing the history, there are algorithms that enhance the noisy speech very well as long as a specific application is concerned such as the In-car noisy environment...

Least Cost Electricity Generation Planning for China with Low GHG Emission Using LEAP and EnergyPLAN

Abstract: As the world's most populated country with a fast growing economy, China presently burns about two times as much coal and fossil fuels as the US and four times as much as India for generating electrical energy, which are the source of GHG emissions. The fossil fuels are expected to unavailable in 50 more years if the consumption rate remains to grow at a high incidence. With the unstable nature of international crude prices, it is important to reduce this dependence and look for alternatives. In this context, renewable energy sources (RES) play an important role in supplying sustainable energy without environmental emissions. This paper investigates the effects of using RES in generation expansion and estimates the CO 2 emissions by developing various scenarios namely Reference scenario and Optimization scenario, under the least cost approach. The Long-Range Energy Alternative Planning (LEAP) model is used to develop these scenarios until the year 2030. The electrical energy demand, capacity to be installed and electrical energy to be produced by each plant are predicted for the year 2030 by LEAP. The output from LEAP are fed into the energy modeling tool EnergyPLAN, to plan the same in monthly and hourly basis.