Anton Bougaev

Pattern recognition based tools enabling autonomic computing.

Fault detection is one of the important constituents of fault tolerance, which in turn defines the dependability of autonomic computing. In presented work several pattern recognition tools were investigated in application to early fault detection. The optimal margin classifier technique was utilized to detect the abnormal behavior of software processes. The comparison with the performance of the quadratic classifiers is reported. The optimal margin classifiers were also implemented to the fault detection in hardware components. The impulse parameter probing technique was introduced to mitigate intermittent and transient fault problems. The pattern recognition framework of analysis of responses to a controlled component perturbation yielded promising results

R-functions based classification for abnormal software process detection

An R-functions based classification approach along with a regularization framework is proposed. The abnormal software process detection problem was used as the test bed. The R-functions based classification method is termed as the R-cloud method. The approach was validated both on synthetic and real-world data. Regularization allows to achieve good generalization and classification performance. In addition, the R-cloud approach gives the benefit of the analytical representation of the decision boundary. The introductory study on practical use of the R-cloud classifiers yielded promising results. The prototyping has shown that application of the R-functions based pattern recognition technique is a significant and practical tool for fault detection in providing fault tolerant computing.

Kernel regression based short-term load forecasting

Electrical load forecasting is an important tool in managing transmission and distribution facilities, financial resources, manpower, and materials at electrical power utility companies. A simple and accurate electrical load forecasting scheme is required. Short-term load forecasting (STLF) involves predicting the load from few hours to a week ahead. A simple non-parametric kernel regression (KR) approach for STLF is presented. Kernel regression is a linear approach with the ability to handle nonlinear information. A Gaussian kernel whose bandwidth selected by the Direct Plug-in (DPI) method is utilized. The performance comparison of the proposed method with artificial neural network (ANN), ordinary least squares (OLS), and ridge regression (RR) predictions on the same data set is presented. Experimental results show that kernel regression performs better than ANN forecaster on the given data set. The method proposed provides analytical solution, features optimal bandwidth selection, which is more instructive compared to ANN architecture and its other parameters.

Tritium measurements in neutron-induced cavitation of deuterated acetone

An attempt to reproduce the tritium measurements in an acoustic cavitation experiment with deuteratad acetone has shown no evidence of tritium production attributed to D-D fusion. The average number of disintegrations per minute observed is within 1σ of zero.

Method Of Key Vectors Extraction Using R-Cloud Classifiers

A novel method for reducing a training data set in the context of nonparametric classification is proposed. The new method is based on the method of R-clouds. The advantages of the R-cloud classification method introduced recently are being investigated. The separating boundary of the R-cloud classifier is represented using Rvachev functions. The method of key vectors extraction uses the value of the R-cloud function to quantify the disturbance of the separating boundary, which is caused by removal of one data vector from the design dataset. The R-cloud method was found instructive and practical in a number of engineering problems related to pattern classification.