Pavlov

pp. 13-26

pp. 215-224

pp. 217-234

pp. 308-327

pp.203-215

The article concerns the harmonic linearization method in the identification of nonlinear dynamical systems problem by the example of a system with dry friction. It describes an algorithm for the systems identification with a known transfer function for an experimental frequency hodograph that contains random measurement accuracy. The transfer function of the system was adopted as a model. The authors suggested finding a solution of the identification problem in the hodograph class, defined by system model. The unknown coefficients transfer function is searched, according to authors’ proposal, by the minimizing a proximity measure of the experimental system hodograph and the hodograph of system model. As a result, the solution of the identification problem was reduced to solving a system of linear equations. An illustrative numerical simulation for the second-order system has shown that the accuracy of determining the values of the transfer function coefficients is comparable with the range of measurement accuracy of experimental samples of this system hodograph.

This article deals with the identification method for a linear dynamic element with known transfer function using an experimental frequency locus with random measurement errors. A transfer function of an element was selected as a model. It was proposed to search for the solution to the identification problem in the class of hodographs, defined by the element’s model. Search for unknown coefficients of a transfer function of the element’s model was carried out by minimizing the proposed proximity measure of the experimental element’s locus and the model’s locus. As a result, the specified problem was reduced to a system of linear equations. An illustrative computing experiment for a second-order element showed that an error of the transfer function’s coefficients was comparable with the range of measurement errors of experimental samples of this element’s locus.

This article describes a method for analyzing working precision of algorithms of automated identification of lexical items of natural language word forms on the basis of virtual texts. Estimation of working precision of this complex automated method was carried out with lexical items of Russian word forms. Expediency of applying this comprehensive solution to the problem was proved. The nature of identification errors was revealed, and on the basis of the results of the conducted analysis recommendations were given on modification of the proposed method in order to increase accuracy of automated identification of Russian lexical items.

 Several prediction models, methods of qualitative assessment of predictions and model selection methodology are presented in this article. Prediction accuracy was evaluated by comparing models with one other. Prediction accuracy of the described models was higher than the prediction of extrapolated average speed. Longtime testing, new models and factors affecting the traffic situation are planned to be considered. 

 The article presents the time stability assessment results of F. Dostoevsky’s and M. Gorky’s vocabularies. The resulting statistical characteristics suggest that frequency dictionaries are unstable. That is why informativity of identifying the authors of the frequency dictionary is poor. The authors also show that a large percentage of words found by these authors are statistically indistinguishable. This fact, in the authors’ opinion, will help to improve authorship examination quality.

This research in the field of extrapolation has begun in 2008 from developing model for extrapolation day ahead price on Russian energy market. Lately we notice that general idea of method is new so we decided to expand it for other pseudo random sequences. Objective of research is to obtain general algorithm of extrapolation of different pseudo random process (market and physical). We would like to implement comparison to any other models so we are open for any kind of cooperation