SEMI-EMPIRICAL NEURAL NETWORK MODELS OF CONTROLLED DYNAMICAL SYSTEMS

Abstract

A simulation approach is discussed for maneuverable aircraft motion as nonlinear controlled dynamical system under multiple and diverse uncertainties including knowledge imperfection concerning simulated plant and its environment exposure. The suggested approach is based on a merging of theoretical knowledge for the plant with training tools of artificial neural network field. The efficiency of this approach is demonstrated using the example of motion modeling and the identification of the aerodynamic characteristics of a maneuverable aircraft. A semi-empirical recurrent neural network based model learning algorithm is proposed for multi-step ahead prediction problem. This algorithm sequentially states and solves numerical optimization subproblems of increasing complexity, using each solution as initial guess for subsequent subproblem. We also consider a procedure for representative training set acquisition that utilizes multisine control signals.

Authors and Affiliations

Mihail Egorchev, Yury Tiumentsev

Keywords

Related Articles

MATHEMATICAL PROGRAMS MODERNIZATION BASED ON RUSSIAN AND INTERNATIONAL STANDARDS

Russian education standards have come a long way from being extremely detailed and regulating the content of education programs at all levels to the current state when there is a great freedom for universities in definin...

EDUCATIONAL PROGRAM ON HPC TECHNOLOGIES BASED ON THE HETEROGENEOUS CLUSTER HYBRILIT (LIT JINR)

The article highlights the issues of training personnel for work with high-performance computing systems (HPC), as well as of support of the software and information environment which is necessary for the efficient use o...

ADAPTIVE TRAINING STRATEGY FOR DIGITAL ECONOMY PROBLEMS IN DUBNA STATE UNIVERSITY

The article is devoted to new educational trends in the transition to the digital economy. The development program of digital economy in Russia requires new approaches to training and the use of modern digital technologi...

FEATURES OF PROGRAMMING IN DSSP FOR THE TERNARY MACHINE

In article characteristic properties of the Dialogue System for Structured Programming (DSSP) in which it significantly differs from the traditional languages (Pascal, C) which are usually used for development of a basic...

QUALITY ASSESSMENT OF ALLOCATION COORDINATES USEFUL RADIATION SOURCE IN CONTROL PROBLEMS

It is offered in control problems to carry out assessment quality of the coordinates of useful radiation source allocated by means of video sensors on set of indicators, in particular, using probability of detection of d...

Download PDF file
  • EP ID EP266388
  • DOI 10.25559/SITITO.2017.4.410
  • Views 118
  • Downloads 0

How To Cite

Mihail Egorchev, Yury Tiumentsev (2017). SEMI-EMPIRICAL NEURAL NETWORK MODELS OF CONTROLLED DYNAMICAL SYSTEMS. Современные информационные технологии и ИТ-образование, 13(4), 241-255. https://europub.co.uk./articles/-A-266388