I.V. Deryabkin, A.A. Kostoglotov, S.V. Lazarenko, A.V. Chebotarev, B.M. Tsennih
Good management of any dynamic system is possible in the case where the known with sufficient accuracy of its state at the current time and the mathematical model. The uncertainty of the system model results, for example, the need to frequently use the procedure assessments and decisions on management, which are valid only for short time intervals. However, the identification of models of dynamical systems can be spent so much money that the resulting gain in the management of a more accurate model can not recoup them. For example, such a situation can arise when testing the rocket and space technology. Therefore, the effectiveness of the solution determines the quality of the designated task management system as a whole.
The most widespread methods of identification were based on the classical theory of the Kalman filter. Their main advantage is the relative ease of use, real-time and high precision of parameter estimation. But in the case of large-scale use of the state vector associated with these procedures need to address issues of sustainability parameter estimation algorithm, which usually increases the computational cost. This limits the ability of the dynamic data.
A method for parameter identification of dynamical systems, based on the use of the true sign of movement in the integral form of the Hamiltonian. This approach allows us to derive equations based on the sequential identification combined with the physical principles of fusion. A study among MathCad mathematical modeling proves constructive approach proposed.