Use of perturbation theory in investigating the engineering and physical characteristics of nuclear power plants

Conclusions and recommendations

Thus, the use of perturbation theory allows an economical, searchless, computation procedure to be constructed for the problem of parametric identification. The method may be extended to the case of a multidimensional model with nonlinear and nonsteady operators.

At high noise levels of the experimental signal, a functional that is more conservative with respect to noise is preferable, in the form$J_j = \int\limits_0^{^\tau j} {y^e (\tau ) d\tau }$, for which perturbation theory is constructed analogously, under the condition P_j=1. Here also, z^e and y^e may be taken in the form of the results of statistical measurements R_zz and R_zy, i.e., the self- and mutual correlations of the noise functions of these signals, respectively.

The perturbation formulas in Eqs. (6) and (9) are useful for the engineering diagnostics of NPP elements performed on the basis of an analysis of slow variations over time in the parametersa ^′_i . In this case, identification by the given algorithm must be repeated periodically in the course of NPP running tests or use.