Abstract <p>The article presents a method for reconstructing the parameters of dynamic model. The models is described by a system of ordinary differential equations with the number of parameters in the right-hand side exceeding the number of sought functions. To determine the parameters, firstly the indefinite system of algebraic equations with a rectangular matrix is constructed which resulting from the approximation of the system of differential equations taking into account the known values of the functions given at two successive moments in time. In addition, secondly the smooth change of parameters over time is involved. Simultaneous minimization of the residual for the underdetermined system and the sum of squares of the parameter differences at two successive moments of time leads to a regularized system of linear algebraic equations with a positive definite matrix and with a unique solution. The presented method is compared with the method of solving the inverse problem of restoring parameters in the frame of the epidemiological model SEIR-HCD. The comparison showed a smaller error in the presented method compared to the traditional method of solving the inverse problem.</p>