Method of predicting the water level in the river Dniester, depending on weather conditions
М. І. Горбійчук, М. А. Шуфнарович · 2013
This article describes a new method of constructing mathematical models of complex oscillatory processes with non-multiple frequencies, based on the ideas of genetic algorithms. Compared with existing methods of creating mathematical models of complex processes considered in the works of Academician O.G.Ivakhnenko and his students, this method allows the synthesis of mathematical models of any complexity. Using the ideas of genetic algorithms to the construction of mathematical models provides an opportunity not only to choose the optimal structure of an adequate model, but also significantly reduce the number of calculations at search patterns. Therefore, the new method can be used to predict how the physical phenomena and complex processes. Effectiveness of the developed method has been confirmed by the model of change of the water level in the river Dniester, depending on weather conditions. The model constructed by the water level in the river Dniester gives sufficiently accurate prediction results. The author suggests that the model can be used for flood forecasting