Algorithm for predicting of the transition of a key point of geodynamic control to the risk zone

Abstract

The article discusses the algorithm for assessing changes in the level of risk of karst hazard of the territory based on the forecasting of the number of holes depending on the water level. Based on the results of the analysis, the values of bifurcation parameters are determined, the transition through which sharply increases the formation of new holes. The article provides the developed block diagram of the neural network for assessing the dynamics of the occurrence of holes, as well as an algorithm for generating a predictive estimate of the number of holes. An analysis of the occurrence of the number of holes is carried out on the basis of water level data in the Oka River. The results of spline interpolation of the data are presented in the dependence on the number of holes on the dynamics of the water level in the river for the current and previous year. Practical verification of the developed algorithm was carried out on the basis of the new set of data on the water level in the river and the number of holes. The developed algorithm can be used in predicting the spread (leaching from the soil) of pollutants.