Open AccessOn the Issue of Neural Modeling of Some Dynamic Parameters of Earthquake
Author(s) -
Larisa Haritonova
Publication year - 2018
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i2.23.15328
Subject(s) - artificial neural network , moment (physics) , magnitude (astronomy) , earthquake prediction , computer science , regression , mode (computer interface) , regression analysis , outcome (game theory) , earthquake magnitude , seismology , econometrics , machine learning , geology , statistics , mathematics , physics , geometry , mathematical economics , classical mechanics , astronomy , scaling , operating system
The author has attempted to refine the methods for modeling seismic effects. It is shown that applying the neural modeling and Artificial Neural Networks (ANN) are very prospective for analyzing such dynamic parameter earthquake foci as the moment magnitude. The further two input signals have been utilized: the power mode as well as the quantity of quakes. The author performed the analysis of regression for the predicted results and the target outputs. The article presents the equations of the regression (empirical dependences) for the outputs and target as well as the correlation factors for learning, assessment, checking, and the overall for the newly developed structure ANN for the moment magnitude. The use of the outcome obtained in this paper for the seismic designing and constructing structures and buildings will provide the conservation from possible consequences of earthquakes, reduce negative consequences for industry, the economy as a whole and human life.