Experience the forecast of engineering geological properties subsiding horizons for designing of neural network method
Abstract
The object of study: Geological environment that is within the influence of the designed structures located on housing estate Topol-3, in the Dnepropetrovsk city. Objective: Justification optimal method forecast of engineering-geological processes on the results of the application of neural network technology. Method of research: quantitative and qualitative analysis of the conditions and factors for the development of engineering-geological processes in the zone of influence of facilities. Mathematical modeling, statistics, correlation and regression analyzes of neural network technology. For the first time in Ukraine had applied neural network technology to forecast changes in soil properties in the zone of influence of facilities that provide new opportunities for building design tasks. It was found that the areas of research in the geological structure of the array participate aeolian-diluvial quaternary sediments. In the geological section, explore to a depth of 26 m, the array is allocated 6 geotechnical elements (GE), within which the thickness is statistically homogeneous in their properties. From the analysis of the conditions and factors of development of engineering-geological processes it is known that it may develop by settling and subsidence. Statistical homogeneity of the sample data on the properties of the soil allows you to combine data. In the future, changes in the properties of the simulation based on the analysis of this set. It was obtained forecast values relative subsidence, strength, modulus of deformation in natural and humid condition on the basis of the application of inductive methods of research. Quantitative analysis of the results indicates a significant difference between the predicted values of indicators obtained by the regulatory procedure, it points to the possible negative effects of the regulatory processes of forecasting techniques. In the future, we recommend the use of neural network technology to forecasting properties of soils in the area with affected structures.Key words: Neural network, Physical and mechanical properties of soils, geological forecasting, stochastic simulation, calculation of rainfall, the calculation of drawdownReferences
Bishop, C., 1995. Neural Networks for Pattern Recognition.
Clarendon Press, Oxford.
Borovikov, V. P., and all, 2008. Nejronnye seti. STATISTICA Neural Networks: Metodologija i tehnologii sovremennogo analiza dannyh.[Neural networks. STATISTICA Neural Networks: Methodology and tehnology of modern data analysis] M.: Hot line, Telekom. (in Russian)
Briaud, J.L. 2013. Introduction to geotechnical engineering
: unsaturated and saturated soils.Wiley. 1022 p.
Chaturvedil, A., Prasad P. R. Ch., 2013.Application of fractal geometry in determining optimal quadrat size for vegetation sampling. Current Science, vol. 105, 9, 1275-1281. cite web |url=http://www.statsoft.ru/home/textbook/default. htm StatSoft. Jelektronnyj uchebnik po statistike.
Dean, E. T. R., 2015. Particle mechanics approach to continuum constitutive modelling. Geotechnical Research, 2(1), 3–34. Hanaor, D.A.H., Gan, Y. Einav, I., 2013. Effects of surface structure deformation on static friction at fractal interfaces.
Geotechnique Letters 3, № 2, 52–58. doi.org/10.1680/ geolett.13.016
Mokritskaja, T.P., 2013. Peculiarities of degradation in loess soils’ deformation and strength properties on the example of Dnipropetrovsk city. Visnyk KNU im. T.Shevchenka, serija Geologija, 62 (3), 66 – 70.
Mokritskaja, T.P., Shestopalov, V.M., Tushev, A.V., 2012. Some facts on relations between stability and fractality of geological structures. Fundamental and Applied Science Problems, Vol 4, M: RAN, 8-16. (in Russian)
Romanova, M. V., 2009. Kompleksnyj podhod k racional’nomu ispol’zovaniju zemel’nyh resursov pri stroitel’nom osvoenii territorij na osnove ocenki geotehnicheskogo riska. [An integrated approach to the rational use of land recources in the construction development of the territory based on the geotechnical risk assessment] Bulletin ChitGU, 1 (52), 100 – 105.
Rumelhart, D.E., Hinton, G.E., Williams, R.J., 1986. Learning representations of back-propagation errors. Nature, 323, 533 – 536.
Schoenball, M., Selzer, M., Kühnlel, N., Nestler, B., Schmittbuhl, J., Kohl, T., 2013. Flow anisotropy in sheared fractures with self-affine surfaces. European Geothermal Congress, Pisa, Italy.
Stepashko, V.S., Bulgakova, O.S., Zosimov, V.V., 2010.
Gibrydni algorytmy samoorganizacii’ modelej dlja prognozuvannja skladnyh procesiv. [Hybrid algorithm self-models to predict complex processes] The induktive modeling of complex systems. Proceedings, is. 2, Kyiv: MNNC ITS, 236 – 246.
Clarendon Press, Oxford.
Borovikov, V. P., and all, 2008. Nejronnye seti. STATISTICA Neural Networks: Metodologija i tehnologii sovremennogo analiza dannyh.[Neural networks. STATISTICA Neural Networks: Methodology and tehnology of modern data analysis] M.: Hot line, Telekom. (in Russian)
Briaud, J.L. 2013. Introduction to geotechnical engineering
: unsaturated and saturated soils.Wiley. 1022 p.
Chaturvedil, A., Prasad P. R. Ch., 2013.Application of fractal geometry in determining optimal quadrat size for vegetation sampling. Current Science, vol. 105, 9, 1275-1281. cite web |url=http://www.statsoft.ru/home/textbook/default. htm StatSoft. Jelektronnyj uchebnik po statistike.
Dean, E. T. R., 2015. Particle mechanics approach to continuum constitutive modelling. Geotechnical Research, 2(1), 3–34. Hanaor, D.A.H., Gan, Y. Einav, I., 2013. Effects of surface structure deformation on static friction at fractal interfaces.
Geotechnique Letters 3, № 2, 52–58. doi.org/10.1680/ geolett.13.016
Mokritskaja, T.P., 2013. Peculiarities of degradation in loess soils’ deformation and strength properties on the example of Dnipropetrovsk city. Visnyk KNU im. T.Shevchenka, serija Geologija, 62 (3), 66 – 70.
Mokritskaja, T.P., Shestopalov, V.M., Tushev, A.V., 2012. Some facts on relations between stability and fractality of geological structures. Fundamental and Applied Science Problems, Vol 4, M: RAN, 8-16. (in Russian)
Romanova, M. V., 2009. Kompleksnyj podhod k racional’nomu ispol’zovaniju zemel’nyh resursov pri stroitel’nom osvoenii territorij na osnove ocenki geotehnicheskogo riska. [An integrated approach to the rational use of land recources in the construction development of the territory based on the geotechnical risk assessment] Bulletin ChitGU, 1 (52), 100 – 105.
Rumelhart, D.E., Hinton, G.E., Williams, R.J., 1986. Learning representations of back-propagation errors. Nature, 323, 533 – 536.
Schoenball, M., Selzer, M., Kühnlel, N., Nestler, B., Schmittbuhl, J., Kohl, T., 2013. Flow anisotropy in sheared fractures with self-affine surfaces. European Geothermal Congress, Pisa, Italy.
Stepashko, V.S., Bulgakova, O.S., Zosimov, V.V., 2010.
Gibrydni algorytmy samoorganizacii’ modelej dlja prognozuvannja skladnyh procesiv. [Hybrid algorithm self-models to predict complex processes] The induktive modeling of complex systems. Proceedings, is. 2, Kyiv: MNNC ITS, 236 – 246.
Published
2016-05-31
How to Cite
Mokritskaya, T., & Kalinina, D. (2016). Experience the forecast of engineering geological properties subsiding horizons for designing of neural network method. Journal of Geology, Geography and Geoecology, 24(1), 102-106. https://doi.org/https://doi.org/10.15421/111615
Section
Статьи
