CITY TRANSPORT SYSTEM ECOLOGICAL STATE FORECASTING WITH THE USE OF NEURAL NETWORKS
DOI:
https://doi.org/10.18372/2306-1472.72.11985Keywords:
city transport system, ecological state, neural network, transit capacity, traffic densityAbstract
Purpose: The purpose of this work is to develop an effective model for city transport system ecological state assessment using neural networks general concept. Methods: The proposed model is based on two neural networks work, taking into account the traffic density effect and the transit capacity level on urban areas. Results: Based on the synthesis of the fuzzy sets theory and neural networks basic principles, the city transport system ecological state assessing model is developed. The graphical representation of the model is given. A forecast reliability high degree is provided even at low learning rates and high dynamics of changing statistical data in the city transit traffic conditions. Conclusions: The use of fuzzy neural networks makes it possible to state a complete correspondence between fuzzy inference procedure mathematical representation and the urban transport system structure. The proposed model allows to formulate well-defined environmental guidelines when making decisions in the transit traffic field, taking into account the interests of enterprises, transport and the urban population, with the subsequent distribution of traffic flows in time and geographical space of the city industrial areas.
References
Chernyshov O. (2016) Vplyv transportu na ekolohiiu mista. Analiz ta stratehii dlia Ukrainy. [The impact of transport on the city environment. Analysis and strategy for Ukraine] Available at http://climategroup.org.ua/wp-content/uploads/2017/02/transport-ukr4_small.pdf. (accessed 26.07.2017)
Hermanova T.V., Kernozhytskaia A.F. (2014) K voprosu ekologicheskoy otsenki transportnoy sistemyi na urbanizirovannyih territoriyah (na primere g. Tyumen). [On the issue of environmental assessment of the transport system in urbanized areas (the example of Tyumen)]. Yzvestyia Samarskoho nauchnoho tsentra RAN [Izvestiya of the Samara Scientific Center of the Russian Academy of Sciences], Tom 16, pp.1713-1716. (In Russian)
Fomenko H.R. (2016) Transportna infrastruktura i problemy mist [Transport infrastructure and cities problems] Problemy rozvytku miskoho seredovyshcha, no.2(16), pp. 177-185. (In Ukrainian)
Rutkovskaia D., Pylynskyi M., Rutkovskyi L.(2013) Neironnye sety, henetycheskye alhorytmy y nechetkye systemy [Neural networks, genetic algorithms and fuzzy systems]. Moscow. Hot line - Telecom., 384 p. (In Russian)
Batyrshyn Y.Z., Nedosekyn Y.Z., Stetsko A.A., Tarasov V.B. eds. (2007) Nechetkye hybrydnye systemy. Teoryia y praktyka [Fuzzy hybrid systems. Theory and practice]. Moscow. FIZMATLIT., 208 p. (In Russian)
Noskov V.Y., Lypchanskyi M.V., Mezentsev N.V., Blyndiuk V.S. (2016) Prymenenye neironnoi sety pry optymyzatsyy protsessov upravlenyia podvyzhnym sostavom metropolytena. [Application of the neural network in the subway rolling stock management processes optimization], Komunalne hospodarstvo mist, no.126, pp. 44-49. (In Russian)
Masalmah Y.M. (2008) Image Compression Using Neural Networks. Title of article. J. Transp. Geogr., vol.10, pp. 63–69.
Liamzyn A.A., Khara M.V. (2013) Otsenka potentsyala systemy «tranzyt» promyshlennykh sytyraionov [Potential assessment of the industrial sites “transit” system], Vestnyk Donetskoi akademyy avtomobylnoho transporta, no.4, pp. 32-43. (In Russian)
Liamzyn A.A., Khara M.V. (2015) Ravnovesnoe sostoianye transportnoho sektora v tranzytnoi srede promyshlennoho raiona [Transport sector balanced condition in the industrial area transit environment]. Zbirnyk naukovykh prats Derzhavnoho ekonomiko-tekhnolohichnoho universytetu transport. no.26-27, pp. 256-261. (In Russian)
Nazemi A. & Omidi F. (2013) An efficient dynamic model for solving the shortest path problem. Title of article. Emerging Technologies. Vol. 26. 2013. P. 1-19.
