Russian Federation
Russian Federation
Objective: collect raw data for building predictive models. Analyze the initial data, identify data outliers and outliers, divide the data into time intervals, calculate correlation coefficients, partial autocorrelation, cross-correlation, analyze the trend and seasonality of the obtained time series. Using autoregressive models, machine learning models, neuro-fuzzy models to build forecasts of time series and determine the quality of the obtained forecasts. Methods: point density, autocorrelation, partial autocorrelation, cross-correlation, Foster-Stewart test, Dickey-Fuller test, ARMA, MLP, Encoder-Decoder LTSM, TSK, Fuzzy-Partitions, SCRG, Transformers. Results: we obtained estimates of the prediction accuracy of the selected models, compared the results of the predictive models trained on different samples of initial data. Conclusions are made about the efficiency and methods of building predictive models. Practical significance: the significance of building accurate predictive models for the key quantitative indicators of stations and nonpublic routes operation is shown. The factors influencing the accuracy of the obtained forecasts are analyzed.
fuzzy neural networks, neural networks, autoregressive models, railway station performance analysis, forecasting
1. Lamehov V. A. Algoritm postroeniya prognoznoy modeli transportno-logisticheskoy deyatel'nosti na osnove primeneniya nechetkih neyronnyh setey / V. A. Lamehov, E.K. Korovyakovskiy // Byulleten' rezul'tatov nauchnyh issledovaniy. 2022. №3. S. 137– 150. DOI:https://doi.org/10.20295/2223-9987-2022-3-137-150.
2. Svidetel'stvo o gosudarstvennoy registracii programmy dlya EVM № 2022684797 Rossiyskaya Federaciya. Programma avtomatizirovannogo opredeleniya kolichestva nechetkih pravil i parametrov antecedentov i konsekventov nechetkih neyronnyh setey tipa TSK: № 2022684298: zayavl. 07.12.2022: opubl. 19.12.2022 / V. A. Lamehov; zayavitel' Federal'noe gosudarstvennoe byudzhetnoe obrazovatel'noe uchrezhdenie vysshego obrazovaniya «Peterburgskiy gosudarstvennyy universitet putey soobscheniya Imperatora Aleksandra I».
3. Dolgopolov P., Konstantinov D., Rybalchenko L., et al. Optimization of train routes based on neuro-fuzzy modeling and genetic algorithms. Procedia Comput Sci, 2019. 149, 11–18. DOI:https://doi.org/10.1016/j.procs.2019.01.101.
4. Ermakova A. V. Application of fuzzy mathematics for choosing maintenance intervals for non-public railway tracks / A. V. Ermakova // Nexo Revista Científica. 2021. Vol. 34, no. 6. P. 1885–1891. DOI: 10.5377/ nexo.v34i06.13194. EDN DBYXRY.
5. Makridakis S., Spiliotis E., Assimakopoulos V. Statistical and Machine Learning forecasting methods : Concerns and ways forward. PLoS ONE, 2018. 13 (3) : e0194889. DOI:https://doi.org/10.1371/journal.pone.0194889.
6. Cárdenas J. J., García A., Romeral J. L., et al. Evolutive ANFIS training for energy load profile forecast for an IEMS in an automated factory. ETFA. 2011. P. 1–8. doi:https://doi.org/10.1109/ETFA.2011.6059079.
7. Zhou Y., Guo S., Chang F. Explore an evolutionary recurrent ANFIS for modelling multi-step-ahead flood forecasts. Journal of Hydrology, 2019.
8. Wei C., Chen T. and Lee S. k-NN Based Neurofuzzy System for Time Series Prediction, 2013 14th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/ Distributed Computing. 2013. P. 569–574. DOI: https://doi.org/10.1109/SNPD.2013.68.
9. Pousinho H. M.I., Mendes V.M. F., Catalão J. P. S. A hybrid PSO-ANFIS approach for short-term wind power prediction in Portugal, Energy Conversion and Management. 2011. Vol. 52, iss. 1. P. 397–402. ISSN 0196–8904.
10. Zeng, A., Chen, M., Zhang, L., et al. Are Transformers Effective for Time Series Forecasting? AAAI Conference on Artificial Intelligence, 2022. DOI: https://doi.org/10.48550/arXiv.2205.13504
