V.N. Klyachkin - Dr. Sc. (Eng.), Professor, Department «Applied Mathematics and Informatics», Ulyanovsk State Technical University. E-mail: v_kl@mail.ru Yu.E. Kuvayskova - Ph. D. (Eng.), Associate Professor, Department «Applied Mathematics and Informatics», Ulya-novsk State Technical University. E-mail: u.kuvaiskova@mail.ru D.S. Bubyr - Post-graduate Student, Department «Applied Mathematics and Informatics», Ulyanovsk State Technical University. E-mail: bubir91@mail.ru

One of the main objectives of the control of the technical object is to predict its state: timely response to possible violations provides object efficiency and may help prevent an emergency. For modeling and forecasting the state of the object is proposed the use of VAR (vector autoregression) and ARCH (autoregressive model with conditional heteroscedasticity in the residuals) approaches to improve the prediction accuracy compared to classical ARIMA models (autoregressive integrated moving average). The experimental results show the efficiency of the use of these approaches in modeling and forecasting of time series the state of the technical object. The accuracy of approximation rises to 4 times whereas prediction accuracy - 6 times as compared with the ARIMA model. VAR model gives results similar to the results of ARCH approach and provides higher prediction accuracy for some time series, due to the joint description of interrelated characteristics of the object.

 

1. Kuvajjskova JU.E., Aleshina A.A.Povyshenie ehffektivnosti sistemy upravlenija tekhnicheskimi obektami pri ispolzovanii adaptivnogo dinamicheskogo regressionnogo modelirovanija vremennykh rjadov // Avtomatizacija processov upravlenija. 2013. № 4(34). S. 77−83.
2. Kljachkin V.N., Kuvajjskova JU.E., Aljoshina A.A., Kravcov JU.A.Informacionno-matematicheskaja sistema rannego preduprezhdenija ob avarijjnojj situacii // Izvestija Samarskogo nauchnogo centra RAN. 2013. № 4(4). S. 919−923.
3. Sims C.A. Macroeconomics and Reality // Econometrica.1980. V. 48.P. 1−48.
4. Engle R.F. Autoregressive Conditional Heteroskedasticity with Estimates of the Variance of U.K. Inflation. Econometrica. 1982. V. 50. P. 987−1008.
5. Boks Dzh., Dzhenkins G. Analiz vremennykh rjadov. Prognoz i upravlenie. M.: Mir. 1974. 242 s.
6. Kljachkin V.N., Bubyr D.S.Prognozirovanie sostojanija tekhnicheskogo obekta na osnove kusochno-linejjnykh regressijj // Radiotekhnika. 2014. № 7. S. 137−140.
7. Kuvajjskova JU.E., Kljachkin V.N., Bubyr D.S. Prognozirovanie sostojanija tekhnicheskogo obekta na osnove monitoringa ego parametrov // Trudy KHII Vseros. soveshhanie po problemam upravlenija. M.: Institut problem upravlenija im. V.A. Tra­peznikova RAN. 2014. S. 7616−7626.