Methodology for choosing methods for forecasting time series of cyber-physical systems of fuel and energy complex enterprises

At the moment, many time series forecasting models have been created. The initial data generated during the operation of cyber-physical systems can be used to predict the future state of the system. The article presents methods for obtaining and processing data on the parameters of cyber-physical systems, their cleaning and building a predictive model for different methods. The methods are compared in terms of the quality of the forecast and the duration of the model building. As a result, the most rational methods for forecasting the time series of cyber-physical systems depending on the predicted parameter were identified.

Authors: D. P. Plakhotnikov

Direction: Informatics, Computer Technologies And Control

Keywords: cyber-physical systems, big data, data analysis, forecasting

View full article