Classification of Random Processes in Industrial Control Systems

Need of random processes´ classification in industrial control systems is considered. Informative signs and existent methods for the classification are analyzed. The new approach suggested. According to a process type (stationary or non-stationary), process kind (additive, multiplicative or additive-multiplicative) and a deterministic constituent´s kind are classification signs. A single realization-based algorithm for classification of random processes is proposed. It implies application of non-parametric criteria, Hurst item, and fuzzy logic.