The State Identification of Industry 4.0 Mechatronic Elements Based on Behavioral Patterns

摘要

Problematic questions of the state of the Industry 4.0 mechatronic elements have been considered. The prerequisites determining the need to use external monitoring systems have been revealed. The type and statistical characteristics of behavioral patterns used for the analysis have been demonstrated. The proposed approach to the analysis of the autonomous object state is based on clustering methods and allows for the identification of the current state based on the processing of digitized signal traces. An experiment aimed at obtaining statistical information on various types of movement of a mechatronic device element has been described. The obtained data were processed using the k-means method. The approach to identifying the state of Industry 4.0 mechatronic elements based on the processing of digitized sequences received through external channels has been proposed. At the minimum time of the statistical information accumulation with the use of the proposed approach, it becomes possible to reveal differences in the manoeuvres performed by the object, with the probability close to 0.7. The proposed approach to the signal information processing can be used as an additional independent element for identifying the state of Industry 4.0 mechatronic elements. The approach can be quickly adapted to achieve the specified quality of the probabilistic assessment.