An Operating Nanoscale Data Segmentation Method Based on Curve Entropy for Close-Loop Identification of Industrial Process Control System

摘要

Data-driven modeling technology for complex industrial processes plays a key role in control system performance assessment, state condition, fault detection, design optimization and etc. This paper provides an automatic segmentation method based on curve entropy for the key problemof nanoscale data segmentation, moreover, gave an simulation application for the typical varying working condition closed-loop control process of super-heated steam temperature. The results showed that the proposed method is strong robust and almost free of the influence of the intensity ofdisturbance signals, and better than the contrast method.