DTWEP METHOD AND DEVICE FOR FAULT DETECTION OF MANUFACTURING PROCESS BASED ON DYNAMIC TIME WARPING AND EXPONENTIOAL PENALTUDTWEP

摘要

A DTWEP-based manufacturing process anomaly detection device including a data management module, a distance calculation module, an anomaly detection module, and a database and a DTWEP-based manufacturing process anomaly detection method using the same are disclosed. The DTWEP-based manufacturing process abnormality detection method includes a sensor data collection step of collecting measured data (x _ij ) from n (n is a natural number of 2 or more) sensors (j), normalizing the collected sensor data (x _ij ) A normalizing step of generating data z _ij and a normalized data z _ij to calculate a DTW distance d _ij with respect to different product data for each product data Selecting reference data r _{j for} each sensor j and determining a DTW distance D _ij between the reference data r _j and other product data as a second DTW distance ); Second DTW to apply a penalty (P _ij) the distance (D _ij) calculating claim 3 DTW distance penalty (P _ij) is applied; Calculating a distance PD _i (hereinafter referred to as a 'penalty applying distance') to which the penalty is applied for each product i by summing all the third DTW distances, calculating a penalty applying distance PD _i ) is out of the anomaly detection criterion, and outputting anomaly detection information when an anomaly is detected.