US 12,066,816 B2
Method for predicting the presence of product defects during an intermediate processing step of a thin product wound in a roll
Alessandro Celli, Bologna (IT)
Assigned to ITALIA TECHNOLOGY ALLIANCE S.R.L., Bologna (IT)
Appl. No. 17/255,889
Filed by ITALIA TECHNOLOGY ALLIANCE S.R.L., Bologna (IT)
PCT Filed Jun. 25, 2019, PCT No. PCT/IB2019/055333
§ 371(c)(1), (2) Date Dec. 23, 2020,
PCT Pub. No. WO2020/003114, PCT Pub. Date Jan. 2, 2020.
Claims priority of application No. 102018000006680 (IT), filed on Jun. 26, 2018.
Prior Publication US 2021/0261374 A1, Aug. 26, 2021
Int. Cl. G05B 19/418 (2006.01); B65H 18/08 (2006.01); B65H 26/02 (2006.01)
CPC G05B 19/41875 (2013.01) [B65H 18/08 (2013.01); B65H 26/02 (2013.01); B65H 2301/4148 (2013.01); B65H 2301/41702 (2013.01); B65H 2301/542 (2013.01); B65H 2301/544 (2013.01); B65H 2553/52 (2013.01); B65H 2801/84 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method for predicting presence of product defects during an intermediate processing step of a thin product wound in a roll, which provides for
receiving a roll of the thin product wound in the roll that has been assigned a unique identification code stored in a database system, the database system comprising at least one of process parameters and product parameters detected in production steps of said thin product wound in said roll upstream of said intermediate processing step, associated with said unique identification code,
accessing said database system,
entering the at least one of the process parameters and product parameters associated with the unique identification code of said roll of the thin product wound in the roll contained in said database system in a model, wherein the model is a predictive model, which uses a correlation, created by means of machine learning logics, from historicized values related to the at least one of the process parameters and the product parameters output from said intermediate processing step and historicized values related to the at least one of the process parameters and product parameters of the roll of the thin product wound in the roll detected in the production steps of said thin product wound in the roll upstream of said intermediate processing step, in order to predict the at least one of the process parameters and the product parameters output from said intermediate processing step,
comparing at least one of predicted process parameters and product parameters with respective predefined limit values,
generating a predictive diagnosis information of at least one thin product defect based on said comparison, wherein, subsequent to said generation of predictive diagnosis information, the method comprises an action, in said intermediate processing step, adapted to modify the at least one of the process parameters and product parameters in order to avoid exceeding said respective predefined limit values, or to reject portions of product potentially defective for any processing downstream of a station.