# US 2021/0261374 A1 · Method for predicting the presence of product defects during an intermediate processing step of a thin product wound in a roll

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0261374 A1
- Document ID: US20210261374A1
- Status: Published application
- Kind: A1
- Application number: 17/255,889
- Filed: 2019-06-25
- Published: 2021-08-26
- Earliest claimed: 2018-06-26
- Assignees: Italia Technology Alliance S.R.L.
- Applicants: Italia Technology Alliance S.R.L.
- Inventors: Alessandro CELLI
- Web page: https://patentgrep.com/patent/US20210261374A1

## CPC

- G05B 19/41875: {characterised by quality surveillance of production}
- B65H 18/08: Web-winding mechanisms
- B65H 26/02: responsive to presence of irregularities in running webs
- B65H 2301/4148: slitting
- B65H 2301/41702: management and organisation of stock and production
- B65H 2301/41704: involving layout of production or storage facility
- B65H 2301/542: Quality control
- B65H 2301/544: Reading; Scanning
- B65H 2511/23: Coordinates, e.g. three dimensional coordinates
- B65H 2553/52: RFID sensor
- B65H 2701/1244: RFID [Radio Frequency Identification Data] transponder
- B65H 2701/1924: Napkins or tissues, e.g. dressings, toweling, serviettes, kitchen paper and compresses
- B65H 2801/84: Paper-making machines
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- B65H: HANDLING THIN OR FILAMENTARY MATERIAL, e.g. SHEETS, WEBS, CABLES

## Abstract

Method for predicting the presence of product defects during an intermediate processing step of a thin product wound in a roll, which provides for—receiving a roll of thin product that has been assigned a unique identification code stored in a database system, this latter containing process and/or product parameters detected in the 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 one or more of the process and/or product parameters associated with the unique identification code of said roll contained in said database system in a predictive model, which uses a correlation, created by means of machine learning logics, from historicized values related to the process and/or product parameters output from said intermediate processing step and historicized values related to process and/or product parameters of the same rolls detected in the production steps of said rolls upstream of said intermediate processing step, in order to predict product parameters output from said intermediate processing step,—comparing said aforesaid product parameters with respective predefined limit values,—generating predictive diagnosis information of thin product defects based on the result of said comparison.

## Claims

### Claim 1

1) Method for predicting the presence of product defects during an intermediate processing step of a thin product wound in a roll, which provides for
receiving a roll of thin product that has been assigned a unique identification code stored in a database system, this latter containing process and/or product parameters detected in the 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 one or more of the process and/or product parameters associated with the unique identification code of said roll contained in said database system in a predictive model, which uses a correlation, created by means of machine learning logics, from historicized values related to the process and/or product parameters output from said intermediate processing step and historicized values related to process and/or product parameters of the same rolls detected in the production steps of said rolls upstream of said intermediate processing step, in order to predict product parameters output from said intermediate processing step,
comparing said aforesaid product parameters with respective predefined limit values,
generating predictive diagnosis information of thin product defects based on the result of said comparison.

### Claim 2 (depends on 1)

2) Method according to claim 1, wherein, subsequent to said generation of predictive information, the method comprises an action, in said intermediate processing step, adapted to modify the processing process parameters in order to avoid exceeding said limit values, or to reject portions of product potentially dangerous for any processing downstream of said station.

### Claim 3 (depends on 1)

3) Method according to claim 1, wherein said predictive model considers at least two different process and/or product parameters correlated with one another by means of machine learning logics and detected upstream of the intermediate processing step, in order to predict an event, or a process and/or product parameter preferably such as a product defect or an inauspicious process event.

### Claim 4 (depends on 1)

4) Method according to claim 1, wherein said predictive model uses a first set of historical data related to a plurality of process and/or product parameters detected in the production steps of the same rolls upstream of the intermediate processing step and a second set of historical data related to a plurality of process and/or product parameters detected output from the intermediate processing step of the prediction; by means of artificial intelligence or machine learning algorithms, a learning function is generated that correlates said two sets, generating a prediction function that allows the prediction, by analyzing one or more of the same process and/or product parameters detected upstream of the intermediate processing step, of one or more process and/or product parameters output from said intermediate processing step.

### Claim 5 (depends on 1)

5) Method according to claim 1, wherein said intermediate processing step is
a rewinding step, comprising unwinding of the thin product of a roll, defined primary roll, and winding of the thin product in one or more rolls, defined secondary rolls, with lateral dimensions the same as or smaller than the primary roll, wherein there is associated with said primary roll a said unique identification code, defined primary code, stored in said database system and with which there are associated process and/or product parameters detected in the production steps of said product upstream of said rewinding step, and there is associated with each secondary roll a related said unique identification code, defined secondary code, with which there are associated both process and/or product parameters detected in the production steps of said thin product upstream of said rewinding step, and process and/or product parameters detected in the rewinding step,
or
a conversion step of the thin product wound in a roll into a finished or semi-finished article, which includes unwinding of a roll of the thin product and insertion of the thin product into a conversion module of the thin product wound in a roll into a finished or semi-finished article.

### Claim 6 (depends on 1)

6) Method according to claim 1, wherein, in said database system, there are associated with said unique identification code of said roll process and/or product parameters detected during inline production of said thin product and/or during subsequent winding of said thin product produced in line into a said roll.

### Claim 7 (depends on 1)

7) Method according to claim 1, wherein in said database system, there are associated with said unique identification code of said roll process and/or product parameters detected during a packaging step of said roll.

### Claim 8 (depends on 1)

8) Method according to claim 1, wherein said product parameters, for a related roll, are stored as a function of the position of the area of the product in the parameter of which it was detected.

### Claim 9 (depends on 8)

9) Method according to claim 8, wherein the position of the area of the product in the parameter of which it was detected is stored by means of a system of coordinates with origin referred to the same roll, comprising
a first axis of coordinates parallel to the axis of the roll, with zero preferably corresponding to a side of the roll, and a second axis of coordinates corresponding to the linear unwinding of the thin product about the axis of the roll on which it is wound, with origin preferably coinciding with the starting point of winding of the thin product on the core of the roll.

### Claim 10 (depends on 1)

10) Method according to claim 1, wherein the assigning of a unique identification code to said at least one roll includes marking at least one said roll by applying an a) graphic, b) electronic, for example of RFID type, or c) magnetic band medium, containing said unique identification code, and storing said unique identification code in said database system.

### Claim 11 (depends on 10)

11) Method according to claim 10, wherein, in the case of an electronic or magnetic medium, it includes a programming step for assigning said primary identification code of said medium, which takes place before applying the medium to said at least one roll, or when said medium is applied to the primary roll.

### Claim 12 (depends on 1)

12) Method according to claim 1, comprising a step of reading the unique code of said roll at the start of said intermediate processing step of the thin product wound in said roll.

### Claim 13 (depends on 1)

13) Method according to claim 1, wherein the product and process parameters that can be associated with a roll and to be detected upstream of the intermediate processing step or during the intermediate processing step, and the historicized parameters used in the predictive model are chosen from two or more of the following: product type identification; machine type, roll length, roll format width, external roll diameter, width profile as a function of wound length, as a function of Y, basis weight of the wound thin product, basis weight tolerance of the wound thin product, type of core on which the thin product is wound to produce the roll, core diameter, spindle identifier, roll weight, roll production start date and time, roll production end date and time, room temperature, relative humidity, product quality grade, roll quality grade, roll density, roll density profile as a function of Y, operator or production manager identifier, production line speed profile as a function of Y, winder speed profile as a function of Y, production line acceleration as a function of Y and/or winder acceleration as a function of Y, surface treatments applied, position of surface treatments, basis weight profile along X and along Y, thickness profile along the X axis and along Y, humidity profile along the X axis and along Y, formation specifications, type of bonding, mix of fibers used, or mix of polymers used, degree of refinement, mix of types of water used, coagulant concentration, flocculant concentration, slurry conductivity, mechanical properties such as load, wet burst, elastic constant, temperatures detected along the formation of the product along Y, oven or hood temperature profile along Y, print pitch along Y, nip force applied per unit of length by the calender along Y, calender temperature along Y, winding tension profile along Y, winder slip profile, type of control applied, production recipe identifier, included defects detected by the detection system, roll growth profile diameter/meters wound, video images collected along the formation of the product with time stamping referred to the Y axis of the roll.

### Claim 14 (depends on 1)

14) Plant for producing rolls of thin products, adapted to implement a method according to claim 1.

### Claim 15

15) Plant for producing rolls of thin products, comprising
a production line of continuous thin product, comprising a control system capable of knowing product and/or process parameters of the same line,
a database system containing product and/or process parameters detected during production of the thin product and of the rolls and identification codes of the rolls produced associated with the product and/or process parameters related to said rolls,
at least an optional winder positioned at the outlet of said production line of continuous thin product, adapted to wind the thin product exiting from said line in a primary roll, comprising a management system of the winder, a reading and/or writing device on graphic, electronic or magnetic media, to be associated with the primary rolls being wound, adapted to write or program said media, operatively connected with said electronic management unit and an evaluation system of the product entering the winder, preferably comprising a quality control module QCS, and/or a visual inspection module of the defects and/or of the type that detects metal particles,
an optional rewinder adapted to produce one or more secondary rolls from said primary rolls coming from the winder, comprising an evaluation system of the product being wound in the secondary rolls, which can for example comprise a visual inspection module of the defects and/or of the type that detects metal particles, and a further optional quality control module QCS, a reading device of the identification code of the roll being unwound, and a reading and/or writing device on graphic, electronic or magnetic media, of identification codes to be associated with the secondary rolls being wound, and a management system adapted to communicate the product and/or process parameters detected by said evaluation system of the product being wound to said database system in association with the identification codes of the secondary rolls to which said parameters refer,
an optional packaging station for primary or secondary rolls, comprising a reading device of the identification codes of the rolls on media associated with said rolls, and a management system of the packaging step, with which there is associated an evaluation system of the rolls that are packaged, which can for example comprise a visual inspection module, said management system being adapted to communicate the product and/or process parameters associated with the packaging and detected by said evaluation system of the rolls being packaged to said database system in association with the identification codes of the rolls to which said parameters refer,
a conversion station for rolls into finished or semi-finished products, comprising one or more unwinding stations for one or more rolls, a conversion machine of said one or more rolls, a reading device of the identification codes of the rolls being unwound, an evaluation system of the product unwound from the rolls and entering the conversion machine, which can for example comprise a visual inspection module, sensors of the conversion machine are present to determine parameters of the product being processed, the process and/or product parameters acquired during the conversion step being associated with the respective rolls being unwound, or with their identification codes in the database system,
a management software of the process and/or product parameters associated with the unique identification codes of the rolls contained in said database system, in which there is implemented a predictive model, which uses a correlation, produced by machine learning logics, between historicized values related to the process and/or product parameters output from a processing step of the plant and historicized values related to process and/or product parameters of the same rolls detected in the production steps of said rolls upstream of said intermediate processing step, said model being adapted to predict product parameters output from said processing step.

### Claim 16 (depends on 15)

16) Plant according to claim 15, comprising means for varying the process parameters of said production line of thin product, and/or of said winder, and/or of said packaging, and/or of said rewinder, and/or of said conversion station, adapted to modify the processing process parameters in order to avoid exceeding said limit values, or to reject portions of product potentially dangerous for any processing downstream of said station.
