Method and system for computer-implemented tracking of production history of continuous web

Abstract

The invention relates to a computer-implemented method for tracking the production history of a continuous web (5), said continuous web (5) being processed in a production system (1), said production system (1) comprising at least one subsystem (10) for performing one or more specific operations. The web (5) is conveyed through the production system (1) along a curved given path (P) by being unwound / wound onto a coil (11, 23) from on the coil (11, 23). The following steps are carried out at each of one or more time points during operation of the production system (1): a) obtaining measured data (MD) of the web (5) while the web (5) is being conveyed along the path (P), the measured data (MD) is currently time-stamped measured data (MD) acquired by at least one of a sensor (31, 32, 33) or a measuring device installed at or near the production system (1) along the given path (P), the spatial position of each sensor (31, 32, 33) and the measuring device relative to the given path (P) has been determined by a one-time reference calibration; b) obtaining movement information (MI) of the web (5); c) determining a production-related feature (PRF) of the web (5) by processing the time-stamped measured data (MD) and the movement information (MI) of the web (5) by mapping the time-stamped measured data (MD) to physical points of a region of the web (5).

Description

Technical Field

[0001] The present invention relates to a computer-implemented method and system for tracking the production history of a continuous web of material. In particular, the present invention relates to a process for manufacturing a battery cell having a continuous electrode carrier material. Background Art

[0002] The background art and the invention are described by way of example only with reference to a process for manufacturing battery cells. It will be appreciated that the problem and the described solution can be applied to many other factory automation fields where the monitoring of multi-step assembly of continuous or discrete products is crucial.

[0003] The process of manufacturing battery cells involves many production steps, all of which have an impact on the quality of the cells at the end of the production line. Since the production of individual cells takes a lot of time and is costly to produce, early identification of faulty cells or cell components can save both money and time. Until now, the battery cell production industry has been faced with relatively high scrap rates of more than 10%. Therefore, there is a need to improve the production process.

[0004] Tracking defects or deviations upstream and / or downstream of a production line is challenging with many individual process steps. In particular, if a battery cell at the end of the production line does not meet the expected quality requirements, it would be desirable to recall the production history of all subcomponents within the battery cell, especially the individual fragments of assembled electrode material in its history. Deviations of individual electrode fragments can now be re-evaluated and process improvements can be performed from the root cause. A known solution for tracking production-related information is to introduce, for example, a spatial index related to the electrode materials of the cell to facilitate full process documentation, such as parameters that are critical to quality / process, recipes used, etc. This enables electronic production records and / or queries based on fragments or positions on the electrode. However, this procedure is time-consuming and complex.

[0005] US2002 / 0030704 A1 discloses a method for measuring a down-web coordinate on a web, the method measuring the down-web coordinate on a web by correlating the time interval elapsed since the detection of a position indicating mark applied to the web with the measured speed. After the position indicating mark is detected, the measured down-web coordinate is synchronized with the down-web coordinate indicated by the mark. Summary of the invention

[0006] The object of the present invention is to provide a method and a system for computer-implemented tracking of the production history of a continuous web of material, which in a simple manner allows certain quality parameters to be tracked and traced.

[0007] These objects are solved by the independent claims. Preferred embodiments are listed in the dependent claims.

[0008] The present invention provides a computer-implemented method for tracking the production history of a continuous coil, which is processed in a production system. The continuous coil may in particular be a continuous electrode carrier material for battery cell production. The production system comprises at least one subsystem for performing one or more specific operations, wherein the continuous coil is conveyed through the production system along a (curved) given path by unwinding from / winding onto a coil.

[0009] According to the method of the invention, the following steps a) to c) are performed at each of one or more points in time during the operation of the production system.

[0010] In step a), measured data of the continuous web is obtained while the continuous web is conveyed along the given path. In the following, the term "measured data" refers to digital data. The term "obtaining measured data" means receiving the data by a processor implementing the method of the present invention. The measured data is current time-stamped measured data (i.e., real-time data) acquired by at least one of the sensors or measuring devices installed at or near the production system along the given path. The spatial position of each sensor and measuring device relative to the given path is known.

[0011] In step b), movement information of the continuous web is obtained. The term "obtaining movement information" means that movement information of the continuous web, such as speed or speed change, is received by a processor implementing the method of the present invention.

[0012] In step c), the time-stamped measured data and the movement information of the continuous web are processed by mapping the time-stamped measured data to physical points of the continuous web to determine production-related characteristics of the continuous web.

[0013] The method of the invention provides a simple and direct method for providing production-related characteristics of each physical point of an area of ​​a continuous web, which, when stored in a database, enables the production history of the continuous web to be tracked. The production-related characteristics of each physical point of an area of ​​the continuous web represent a virtual electrode web representation containing the production-related characteristics of the physical points of the area of ​​the continuous web.

[0014] Knowledge of the production history of the continuous web allows a search for production anomalies to identify faults at a later point in time and to analyze the root causes. Furthermore, cause-effect relationships can be identified. As a further advantage, an increase in the localization resolution of web segments can be achieved. Furthermore, by providing an information feature (fingerprint), spatial segments downstream of the production process can be (re)identified and linked without the use of markers as in the past / or with significantly less intrusive markers.

[0015] Tracking the production history enables online adjustment of control parameters depending on the characteristic information of the current coil segment. In addition, it also allows early stopping of the production cycle due to quality alarms or decision support for recycling, so that parts with a large number of defects do not need to be scrapped in their entirety. All these items help reduce scrap, improve efficiency, and thereby save costs or energy during battery cell production. Therefore, the life and quality of the battery cells produced at the end of production can be optimized.

[0016] The methods described herein can be derived from existing production equipment using standard factory automation systems. They already allow continuous, i.e. high-resolution spatial feature mapping. The acquired measured data allow the generation of unique characteristics through spatial alignment and consistent feature vectors.

[0017] According to a preferred embodiment, the spatial position of each sensor and measuring device relative to the given path is determined by a one-time reference calibration. A one-time reference calibration is performed before performing steps a) to c) as described above. For the reference calibration, the following steps are performed.

[0018] In step a1), a spatial mark is attached to the continuous web. The spatial mark can be a removable mark or a laser code printed mark on the edge of the continuous web.

[0019] In step a2), the continuous web is moved at a given speed during calibration. Preferably, the speed is constant throughout the reference calibration. Further preferably, the speed is slow, i.e. slower than the speed at which the continuous web moves along the given path during a normal production process.

[0020] In step a3), a time step is acquired each time the position of a certain sensor or measuring device is crossed. Such a sensor may be an optical sensor, such as a camera. The camera may operate in the infrared spectrum. The measuring device may be a device for determining the thickness of a coating, a temperature sensor, a humidity sensor, etc. Certain sensors are installed along the production path of a given process.

[0021] In step a4), the time interval between two adjacent positions is converted into absolute coordinates or relative spatial distances of a first frame using a predefined first origin. The first frame can be considered as the coordinate system of the production system. In this article, it is also called the machine (production) coordinate system.

[0022] The first origin may correspond to a portion of the production system, optionally marked with a unique marker placed within the field of view of an optical sensor.

[0023] Preferably, the first origin corresponds to a fixed mark which can be observed in the field of view of the sensor positioned at the end of the given path. The term "end of a given path" is to be understood as all production steps before winding onto a coil have been completed. The sensor observing the fixed mark may be positioned close to the winding coil. However, the sensor may be arranged at a different position in the end section of a given path after a specific production step has been completed.

[0024] According to another preferred embodiment, the projected position of each sensor and measuring device along said given path is stored.The protected positions are related to a first frame in which the (curved) given path is unfolded.

[0025] According to another preferred embodiment, a predefined second origin of a second frame of the continuous web is obtained by processing measured data of a certain sensor or measuring device. The second frame can be regarded as a web coordinate system.

[0026] In particular, the second origin corresponds to a visible feature of the continuous web that can be observed in the field of view of the sensor positioned at the end of the given path. A practical choice for defining such an origin of the continuous web is the leading edge of the coated area (in the case of a production system and its subsystems performing coating), as it can be easily identified by optical image processing. Furthermore, the coated area is less likely to be cut off later during the production process.

[0027] According to another preferred embodiment, the second frame refers to the continuous web fully unfolded along the length dimension of the continuous web, wherein each physical point can be identified by a coordinate along the unfolded length dimension and perpendicular to the length dimension, and the coordinate perpendicular to the length dimension extends along the width direction of the continuous web.

[0028] The coordinates in the second frame may include other information about the side (ie, the upper side or the lower side) of the continuous web.

[0029] According to another preferred embodiment of the present invention, the following steps are performed in step c).

[0030] In step c1), for each physical point or point region on the continuous web, the moment of crossing over a corresponding sensor or measuring device positioned along the direction of movement is calculated.

[0031] In step c2), the time-stamped measured data are queried to obtain characteristics associated with the corresponding sensor or measuring device. Querying the time-stamped measured data may be performed by retrieving them from a database.

[0032] In step c3), the set of features retrieved is stored in a database for each point or area on the continuous web.The measured data with time stamp and the information about the physical position on the continuous web constitute a "feature vector".

[0033] The spatial position of a certain sensor or measuring device or the spatial interval between two adjacent sensors and, optionally, the side information can be processed into a unique identifier.

[0034] According to another preferred embodiment, a method according to one or more embodiments as described herein above is performed for a second subsequent production step, wherein a particular physical point or point region on the continuous web is re-identified by matching the resulting feature vectors which exhibit the highest similarity score in their corresponding feature fields. This enables identification and tracking of web positions or segments across separate process steps. Improving the accuracy of position determination or (re-identification of feature patterns) provides a link between different production processes and can be achieved via Bayesian methods (e.g. "Kalman filter" or "particle filter").

[0035] In addition to the above method, the invention relates to a device for computer-implemented tracking of the production history on a continuous web, wherein the device is configured to perform the method according to the invention or one or more preferred embodiments of the method according to the invention.

[0036] Furthermore, the invention relates to a production system comprising a device as described above.

[0037] Furthermore, the invention relates to a computer program product with a program code stored on a non-transitory machine-readable carrier for performing the method according to the invention or one or more preferred embodiments thereof when the program code is executed on a computer. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0039] Figure 1is a schematic diagram of a production system including a controller for performing an embodiment of the present invention.

[0040] Figure 2 is a schematic diagram showing steps for mapping time-stamped measured data and movement information of a continuous web to determine production-related characteristics.

[0041] Figure 3 A schematic flow chart of a method according to the invention is shown. DETAILED DESCRIPTION

[0042] Typically, web production machines are used to print, coat or laminate continuous flat materials such as paper, textiles and plastics. They are used in many products and industry sectors, such as packaging and foil materials, printing services or adhesive coating. Web machines combine several process steps, such as cutting, rolling and positioning, which are usually spread across the entire workshop.

[0043] The following description will be based on a coating process by way of example, in which the active material (the so-called battery slurry) is applied to a continuous web (i.e., the electrode carrier material). It should be understood that the same or similar concepts can also be transferred to other processes, such as subsequent calendering, slitting or cell assembly steps.

[0044] Figure 1 A schematic diagram of a system 1 according to an embodiment of the present invention is shown. The production system 1 consists of a coating machine for coating a continuous web 5 in the form of a continuous electrode material. The continuous web 5 is conveyed through the production system 1 along a given curved path P, unwinding from an unwinding coil 11 and winding onto a winding coil 23. In between, the continuous web 5 passes through a plurality of subsystems 10 of the production system 1. In order starting from the unwinding coil 11 to the winding coil 23, the following subsystems 10 are passed along the path P, which are related to coating and provide measurements of relevant production and condition parameters: a crown 12, a moving roller 13, an edge control device 14, a coating device (coating machine) 15, a wet layer thickness measuring device 16, a dryer unit 17, a dryer unit 18, a dryer unit 19, a surface weight measuring device 20, a moving roller 21 and an edge control device 22. Each of these subsystems 10 may include a plurality of rollers along which the web 5 is guided and redirected.

[0045] The production system 1 also includes a plurality of sensors 31, 32, 33 and measuring devices, wherein: Figure 1Only the sensor 33 at the end of a given path P is illustrated. The sensor 33 positioned at the end of the path P is an optical and / or infrared camera and is placed so that the unique mark 24 is within the field of view of the sensor 33. The mark 24 represents the first origin of the first frame MF, as will be described below. The first frame MP corresponds to a machine coordinate system, wherein the machine is represented by the production system 1.

[0046] The sensors 31, 32, 33 and the measuring devices are installed along the path P of a given process of the production system 1. Typically, the sensors may be optical and / or infrared cameras or the like. The measuring devices may be adapted to determine the coating thickness or the humidity of the coating material in one or more of the dryer units 17, 18, 19. The sensors and / or measuring devices may also be adapted to acquire data about the surroundings of the production system 1, such as temperature or humidity.

[0047] Each of the sensors 31, 32, 33 and the measuring means is communicatively connected to a device 40, which comprises a processor PR. The measured data MD acquired by the sensors 31, 32, 33 and the measuring means are transferred to the device 40 and stored in a database 50, which is connected to the device 40.

[0048] In order to be able to track the production history of the continuous web 5, a one-time preparatory step must be performed in which information is collected about the spatial position and other measured properties of each of the sensors 31, 32, 33 and the measuring devices along a given path P. This preparatory step consists of determining a one-time reference calibration for measuring / determining the spatial position of the sensors 31, 32, 33 and the measuring devices along the curved path P.

[0049] First, a spatial mark (not shown) is attached to the continuous web 5. The spatial mark may be removable or may be generated by a semi-invasive technique, for example by applying a laser code print mark on an edge on the continuous web 5. Next, the web 5 is moved on a path P at a given speed during this reference calibration. Preferably, the speed is constant during the calibration process. In order to be able to obtain accurate calibration information, it is preferred to move the web 5 at a slow speed (i.e., at a speed that is less than the usual movement speed of the web 5 during normal operation of the production system 1). A timestamp is acquired each time a certain sensor position is crossed. The time interval between two adjacent positions can then be converted into absolute coordinates of the first frame MF (machine frame) or relative spatial distance using the predefined first origin (marker) 24 already mentioned.

[0050] Figure 2 This setup and definition is illustrated in a simplified manner. Figure 2At the upper part of the , the deployment path UP along its length dimension is illustrated. Along the deployment path UP, the subsystems 10 of the production system 1 as well as the sensors 31, 32, 33 and the measuring devices are arranged. Figure 2 In the illustration of , for reasons of simplicity, only the unwinding coil 11, the coating device 15 and the sensors 31, 32, 33 are shown. The sensor 33 corresponds to the above-mentioned sensor at the end of the path P. As already mentioned elsewhere, the fixed mark 24 observable in the field of vision of the sensor 33 constitutes a first origin of the first frame MF (machine frame).

[0051] Each of the sensors and subsystems has a corresponding position P0, P1, P2, wherein if the corresponding position of a certain sensor or measuring device is crossed, a timestamp is acquired. Figure 2 , position P2 is crossed at time T_0. Position P1 of sensor 32 is crossed at time t=T_1. Since the speed of the moving web 5 is known, the distance d between each two adjacent positions can be determined. The distance d between positions P1 and P2 is denoted d(T_1, T_0). Starting from the predefined first origin 24, the spatial position of each sensor 31, 32, 33 and the measuring device can be obtained.

[0052] The spatial positions are converted into absolute coordinates p(T_0), p(T_1) relative to the origin of the marking 24. The projected position P0, P1, P2 of each sensor 31, 32, 33 and measuring device along the path is then stored in the database 50.

[0053] In order to be able to track the production-related feature PRF of the web 5, a predefined second origin 6 of a second web frame WF corresponding to the web coordinate system of the web 5 is obtained by processing the measured data MD of a certain sensor 31, 32, 33 or measuring device. A practical choice for designing such an origin on the web 5 is the leading edge of the coating area CA, since it is easily identifiable by optical image processing of the measured data MD by the camera (sensor) 33. More generally, the second origin is a visible feature of the web 5.

[0054] exist Figure 2 , an unwound web 5 is illustrated along an unwound path UP at two different times T_0 and T_1 along the time axis t. The coating area is denoted by CA on the web 5. It can be seen from the unwound web 5 at time t=T_0 that the leading edge of the coating area CA passes over the sensor 33 and the mark 24 in the field of view of the sensor 33, respectively. At t=T_1, the web 5 has moved according to the measured conveying speed of the web 5, so that the physical point PPW of the web, which passed the sensor 31 at time t=T_0, now passes over the sensor 32 at t=T_1.

[0055] In order to identify the physical point of the region of the web 5, it must be taken into account that the electrode carrier material usually consists of two sides, namely an upper side 5u and a lower side 5l. Figure 1 In the coating process of the web 5, the two sides are usually processed in succession. The additional identifier s is used to distinguish between the two sides 5u, 51. Furthermore, for some sensor measurements, such as optical images, the position on the web 5 is defined not only by coordinates along the direction of movement (x-axis), but also by coordinates along the orthogonal axis (y-axis).

[0056] The second frame DF refers to the continuous web fully unrolled along the length direction of the continuous web. Each physical point PPW can be identified by a coordinate along the unrolled length dimension x and perpendicular to the length dimension x, and the coordinate perpendicular to the length dimension x extends along the width direction y of the continuous web 5. Therefore, the coordinates in the second frame WF consist of three values ​​x, y, s, as shown in Figure 2 The origin of the second frame thus has coordinates (0, 0, s).

[0057] Computer-implemented tracking of the production history of the coil 5 is performed by performing the following steps at each of one or more points in time during operation of the production system.

[0058] In the first step ( Figure 3 In S1), when the coil 5 is conveyed along the path P (see Figure 1 ) obtains measured data MD of the coil 5. The measured data is the current timestamp measured data MD, which is obtained by the sensors 31, 32, 33 and any measuring devices installed at or near the production system 1 along the path P. The measurement results can include complex data structures, such as RGB images or hyperspectral infrared measurements. They can also include simple scalar values, such as temperature, humidity readings or coating thickness, or any combination thereof. It should be noted that some of the sensors and measuring devices can be arranged in the room where the production system is built. For example, temperature and humidity values ​​can be measured around the production system 1.

[0059] When the continuous web 5 is conveyed along a given path, the movement information MI of the continuous web 5 is obtained. Figure 3 The movement information MI of the web 5 is continuously measured, for example, via an angular encoder or a barcode scanning system. The movement information is also stored as high-resolution time series data. This allows the calculation of the distance traveled by the web 5 along the path P at any time.

[0060] With these prerequisites, the production-related characteristics PRF ( Figure 3The determination is performed by processing the acquired time-stamped measured data MD and the movement information MI of the web 5 by mapping the time-stamped measured data MD to physical points or areas of the continuous web. This mapping corresponds to a coordinate transformation between the coordinates of the first frame MF and the coordinates of the second frame WF.

[0061] Thus, a virtual electrode coil representation may be achieved, which contains the production-related features PRF of the physical points of the areas of the continuous coil 5. The production-related features PRF may be stored in the database 50.

[0062] To do this, for each physical point or area of ​​points on the web 5, the moment of crossing the position of the corresponding sensor 31, 32, 33 or measuring device along the movement direction of the web 5 can be calculated. This can be achieved by applying the measurement of the movement as a function of time. Next, the previously acquired time series measured data MD can be queried to obtain the characteristics associated with the corresponding sensor 31, 32, 33 or measuring device. Finally, for each point or area on the continuous web 5, the group of retrieved features is stored in the database 50. The database structure can be centralized or "shop-floor-level". The spatial position or spatial interval plus the (upper / lower) side information can be used as a unique identifier.

[0063] The mapping of the movement of the web 5 along the path P as observed from the web frame is performed as follows: First, an origin T0 along the time axis t is introduced. T0 is conveniently defined as the moment when the origin 6 of the x-axis in the web frame WF (i.e., the leading edge of the coating area CA) coincides with the origin 24 along the reel path in the machine frame MF (i.e., the position of the mark 24 in the field of view of the sensor 33 at the end of the path P). At this moment, all coordinate values ​​of x correspond to positions along the path P.

[0064] Assuming that a measurement of the web velocity v(t) in the machine frame MF is available, by using a sensor measuring the velocity of the web 5 it is possible to calculate how much the web 5 has moved in the time interval Δt=t-T0:

[0065]

[0066] Therefore, for any moment, the position of a point x0, y0, s0 of the web 5 along the path P can be calculated by the following formula:

[0067]

[0068] Therefore, assuming Since by definition the x-axis and the path P overlap at t = T0. In practice, the integral of the velocity over time can be approximated by summing discrete measurements of the velocity v(t) of the web.

[0069] In order to map features such as temperature or camera images according to the above calculations, the positions of the corresponding sensors 31 , 32 , 33 and the measuring device relative to the origin of the path P are used. As described above, a calibration run is used to calculate relative distances.

[0070] Using the above method, production features can be recorded and mapped onto the electrode coil within each dedicated process. Coil positions or fragments can be (re)identified and tracked across different processes in the following manner. Assuming that a set of features have been correctly mapped to physical points or regional points of the coil 5 in a certain production step A) (e.g., coating as described above), the specific fragment can be re-identified by applying a similar procedure along a different production step B (e.g., calendering) and then matching the resulting feature vectors. In particular, given the characterized spatial positions and corresponding feature vectors of step B, the spatial positions that show the highest similarity in the corresponding feature field during step A can be searched. This solution can be used as a refinement or fine calibration of physical positions across different processes, while other solutions (e.g., laser marking as defined positions) can only provide coarse spatial resolution. However, laser marking can simplify the search complexity in feature space by providing reasonable position spacing restrictions.

[0071] Via Bayesian methods (such as "Kalman filters" or "particle filters"), it is possible to improve the accuracy of position determination or (re)identification of feature patterns to provide a link between the frame production process: starting from rough prior knowledge, such as the approximate position of the web segment according to the printed barcode, each additional feature provides further information to recalibrate the matching scores between the web segments observed at different stages along production, ultimately allowing positioning accuracy in the range of a few centimeters or better.

Claims

1. A computer-implemented method for tracking the production history of a continuous web (5), in particular a continuous electrode carrier material, which is processed in a production system (1), the production system (1) comprising at least one subsystem (10), the at least one subsystem (10) being configured to perform one or more specific operations, wherein: The continuous web (5) is conveyed along a given path (P) through the production system (1) by unwinding from / winding onto a coil (11, 23), wherein the following steps are performed at each of one or more points in time during the operation of the production system (1): a) obtaining measured data (MD) of the continuous web (5) while the continuous web (5) is being conveyed along the given path (P), the measured data (MD) being current time-stamped measured data (MD) acquired by at least one of sensors (31, 32, 33) or measuring devices installed at or near the production system (1) along the given path (P), wherein the spatial position of each sensor (31, 32, 33) and measuring device relative to the given path (P) has been determined by a one-time reference calibration; b) obtaining movement information (MI) of the continuous web (5); c) processing the time-stamped measured data (MD) and the movement information (MI) of the continuous web (5) by mapping the time-stamped measured data (MD) to physical points of areas of the continuous web (5) to determine production-related characteristics (PRF) of the continuous web (5).

2. The method according to claim 1, wherein: The spatial position of each sensor (31, 32, 33) and measuring device relative to the given path (P) is determined by means of a one-time reference calibration, in which the following steps are performed: a1) attaching a space mark on the continuous web (5); a2) moving the continuous web (5) at a given speed during calibration; a3) obtaining a timestamp whenever a position (P0, P1, P2) of a sensor (31, 32, 33) or a measuring device is crossed; a4) using a predefined first origin (24) the time interval between two adjacent positions (P0, P1, P2) is converted into absolute coordinates (p(T_0), p(T_1)) or relative spatial distances of a first frame (MF), wherein the first frame (MF) is the coordinate system of the production system (1).

3. The method according to claim 2, wherein: The first origin (24) corresponds to a portion of the production system (1), optionally marked with a unique marker placed within the field of view of an optical sensor (31, 32, 33).

4. The method according to claim 2 or 3, wherein: The first origin (24) corresponds to a fixed mark observable in the field of view of the sensor (33) positioned at the end of the given path (P).

5. The method according to claim 2, wherein: The projected position (P0, P1, P2) of each sensor (31, 32, 33) and measuring device along the given path (P) is stored.

6. The method according to one of the preceding claims, wherein: A predefined second origin (6) of a second frame (WF) of the continuous web (5) is obtained by processing measured data (MD) of a sensor (31, 32, 33) or a measuring device, wherein the second frame (WF) is a web coordinate system.

7. The method according to claim 6, wherein: The second origin (6) corresponds to a visible feature of the continuous web (5) that can be observed in the field of view of the sensor (33) positioned at the end of the given path (P).

8. The method according to claim 6 or 7, wherein: The second frame (WF) refers to the continuous web (5) fully unfolded along the length dimension of the continuous web (5), wherein each physical point (PPW) can be identified by a coordinate along the unfolded length dimension (x) and perpendicular to the length dimension (x), and the coordinate perpendicular to the length dimension (x) extends along the width direction (y) of the continuous web (5).

9. The method according to claim 8, wherein: The coordinates (x, y) in the second frame (WF) comprise information (one or more) about the upper side (5u) or the lower side (51) of the continuous web (5).

10. The method according to one of the preceding claims, wherein: In step c) the following steps are performed: c1) for each physical point or point area on the continuous web (5), calculating the moment of crossing the corresponding sensor (31, 32, 33) or measuring device position (P0, P1, P2) along the moving direction; c2) querying said time-stamped measured data (MD) to obtain said characteristics (PRF) associated with said corresponding sensor (31, 32, 33) or measuring device; c3) storing the retrieved set of said features (PRF) in a database for each point or area on said continuous web (5).

11. The method according to claim 10, wherein: The spatial position (P0, P1, P2) of the certain sensor (31, 32, 33) or measuring device or the spatial interval between two adjacent sensors (31, 32, 33), and optionally, the (one or more) side information are processed into a unique identifier.

12. The method according to one of the preceding claims, wherein: For a second subsequent production step a method according to one of the preceding claims is carried out, wherein specific physical points (PPW) or point regions on the continuous web (5) are re-identified by matching resulting feature vectors which exhibit the highest similarity scores in their corresponding feature fields.

13. A computer-implemented device for tracking the production history of a continuous web (5), in particular a continuous electrode carrier material, which is processed in a production system (1), the production system (1) comprising at least one subsystem (10), the at least one subsystem (10) being used to perform one or more specific operations, wherein: The continuous web (5) is conveyed along a given path (P) through the production system (1) by unwinding from / winding onto a coil (11, 23), wherein the device (40) comprises a processor (PR) configured to execute a method comprising the following steps: a) obtaining measured data (MD) of the continuous web (5) while the continuous web (5) is being conveyed along the given path (P), the measured data (MD) being current time-stamped measured data (MD) acquired by at least one of sensors (31, 32, 33) or measuring devices installed at or near the production system (1) along the given path (P), wherein the spatial position of each sensor (31, 32, 33) and measuring device relative to the given path (P) has been determined by a one-time reference calibration; b) obtaining movement information (MI) of the continuous web (5); c) processing the time-stamped measured data (MD) and the movement information (MI) of the continuous web (5) by mapping the time-stamped measured data (MD) to physical points of areas of the continuous web (5) to determine production-related characteristics (PRF) of the continuous web (5).

14. The device according to claim 13, wherein: The device (40) is configured to perform the method according to one of claims 2 to 12.

15. A production system (1), comprising at least one subsystem (10), wherein the at least one subsystem (10) is used to perform one or more specific operations, wherein: The continuous web (5) is conveyed along a given path (P) through the production system (1) by being unwound from / wound onto a coil (11, 23) and thereby passing through at least one of sensors or measuring devices, each of which is suitable for acquiring measured data (MD), wherein the production system (1) comprises an apparatus according to claim 13 or 14.

16. A computer program product having a program code stored on a non-transitory machine-readable carrier for performing the method according to one of claims 1 to 12 when the program code is executed on a computer.

Images