Method for determining at least one area of ​​the surface of a glass plate

Abstract

Method for Determining At Least One Area of ​​the Surface of a Glass Tray. The invention relates to a method for determining at least one area of ​​the surface of a glass tray. This method comprises the steps of: - obtaining (E10) a plurality of cutting planes of the glass tray, each cutting plane being optimized to eliminate a maximum number of defects from a set of defects, - for each defect, determining (E20) an indicator representing the number of cutting planes capable of eliminating said defect, - discretizing (E30) the surface of the glass tray into a plurality of slabs, - for each slab and according to the determined indicators, determining (E40) a score representing the capacity to eliminate a defect located in said slab, - determining (E50) at least one area of ​​the surface of the glass tray according to the scores respectively associated with the slabs forming said area. Figure for the abstract: Fig. 3

Description

Title of the invention: Method for determining at least one area of ​​the surface of a glass plate Prior art

[0001] The present invention belongs to the general field of cutting pieces of glass from glass trays.

[0002] The invention relates more particularly to a method for determining at least one area of ​​the surface of a glass plate, a first method for cutting at least one glass plate from a continuous ribbon of glass from at least one area thus determined, and a second method for cutting a given set of pieces of glass from at least one glass plate obtained using the first cutting method. The invention also relates to systems configured to implement these methods.

[0003] Glass is traditionally manufactured in the form of a continuous ribbon of glass called "float" (or "float" in English). This ribbon is then cut into glass plates called "motherglass" (literally "mother glass" in French, even if this term is not used in practice), this cutting operation being still known under the name "quarrel".

[0004] These are, for example, large glass trays called "PLF" (acronym for the expression "Large Format Glass Trays"), typically measuring 3.21 m by approximately 6 m, or even smaller glass trays called "DLF" measuring approximately 2.55 m by 3.21 m.

[0005] A defect analysis step may be performed prior to this cutting to verify whether the glass ribbon complies with specifications in terms of the presence of defects on the glass. If there are out-of-specification defects, the glass plates may then be cut by excluding a certain length of the ribbon corresponding to the out-of-specification portion of the ribbon.

[0006] This way of proceeding is nevertheless far from satisfactory insofar as it tends to significantly reduce the glass production yield sought by glass plate manufacturers.

[0007] Therefore, in order to maintain good performance for glass tray manufacturers while minimizing the presence of defects in the pieces (substrates) of glass, also called "primitives", intended to be cut from each of the glass trays, alternative solutions have been proposed. These consist first of all in limiting the elimination of defects during the cutting of glass trays, and are also based on the determination of cutting plans making it possible to obtain the primitives in question and which are optimized to eliminate a maximum of defects when cutting said primitives.

[0008] It is understood that by "elimination of a defect" in the case of cutting primitives, we are here conventionally referring to the fact that a cutting plane allows the defect to be placed in a piece of glass rather than in one of the primitives.

[0009] By way of example, document WO 2014 / 128424 describes a method for cutting pieces of glass in at least one glass plate in which such optimized cutting plans are generated and used. The optimization of a cutting plan according to this method can take into account different parameters, including in particular parameters specific to the defects present on the glass plates, such as typically the position, size or type of defects. Cutting plans thus generated can for example be combined with a plurality of glass plates so as to achieve an acceptable number of cut primitives while reducing the presence of defects in the latter.

[0010] These alternative solutions, although acceptable to glass tray manufacturers, nevertheless place a substantial burden of eliminating defects on the operators (also called "processors") who cut the blanks.

[0011] What is more, these solutions remain relatively complex to implement (and are therefore more expensive) since they require significant computing resources in order to be able to generate the optimized cutting plans, in particular when the glass plates contain a relatively large number of defects that are unacceptable with respect to the production specifications of the primitives, but also because it is necessary to be able to respect certain production rates.

[0012] In practice, an intermediate solution is generally adopted. This consists of allowing, to a certain extent, the presence of defects in the glass plates, which therefore reduces the efficiency of the glass plate manufacturers but nevertheless makes it possible to limit the impact at the level of the processors. This alternative solution is nevertheless far from optimal because there is no guarantee that the defects accepted during the production phase of the glass plates are localized so as to be able to be eliminated during the cutting of the primitives in sufficient proportions to satisfy the production specifications.

[0013] Statement of the invention

[0014] The present invention aims to overcome all or part of the drawbacks of the prior art, in particular those set out above, by proposing a solution which makes it possible to produce glass plates as well as pieces of glass cut from these glass plates in a much more efficient manner than the solutions of the state of the technique.

[0015] To this end, and according to a first aspect, the invention relates to a method for determining at least one area of ​​the surface of a glass plate, said method comprising steps of: - obtaining a plurality of cutting plans for the glass plate, each cutting plan being configured for cutting a given set of pieces of glass in the glass plate, each cutting plan being further optimized to eliminate during cutting a maximum number of defects from a set of defects associated with said cutting plan, - for each defect, determination of an indicator representative of the number of cutting plans capable of eliminating said defect, - discretization of the surface of the glass plate into a plurality of slabs, - for each slab and based on the indicators determined, determination of a score representative of a capacity to eliminate a localized defect in said slab when cutting, - determination of at least one zone of the surface of the glass plate, said at least one zone comprising at least one slab and satisfying at least one criterion consisting in that the score of said at least one zone is greater than a given threshold, the score of a zone being representative of the scores respectively associated with the slabs forming said zone.

[0016] The determination method according to the invention therefore makes it possible to identify one or more zones of the surface of a glass plate, this or these zones having the particularity of being configured in an appropriate manner to eliminate defects at the time of cutting the pieces of glass (primitives).

[0017] This way of proceeding therefore represents a considerable advantage because, once the zone(s) have been determined, it is possible to communicate them to the glass plate manufacturers. The latter can then configure the glass ribbon cutting process so that a maximum number of defects are located in the zone(s) thus identified. In this way, the efficiency of the glass plate manufacturers is very little impacted, because the elimination of defects is taken care of by the processors, it being understood that a very large number of defects will be able to be eliminated by the processors (due to their locations in the identified zones) without this requiring significant resources (in terms of determining the cutting plan in particular).

[0018] In other words, the determination method according to the invention does not aim to seek to optimize the operations of glass tray manufacturers and processors independently of each other, as is the case in the state of the art, but on the contrary allows the optimization of the operations of one to be beneficial to the other, so that the entire production chain (cutting of glass plates, cutting of primitives) is optimized.

[0019] In particular embodiments, the determination method may further comprise one or more of the following characteristics, taken in isolation or in all technically possible combinations.

[0020] In particular embodiments, the combination of the sets of defects respectively associated with the cutting planes forms a distribution of defects distributed homogeneously on the surface of the glass plate.

[0021] In particular embodiments, the pieces of glass to be cut are characterized by a set of parameters comprising: - the respective dimensions of the pieces of glass, - a number of pieces of glass to be cut from the glass tray.

[0022] In particular embodiments, the set of parameters comprises respective quality indicators of the pieces of glass.

[0023] In particular modes of implementation: - at least one parameter in the parameter set comes from a history of parameters used during at least one past glass production campaign, and / or - at least one parameter in the parameter set is a simulated parameter.

[0024] In particular embodiments, the number of slabs is between 1 and 100.

[0025] In particular modes of implementation, the indicator associated with a defect is equal to the number of cutting planes capable of eliminating said defect, the score of a slab corresponding to the sum of the indicators respectively associated with the defects located in said slab.

[0026] In particular embodiments, the step of determining said at least one zone is implemented so as to satisfy yet another criterion consisting in the size of said at least one zone being included in a given interval.

[0027] In particular embodiments, the step of determining said at least one zone is implemented so as to satisfy yet another criterion consisting in the number of zones determined being less than a given threshold.

[0028] In particular modes of implementation, the step of determining said at least one zone comprises an optimization of a cost function parameterized by said at least one criterion.

[0029] According to a second aspect, the invention relates to a method for cutting at least one glass plate from a continuous ribbon of glass, said cutting method being carried out from at least one determined area in accordance with a method of cutting termination according to the invention, and in which the cutting of said at least one glass plate is carried out so that the number of defects located in said at least one zone is greater than a given threshold.

[0030] According to a third aspect, the invention relates to a method of cutting a given set of pieces of glass in at least one glass plate obtained by a method of cutting at least one glass plate according to the invention.

[0031] According to a fourth aspect, the invention relates to a computer program comprising instructions for implementing steps of: - a determination method according to the invention, or - a method of cutting at least one glass plate according to the invention, or - a method of cutting a given set of pieces of glass according to the invention, when said computer program is executed by a computer.

[0032] This program may use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0033] According to a fifth aspect, the invention relates to a computer-readable information or recording medium on which a computer program according to the invention is recorded.

[0034] The information or recording medium may be any entity or device capable of storing the program. For example, the medium may comprise a storage means, such as a ROM, for example a CD ROM or a microelectronic circuit ROM, or a magnetic recording means, for example a hard disk.

[0035] On the other hand, the information or recording medium may be a transmissible medium such as an electrical or optical signal, which may be conveyed via an electrical or optical cable, by radio or by other means. The program according to the invention may in particular be downloaded from a network such as the Internet.

[0036] Alternatively, the information or recording medium may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the method in question.

[0037] According to a sixth aspect, the invention relates to a system comprising means configured to implement: - a determination method according to the invention, or - a method of cutting at least one glass plate according to the invention, or - a method of cutting a given set of pieces of glass according to the invention.

[0038] Brief description of the drawings

[0039] Other characteristics and advantages of the present invention will emerge from the description given below, with reference to the appended drawings which illustrate an exemplary embodiment thereof without any limiting character. In the figures:

[0040] [Fig-1] [Fig.l] schematically represents a system configured to determine one or more areas of the surface of a glass plate according to a particular embodiment of the invention;

[0041] [Fig.2] [Fig.2] schematically represents an example of hardware architecture of a processing device belonging to the system of [Fig.l];

[0042] [Fig.3] [Fig.3] represents, in the form of a flowchart, a particular mode of implementation of a method for determining at least one zone according to the invention, as executed by the processing device of [Fig.2];

[0043] [Fig.4] [Fig.4] schematically represents an example of discretization of a glass plate by means of rectangular slabs, as envisaged in the determination method of [Fig.3];

[0044] [Fig.5] [Fig.5] schematically represents two areas of the glass plate of [Fig.4], as determined by means of a specific example of implementation of the determination method of [Fig.3];

[0045] [Fig.6] [Fig.6] schematically represents a particular embodiment of a system for cutting at least one glass plate according to the invention;

[0046] [Fig.7] [Fig.7] schematically represents an example of hardware architecture of a processing device belonging to the system of [Fig.6];

[0047] [Fig.8] [Fig.8] represents, in the form of a flowchart, a particular mode of implementation of a cutting method according to the invention, as executed by the system of [Fig.6].

[0048]

[0049] Description of embodiments

[0050] [Fig.l] schematically represents a particular embodiment of a system SYS_1 according to the invention, said system SYS_1 being configured to determine one or more zones of the surface of a glass plate.

[0051] It should be noted that the determination of the zone(s) is carried out here in an automated manner and using means configured in software and hardware, as described in more detail later. Therefore, it should be understood that the “glass plate” referred to for the determination of the zone(s) is in fact a digital model of a glass plate (i.e. a modeled glass plate). The fact remains that this digital model takes into account all the classic characteristics of a physical glass top, so that, in what follows and with regard to the determination of the zone(s), only a "glass top" (and not a "modeled glass top") is referred to without this causing confusion.

[0052] For the remainder of the description, it is considered in a non-limiting manner that the glass top is of the “PLF” type. These provisions are however not limiting of the invention, and nothing excludes for example the consideration of a “DLF” type glass top. In any event, whatever the type of glass top, it corresponds to the tops conventionally produced by cutting (by means of a guillotine) in a continuous ribbon of glass (i.e. float glass or “float” glass in English).

[0053] In the present embodiment, the glass plate itself is intended to be cut according to a plurality of cutting planes PL_DEC_{i,j], i (respectively j) being an integer index between 1 and M (respectively between 1 and N). No limitation is attached to the values ​​of the numbers M and N. As a non-limiting example, the numbers M and N can both be set to 100.

[0054] Each cutting plane PL_DEC_{i,j} is configured for cutting a given set of glass pieces ENS_PRIM_{i,j] (primitives) in the glass plate. In the present case (i.e. with regard to the determination of the zone(s)), and with regard to the elements discussed above, the cutting of the glass pieces is a virtual cutting carried out using said cutting planes PL_DEC_{i,j}. In addition, each cutting plane PL_DEC_{i,j] (i.e. for fixed indices i and j) is optimized to eliminate a maximum number of defects during cutting, these aspects being described in more detail later.

[0055] It is important to note that said cutting planes PL_DEC_{i,j] are distinct from each other. As regards the sets of pieces of glass ENS_PRIM_{i,j] respectively associated with the cutting planes PL_DEC_{i,j], these may also be distinct from each other in whole or in part.

[0056] No limitation is attached to the industrial application for which the pieces of glass are intended. For example, all or part of the pieces of glass can be used in the manufacture of automotive glazing, glazing for solar applications, for example photovoltaic, glazing for a projection screen, for example an OLED screen, glazing for a mirror or even building glazing.

[0057] In a manner known per se, the pieces of glass of the set ENS_PRIM_{i,j] are characterized by a set of parameters ENS_PAR_{i,j] from which it is possible to derive characteristics (geometric, qualitative, etc.) for the glass plates (and therefore a fortiori for the modeled glass plate considered in the framework for determining the area(s) intended for the production of said pieces of glass.

[0058] In the embodiment described here, said set ENS_PAR_{i,j} of parameters comprises: - the respective dimensions of the pieces of glass in the set ENS_PRIM_{i,j], - a number of pieces of glass to be cut from the glass plate.

[0059] The dimensions make it possible in particular to define the respective shapes of the pieces of glass. Regarding these aspects, it should be noted that the invention is not limited by the shapes of the pieces of glass to be cut from the glass panel. Thus, a piece of glass may have the shape of a rectangle, a polygon, a circle, an ellipse, or any other more complex shape. Depending on the shape considered for a piece of glass, the associated dimensions may refer to lengths / widths / thicknesses, etc.

[0060] The invention is not further limited by the number of pieces of glass that can be envisaged for cutting from the glass plate, it being understood that this number is nevertheless limited above as a function of the respective dimensions of the glass plate and the pieces of glass to be cut.

[0061] Furthermore, nothing precludes considering other parameters in addition to the dimensions of the pieces of glass as well as their number. For example, the set of parameters ENS_PAR_{i,j] characterizing the pieces of glass to be cut may also include respective quality indicators of the pieces of glass. Such quality indicators may for example be representative of criteria for accepting defects in each of the pieces of glass, these acceptance criteria being able to differ from one piece of glass to another. By way of illustration, a defect which is for example not acceptable in the center of a piece to be cut may for example be unacceptable at its periphery, and vice versa for another piece of glass.

[0062] It should be noted that the term "defect" here conventionally refers to an imperfection of the glass. An imperfection may for example be a "pinhole" type defect (a defect on the coating), a bubble defect, a scratch defect on the glass, a surface defect, a thickness defect, a layer defect, etc. In addition, an imperfection may be acceptable in certain cases and not in other cases, for example depending on the application envisaged for the pieces of glass (i.e. depending on the specifications required for the glazing that one wishes to produce). Glass plates intended for the field of semiconductors will for example be more sensitive to surface defects, while in the field of automotive or building glazing, transmission or reflection defects will be more significant.

[0063] In the present embodiment, the set of parameters ENS_PAR_{i,j} de ending the parameters of the set of glass pieces ENS_PRIM_{i,j} comes more particularly from a history of parameters used during at least one past glass production campaign.

[0064] Considering previously used parameters is advantageous insofar as the determination of the zone(s) in the glass plate using the SYS_1 system takes place in a real operating and production context.

[0065] It should be noted that a parameter of the set ENS_PAR_{i,j} can for example be directly equal to a parameter of the parameter history, or be the result of a statistical calculation applied to parameters of the history.

[0066] It should also be noted that the parameter history, and therefore a fortiori the set of parameters ENS_PAR_{i,j], can be stored by any known storage means. As a non-limiting example, and as illustrated by [Fig.l], it can be a database BDD belonging to the system SYS_1.

[0067] Although it is considered here that the parameters of the set ENS_PAR_{i,j] are all derived (directly or indirectly) from the parameter history, other variants can still be considered. For example, the parameters of the set ENS_PAR_{i,j] can be simulated parameters.

[0068] Such an alternative is advantageous in that it makes it possible to envisage determining one or more zones in the glass plate following an exploratory approach, i.e. without any link to past operating and production conditions.

[0069] Finally, it is also understood that nothing excludes having a configuration in which only part of the parameters of the set ENS_PAR_{i,j] comes from the history, the remaining part coming from simulated parameters. In particular, no limitation is attached to the cardinal of these two parts.

[0070] In the present embodiment, and in addition to the database BDD, the system SYS_1 also comprises a processing device DISP_1. Said processing device DISP_1 is configured to carry out processing operations aimed at determining one or more zones of the glass plate, by implementing steps of a method for determining said zone(s).

[0071] [Fig.2] schematically represents an example of hardware architecture of the processing device DISP_1 according to the invention.

[0072] As illustrated by [Fig.2], the processing device DISP_1 has the hardware architecture of a computer. Thus, such a processing device DISP_1 comprises, in particular, a processor 1_1, a random access memory 2_1, a read only memory 3_1 and a non-volatile memory 4_1. It also has communication means 5_1.

[0073] The read-only memory 3_1 of the processing device DISP_1 constitutes a support recording according to the invention, readable by the processor 1_1 and on which is recorded a computer program PROG_1 according to the invention, comprising instructions for executing steps of the determination method according to the invention.

[0074] The program PROG_1 defines functional modules of the processing device DISP_1, which rely on or control the hardware elements 1_1 to 5_1 of the processing device DISP_1 mentioned above. These functional modules are illustrated in [Fig.l] in a non-limiting manner, and are described in more detail below with reference to particular modes of implementation of the determination method.

[0075] The communication means 5_1 allow the processing device DISP_1 to receive data, in particular the set of parameters ENS_PAR_{i,j] defining the parameters of the set of pieces of glass ENS_PRIM_{i,j], from the database BDD. These communication means 5_1 rely, in a manner known per se, on a communication interface capable of exchanging data between the processing device DISP_1 and the database BDD. No limitation is attached to the nature of this communication interface, which may be wired or wireless, so as to allow the exchange of data according to any protocol known to the person skilled in the art (Ethernet, Wifi, Bluetooth, 3G, 4G, 5G, Modbus, TCP-IP, analog, etc.).

[0076] In its general principle, the determination method according to the invention aims to identify on the surface of the glass plate at least one zone in which, for each of the sets of given pieces of glass ENS_PRIM_{i,j] intended to be cut from said glass plate, there is a sufficiently large number of defects which can be eliminated during cutting (via one or more appropriate cutting planes).

[0077] Knowing how to identify such areas represents a considerable advantage because, once this information is obtained, it is possible to communicate it to the glass plate manufacturers. The latter can then configure the glass ribbon cutting process so that a maximum number of defects are located in the areas thus identified. In this way, the efficiency of the glass plate manufacturers is very little impacted, because the elimination of defects is taken care of by the processors, it being understood that a very large number of defects will be able to be eliminated by the processors (due to their locations in the identified areas) without this requiring significant resources (in terms of determining the cutting plan in particular).

[0078] [Fig.3] represents, in the form of a flowchart, a particular mode of implementation of the determination method according to the invention, as executed by the device DISP_1 of [Fig.2],

[0079] As mentioned above, the execution of the determination method is carried out for the sets ENS_PRIM_{i,j] of pieces of glass, whose associated parameters ENS_PAR_{i,j] are stored in the database BDD.

[0080] For the description of the mode of [Fig.3], and in order to simplify it, it is considered in a non-limiting manner that the glass plate from which the pieces of glass ENS_PRIM_{i,j] can be cut has already been modeled. The data useful for modeling the glass plate have in particular been determined from the parameters ENS_PAR_{i,j}.

[0081] The modeling of the glass plate was for example executed by the processing device DIPS_1. In other words, in this example, the processing device DISP_1 obtained, during a data exchange with the database BDD, the parameters ENS_PAR_{i,j] and then carried out the modeling in question. This data exchange is carried out using the communication means 5_1, for example being initiated by the processing device DISP_1 by means of an appropriate request.

[0082] Alternatively, a third-party entity was responsible for the modeling, then transmitted the modeled tray to the processing device DISP_1 (typically in the form of a computer file).

[0083] In any event, when the process of [Fig.3] begins, the processing device DISP_1 stores the modeled tray in memory, for example in its non-volatile memory 4_1.

[0084] Although obtaining the modeled tray by the processing device DISP_1 is not described here as being an integral part of the method for determining [Fig.3], it should however be noted that all or part of the steps leading to this obtaining (transmission of the parameters ENS_PAR_{i,j], modeling of the glass tray, transmission of the modeled tray) can be integrated into said method according to other modes of implementation not detailed here.

[0085] As illustrated by [Fig.3], the determination method firstly comprises a step E10 of obtaining the plurality of cutting planes PL_DEC_{i,j] of the glass plate.

[0086] As mentioned above, each cutting plan PL_DEC_{i,j] (i.e. for fixed indices i and j) is optimized to eliminate a maximum number of defects during cutting from a set of defects ENS_DEF_{i,j] associated with said cutting plan.

[0087] For the remainder of the description, we adopt the notation DEF_{i,j,k] (k being an integer index) to designate a given defect within the set ENS_DEF_{i,j}. The position of the defect DEF_{i,j,k] on the glass plate is denoted POS_DEF_{i,j,k}.

[0088] The number of defects considered within the set ENS_DEF_{i,j] does not constitute a limitation of the invention. In addition, two sets of defects ENS_DEF_{i_l,j_l} and ENS_ DEF_{i_2,j_2} may have the same cardinality or not, and may or may not have empty intersection.

[0089] More particularly, in the implementation mode of [Fig.3], obtaining the cutting plans PL_DEC_{i,j] corresponds to a determination of the latter by the processing device DISP_1, by means of appropriate digital (algorithmic) processing. For this purpose, said step E10 is implemented by a first determination module MOD_DET_1 equipping the processing device DISP_1.

[0090] The digital processing operations are, for example, those described in the document WO 2014 / 128424 already mentioned previously. As a reminder, and in summary, the procedure taught by the document WO 2014 / 128424 describes implementation modes in which generic cutting plans PL_DEC_GEN_{1},..., PL_DEC_GEN_{M] are first determined. These generic cutting plans PL_DEC_GEN_{1},..., PL_DEC_GEN_{M] are optimized plans of the glass plate when the latter is considered to be defect-free. Furthermore, in order to simulate the presence of defects, random defect maps MAP_RAND_{1},..., MAP_RAND_{N] are considered which are combined with the glass plate, each defect map MAP_RAND_{j} making it possible to simulate a distribution of defects on the surface of the glass plate. Each cutting plane PL_DEC_{i,j] (i.e.for indices i and j fixed between 1 and M as well as 1 and N respectively) then results from an optimization procedure of the generic cutting plan PL_DEC_GEN_{i] for which the defect map MAP_RAND_{j] has been taken into account (the set of defects ENS_DEF_{i,j] therefore corresponding to the association of the defects of the map MAP_RAND_{j] with said generic cutting plan PL_DEC_GEN_{i}).

[0091] In addition, the randomness associated with the distribution of defects in the defect maps MAP_RAND_{j} can be controlled, so that the combination of said sets of defects ENS_DEF_{i,j] conforms to a distribution of a particular type. Thus, according to a preferred example of implementation, the combination of the sets of defects ENS_DEF_{i,j] forms a distribution of defects distributed homogeneously on the surface of the glass plate.

[0092] Finally, if the obtaining of the cutting plans is described here as being the implementation of processing such as those detailed in the document WO 2014 / 128424, nothing excludes the consideration of other alternatives. For example, the term “obtaining” may refer to a transmission of the cutting plans PL_DEC_{i,j], it being understood that these would have been determined prior to the execution of the determination method by another entity. In this case, the obtaining step would be implemented using the communication means 5_1.

[0093] Generally speaking, no limitation is attached to the way in which the processing device DISP_1 obtains said cutting plans PL_DEC_{i,j], these aspects being known to the person skilled in the art.

[0094] Therefore, once the cutting plans PL_DEC_{i,j] have been obtained, and as illustrated by [Fig.3], the determination method comprises a step E20 of determining, for each defect DEF_{i,j,k], an indicator IND_DEF_{i,j,k] representative of the number of cutting plans capable of eliminating said defect DEF_{i,j,k}. Said step E20 is implemented by a second determination module MOD_DET_2 equipping the processing device DISP_1.

[0095] Said indicator IND_DEF_{i,j,k] therefore corresponds to a metric making it possible to quantify a capacity for eliminating the defect DEF_{i,j,k] (located according to the position POS_ DEF_{i,j,k] on the glass plate) when cutting the glass plate to obtain the pieces of glass of the set ENS_PRIM_{i,j}.

[0096] More particularly, in the implementation mode described here, the indicator IND_DEF_{i,j,k] associated with a defect DEF_{i,j,k] is equal to the number of cutting planes capable of eliminating said defect DEF_{i,j,k}.

[0097] However, considering such a value of the indicator IND_DEF_{i,j,k] only constitutes a variant of implementation of the invention, other variants being conceivable. For example, the indicator IND_DEF_{i,j,k] can be expressed in the form of a ratio between the number of cutting plans capable of eliminating said defect DEF_{i,j,k] and the total number of cutting plans PL_DEC_{i,j] (i.e. this total number here being equal to M x N). According to yet another variant, this ratio can be expressed in the form of a percentage.

[0098] Once the indicators IND_DEF_{i,j,k] have been determined for each of the defects DEF_{i,j,k], the determination method comprises a step E30 of discretizing the glass plate into a plurality of slabs TIL_{p] (p being an integer index greater than 1). Said step E30 is implemented by a discretization module MOD_DISC equipping the processing device DISP_1.

[0099] Said discretization amounts to making a partition (a mesh) of the surface of the glass plate by said TIL_{p} slabs.

[0100] In the present embodiment, the TIL_{p] slabs used for the discretization of the surface of the glass plate are rectangular slabs. More specifically, it is considered here, without limitation, that the width and the length of the glass plate are both divided by 10, so as to obtain a discretization of the surface of the glass plate by 100 slabs.

[0101] However, considering rectangular slabs only constitutes a variant implementation of the invention. Generally speaking, no limitation is attached to the respective shapes of said slabs, nor even to their respective sizes, since they make it possible to create a partition (mesh) of the surface of the glass plate. Furthermore, nothing excludes having a mixture of different shapes.

[0102] It should also be noted that the slabs may be of different or identical respective sizes, in whole or in part. Finally, the number of slabs does not constitute a limitation of the invention. Thus, the number of slabs may, for example, be between 1 and 100.

[0103] It is of course understood that the number of slabs can be chosen according to different criteria. Thus, such a criterion can for example concern a calculation time to check whether or not a defect belongs to a slab (it being understood that the calculation time in question must ideally be short in order to be able to quickly make a decision as to whether or not to cut to the square given that the strip arrives continuously). In addition or as an alternative, a criterion that can be taken into account to decide on the number of slabs can concern the size of the latter (e.g. 10x10cm or 5x5cm) in order to ensure minimal waste for the transformer.

[0104] As illustrated by [Fig.3], the determination method then comprises a step E40 of determining, for each slab TIL_{p] and as a function of the indicators IND_DEF_{i,j,k] determined during step E20, a score SCO_TIL_{p] representative of a capacity to eliminate a localized defect in said slab when cutting. Said step E40 is implemented by a third determination module MOD_DET_3 equipping the processing device DISP_1.

[0105] The SCO_TIL_{p] score therefore makes it possible to quantify the capacity to eliminate localized defects in the TIL_{p} slab.

[0106] It is of course understood that the determination of the score SCO_TIL_{p] associated with a TIL_{p] slab depends on the way in which the indicators IND_DEF_{i,j,k} are expressed. In the implementation mode described here, the indicator IND_DEF_{i,j,k] associated with a defect DEF_{i,j,k] being equal to the number of cutting planes capable of eliminating said defect DEF_{i,j,k], the score SCO_TIL_{p] of a TIL_{p] slab corresponds (is equal) to the sum of the indicators respectively associated with the defects located in said TIL_{p} slab.

[0107] In more detail, the determination of the SCO_TIL_{p] score may for example include, initially, an identification of the defects located in the TIL_{p} slab. This location is carried out from the POS_ DEF_{i,j,k] positions respectively associated with the DEF_{i,j,k} defects.

[0108] When the defects located in the TIL_{p] slab have been identified, their respective indicators are then summed to obtain the SCO_TIL_{p} score.

[0109] It is important to note, however, that considering a sum of the indicators respectively associated with the defects located in a slab does not constitute a limitation of the invention. Indeed, nothing excludes considering other modes of implementation in which, for example, the score SCO_TIL_{p] is a statistic by particular (e.g. an average) obtained from the indicators respectively associated with the defects located in the TIL_{p} slab.

[0110] Of course, if the indicators IND_DEF_{i,j,k] are expressed differently, for example in the form of a ratio as mentioned above, the person skilled in the art knows how to adapt the implementation of step E40 so that the score SCO_TIL_{p] makes it possible to quantify a capacity for eliminating localized defects in the slab TIL_{p}.

[0111] [Fig.4] schematically represents an example of discretization of the plateau of glass by means of rectangular slabs, as envisaged in the determination method of [Fig.3].

[0112] In [Fig.4], the indexing of the TIL_{p] tiles is done by rows, starting with the topmost row, and, on each row, going through the tiles from left to right. The tiles TIL_{1}, TIL_{2], TIL_{10], TIL_{11} and TIL_{100} are specifically indicated.

[0113] Furthermore, as illustrated by [Fig.4], each TIL_{p] slab contains a number which corresponds to the SCO_TIL_{p] score determined during step E40.

[0114] As can be seen in [Fig.4], there may be significant differences between the SCO_TIL_{p] scores of the TIL_{p] slabs, even when two slabs are contiguous. This specific distribution of SCO_TIL_{p] scores over the surface of the glass plate is therefore used to identify one or more areas (each area grouping one or more slabs) for which a sufficiently large removal capacity can be determined.

[0115] For this purpose, the determination method comprises a step E50 of determining at least one zone Z_{q] of the surface of the glass plate, said at least zone Z_{q] comprising at least one slab TIL_{p] and satisfying at least one criterion CRIT_1 consisting in that the score SCO_Z_{q] of said at least one zone Z_{q] is greater than a given threshold S_l, the score of a zone being representative of the scores respectively associated with the slabs forming said zone.

[0116] It should be noted that by "zone" we refer here to a portion of the surface of the glass plate which, if it comprises a plurality of slabs, is such that each of said slabs is contiguous to at least one other of said slabs (in other words, a zone is a geometric space of a single piece, that is to say connected).

[0117] In the manner described above with regard to step E40, the score of a zone Z_{q] may correspond to a sum of the scores respectively associated with the slabs forming said zone Z_{q], or to a particular statistic obtained from the scores respectively associated with the slabs forming said zone Z_{q] (e.g. an average, more particularly an average calculated as a ratio between the scores of the slabs considered along a first direction in which the plateau extends and the scores of the slabs considered along a second direction in which the plateau extends, the first direction being perpendicular to the second direction). Following yet another example, it may be the minimum score among the scores respectively associated with the slabs forming said zone Z_{q}.

[0118] No limitation is attached to the value of the threshold S_l. In particular, it may be taken into account that the square cutting can be optimized (as detailed below) by modifying the cutting location of the glass ribbon in the direction perpendicular to the direction of flow of the glass. Consequently, it may be advantageous to determine zones Z_{q] extending more in said perpendicular direction rather than in said direction of flow of the glass. Thus, and with reference to the scores of the slabs TIL_{p] of [Fig.4], the threshold S_1 may for example be set equal to 2800, so as to favor the creation of zones which comprise an entire strip of slabs in the direction perpendicular to the flow of the glass.

[0119] The fact of considering only the criterion CRIT_1 to determine said at least one zone Z_{q] does not constitute a limitation of the invention, and nothing excludes taking into account still other criteria.

[0120] For example, it is possible to take into account (in addition to the criterion CRIT_1) another criterion consisting in that the size of said at least one zone Z_{q] is included in a given interval. By way of illustration, the size of said at least one zone Z_{q] can be determined according to the number of slabs which compose it, this number being able for example to be between 1 and 10. It is of course understood that this number of slabs can also depend itself on the size of the slabs concerned.

[0121] Alternatively or in addition, it is possible to take into account (in addition to the criterion CRIT_1) another criterion consisting of the number of determined zones being less than a given threshold. For example, the number of determined zones may be less than or equal to 4.

[0122] Any method known to the person skilled in the art for determining one or more zones Z_{q] satisfying at least the criterion CRIT_1 may be used. By way of non-limiting example, step E50 of determining said at least one zone Z_{q] may comprise an optimization of a cost function parameterized by the criterion(s) taken into account (and therefore at least by the criterion CRIT_1).

[0123] [Fig.5] schematically represents two zones Z_{1} and Z_{2] of the glass plate of [Fig.4], as determined by means of a specific example of implementation of the determination method of [Fig.3].

[0124] The invention has been described so far by considering only the determination of said at least one zone Z_{q}. That being said, and according to another aspect, the invention also relates to a system SYS_2 for cutting at least one glass plate in a continuous ribbon of glass. [Fig.6] schematically represents a mode par- particular implementation of said SYS_2 system.

[0125] In a manner known per se, the SYS_2 system comprises cutting means MOY_SLICE configured materially to cut at least one glass plate in a continuous ribbon of cooled glass. Said means conventionally comprise a guillotine.

[0126] In addition to said cutting means MOY_SLICE, the system SYS_2 also comprises a control device DISP_2 configured to carry out processing operations making it possible to generate control commands intended for the cutting means MOY_SLICE, by implementing steps of a first method for cutting at least one glass plate. Said commands are more particularly generated so as to cut at least one glass plate according to said at least one zone Z_{q] determined using the determination method described above.

[0127] [Fig.7] schematically represents an example of hardware architecture of the processing device DISP_2 according to the invention.

[0128] As illustrated by [Fig.7], the device DISP_2 has the hardware architecture of a computer. Thus, the device DISP_2 comprises, in particular, a processor 1_2, a random access memory 2_2, a read only memory 3_2 and a non-volatile memory 4_2. It also has communication means 5_2.

[0129] The read-only memory 3_2 of the device DISP_2 constitutes a recording medium in accordance with the invention, readable by the processor 1_2 and on which is recorded a computer program PROG_2 in accordance with the invention, comprising instructions for the execution of steps of the first cutting method according to the invention.

[0130] The program PROG_2 defines functional modules of the device DISP_2, which rely on or control the hardware elements 1_2 to 5_2 of the device DISP_2 mentioned above. These functional modules are illustrated in [Fig.2] in a non-limiting manner, and are described in more detail below with reference to particular modes of implementation of the first cutting method.

[0131] The communication means 5_2 allow the device DISP_2 to receive data, in particular said at least one zone Z_{q], from the processing device DISP_1. These communication means 5_2 rely, in a manner known per se, on a communication interface capable of exchanging data between the device DISP_2 and the processing device DISP_1. No limitation is attached to the nature of this communication interface, which may be wired or wireless, so as to allow the exchange of data according to any protocol known to the person skilled in the art (Ethernet, Wifi, Bluetooth, 3G, 4G, 5G, Modbus, TCP-IP, analog, etc.).

[0132] [Fig. 8] represents, in the form of a flowchart, a particular mode of implementation of the first cutting method according to the invention, as executed by the system SYS_2 of [Fig.6].

[0133] As illustrated by [Fig.8], the first cutting method initially comprises a step F10 of obtaining said at least one zone Z_{q] determined following the execution of the determination method.

[0134] The term “obtaining” here refers to a reception of one or more data files describing said at least one zone Z_{q}. Consequently, said step F10 is implemented by a reception module M0D_RX equipping the device DISP_2 and integrated into the communication means 5_2.

[0135] The reception of the data file(s) describing said at least one zone Z_{q] is for example consecutive to an appropriate request transmitted by the device DISP_2 to the system SYS_1, more particularly to the processing device DISP_1.

[0136] It should be noted that this implementation of step F10 is not limiting of the invention, and nothing excludes the possibility that the device DISP_2 is already in possession of the data file(s) describing said at least one zone Z_{q] when the first cutting process begins, this or these files being able for example to be stored in its non-volatile memory 4_2.

[0137] Therefore, and as illustrated by [Fig.8], the first cutting method comprises a step F20 of generating at least one cutting command COM_SLICE of at least one glass plate in the continuous ribbon of glass as a function of said at least one zone Z_{q}. More particularly, said at least one command COM_SLICE is generated so that the number of defects located in said at least one zone is greater than a given threshold. Said step F20 is implemented by a generation module MOD_GEN equipping the device DISP_2.

[0138] Taking into account said at least one zone Z_{q] to generate said at least one COM_SLICE command is particularly advantageous insofar as said at least one zone Z_{q] has been precisely identified as being an area in which the defect removal capacity is significant. In other words, and with regard to the first cutting method, it is therefore a matter of taking advantage of this knowledge of said at least one zone Z_{q] to generate at least one appropriate COM_SLICE command, so that the cutting of the continuous ribbon of glass allows the placement of a sufficient number of defects in said at least one zone Z_{q}.

[0139] Conventionally, the generation of a cutting command at the canting level constitutes the purpose of a series of steps. Thus, a determined length of the glass ribbon is continuously analyzed to detect the defects present (position and size of the defects). This length is for example equal to 10 meters. Subsequently, depending on a priority of pieces of glass to be produced, a decision is made as to whether or not to cut the glass ribbon. More particularly, we consider firstly the highest priority for production, and it is determined whether a suitable glass plate can be cut by being placed at the end of the ribbon (i.e. it is checked that the cutting of this glass plate respects the imposed defect thresholds). If the cutting is possible, it takes place and the glass ribbon is allowed to run again along the said determined length to restart the operations. If the cutting is not possible for this highest priority, it is evaluated (as detailed previously) whether it can be done for the priority whose order follows that of the said highest priority. The principle is thus iterated, it being understood that if it is determined that no cutting of a glass plate is possible to produce expected pieces of glass, a cutting command is generated in order to produce a piece of cullet.

[0140] With regard to this conventional method of cutting the glass ribbon, the invention advantageously makes it possible to take into account the zone(s) Z_{q] when evaluating the possibility of cutting (for a given production priority) a glass plate. In this way, it is possible to considerably reduce the production of cullet.

[0141] Once said at least one COM_SLICE command has been generated, it is transmitted to the cutting means MOY_SLICE during a step F30. Said step F30 is implemented by a transmission module M0D_TX equipping the device DISP_2 and integrated into the communication means 5_2.

[0142] Upon receipt of said at least one COM_SLICE command by the cutting means MOY_SLICE (step F40), the latter carry out a step F50 of cutting said at least one glass plate in the continuous ribbon of glass (step F50 is referenced “DEC” in [Fig.8]).

[0143] Finally, and according to yet another aspect, the invention also relates to a system SYS_3 (not shown in the figures) for cutting the set ENS_PRIM_{i,j] of pieces of glass in at least one glass plate obtained by means of the first cutting method. To this end, said system SYS_3 is configured in a known manner to implement a second cutting method (not shown in the figures) of the pieces of glass from said set ENS_PRIM_{i,j] in said at least one glass plate glass.

Claims

Claims

1. A computer-implemented method for determining at least one area of the surface of a glass plate, said method comprising steps of: - obtaining (E10) a plurality of cutting plans for the glass plate, each cutting plan being configured for cutting a given set of pieces of glass in the glass plate, each cutting plan being further optimized to eliminate, during cutting, a maximum number of defects from a set of defects associated with said cutting plan, - for each defect, determining (E20) an indicator representative of the number of cutting plans capable of eliminating said defect, - discretizing (E30) the surface of the glass plate into a plurality of slabs, - for each slab and as a function of the determined indicators, determining (E40) a score representative of a capacity to eliminate, during cutting, a defect located in said slab, - determining (E50) at least one area of the surface of the glass plate,said at least one zone comprising at least one slab and satisfying at least one criterion consisting in that the score of said at least one zone is greater than a given threshold, the score of a zone being representative of the scores respectively associated with the slabs forming said zone.,

2. Method according to claim 1, in which the combination of the sets of defects respectively associated with the cutting planes forms a distribution of defects distributed homogeneously on the surface of the glass plate.

3. Method according to any one of claims 1 to 2, in which the pieces of glass to be cut are characterized by a set of parameters comprising: - respective dimensions of the pieces of glass, - a number of pieces of glass to be cut in the glass plate.

4. The method of claim 3, wherein the set of parameters comprises respective quality indicators of the pieces of glass.

5. Method according to any one of claims 3 to 4, in which: - at least one parameter of the set of parameters comes from a history of parameters used during at least one campaign of past glass production, and / or - at least one parameter in the parameter set is a simulated parameter.

6. A method according to any one of claims 1 to 5, wherein the number of slabs is between 1 and 100.

7. Method according to any one of claims 1 to 6, in which the indicator associated with a defect is equal to the number of cutting planes capable of eliminating said defect, the score of a slab corresponding to the sum of the indicators respectively associated with the defects located in said slab.

8. Method according to any one of claims 1 to 7, in which the step (E50) of determining said at least one zone is implemented so as to satisfy yet another criterion consisting in that the size of said at least one zone is included in a given interval.

9. Method according to any one of claims 1 to 8, in which the step (E50) of determining said at least one zone is implemented so as to satisfy yet another criterion consisting in that the number of zones determined is less than a given threshold.

10. Method according to any one of 1 to 9, in which the step (E50) of determining said at least one zone comprises an optimization of a cost function parameterized by said at least one criterion.

11. A method of cutting at least one glass plate from a continuous ribbon of glass, said cutting method being carried out from at least one area determined in accordance with a determination method according to any one of claims 1 to 10, and in which the cutting of said at least one glass plate is carried out so that the number of defects located in said at least one area is greater than a given threshold.

12. A method of cutting a given set of pieces of glass from at least one glass plate obtained by a method according to claim 11.

13. A computer program comprising instructions for implementing steps of: - a method according to any one of claims 1 to 10, or - a cutting method according to claim 11, or - a cutting method according to claim 12, when said computer program is executed by a computer.

14. A computer-readable recording medium on which is en- registered a computer program according to claim 13.

15. System comprising means configured to implement: - a method according to any one of claims 1 to 10, or - a cutting method according to claim 11, or - a cutting method according to claim 12.

Images