METHOD FOR PRODUCING MATERIAL BOARDS IN A PRODUCTION PLANT, PRODUCTION PLANT, COMPUTER-PROGRAM PRODUCT AND USE OF A COMPUTER-PROGRAM PRODUCT

Abstract

A method for producing material boards in a production plant in which apparatuses form a material into a mat that is pressed to obtain the material board which has specific quality parameters. The production plant and/or the apparatuses are controlled in an open- or closed-loop manner by a controller, which preferably includes a programmable logic controller, and input parameters are received, processed and/or output by the controller. The input parameters are formed at least from settable product parameters for the material board to be produced, from settable and/or recorded plant parameters of the production plant and/or the apparatuses and/or from recorded material parameters. A quality value of at least one quality parameter of the material board to be produced is determined based on the input parameters by an algorithm based on artificial intelligence. The algorithm is trained or formed by a database which has at least one quality parameter and input parameters correlating with the quality parameter.

Claims

1. A method for producing a material board (12) in a production installation (1), comprising: forming a material (10) in apparatuses (2, 3, 4, 5, 6, 7, 8, 9) of the production installation (1) into a mat (11) and pressing the mat to obtain the material board (12), which has specific quality parameters, forming the material (10) to obtain the mat (11), controlling or regulating at least one of the production installation (1) or the apparatuses (2, 3, 4, 5, 6, 7, 8, 9) via at least one controller (20), with the controller (20) at least one of receiving, processing, or outputting input parameters (23), forming the input parameters (23) from at least one of variable product parameters for the material board (12) that is to be produced, at least one of variable or acquired installation parameters of at least one of the production installation (1) or the apparatuses (2, 3, 4, 5, 6, 7, 8, 9) or acquired material parameters, and using an artificial intelligence-based algorithm (32) to ascertain a quality value (24) of at least one quality parameter of the material board (12) that is to be produced based on the input parameters (23), the algorithm (32) having been trained, or formed, by a database (30) that includes at least one quality parameter and input parameters correlating with the quality parameter.

2. The method as claimed in claim 1, further comprising acquiring the input parameters (23) with at least one of a time value or position value.

3. The method as claimed in claim 1, further comprising at least one of normalizing or aggregating the input parameters (23).

4. The method as claimed in claim 3, wherein at least one of (a) the normalizing of the input parameters (23) includes normalizing the input parameters over at least one of time or a length of at least one of the production installation (1) or one or more of the apparatuses (2, 3, 4, 5, 6, 7, 8, 9), or (b) the aggregating of the input parameters (23) includes forming clusters within one or more of the apparatuses (2, 3, 4, 5, 6, 7, 8, 9).

5. The method as claimed in claim 1, wherein the database (30) is formed at least from a dataset formed from data relating to the at least one quality parameter and the input parameters correlating with the at least one quality parameter, and the database includes at least an additional dataset, and the dataset differs from the additional dataset of the database (30) in at least one of the quality parameter or the input parameter correlating with the quality parameter.

6. The method as claimed in claim 5, wherein the data or the dataset are at least one of (a) normalized over at least one of time or to at least one of a length of the production installation (1) or one or more of the apparatuses (2, 3, 4, 5, 6, 7, 8, 9) or (b) aggregated by forming data clusters.

7. (canceled)

8. (canceled)

9. The method as claimed in claim 1, further comprising checking the algorithm (32) using a test database (33), the test database (33) includes test datasets containing quality parameters and also input parameters correlating with the quality parameters, and the test datasets differ from datasets of the database (30).

10. (canceled)

11. The method as claimed in claim 1, further comprising first checking an alteration in the input parameters (23) by the algorithm (32) for an effect on the ascertained quality value (24).

12. The method as claimed in claim 1, further comprising optimizing the quality value (24) in an optimization computer (22) of the controller (20) by at least one of the algorithm (32) or at least one other algorithm (34) based on artificial intelligence, which has been formed by training using at least one of the database (30) or another database.

13. The method as claimed in claim 12, wherein the quality value (24) is optimized in the optimization computer (22) by determining and optionally visualizing altered input parameters (28).

14. The method as claimed in claim 12, wherein the optimization of the quality value (24) is performed at least one of before, at a start of production, during production, or after a time interval.

15. The method as claimed in claim 1, further comprising at least one of outputting a warning or switching on the optimization computer (22) upon a difference between the ascertained quality value (24) and a predefined setpoint value, or a setpoint value range, being detected, and/or outputting a message if adherence to the quality value (24) in the setpoint value range given altered the input parameters (28), and displaying the altered input parameters (28) to a user.

16. The method as claimed in claim 1, wherein the algorithm (32, 34) is based on or includes at least one method or a modification thereof from one of the following groups: linear regression, polynomial regression, functional regression, K nearest neighbors regression, random forest regression, support vector regression, neural networks, recurrent neural networks, convolutional neural networks, residual networks, Bayesian networks, example K nearest neighbors classification, decision trees, random forests, naive Bayes, or support vector machines.

17. The method as claimed in claim 1, wherein a physical model is incorporated into the algorithm (32) to ascertain the quality value (24).

18. The method as claimed in claim 1, wherein the algorithm includes multiple algorithms (32, 34) to ascertain the quality value (24) for the at least one quality parameter.

19. The method as claimed in claim 18, wherein the quality values (24) ascertained by the different algorithms (32, 34) are at least one of offset against one another or compared, and a combined result is used as a basis for the optimization.

20. (canceled)

21. The method as claimed in claim 1, further comprising at least one of (a) deriving information relating to at least one of wear or abnormal behavior from the input parameters or forecasting the information with the algorithm (32, 34) for at least one of state monitoring, predictive maintenance detection, or failure detection.

22. The method as claimed in claim 1, further comprising checking the algorithm (32, 34) at least one of before or during ongoing production by comparing the ascertained quality value (24) with a measured quality value of the quality parameter of the material board (12), and if a difference between the ascertained and measured quality value (24) is detected that is above a stipulated threshold value, retraining the algorithm (32, 34).

23. The method as claimed in claim 1, further comprising at least one of transmitting or receiving the algorithm (32, 34), at least one dataset of the production installation (1), and/or the database (30) by an interface (25).

24. (canceled)

25. A production installation (1) for producing material boards (12) that have specific quality parameters, the production installation (1) comprises: apparatuses (2, 3, 4, 5, 6, 7, 8, 9) for forming a mat (11) from a material (10) and pressing said material to obtain material boards (12) and for gluing the material (10) before being formed to obtain the mat (11); a controller (20) configured for controlling or regulating at least one of the production installation (1) or the apparatuses (2, 3, 4, 5, 6, 7, 8, 9), the controller (20) being configured to at least one of receive, process or output input parameters (23); wherein the input parameters (23) are formed from at least one of: variable product parameters for the material board (12) that is to be produced, variable and/or acquirable installation parameters of at least one of the production installation (1) or the apparatuses (2, 3, 4, 5, 6, 7, 8, 9), or acquirable material parameters; and the controller is further configured to use an algorithm (32) based on artificial intelligence to ascertain a quality value (24) relating to at least one quality parameter based on the input parameters (23), the algorithm (32) being trainable, or formable, by a database (30) that includes at least one quality parameter and input parameters correlating with the quality parameter.

26. The production installation (1) as claimed in claim 25, wherein the input parameter (23) has at least one of an attributed time value or position value.

27. (canceled)

28. The production installation (1) as claimed in claim 25, wherein the controller is further configured to at least one of normalize or aggregate the input parameters (23) over time, to a length of the production installation (1) and/or one of the apparatuses (2, 3, 4, 5, 6, 7, 8, 9).

29.-43. (canceled)

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0102] Further advantageous embodiments of the method for producing material boards and of the production installation for producing material boards are presented in the figures that follow, in which:

[0103] FIG. 1 shows a basic representation of a production installation for producing material boards;

[0104] FIG. 2 shows a flow diagram for training an algorithm by means of a database;

[0105] FIG. 3 shows a flow diagram for the method according to the invention for ascertaining and visualizing an ascertained quality value; and

[0106] FIG. 4 shows a flow diagram for optimizing the quality value in the course of the production of material boards.

DETAILED DESCRIPTION

[0107] In the text that follows, identical reference signs denote identical or at least identically acting parts.

[0108] FIG. 1 shows, merely schematically, a production installation 1 according to the invention for producing material boards 12. The production installation 1 in this case includes multiple apparatuses 2, 3, 4, 5, 6, 7, 8, 9 through which the material 10 passes in order to be ultimately pressed to obtain a material board 12.

[0109] The material 10 consists primarily of a plant-based primary material, in particular wood, that is delivered to the production installation 1. Besides wood as the material 10, other lignocellulosic materials, recycling materials, for example recycling wood or recycled plastic, or plastics may be processed directly as the material 10 in the installation. In addition, the material 10 may also consist of a mixture of different materials, for example green wood and recycling wood and/or plastic.

[0110] In the production installation 1 shown, the material 10 provided is first of all added to an apparatus 5 for comminuting the material 10. By way of example, the apparatus 5 for comminuting may be a cutting device, a wood chipper or a knife ring flaker. The comminuted material 10 is subsequently delivered to an apparatus 6 for drying, in which the material 10 is dried to a predefined residual moisture for the further process. Upstream of the apparatus 6 for drying, the material 10 is mixed with a binder in an apparatus 3 for gluing. The use of a binder is dependent on the respective material 10. When plastics are used as the material 10, it may be possible to dispense with addition of a binder if appropriate, which means that a corresponding apparatus 3 for gluing is not required. Alternatively or additionally, as shown in FIG. 1, an apparatus 3′ for gluing may also be arranged downstream of the apparatus 6 for drying. If appropriate, the material 10 may be re-glued once more in the downstream apparatus 3′ for gluing after the apparatus 6 for drying.

[0111] The glued material 10 subsequently passes through an apparatus 7 for sorting, in which the material 10 is divided up into multiple fractions. In the present case, the material 10 is fractionated on the basis of the size of the comminuted material 10, with material 10 with a smaller particle size being separated from material 10 with a larger particle size and being delivered to separate processing lines or temporary stores. Fractionation of the material 10 in an apparatus 7 for sorting is aimed at improved production of the material board 12, which involves for example the material 10 with a smaller particle size being arranged on the outer faces of the flat sides of the material board 12 and coarser material 10 forming a middle layer of the material board 12.

[0112] Subsequently, the material 10 is delivered to an apparatus 2 for spreading the material 10. In this apparatus 2, the material 10 is spread on a forming belt to obtain a mat 11 or a web, the fractions of the material 10 being spread in layers, as a result of which the mat 11 has a layered structure. Alternatively, the material 10 may also be spread only in one layer on a forming belt. The mat 11 spread in this way subsequently passes through an apparatus 8 for pre-pressing, as shown in FIG. 1, in which the mat 11 is pre-pressed. The mat 11 is thus firstly compressed and secondly vented in the apparatus 8, allowing improved pressing of the mat 11 to obtain the material board 12 in the apparatus 4 for pressing. Depending on the material 10 used, for example the use of plastic as the material 10, an apparatus 8 for pre-pressing may not be required.

[0113] In the apparatus 4 for pressing, which is in the form of a continuously operating press in the present case, the mat 11 is exposed to pressure and heat, as a result of which the binder in the mat 11 sets and the material board 12, or a continuous material board strand, is formed at the end of the apparatus 4 for pressing. Arranged upstream of the apparatus 4 for pressing, there may also be a further apparatus for preheating the mat 11, for example by means of steam or microwave radiation, if appropriate. The continuous material board strand exiting the apparatus 4 is subsequently divided up by means of diagonal saws in an apparatus 9 for separating, as a result of which material boards 12 of desired length are formed. The material boards 12 thus formed are subsequently cooled, stacked and delivered to a storage site or to immediate further processing.

[0114] The production installation 1 further includes a controller 20 that is directly connected to the individual apparatuses 2 to 9 of the production installation 1 and to further apparatuses, for example measuring apparatuses or temperature sensors. The controller 20 in this case includes a programmable logic controller 21, by means of which the production installation 1 and the apparatuses 2 to 9 are controlled or regulated. The operational connection of the controller 20 to the individual apparatuses 2 to 9 of the production installation 1 is used to acquire input parameters 23 in the controller 20 and to transfer altered input parameters 28 to the production installation 1, or to the individual apparatuses 2 to 9. The input parameters 23 are setpoint values and actual values of installation parameters of the production installation 1 and of the apparatus 2 to 9. In the present case, these are for example the temperature in the apparatus 6 for drying, the fill levels of temporary stores, the pressing speed at which the mat 11 is pressed in the apparatus 4 or the temperatures and pressures in the apparatus 4. Some of the installation parameters are acquired by sensors on the individual apparatuses 2 to 9 or are formed by fixed parameters that characterize the individual apparatuses 2 to 9 and the production installation 1. Fixed installation parameters are for example the press length of the continuously operating press as the apparatus 4 for pressing, the number of spreading units in the apparatus 2 for spreading or the number of fractions in the apparatus 7 for sorting.

[0115] Besides the installation parameters, the material parameters also form further input parameters 23 that are incorporated into the controller 20. The material parameters are the type of material 10 used, the residual moisture of the material 10 after the apparatus 6 for drying, the binder used in the apparatus 3 for gluing, or the height and width of the spread mat 11 in and/or after the apparatus 2 for spreading.

[0116] Finally, other input parameters 23 may also be predefined by a user. As such, product parameters characterizing the material board 12 are likewise input parameters 23. The product parameters are predefined parameters of the material board 12 that is supposed to be produced in the production installation 1. Certain minimum requirements for the material board 12 are thus stipulated using the product parameters, which are likewise incorporated into the controller 20. The product parameters are the type of material board to be produced or the thickness of the material board 12 to be produced.

[0117] The controller 20 is designed such that it is able to receive and process input parameters 23 and also to output altered input parameters 28 to the apparatuses 2 to 9 of the production installation 1. In addition, the controller 20 is designed such that a quality value 24 is ascertained in the controller 20 on the basis of the input parameters 23 and is displayed to a user or technologist of the production installation 1 on a display 26. In addition, the controller 20 includes an optimization computer 22 that is used to optimize the ascertained quality value 24. An interface 25 is used to send the parameters acquired in the controller 20 to other apparatuses, such as for example a memory 27, other displays, or a cloud data storage. Furthermore, the interface 25 is also used to make a connection to a system of the manufacturer of the production installation 1 and/or of apparatuses 2 to 9 of the production installation 1, which connection is used to send data of the production installation 1 and/or of the apparatuses 2 to 9 and to transmit data such as the database 30 or a pre-trained algorithms 32 to the controller 20. The interface 25 may also be used by the controller 20 to receive data, in particular datasets or test datasets, information and/or parameters from other production installations 1, in particular if the manufacturer of the material boards 12 operates multiple production installations 1.

[0118] FIG. 2 schematically shows the sequence for forming and training an algorithm 32 based on artificial intelligence to ascertain at least one quality value 24 of a quality parameter. To produce the algorithm 32, a method on which the algorithm 32 is based is first of all stipulated in the controller 20 or selected from a list. In the present case, this involves resorting to the method of neural networks as the basis for the algorithm 32. In order to be able to use the algorithm 32 to ascertain, or forecast, a quality value 24 for a quality parameter, it is necessary for the algorithm 32 to be trained. A database 30 that includes a multiplicity of datasets, in the present case at least over 500, is provided for training the algorithm 32. The individual datasets of the database 30 include details relating to quality parameters and input parameters correlating with the quality parameters. As such, a dataset of the database 30 reflects information for a material board 12 for which the input parameters and the quality parameters are known. The multiplicity of datasets in the database 30 are in a form such that they differ in at least one parameter, a quality parameter or an input parameter correlating with the quality parameter. Moreover, the datasets of the database 30 are designed such that they cover as broad as possible a spectrum of product parameters in regard to data relating to the product parameters, which form a subgroup of the input parameters. For material boards 12, the product parameters, which are supposed to cover as broad as possible a spectrum of product parameters, are provided by the board thickness of the material board 12 and the type of material boards 12. It is thus possible to ensure that the algorithm 32 trained by way of the database 30 allows reliable ascertainment of a quality value 24 for a quality parameter, and beyond, over as broad as possible a spectrum of product parameters.

[0119] Before the algorithm 32 is trained using the database 30, the data of the database 30 are normalized and aggregated in a processing step 31, as shown schematically by the dashed arrow in FIG. 2. Processing of the data of the database 30 by means of normalization and aggregation may also be carried out for particular individual datasets before they are delivered to the database 30.

[0120] The datasets, or data, in the database 30 are based on historical data of the production installation 1 that are stored in particular locally in situ in a memory 27 of the controller 20. Besides the locally available data for forming a database 30, the database 30 also includes datasets from other production installations, which are sent by way of an interface 25. The datasets from other production installations are provided and sent by way of the interface 25 by the manufacturer of the production installation 1. If the manufacturer of material boards 12 has multiple production installations 1, datasets for the database 30 may also be provided by these production installations 1.

[0121] In the processing 31, data clusters are formed within the available data in order to reduce the volume of data. The aggregation is carried out by way of interpolation, extrapolation or averaging. Moreover, the data are normalized to a statistical mean of zero and a standard deviation of one. After the processing 31 of the database 30, the algorithm 32 is finally trained using the database 30, which includes aggregated and normalized datasets after the processing 31.

[0122] When the training of the algorithm 32 using the database 30 has concluded, a further step involves the algorithm 32 formed being verified using a test database 33. The test database 33 in turn includes test datasets, which are of analogous design to the datasets of the database 30. Prior to their being used, the test datasets of the test database 33 have likewise undergone processing in which the test datasets of the test database 33 were normalized and aggregated.

[0123] The input parameters recorded in the test datasets of the test database 33 and correlating with the quality parameters are predefined for the algorithm 32 as input parameters, as a result of which said algorithm takes these input parameters as a basis for ascertaining a quality value 24 for at least one quality parameter. In particular, the algorithm 32 is used to ascertain quality values 24 for multiple quality parameters. While the algorithm 32 is being checked, quality values 24 are ascertained for quality parameters recorded in the test datasets of the test database 33.

[0124] The quality values 24 now ascertained by means of the algorithm 32 are subsequently compared with the data relating to the quality parameters recorded for the applicable test dataset in the test database 33. If the quality values are in a predefined range, in the present case differ from the test dataset by less than 5% of the data of the quality values, the algorithm 32 continues to be checked using further test datasets of the test database 33. If only slight differences between the ascertained quality value 24 and the data recorded in the test database 33, or in the test datasets, are found for the entire test database 33 overall, the checking of the algorithm 32 is complete and said algorithm is used in the controller 20 of the production installation 1 to ascertain quality values 24.

[0125] If, by means of the test database 33, there are larger differences between the data recorded in the test database 33, or in the test datasets, for the quality parameters and the quality values 24 ascertained by the algorithm, the algorithm 32 is retrained or even fundamentally reconstructed by means of an extended database 30, reconstruction of the algorithm 32 possibly involving the latter also being based on a different method, for example K nearest neighbors regression.

[0126] After the algorithm 32 has been retrained or formed from scratch, it is again checked using a test database 33 and finally used in the controller 20 to produce material boards 12 in a production installation 1, provided that the check using the test database 33 has in turn yielded the smallest possible difference in the quality values 24.

[0127] FIG. 3 schematically shows the sequence for ascertaining a quality value 24 in the controller 20 of the production installation 1 for at least one quality parameter in the course of the production of material boards 12. To ascertain the quality value 24 for multiple quality parameters, input parameters 23 are first of all input into the controller 20 or flow into the latter from the production installation 1 and the apparatuses of the production installation 2 to 9. The input parameters 23 include preset product parameters, including the board thickness and the type of material board 12 to be produced, material parameters, including the mix ratio of green and recycling material for the material 10 used and the type of binder, in the present case PMDI, and installation parameters, including sensor data from the individual apparatuses 2 to 9 of the production installation 1 and preset parameters of specific apparatuses 2 to 9.

[0128] On the basis of these input parameters 23, the algorithm 32, which has been trained and checked as previously illustrated in FIG. 2, ascertains quality values 24 for different quality parameters, which are displayed to a user of the production installation 1 on a display 26. The quality values 24 are ascertained by the algorithm 32 on an ongoing basis in this case, which means that the quality values 24 for the material board 12 being produced or needing to be produced are constantly displayed on the display 26. A message or warning is furthermore output on the display 26 if the ascertained quality value 24 drops below or at least approaches a minimum quality value. The user of the production installation 1 is thus informed if there is the risk of material boards 12 being produced that no longer comply with the predefined values for the quality parameters, which ultimately means rejects. The user should react to the warning or message by manually modifying input parameters 23 such as the production speed or the pressure in the apparatus 4 for pressing.

[0129] Besides a display 26 for a user of the production installation 1 that is arranged in the control center of the production installation 1, the quality values 24 may also be retrieved and displayed on other displays, for example a display on the installation or on a tablet or smartphone.

[0130] The input parameters 23 provided for ascertaining the quality values 24, and also the ascertained quality values 24, are stored as a production dataset in a memory 27, which allows the ascertained quality values 24 to be documented for the input parameters and, if necessary, checked by way of values that are later measured for the quality parameters. A corresponding production dataset, including the ascertained quality values 24 and the correlating input parameters 23 for this ascertained quality value 24, is stored manually as required or on an ongoing basis after a predetermined interval, for example every second or before and/or after a specific event, such as for example a change of product, the production datasets being sent to the manufacturer of the production installation 1 by way of the interface 25. The interface 25 may also be used to send the applicable production datasets to other production installations 1 of the manufacturer. Moreover, storage of the production datasets, including the quality values 24 and input parameters 23 correlating with these quality values 24, results in logging being performed for the quality of the material boards 12 produced. The data of the production datasets may furthermore be normalized and/or aggregated. Aggregation and/or normalization also allows the applicable data of the production datasets to be ported to production installations of identical or similar type.

[0131] FIG. 4 schematically shows the optimization of an ascertained quality value 24 in the controller 20. As already presented in FIG. 3, input parameters 23 flow into the algorithm 32, which the algorithm 32 takes as a basis for ascertaining one or more quality values 24. This ascertained quality value 24 is now transferred to an optimization computer 22, which includes a further algorithm 34 by means of which the ascertained quality value 24 is optimized. In the present case, the algorithm 34 is an algorithm that is of identical design and form to the algorithm 32.

[0132] Alternatively, the algorithm 34 may be different than the algorithm 32. For this, the algorithm 34 is, as already presented above in FIG. 2 in relation to the algorithm 32, trained by means of a database and, if necessary, checked and verified using a test database.

[0133] The optimization of the quality value 24 in the optimization computer 22 is subject to specific boundary conditions in this case that are predefined by a user or are already recorded in the algorithm 34.

[0134] Optimization of the quality value 24 in this instance may be based on diverse aspects. In the present case, the optimization computer 22 is designed such that the quality value 24 per se is optimized and improved. The aim is therefore to produce a material board 12 of the highest possible quality.

[0135] The quality value 24 may also be optimized so that the quality value 24 is maintained but alteration of the input parameters 23 to achieve the quality value 24 means that lower use of primary materials, auxiliaries and consumables is necessary or a higher output of material boards 12 per unit time becomes possible. Thus, for constant quality, indicated by an approximately constant quality parameter 24, the use of resources will be decreased, rendering the production of material boards 12 cheaper.

[0136] Furthermore, the quality value 24 may also be optimized in the optimization computer 22 so that the quality value 24 is reduced to a minimum value, allowing the use of primary materials, auxiliaries and consumables to be reduced further. The material boards 12 therefore meet minimum requirements, while at the same time the use of resources is reduced and the material boards 12 are able to be produced more cheaply.

[0137] Optimization of the quality value 24 in the optimization computer 22 involves the algorithm 34 ascertaining altered input parameters 28 on the basis of which the desired quality value is supposed to be achievable. The ascertained altered input parameters 28 are then displayed to a user of the production installation 1 on the display 26 by virtue of a message being output on the display 26. The user will then check the altered input parameters 28, correct them if necessary and send them to the programmable logic controller 21, or enter them manually there. The programmable logic controller 21 transfers the ascertained altered input parameters 28 to the production installation 1, or the apparatuses 2 to 9 of the production installation 1, and engages them there as appropriate. A transfer of the applicable altered input parameters 28 is shown by dashed lines in FIG. 4. However, the user may also refrain from optimizing the quality value 24 by way of the ascertained altered input parameters 28, for example if he is aware of circumstances such as imminent maintenance or a change of product and optimization would therefore no longer be expedient.

[0138] Alternatively, the ascertained altered input parameters 28 may be altered by the algorithm 34 directly in the programmable logic controller 21 and finally transferred to the production installation 1, or the apparatuses 2 to 9 of the production installation 1, to alter the respective settings. The altered input parameters 28 ascertained in the optimization computer 22 are thus used to continually and fully automatically optimize production of the material boards 12 in the production installation 1.

LIST OF REFERENCE SIGNS

[0139] 1 production installation [0140] 2 apparatus for spreading [0141] 3, 3′ apparatus for gluing [0142] 4 apparatus for pressing [0143] 5 apparatus for comminuting [0144] 6 apparatus for drying [0145] 7 apparatus for sorting [0146] 8 apparatus for pre-pressing [0147] 9 apparatus for separating [0148] 10 material [0149] 11 mat [0150] 12 material board [0151] 20 controller [0152] 21 programmable logic controller [0153] 22 optimization computer [0154] 23 input parameter [0155] 24 quality value [0156] 25 interface [0157] 26 display [0158] 27 memory [0159] 28 altered input parameters [0160] 30 database [0161] 31 processing [0162] 32 algorithm [0163] 33 test database [0164] 34 algorithm