METHOD AND DEVICE FOR DETERMINING A PARAMETER OF A SIZE DISTRIBUTION OF A MIXTURE, PLANT FOR THE MANUFACTURE OF MATERIAL BOARDS AND COMPUTER PROGRAM PRODUCT

Description

[0001] The invention relates to a method for determining a parameter of a size distribution of a mixture according to claim 1 and a device for determining a parameter of a size distribution of a mixture according to claim 11. The invention further relates to a system for producing a material plate from a mixture of particles according to claim 13 and a computer program product according to claim 16.

[0002] Mixtures of one or more materials form the basis for the production of further products in a wide variety of applications. For example, in the production of engineered sheets, mixtures of particles of a starting material form the essential basis for the formation of such a sheet. The composition and properties of the mixture have a significant influence on the quality and properties of the engineered sheet produced.

[0003] In the production of wood-based panels, especially particleboard, mixtures containing lignocellulose particles are used. These particles exhibit different properties, particularly varying sizes and lengths. The particle size, especially in particleboard, significantly influences the panel's strength, such as its bending strength, as well as the required amount of adhesive. This amount correlates with the particle surface area, which is itself largely determined by the particle size and, in particular, length. Therefore, determining the particle size distribution of such a mixture is of paramount importance.

[0004] To determine this size distribution, a sieve analysis is a known method. For this, a sample of the mixture from the process for manufacturing material sheets is taken and separated into multiple fractions, each representing a specific size range, using a sieve system. The ratio of these fractions to each other is then determined and used as the basis for the subsequent process until the next analysis. Sieve analysis is not only very complex and time-consuming for sieving, analyzing, and determining the size distribution, but it also only provides a snapshot in time.

[0005] US Patent 2007 / 0222100 A1 discloses a method and system that uses near-infrared spectroscopy for inline monitoring and control of the proportion of resin solids and other additives in a binder mixture during the continuous production of composite sheets. Process variables of the binder mixture are dynamically adjusted in real time to maintain desired resin load levels and thus improve the uniformity and quality of the final product.

[0006] The object of the invention is to provide a method and a device as well as a system and a computer program product which makes it possible to easily determine the parameter of a size distribution, in particular during a production process.

[0007] This problem is solved by a method according to claim 1 and by a device according to claim 11. Furthermore, the problem is also solved by a system according to claim 13 and a computer program product according to claim 16.

[0008] Advantageous embodiments of the method according to the invention are found in claims 2 to 10. Advantageous embodiments of the device according to the invention are found in claim 12. Advantageous embodiments of the system according to the invention are found in claims 14 and 15. The wording of all claims is hereby explicitly incorporated into the description by reference.

[0009] The method according to the invention is preferably designed to be carried out using the device according to the invention or a preferred embodiment of the device. The device according to the invention is preferably designed to carry out the method according to the invention or a preferred embodiment of the method according to the invention.

[0010] The problem is solved by a method for determining at least one parameter of a size distribution of a mixture of particles during the production of material plates, comprising the following steps: A. Creating at least one image of the mixture with an optical system; B. Transmitting the at least one image to an evaluation device and generating at least one output characteristic curve from the at least one image, wherein the at least one output characteristic curve is formed by a mathematical transformation of the at least one image, preferably by a Fourier transform, and represents a profile of an intensity, preferably a normalized intensity, over a spatial frequency; C. Determining at least one approximate characteristic curve in the evaluation device by approximation such that the at least one approximate characteristic curve essentially coincides with the at least one output characteristic curve or at least in some areas essentially agrees, wherein the at least one approximate characteristic curve is based on or depends on at least one geometry of the particles; and D.Determination of at least one parameter of the mixture's size distribution from at least one approximate characteristic curve and / or initial characteristic curve.

[0011] To determine at least one parameter of the size distribution, such as the size distribution itself, it is no longer necessary to isolate and analyze a subset of the mixture. Instead, an initial characteristic curve can be generated directly from the mixture sample, allowing at least one parameter of the mixture's size distribution to be derived or determined directly. This saves considerable time, as the analysis can be performed much faster. Furthermore, determinations can be performed much more frequently, since the data can now be acquired much more quickly and with less effort. Consequently, subsequent processes can be influenced much more rapidly, for example, by adjusting other parameters based on the determined parameter of the mixture's size distribution.Furthermore, faster and more frequent analysis allows for the detection of errors in the preceding processes during the mixture's production, such as dull tools used in particle manufacturing. Mathematical transformation, particularly Fourier transformation, also enables the identification of small particles within the mixture, especially those smaller than 0.5 mm, which is not easily possible with direct analysis of the image due to particle size limitations.

[0012] A parameter of the size distribution of a mixture of particles could be, for example, the size distribution of the particles in the mixture itself, the spread of the size distribution of the particles, or the mean or a calculated mean particle size of the mixture.

[0013] The optical system can be, in particular, any device for capturing an image of at least the surface of the mixture, preferably only the surface of the mixture. Thus, the optical system can be a camera, especially a black and white camera. Preferably, the optical system operates in the visible spectral range or at least partially encompasses it. The use of radiation outside the visible range to capture an image of the mixture, for example, X-rays, can also serve as the basis for a corresponding optical system.

[0014] The evaluation unit can preferably form an independent unit or be integrated into or included in a control device of a plant, in particular a plant for the production of material panels.

[0015] Preferably, at least one parameter of the size distribution of the mixture can be determined directly from the initial characteristic curve, without the need to determine an approximate characteristic curve.

[0016] In a preferred embodiment, the particles of the mixture are chips, shavings, and / or fibers. With this type of mixture, which, for example, forms the starting material for the production of a composite panel, particularly a wood-based panel, the variation in the size distribution parameter can be very large. A large variation in at least one parameter in the size distribution of a mixture, which is also to be determined, can now be determined more quickly than with methods known from the prior art, without requiring additional equipment for separate analysis of the mixture.

[0017] Preferably, the mixture consists at least partially of particles made from a lignocellulose-containing material. Lignocellulose-containing material is usually a natural product, so the corresponding starting material for particle production already exhibits variance in its size. During the production of a mixture, one parameter of the size distribution is inherently subject to a certain range, which is why determining at least one parameter of the size distribution is of particular interest for this material.

[0018] A preferred embodiment is characterized by the fact that the creation of the at least one sample is carried out inline, particularly in a system for the production of material sheets, and preferably no discharge of the mixture or a portion thereof takes place for the at least one sample. With inline sampling, the sample of the mixture is created directly and immediately during the process. The creation of a sample of the mixture can preferably be continuous or on demand. In particular, a sample can be taken at regular intervals, preferably in the range of seconds to minutes. For example, a sample of the mixture is taken every second. Alternatively, the intervals between samples can also be longer, for example, 10 seconds, 30 seconds, 1 minute, 5 minutes, or even 15 minutes.A recording can be made if necessary, for example, if there is a change in the starting material of the mixture or if defective components in the production of the mixture have been repaired.

[0019] Alternatively, or preferably additionally, the mixture is directly sampled, particularly without prior singulation, sieving, and / or fractionation. The sample is taken directly from the mixture during the process of manufacturing material sheets at a predetermined location within the system. Preferably, multiple samples of the mixture are taken at different locations within the system.

[0020] Preferably, the at least one approximation characteristic curve is based on a parametric base curve which includes the at least one parameter of the size distribution as a variable, or from whose variables the at least one parameter of the size distribution is determined. The approximation characteristic curve can thus also be formed by a parametric base curve, wherein the variables and the parameterization preferably depend on the type of mixture. The at least one parameter of the size distribution of the mixture can be determined from the parametric base curve, since it depends directly or indirectly on the individual variables of the parametric base curve or itself forms a variable of the parametric base curve.

[0021] In a preferred embodiment, the at least one approximate characteristic curve is determined by weighting at least two reference characteristic curves. The reference characteristic curves are provided in the evaluation unit, and each reference characteristic curve represents a characteristic curve for a reference mixture of particles. The at least one parameter of the size distribution is determined from the weighting of the reference characteristic curves. The reference characteristic curves thus form the starting point for determining the approximate characteristic curve. The weighting can be performed manually by a user, but in particular, it can also be performed automatically in the evaluation unit, using mathematical approximation methods. The reference mixture is a mixture for which parameters of the particles and preferably also of the composition of the mixture are known.

[0022] Preferably, the reference mixture is a mixture of particles of a certain size or particle size category, in which the size of the particles in the reference mixture, particularly their length, lies within a range of known particle sizes. The width of the range, which determines the difference between the largest and smallest particle sizes within the range, is preferably kept small. Preferably, the width of the range is less than 20 mm, more preferably less than 10 mm, more preferably less than 5 mm, and most preferably less than or equal to 1 mm. In particular, the smaller the particles in a particle size category, the smaller the range and, more importantly, their width.

[0023] Preferably, the reference characteristic curve is determined from at least one reference image, wherein the reference image is preferably an image of a real reference mixture or an artificially generated image of a reference mixture. The real reference mixture can be obtained, for example, by sieving a mixture, whereby the real reference mixture, due to a certain variation in particle size, can usually only be assigned to one particle size category. The artificially generated image of a reference mixture has the advantage that modeled particles of a single size, in particular a predetermined length, are used to generate an artificially created reference mixture in which the modeled particles are preferably arranged chaotically.

[0024] Alternatively or preferably additionally, at least one reference characteristic curve is formed by a mathematical transformation of the reference recording, in particular by a Fourier transform, wherein the reference characteristic curve represents a profile of an intensity, preferably a normalized intensity, against the spatial frequency. Thus, the reference characteristic curve is preferably formed on the basis of the same mathematical transformation as the original characteristic curve.

[0025] In an advantageous embodiment, at least five, preferably at least seven, and particularly preferably at least ten reference curves are provided or determined in the evaluation unit. A higher number of reference curves, each assigned to a different reference mixture, allows for a better reproduction of the original characteristic curve and thus increases the quality of the determination of at least one parameter of the mixture's size distribution. In a preferred embodiment, seven reference curves are generated for the following particle size categories: 0 mm to 0.125 mm, 0.125 mm to 0.315 mm, 0.315 mm to 0.63 mm, 0.63 mm to 1.0 mm, 1.0 mm to 2.0 mm, 2.0 mm to 4.0 mm, and greater than 4.0 mm.

[0026] Preferably, a portion, or preferably all, of the reference characteristic curves are derived from a single determined reference characteristic curve. The influence of particle size on the reference characteristic curve is particularly important in this regard.

[0027] In a preferred embodiment, at least one sample is taken when the mixture is conveyed, particularly when the mixture is scattered onto a conveying device and / or forms a particle curtain. Thus, the mixture can be taken during the manufacturing process of the material sheet, in which the mixture is transported by means of a conveying device or forms a corresponding particle curtain. In particular, no discharge of the mixture is required to determine at least one parameter of the size distribution. This allows the method to be easily integrated into an existing process or plant for the production of material sheets.

[0028] Alternatively, or preferably additionally, the mixture is spread onto a conveying device for sampling. A separate scattering of the mixture, specifically designed to determine at least one parameter of the size distribution, allows the mixture to be dispensed in the manner necessary or preferred for the determination. In particular, a separate scattering process enables idealized and consistent conditions for sampling, thereby improving the comparability of the samples and the resulting characteristic curves.

[0029] Preferably, the mixture has a substantially flat surface in the at least one image and is smoothed, in particular, before the at least one image is captured. The flat surface ensures that the particles are not distorted by the depth of the image. Rather, the particles of the mixture should lie in almost the same plane. Smoothing the mixture before capture, for example with a smoothing roller, can also contribute to this.

[0030] Alternatively, or preferably additionally, the mixture is illuminated, preferably homogeneously, particularly by an external light source. This illumination improves the definition of the individual particles in the mixture, making them more visible in the image. Homogeneous illumination, for example, reduces shadows cast on the mixture, further clarifying the particle contours. The light source can preferably be integrated into the optical system itself or form a separate unit independent of the optical system. This independence allows for more precise and improved illumination of the mixture, particularly enabling individual adjustment.

[0031] Preferably, the mixture is illuminated using a particularly homogeneous dark-field technique. This makes the surface of the mixture more clearly visible. Furthermore, particles located deeper below the surface are not illuminated, thus enabling a more precise determination of at least one parameter of the size distribution.

[0032] Preferably, the light source can be designed to illuminate the mixture continuously or at regular or irregular intervals. Sequential illumination can be configured, in particular, to occur only when needed, especially when taking a photograph. For this purpose, this device is synchronized with the optical system used to take the photograph.

[0033] The light source itself can preferably be composed of several light elements and / or have a special light color, by means of which, for example, the contours of the individual particles of the mixture are more clearly visible or depicted in the recording.

[0034] Preferably, the light source can be positioned directly above the mixture. Alternatively, or more preferably, the light source can be positioned laterally, so that the light from the light source strikes the surface of the mixture at an oblique angle. In particular, homogeneous dark-field illumination is advantageous.

[0035] In a preferred embodiment, the at least one image is processed before the at least one output characteristic curve is created, and / or the reference image is processed before the reference characteristic curve is created, and / or only a section of the at least one image and / or the reference image is used for creating the at least one output characteristic curve and / or the reference characteristic curve. Processing the image before creating the corresponding characteristic curve allows image errors in the image to be detected and corrected in advance. This effectively smooths the image. Erroneous images can be identified and discarded beforehand, as these would otherwise lead to error messages during evaluation.

[0036] Processing the recording of the initial characteristic curve and / or reference characteristic curve can also serve to further emphasize particularly relevant aspects for its creation, such as the contour of the particles. This involves, for example, sharpening the edges of individual particles in the recording, thereby making it easier to determine their length or size within the mixture. The recording and / or reference recording is preferably processed before mathematical transformation so that it shows a square image section whose side length is normalized or known.

[0037] Preferably, at least one approximate characteristic curve is determined using an optimization algorithm. This optimization algorithm can be a simple method that determines the shape of the approximate characteristic curve based on its deviation from the original characteristic curve. The deviation can be considered and minimized over the entire approximate characteristic curve or only over a specific segment of it. In particular, this is a mathematical approximation method.

[0038] Alternatively, or preferably additionally, the optimization algorithm can also be an advanced algorithm based on one or a combination of the following methods: 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, K nearest neighbors classification, decision trees, random forests, Naive Bayes, and support vector machines. The aforementioned advanced methods are essentially based on machine learning and can enable faster and better determination of the output characteristic curve, thereby allowing for faster and more accurate determination of at least one parameter of the mixture's size distribution.

[0039] Preferably, the optical system comprises a camera which, in particular, has a resolution of at least 640 x 640 pixels. Most preferably, the camera has a resolution of 2048 x 2048 pixels.

[0040] A preferred embodiment is characterized by the fact that the at least one output characteristic curve, the at least one approximation characteristic curve, and / or the at least one parameter of the size distribution are displayed on a screen. By displaying, in particular, the at least one output characteristic curve and the at least one approximation characteristic curve, it is also possible to visually verify whether the determination of the approximation characteristic curve is substantially correct, and an operator can intervene in the process if necessary. The display can show variables of the approximation characteristic curve, for example, if it is a parametric base curve. Furthermore, the weighting and / or the at least one parameter of the size distribution can also be displayed directly, with or without showing the output characteristic curve and / or the approximation characteristic curve.

[0041] The problem is further solved by a device for determining a parameter of a particle size distribution of a mixture during the production of material sheets, comprising an optical system for acquiring at least one image of the mixture and an evaluation device, wherein the optical system is operatively connected to the evaluation device, wherein the evaluation device is configured to generate at least one output characteristic curve from the at least one image of the mixture, which is formed by a mathematical transformation of the at least one image, preferably by a Fourier transform, and represents a profile of an intensity, preferably a normalized intensity, over a spatial frequency, and wherein the evaluation device is configured to generate at least one approximate characteristic curve in the evaluation device by approximation in such a manner as follows:that the at least one approximation characteristic curve essentially coincides with the at least one output characteristic curve or at least agrees in certain areas, wherein the approximation characteristic curve is based on or depends on at least one geometry of the particles, wherein at least one parameter of the size distribution of the mixture can be determined from the at least one approximation characteristic curve, and wherein the device further comprises an output unit for outputting at least one parameter of the size distribution.

[0042] Preferably, the optical system of the device is a camera, in particular a black and white camera, wherein the camera preferably has a resolution of at least 640 x 640 pixels. Particularly preferably, the camera has a resolution of 2048 x 2048 pixels. The high resolution ensures that the contours of the particles are displayed sharply in the image, which has a positive effect on the output characteristic curve and its evaluation.

[0043] Preferably, the device comprises a device for smoothing the mixture before the at least one intake and / or an external light source, in particular for homogeneous illumination of the mixture.

[0044] Another solution to the problem is given by a plant for producing a material sheet from a mixture of particles, comprising devices for forming a nonwoven fabric from the mixture or the particles and pressing it into material sheets, and preferably for applying adhesive to the mixture or the particles before forming the nonwoven fabric, characterized in that the plant comprises a device as described above or a preferred embodiment thereof.

[0045] Using the device within the system, at least one parameter of the size distribution can be determined, and based on this parameter, the other components can be controlled accordingly, thus improving the manufacturing process. Alternatively, or preferably additionally, the quality of the material sheet can be determined or predicted based on at least one parameter of the size distribution and other system parameters.

[0046] A preferred embodiment of the device is characterized by the optical system being arranged at a fixed distance from a surface of the mixture. This fixed distance ensures that the recorded section is always consistent, thus guaranteeing comparability.

[0047] Alternatively or preferably additionally, the distance between the optical system and a surface of the mixture can be detected and / or adjusted, particularly for determining and / or depending on the recording area. This allows for comparability of the recordings, as the size of the recording area correlates with the distance. Especially with a mixture containing small particles, a recording can be made closer to the mixture for a more precise examination of the particles, thus enabling a more accurate determination of at least one parameter of the size distribution. Preferably, the device for determining a parameter of a mixture's size distribution includes appropriate means for adjusting and / or detecting the distance.

[0048] Preferably, the optical system has a distance of 1 cm to 250 cm, preferably of 5 cm to 150 cm, particularly preferably of 10 cm to 100 cm to the mixture, the distance depending in particular on the expected size of the particles so that they are clearly shown in the recording.

[0049] Preferably, the optical system is arranged downstream of a device for generating particles from a starting material, upstream and / or downstream of a sieving device, upstream and / or downstream of the sizing device, and / or upstream and / or downstream of the nonwoven forming device. The arrangement can therefore be located virtually anywhere within the system. An arrangement immediately downstream of particle generation has the advantage that problems in the production of the particles and / or the mixture can be detected early, allowing for a correspondingly quick response to issues such as dull blades or incorrectly set machine parameters.

[0050] During sieving, different fractions are formed from an initial mixture after particle production. These fractions are then processed differently and used for different purposes, for example, in different areas of the material sheet production process. Determining at least one parameter of the size distribution after fractionation allows for the parameter to be determined specifically for the respective mixture, whereas determination before sieving only provides information about the mixture as a whole.

[0051] In a preferred embodiment, the evaluation unit is integrated into a control unit of the system. The data determined in the evaluation unit can then be fed into the system's control unit. Thus, based on the determination of at least one parameter of the particle size distribution, for example, during the production of a sheet from the mixture, the amount of adhesive can be adjusted according to the particle size distribution. Furthermore, other parameters during the production of the sheets, such as the pressure and / or temperature when pressing the mixture into a sheet, can be better controlled, thereby improving the quality of the sheet. Finally, determining at least one parameter of the particle size distribution allows for a preliminary prediction of the expected quality of the sheet.Furthermore, changes can be initiated in the particle manufacturing process if the size distribution deviates from a specification. Errors in particle production can thus be detected and corrected quickly.

[0052] Preferably, the plant includes facilities for comminution, sorting, provision of a binder, storage, air conditioning, weighing, monitoring and / or testing of the particles and / or a starting material of the particles, pre-pressing, adjusting the moisture, adjusting the temperature, adjusting the width and / or height of the nonwoven, measuring parameters of a feedstock and product parameters and / or quality parameters of the material sheet, dividing and / or stacking and / or tempering the material sheet.

[0053] Finally, the problem is also solved by a computer program product with a computer-readable storage medium on which instructions are embedded which, when executed by an evaluation unit, cause the evaluation unit to be configured to carry out a method for determining at least one parameter of a size distribution of a mixture of particles in the course of manufacturing material plates as described above or a preferred embodiment thereof, in particular with a device as described above or a preferred embodiment thereof, preferably in a plant as described above or a preferred embodiment thereof.

[0054] Further advantageous embodiments of the process for manufacturing material panels as well as of the production plant for manufacturing material panels are shown in the following figures.

[0055] They show: Figure 1 shows a basic representation of a production plant for manufacturing material sheets; and Figure 2 shows a flowchart for determining at least one parameter of a size distribution of a mixture.

[0056] In Figure 1The figure shown is merely a schematic representation of a plant 1 according to the invention for the production of material panels 12. Plant 1 comprises several units 2, 3, 4, 5, 6, 7, 8, 9 in which the starting material 16 is ultimately processed into a material panel 12. In this case, the starting material 16 is wood, which is processed into particles, in particular wood chips, and serves as the basis for the production of a particleboard as a material panel 12. The particles are produced, for example, in a knife-ring chipper. Besides wood, the starting material 16 can also consist of other lignocellulosic materials, recycled materials, for example, recycled wood or recycled plastic, or plastics. Furthermore, the starting material 16 can also consist of a mixture of different materials, for example, virgin wood and recycled wood and / or plastic.

[0057] In the schematically depicted Annex 1, the starting material 16 is first processed into particles in a device 5 for comminuting the starting material 16. For example, the comminuting device 5 could be a cutting device, a chipper, or a ring shredder. The starting material 16 is processed in the comminuting device 5 in such a way that a mixture 10 of particles is formed. The mixture 10 exhibits a size distribution within the mixture 10 with respect to the size of the particles, in this case, in particular with respect to the length of the individual particles. Not only one mixture 10, but also several different mixtures 10, 10' can be formed if the material sheet 12 is to be formed from several layers of different properties.Particularly in the production of particleboard, this can lead, for example, to the formation of a mixture 10 for the core layer and a mixture for the surface layers, which are produced by using different machines and / or processes in the comminution facilities 5. Alternatively, different mixtures 10, 10' can also be produced at a later point in time in the plant 1, for example by a device 7 for sorting the particles.

[0058] After comminution, the particles of the mixture 10 are then fed to a drying facility 6, in which it is dried to a predetermined residual moisture content for the further process.

[0059] The particles then pass through a sorting unit 7, in which they are separated into several mixtures 10, 10'. In this case, the particles are fractionated based on their size, whereby smaller particles are separated from larger particles of the comminuted starting material 16, each forming a mixture 10, 10'. Further processing of the mixtures 10, 10' takes place in separate processing lines or they are fed to separate intermediate storage areas. The formation of mixtures 10, 10' in the sorting unit 7 aims to produce a high-quality particleboard 12, in which, for example, the mixture 10, 10' with smaller particles is arranged on the outer surfaces of the particleboard 12, while the mixture 10, 10' with the larger particles forms a middle layer of the particleboard 12.

[0060] The sorting or sieving of the particles serves solely to form several mixtures 10, 10' for the corresponding layers of the material plate 12. This only involves defining the maximum dimensions, in particular the maximum length, that the particles in the mixtures 10, 10' should have. However, the sorting or sieving process does not analyze the size distribution within the mixture 10, 10'.

[0061] After drying in unit 6, the mixtures 10, 10' are mixed with a binder in unit 3 for gluing. The choice of binder depends on the specific mixture 10, 10', particularly on the particle size distribution within the mixture 10, 10', which is a parameter of the size distribution. The average particle size of the mixture 10, 10' can also be a parameter of the size distribution and serve as a basis for determining the amount of glue to be used.

[0062] To determine the parameter of the size distribution, a device 13 for determining a parameter of the size distribution of the mixture 10, 10' of particles is arranged upstream of the gluing unit 3. This device 13 comprises an optical system 17 in the form of a camera for capturing at least one image of the mixture. The image is captured during the production process and while the mixture 10, 10' is resting on a conveying unit. Therefore, no material removal is necessary to create the image. The image shows the surface of the mixture 10, 10' resting on the conveying unit. The device 13 further comprises an evaluation unit 14, to which the optical system 17 is operatively connected.The evaluation unit 14 is designed to generate at least one output characteristic curve from at least one image of the mixture 10, 10', which is formed by a Fourier transform of the at least one image. For the Fourier transform, the image is processed so that it shows a square image section. Further processing is carried out to sharpen the edges of the particles in the image. The Fourier transform generates the output characteristic curve from the image, which represents a normalized intensity curve as a function of a spatial frequency.

[0063] The evaluation unit 14 is designed to generate an approximate characteristic curve by approximation such that at least one approximate characteristic curve essentially coincides with at least one original characteristic curve. The approximate characteristic curve is formed by weighting seven reference characteristic curves, which are stored in the evaluation unit 14. These reference characteristic curves are characteristic curves of mixtures of particles of a particle size category, which were also generated from a reference image by Fourier transformation. The reference mixtures for the particle size categories were created by sieving a starting material according to the particle size of the respective category.The seven reference curves represent the characteristic curves for the following particle size categories: 0 mm to 0.125 mm, 0.125 mm to 0.315 mm, 0.315 mm to 0.63 mm, 0.63 mm to 1.0 mm, 1.0 mm to 2.0 mm, 2.0 mm to 4.0 mm, and larger than 4.0 mm.

[0064] From the weighting of the individual reference characteristic curves or from the determined approximate characteristic curve, one parameter of the size distribution of the mixture 10, 10', in this case the size distribution itself and / or the mean chip size, can finally be determined. For further processing and visualization of the determined values, the device 13 further comprises an output unit 15, which transmits the determined parameter to the control unit 20 of the system 1. The evaluation unit 14 and output unit 15 can also be part of the control unit 20 and integrated within it. In addition, the determined parameter(s) can also be displayed on a display 21.

[0065] After the mixture 10, 10' is coated with sizing agents, it is fed to a spreading device 2. In this device 2, the mixture 10, 10' is spread onto a forming belt to form a fleece 11 or a mat. The mixture 10, 10' containing smaller particles is used as the top layer, and the mixture 10, 10' containing larger particles as the middle layer, so that the fleece 11 has a layered structure of top layer - middle layer - top layer. The spread, three-layered fleece 11 then passes through, as described in the Figure 1 The figure shows a pre-pressing device 8 in which the nonwoven fabric 11 is pre-pressed. In the device 8, the nonwoven fabric 11 is thus compressed on the one hand and de-aerated on the other, thereby enabling improved compression of the nonwoven fabric 11 against the material sheet 12 in the pressing device 4.

[0066] In the pressing unit 4, which is designed as a continuous press, the nonwoven fabric 11 is subjected to pressure and heat. This causes the binder in the layered nonwoven fabric 11 to cure, resulting in the formation of a material sheet 12 or a continuous strand of material sheets at the end of the pressing unit 4. Optionally, a further device for preheating the nonwoven fabric 11, for example by means of steam or microwave radiation, may be arranged upstream of the pressing unit 4. The continuous strand of material sheets exiting the pressing unit 4 is then cut in a separating unit 9 using diagonal saws, producing material sheets 12 of the desired length. The resulting material sheets 12 are then cooled, stacked, and sent to storage or for immediate further processing.

[0067] The individual units 2, 3, 4, 5, 6, 7, 8, 9 and the device 13 of Annex 1 are interconnected via a central control unit 20. The control unit 20 comprises a programmable logic controller (PLC) by means of which the Annex 1, as well as the units 2, 3, 4, 5, 6, 7, 8, 9 and the device 13, are controlled or regulated. Through the interconnection of the control unit 20 with the individual units 2, 3, 4, 5, 6, 7, 8, 9 and the device 13 of Annex 1, parameters are acquired by sensors in the control unit 20, and changed parameters, for example, depending on the determined size distribution of the mixture 10, 10', are transmitted to the Annex 1 or the individual units 2, 3, 4, 5, 6, 7, 8, 9.

[0068] In Figure 2The individual steps for determining at least one parameter of a size distribution of the mixture 10 during the production of material plates 12 are clearly shown again.

[0069] In step A, an image of the mixture 10 is first acquired using an optical system 17 in the form of a camera. The image is captured under homogeneous dark-field illumination of the mixture 10, which makes the surface of the mixture 10 particularly clear in the image. This image is then transmitted to an evaluation unit 14 in step B.

[0070] In the evaluation unit 14, the image is first processed, specifically by selecting a square image section and sharpening the edges of the particles shown in the image. An output characteristic curve is then generated from this processed image. Alternatively, several images taken in quick succession or generated from sections of a single image can be used to generate one or more output characteristic curves. Alternatively, output characteristic curves can be generated from several images or from several sections of a single image, which can then be viewed separately or combined into an averaged output characteristic curve. The characteristic curve is generated by Fourier transformation of the image and represents the intensity profile as a function of a spatial frequency.The transformation also enables, in particular, the determination of small particle sizes or particle size categories within mixture 10, especially for particle sizes smaller than 0.5 mm. For small particle sizes, methods based on a direct evaluation of the recording do not allow for a meaningful determination of the proportion in the mixture.

[0071] In a further step C, an approximate characteristic curve is determined in the evaluation unit 14 by approximation such that the approximate characteristic curve essentially coincides with the original characteristic curve or at least corresponds substantially to it within a predefined range. The approximate characteristic curve is based on the geometry of the particles, in this case, the particle size. The approximate characteristic curve is further formed by weighting seven reference characteristic curves, as explained above. Alternatively, the approximate characteristic curve can also be based on a parametric base curve, the formula of which is stored in the evaluation unit.The individual variables of the parametric base curve can be determined either manually by an operator of the system or automatically by an optimization algorithm, whereby the parameter of the size distribution is derived from the variable(s) of the base curve or even the variable of the parametric base curve itself represents the parameter of the size distribution.

[0072] Finally, in step D, a parameter of the size distribution of the mixture 10, in this case the size distribution itself or the mean chip size, is determined from at least one approximate characteristic curve and / or output characteristic curve. This can then be transmitted to the control unit 20 of the system 1, whereby the other units 2, 3, 4, 5, 6, 7, 8, 9 of the system 1 can be adjusted accordingly depending on the size distribution, so that the material plate 12 has optimal properties. Furthermore, the device 13 can also be controlled based on the determined parameter by initiating more frequent measurements, since the parameter is, for example, in a critical range or exhibits a wide variance compared to previous measurements. Reference symbol list

[0073] 1 Plant 2 Spreading device 3 Gluing device 4 Pressing device 5 Shredding device 6 Drying device 7 Sorting device 8 Pre-pressing device 9 Separating device 10 Mixture 11 Nonwoven fabric 12 Material board 13 Device 14 Evaluation device 15 Output unit 16 Input material 17 Optical system 20 Control device 21 Display

Claims

1. Method for determining at least one parameter of a size distribution of a mixture (10, 10') of particles during the production of material boards (12), comprising the steps of: A. producing at least one recording of the mixture (10, 10') with an optical system (17); B. transmitting the at least one recording to an evaluation apparatus (14) and producing at least one output characteristic line from the at least one recording, wherein the at least one output characteristic line is formed by a mathematical transformation of the at least one recording, preferably by means of a Fourier transformation, and represents a path of an intensity, preferably a standardized intensity, over a spatial frequency, C. establishing at least one approximation characteristic line in the evaluation apparatus (14) by means of approximation in such a manner that the at least one approximation characteristic line substantially coincides with the at least one output characteristic line or at least in areas substantially corresponds, wherein the at least one approximation characteristic line is based at least on a geometry of the particles or is dependent thereon, and D. determining the at least one parameter of the size distribution of the mixture (10, 10') from the at least one approximation characteristic line and / or output characteristic line.

2. Method according to claim 1, characterized in that the particles of the mixture (10, 10') are chips, shreds and / or fibers, and / or in that the mixture (10, 10') at least partially consists of particles of a lignocellulose-containing material.

3. Method according to any one of the preceding claims, characterized in that the production of the at least one recording is carried out inline, in particular in an installation (1) for producing material boards (12), and preferably for the at least one recording no discharge of the mixture (10, 10') or a portion thereof is carried out and / or in that a direct recording of the mixture (10, 10'), in particular without previous separation, sieving and / or fractionation of the mixture (10, 10'), is carried out.

4. Method according to any one of the preceding claims, characterized in that the at least one approximation characteristic line is based on a parametric base line which has the at least one parameter of the size distribution as a variable or from whose variables the at least one parameter of the size distribution is determined.

5. Method according to any one of the preceding claims, characterized in that the at least one approximation characteristic line is established by means of a weighting of at least two reference characteristic lines, wherein the reference characteristic lines are provided in the evaluation apparatus and each reference characteristic line represents a characteristic line for a reference mixture of particles, wherein the at least one parameter of the size distribution is determined from the weighting of the reference characteristic lines.

6. Method according to claim 5, characterized in that the reference characteristic line is established from at least one reference recording, wherein the reference recording is preferably a recording of a real reference mixture or an artificially produced image of a reference mixture, wherein in particular the reference mixture is formed as a mixture with particles of only one particle size or a particle size category, and / or in that at least one reference characteristic line is formed by a mathematical transformation of the reference recording, in particular by means of a Fourier transformation, wherein the reference characteristic line represents a path of an intensity, preferably a standardized intensity, over the spatial frequency.

7. Method according to any one of the preceding claims 5 or 6, characterized in that at least five reference characteristic lines, preferably at least seven reference characteristic lines, in a most preferred manner at least ten reference characteristic lines are provided or established in the evaluation apparatus.

8. Method according to any one of the preceding claims, characterized in that the at least one recording is carried out when the mixture (10, 10') is conveyed, in particular when the mixture (10, 10') is scattered on a conveyor apparatus and / or forms a particle curtain, and / or in that the mixture (10, 10') is scattered for the recording onto a conveyor apparatus and / or in that the mixture (10, 10') during the at least one recording has a substantially planar surface and in particular is smoothed prior to the at least one recording and / or in that the mixture (10, 10') is illuminated, preferably illuminated in a homogeneous manner, in particular by means of an external light source.

9. Method according to any one of the preceding claims, characterized in that the at least one recording is processed before the at least one output characteristic line is produced and / or the reference recording is processed before the reference characteristic line is produced and / or in that only a portion of the at least one recording and / or the reference recording is used to produce the at least one output characteristic line and / or the reference characteristic line.

10. Method according to any one of the preceding claims, characterized in that the establishment of the at least one approximation characteristic line is carried out by means of an optimization algorithm, and / or in that the optical system (17) comprises a camera which in particular has a resolution of at least 640 x 640 pixel and / or in that the at least one output characteristic line, the at least one approximation characteristic line and / or the at least one parameter of the size distribution are displayed on a display (21) of an output unit (15).

11. Device (13) for determining a parameter of a size distribution of a mixture (10) of particles during the production of material boards (12), comprising an optical system (17) for detecting at least one recording of the mixture (10, 10') and an evaluation apparatus (14), wherein the optical system (17) is actively connected to the evaluation apparatus (14), wherein the evaluation apparatus (14) is constructed to produce from the at least one recording of the mixture (10, 10') at least one output characteristic line which is formed by a mathematical transformation of the at least one recording, in particular by means of a Fourier transformation, and represents a path of an intensity, preferably a standardized intensity, over a spatial frequency, and wherein the evaluation apparatus (14) is constructed to produce at least one approximation characteristic line in the evaluation apparatus (14) by means of approximation in such a manner that the at least one approximation characteristic line substantially coincides with the at least one output characteristic line or at least in areas corresponds, wherein the approximation characteristic line is based at least on a geometry of the particles or is dependent thereon, wherein from the at least one approximation characteristic line one parameter of the size distribution of the mixture can be determined, and wherein the device (13) further comprises an output unit (15) at least for output of the at least one parameter of the size distribution.

12. Device (13) according to claim 11, characterized in that the optical system (17) is a camera, in particular a black / white camera, wherein the camera preferably has a resolution of at least 640 x 640 pixel and / or in that the device (13) comprises an apparatus for smoothing the mixture (10, 10') prior to the at least one recording and / or an external light source, in particular for homogeneous illumination of the mixture (10, 10').

13. Installation (1) for producing a material board (12) from a mixture (10) of particles, comprising apparatuses (2, 3, 4, 5, 6, 7, 8, 9) for forming a nonwoven (11) from the mixture (10, 10') or the particles and the pressing thereof to form material boards (12) and preferably for glueing the mixture (10, 10') or the particles prior to forming as the nonwoven (11), characterized in that the installation (1) comprises a device (13) according to any one of claims 11 or 12.

14. Installation (1) according to claim 13, characterized in that the optical system (17) of the device (13) is arranged with a fixed spacing relative to a surface of the mixture (10, 10') and / or a spacing of the optical system (17) with respect to a surface of the mixture (10, 10') can be detected and / or adjusted, in particular for establishment and / or depending on a recording surface of the recording and / or in that the optical system (17) is arranged after an apparatus (5) for producing the particles from a starting material (16), in front of and / or after an apparatus (7) for sieving, in front and of and / or after the apparatus (3) for glueing, and / or in front of and / or after the apparatus (2) for forming the nonwoven.

15. Installation (1) according to any one of claims 13 or 14, characterized in that the evaluation apparatus (14) is integrated in a control apparatus (20) of the installation (1) and / or in that in the installation (1) apparatuses (2, 3, 4, 5, 6, 7, 8, 9) for crushing, sorting, providing a binding agent, storage, climate-control, for weighing, for monitoring and / or checking the particles and / or a starting material (16) of the particles, for pre-pressing, for adjusting the moisture, for adjusting the temperature, for adjusting the width and / or the height of the nonwoven, for measuring parameters of an educt and product parameters and / or quality parameters of the material board, for dividing and / or for stacking and / or for temperature control of the material board (12) are provided.

16. Computer program product having a computer-readable storage medium on which commands are embedded and, when they are carried out by an evaluation apparatus, result in the evaluation apparatus (14) being configured to carry out a method for determining at least one parameter of a size distribution of a mixture (10, 10') from particles during the production of material boards (12) according to any one of claims 1 to 14, in particular with a device (13) according to either of claims 11 or 12, preferably in an installation (1) according to any one of claims 13 to 15.