Waterproofing agent mixing uniformity detection device and control device in the production process of artificial board

Abstract

The present invention provides a waterproofing agent mixing uniformity detection device and control device in the production process of artificial boards, wherein the waterproofing agent mixing uniformity detection device comprises a spectrum detection module and a processor; the spectrum detection module is arranged on the artificial board production line and is located after the waterproofing agent adding process, the spectrum detection module detects the artificial board raw material and outputs a spectrum data set to the processor in real time; the processor receives the spectrum data set, and converts the spectrum data collected by each spectrum probe in real time or converts the spectrum data collected by each spectrum probe into two-dimensional data and inputs it into a pre-trained multi-layer convolutional neural network model, obtains the waterproofing agent addition amount detection value of each spectrum probe corresponding to the raw material detection area, and performs uniformity judgment according to the waterproofing agent addition amount detection value of each spectrum probe corresponding to the raw material detection area and outputs the detection result. The analysis efficiency is high, and the waterproofing agent mixing uniformity in the artificial board can be accurately detected online on the production line.

Description

Technical Field

[0001] The invention relates to the technical field of waterproofing agent detection for artificial boards, and in particular to a waterproofing agent mixing uniformity detection device and a control device in the production process of artificial boards. Background Art

[0002] In the production process of particleboard and fiberboard, the purpose of applying waterproofing agent is to improve the water resistance of board products, improve dimensional stability, reduce or eliminate warping deformation caused by swelling and shrinkage, and prevent board products from mildew and increased conductivity due to moisture absorption. It is one of the simplest, low-cost and most effective measures. The essence of applying waterproofing agent is to apply hydrophobic substances on the wood particle raw materials or fiber raw materials. After the hydrophobic substances are adsorbed on the surface of the wood particle raw materials or fiber raw materials, the following effects can be produced: the hydrophobic substances partially block the gaps between monomer wood particles or monomer fibers, cutting off the channel for water transfer; increasing the contact angle between water and wood particles or fibers, reducing the contact area; partially covering the polar functional groups (such as hydroxyl groups) on the surface of the material, reducing the adsorption effect.

[0003] Practice has proved that as long as the application process is reasonable and the waterproofing agent is evenly distributed in the wood-based panel, a good waterproofing effect can be obtained. On the contrary, if the amount of waterproofing agent added to the wood-based panel is too much or the mixing is uneven, the strength of the product will be reduced. At present, the uniformity of the waterproofing agent mixing of wood shavings and fiber raw materials is mainly judged and controlled by manual experience, and there is no available waterproofing agent mixing uniformity detection device in continuous automated production.

[0004] Therefore, there is an urgent need for a device that can accurately detect the mixing uniformity of the waterproofing agent in the artificial board on the production line. Summary of the invention

[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a waterproofing agent mixing uniformity detection device and a control device in the production process of artificial boards, which have high analysis efficiency and can accurately detect the mixing uniformity of waterproofing agents in artificial boards online on the production line.

[0006] In order to achieve the above object, the main technical solutions adopted by the present invention include:

[0007] In a first aspect, the present invention provides a device for detecting the uniformity of mixing of waterproofing agents in a production process of artificial boards, comprising a spectral detection module and a processor; the spectral detection module is arranged on the artificial board production line and is located after the waterproofing agent adding process, the detection area of ​​the spectral detection module covers the conveying width of the artificial board raw materials, the spectral detection module comprises at least one spectral probe, the spectral detection module detects the artificial board raw materials and outputs a spectral data set to the processor in real time, the spectral data set comprises spectral data collected by each spectral probe; the processor receives the spectral data set, inputs the spectral data collected by each spectral probe into a pre-trained machine learning detection model or a multi-layer convolutional neural network model in real time, or converts the spectral data collected by each spectral probe into two-dimensional data and inputs it into a pre-trained multi-layer convolutional neural network model, obtains the waterproofing agent addition amount detection value of each spectral probe corresponding to the raw material detection area, performs uniformity judgment according to the waterproofing agent addition amount detection value of each spectral probe corresponding to the raw material detection area, and outputs the artificial board waterproofing agent mixing uniformity detection result according to the judgment result.

[0008] Optionally, the multi-layer convolutional neural network model includes an input layer, a first convolutional layer, a first maximum pooling layer, a first residual connection module, a second residual connection module, a third residual connection module, a second convolutional layer, a second maximum pooling layer, a fully connected layer and an output layer connected in sequence; each residual connection module includes a first convolution unit and a second convolution unit arranged in parallel, the first convolution unit includes two convolutional layers connected in sequence, the second convolution unit includes a convolution layer, each convolution layer is connected to a batch normalization layer, and the output features of the first convolution unit and the output features of the second convolution unit are added as the input of the next layer.

[0009] Optionally, the detection area of ​​the spectrum detection module completely overlaps with the conveying width of the wood-based panel raw material.

[0010] Optionally, the device for detecting uniformity of waterproofing agent mixing in the production process of artificial boards also includes an alarm, and the processor is connected to the alarm; the processor makes a uniformity judgment based on the detection value of the amount of waterproofing agent added in the raw material detection area corresponding to each spectral probe, including: judging whether the detection value of the amount of waterproofing agent added in the raw material detection area corresponding to each spectral probe is between a preset upper limit value of the amount of waterproofing agent added and a preset lower limit value of the amount of waterproofing agent added; if so, the uniformity of waterproofing agent mixing in the raw material detection area corresponding to the spectral probe is uniform; if it is greater than the upper limit value of the amount of waterproofing agent added, the uniformity of waterproofing agent mixing in the raw material detection area corresponding to the spectral probe is uneven, the amount of waterproofing agent added is too high, and the alarm is instructed to send an alarm message; if it is less than the lower limit value of the amount of waterproofing agent added, the uniformity of waterproofing agent mixing in the raw material detection area corresponding to the spectral probe is uneven, the amount of waterproofing agent added is too low, and the alarm is instructed to send an alarm message.

[0011] Optionally, the device for detecting the uniformity of mixing of waterproofing agents in the production process of artificial boards also includes a model trainer; the spectral detection module sequentially detects the same type of artificial board raw materials with different amounts of waterproofing agents added, and outputs a spectral data set to the model trainer; the model trainer divides the spectral data set of the same type of artificial board raw materials with different amounts of waterproofing agents added according to the received data, obtains a training set, a validation set and a test set, and uses an optimizer to train a machine learning detection model or a multi-layer convolutional neural network model according to the training set, the validation set and the test set to obtain a trained machine learning detection model or a multi-layer convolutional neural network model; the model trainer is connected to the processor, and the model trainer outputs the trained machine learning detection model or the multi-layer convolutional neural network model to the processor for model updating.

[0012] Optionally, the device for detecting the uniformity of waterproofing agent mixing during the production process of artificial boards also includes a model library, which pre-stores trained machine learning detection models or multi-layer convolutional neural network models for different types of artificial boards; the processor is connected to the model library, and the processor selects a corresponding machine learning detection model or multi-layer convolutional neural network model from the model library for model update according to the type of artificial board to be tested.

[0013] Optionally, the model trainer is connected to the model library, and the model trainer outputs the trained multi-layer convolutional neural network model to the model library for storage.

[0014] Optionally, the spectral probe is stationary on the wood-based panel production line; or, the spectral probe is movably arranged on the wood-based panel production line along the raw material conveying width direction. During the movement of the spectral probe, the detection areas of all spectral probes always cover the conveying width of the wood-based panel raw materials.

[0015] Optionally, the spectrum probe collects spectrum wavelength ranges including visible spectrum wavelengths, combined spectrum wavelengths of visible light and near-infrared light, near-infrared spectrum wavelengths, mid-infrared spectrum wavelengths, combined spectrum wavelengths of near-infrared light and mid-infrared light, or combined spectrum wavelengths of visible light, near-infrared light and mid-infrared light.

[0016] In a second aspect, the present invention provides a device for controlling the mixing uniformity of a waterproofing agent in a production process of artificial boards, comprising a controller and the waterproofing agent mixing uniformity detection device as described above, wherein the controller is connected to a processor, and the controller controls the mixing uniformity of the artificial board waterproofing agent according to the artificial board waterproofing agent mixing uniformity detection result output by the processor.

[0017] The beneficial effects of the present invention are:

[0018] The waterproofing agent mixing uniformity detection device provided by the present invention is provided with a spectrum detection module including at least one spectrum probe, and the detection area of ​​the spectrum detection module covers the conveying width of the artificial board raw material, so as to ensure that representative raw materials of different width areas on the artificial board production line are detected, and the comprehensiveness of the spectrum data is guaranteed, thereby improving the accuracy of the waterproofing agent mixing uniformity detection in the artificial board. By collecting the spectrum data of the artificial board raw material and analyzing the spectrum data using a specific model, the waterproofing agent mixing uniformity of the artificial board can be detected non-destructively, quickly and in real time, and the waterproofing agent mixing uniformity in the artificial board can be accurately detected online on the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a structural schematic diagram of a device for detecting the mixing uniformity of a waterproofing agent in a production process of an artificial board according to a specific embodiment of the present invention;

[0020] Figure 2 A schematic diagram of the structure of a multi-layer convolutional neural network model according to a specific embodiment of the present invention;

[0021] Figure 3 It is a structural schematic diagram of a spectrum detection probe statically arranged on a wood-based panel production line according to a specific embodiment of the present invention;

[0022] Figure 4 It is a schematic structural diagram of a spectrum detection probe according to a specific embodiment of the present invention, which is movably arranged on a wood-based panel production line along the raw material conveying width direction.

[0023] Description of Reference Numerals

[0024] 1: Spectral detection module; 2: Processor; 3: Alarm; 4: Model trainer; 5: Model library. DETAILED DESCRIPTION

[0025] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below with reference to the accompanying drawings through specific embodiments. It should be noted that the directions "upper", "lower", "left", "right", "front", "back" and so on mentioned herein are in the sense of Figure 3 The left and right directions are also the conveying width directions of the raw materials.

[0026] Figure 1 This is a schematic diagram of the structure of a device for detecting the mixing uniformity of a waterproofing agent in the production process of artificial boards.

[0027] like Figure 1 As shown, the device for detecting the mixing uniformity of a waterproofing agent in the production process of artificial boards provided by the present invention comprises a spectrum detection module 1 and a processor 2 .

[0028] Among them, the spectral detection module 1 is arranged on the artificial board production line and is located after the waterproofing agent adding process. The detection area of ​​the spectral detection module 1 covers the conveying width of the artificial board raw materials. The spectral detection module 1 includes at least one spectral probe. The spectral detection module 1 detects the artificial board raw materials and outputs a spectral data set to the processor 2 in real time. The spectral data set includes spectral data collected by each spectral probe.

[0029] Among them, processor 2 receives a spectral data set, and inputs the spectral data collected by each spectral probe into a pre-trained machine learning detection model or a multi-layer convolutional neural network model in real time, or converts the spectral data collected by each spectral probe into two-dimensional data and inputs it into a pre-trained multi-layer convolutional neural network model, obtains the waterproofing agent addition detection value of each spectral probe corresponding to the raw material detection area, performs uniformity judgment based on the waterproofing agent addition detection value of each spectral probe corresponding to the raw material detection area, and outputs the artificial board waterproofing agent mixing uniformity detection result based on the judgment result.

[0030] Preferably, if Figure 2 As shown, the multi-layer convolutional neural network model includes an input layer, a first convolutional layer, a first maximum pooling layer, a first residual connection module, a second residual connection module, a third residual connection module, a second convolutional layer, a second maximum pooling layer, a fully connected layer and an output layer connected in sequence; each residual connection module includes a first convolution unit and a second convolution unit arranged in parallel, the first convolution unit includes two convolutional layers connected in sequence, the second convolution unit includes a convolution layer, each convolution layer is connected to a batch normalization layer, and the output features of the first convolution unit and the output features of the second convolution unit are added as the input of the next layer.

[0031] Optionally, the machine learning detection model is a ridge regression model, a BP neural network model, an extreme learning machine model, a support vector machine model, a decision tree model, a random forest model, an XGBoost model, a KNN model, a principal component regression model, a linear regression model or an artificial neural network.

[0032] The waterproofing agent mixing uniformity detection device thus configured, by configuring a spectrum detection module 1 including at least one spectrum probe, and the detection area of ​​the spectrum detection module 1 covers the conveying width of the artificial board raw material, ensures that representative raw materials of different width areas on the artificial board production line are detected, ensures the comprehensiveness of the spectrum data, and thus improves the accuracy of the waterproofing agent mixing uniformity detection in the artificial board. By collecting the spectrum data of the artificial board raw material and analyzing the spectrum data using a specific model, the waterproofing agent mixing uniformity of the artificial board can be detected non-destructively, quickly, and in real time, and thus the waterproofing agent mixing uniformity in the artificial board can be accurately detected online on the production line.

[0033] It should be noted that the conveying width direction of the man-made board raw material is perpendicular to the conveying forward direction of the man-made board raw material. The man-made board raw material can be wood shavings or wood fibers prepared from wood species such as poplar, eucalyptus, pine, and fir, and the wood shavings or wood fibers can be wood shavings or wood fibers of a single tree species, or wood shavings or wood fibers of mixed tree species, or bamboo shavings or bamboo fibers prepared from bamboo. The waterproofing agent can be paraffin, rosin, asphalt, synthetic resin, drying oil, silicone resin, and other commonly used waterproofing agents in the production of man-made boards.

[0034] It should be noted that the interval time for the spectrum detection module 1 to collect spectrum data is set according to the detection area of ​​the spectrum detection module 1 and the running speed of the artificial board production line.

[0035] Preferably, the spectrum detection module 1 includes 1 to 10 spectrum detection probes.

[0036] Preferably, the spectrum probe collects spectrum wavelength ranges including visible spectrum wavelengths, combined spectrum wavelengths of visible light and near-infrared light, near-infrared spectrum wavelengths, mid-infrared spectrum wavelengths, combined spectrum wavelengths of near-infrared light and mid-infrared light, or combined spectrum wavelengths of visible light, near-infrared light and mid-infrared light.

[0037] Preferably, the spectrum probe is arranged on the artificial board raw material conveying pipeline, above the artificial board raw material after paving is completed, or above the artificial board raw material after pre-pressing is completed.

[0038] Preferably, the spectrum probes are stationary on the wood-based panel production line, and the detection areas of all spectrum probes cover the conveying width of the wood-based panel raw materials; or, the spectrum probes are movably arranged on the wood-based panel production line along the direction of the raw material conveying width, and during the movement of the spectrum probes, the detection areas of all spectrum probes always cover the conveying width of the wood-based panel raw materials.

[0039] Specifically, the spectrum probe is statically set on the artificial board production line, and the intervals between each adjacent spectrum probe are equal. The spectrum probes are evenly distributed on the artificial board production line to ensure that representative raw materials of different widths on the artificial board production line are detected. It is suitable for being set on the artificial board raw material delivery pipeline, above the artificial board raw material after paving, or above the artificial board raw material after pre-pressing. When the spectrum probe is set on the artificial board raw material delivery pipeline, a quartz window is opened on the wall of the artificial board raw material delivery pipeline, and the spectrum probe detects the raw materials in the pipeline through the quartz window.

[0040] As an example, Figure 3As shown, the spectral detection module 1 includes 4 spectral probes, which are arranged statically above the raw materials after pre-pressing along a straight line perpendicular to the conveying direction of the artificial board raw materials. The 4 spectral probes are arranged from left to right as spectral probe 1, spectral probe 2, spectral probe 3 and spectral probe 4. Spectral probe 1 is 0.16 meters away from the left end of the raw material conveying area of ​​the artificial board production line, spectral probe 2 is 0.3 meters away from spectral probe 1, spectral probe 3 is 0.3 meters away from spectral probe 2, spectral probe 4 is 0.3 meters away from spectral probe 3, and spectral probe 4 is 0.16 meters away from the right end of the raw material conveying area of ​​the artificial board production line. The distance between the four spectral probes and the surface of the artificial board raw materials on the artificial board production line is 0.025 meters.

[0041] Specifically, the spectral probe is movably arranged on the wood-based panel production line along the width direction of the raw material conveying, so that the detection trajectory of the spectral probe relative to the wood-based panel raw material is "S" shaped, ensuring the detection of wood-based panel raw materials in different width areas to the maximum extent. It is suitable for being arranged above the raw materials after paving is completed or above the raw materials after pre-pressing is completed.

[0042] As an example, Figure 4 As shown, the spectral detection device includes two spectral probes, which are arranged in sequence from front to back and from left to right above the raw material after pre-pressing, and the two spectral probes are movably arranged on the wood-based panel production line along the raw material conveying width direction. The two spectral probes are arranged in sequence from front to back as spectral probe 1 and spectral probe 2, and the distance between spectral probe 1 and spectral probe 2 and the surface of the wood-based panel raw material on the wood-based panel production line is 0.025 meters.

[0043] Preferably, the detection area of ​​the spectrum detection module 1 completely overlaps with the conveying width of the artificial board raw material. In this way, the spectrum data collected by the spectrum probes at the left and right ends of the artificial board production line are more comprehensive, and the detection value of the amount of waterproofing agent added obtained after the spectrum data is processed by the processor 2 is more accurate.

[0044] In processor 2, when spectral data is input into the multi-layer convolutional neural network model, the multi-layer convolutional neural network model is a one-dimensional convolutional neural network, and when two-dimensional data converted from spectral data is input into the multi-layer convolutional neural network model, the multi-layer convolutional neural network model is a two-dimensional convolutional neural network.

[0045] Preferably, in the one-dimensional convolutional neural network, the first convolution layer uses the tanh activation function, the number of convolution kernels is 128, the size of the convolution kernel is 1*5, and the step size of the convolution kernel is 1; the second convolution layer uses the tanh activation function, the number of convolution kernels is 256, the size of the convolution kernel is 1*5, and the step size of the convolution kernel is 1; the convolution layer of the residual connection module uses the tanh activation function, the number of convolution kernels is 128, the size of the convolution kernel is 1*3, and the step size of the convolution kernel is 1; the size of the pooling kernels of the first maximum pooling layer and the second maximum pooling layer are both 1*2, and the step size of the pooling kernels is 2.

[0046] Preferably, in the two-dimensional convolutional neural network, the second convolutional layer uses the relu activation function, the number of convolution kernels is 128, the size of the convolution kernel is 5*5, and the step size of the convolution kernel is 1; the second convolutional layer uses the relu activation function, the number of convolution kernels is 256, the size of the convolution kernel is 5*5, and the step size of the convolution kernel is 1; the convolution layer of the residual connection module uses the relu activation function, the number of convolution kernels is 128, the size of the convolution kernel is 3*3, and the step size of the convolution kernel is 1; the size of the pooling kernel of the first maximum pooling layer and the second maximum pooling layer is both 2*2, and the step size of the pooling kernel is both 2.

[0047] Preferably, converting the spectral data collected by each spectral probe into two-dimensional data includes: converting the spectral data collected by each spectral probe into two-dimensional data using a recurrence plot method.

[0048] Furthermore, the spectral data collected by each spectral probe is converted into two-dimensional data using a recursive graph method, including:

[0049]

[0050] Where p(i,j) is the pixel with coordinates (i,j) in the recursive graph, i = 1,2,...n, j = 1,2,...,n; S = {s 1 ,s 2 ,……s (n-1) ,s n} represents the spectral data with n wavelength points. The data is normalized by dividing by the maximum value of S. The recurrence graph is obtained by visualizing the pixel value of each coordinate.

[0051] Preferably, the processor 2 receives a spectral data set, preprocesses the spectral data set, converts the spectral data collected by each preprocessed probe or converts the spectral data collected by each preprocessed spectral probe into two-dimensional data and inputs it into a pre-trained multi-layer convolutional neural network model, obtains the waterproofing agent addition amount detection value of each spectral probe corresponding to the raw material detection area, performs uniformity judgment based on the waterproofing agent addition amount detection value of each spectral probe corresponding to the raw material detection area, and outputs the artificial board waterproofing agent mixing uniformity detection result based on the judgment result.

[0052] Furthermore, preprocessing the spectral data set includes: preprocessing the spectral data set by using one or a combination of two or more methods such as data smoothing, first-order differential, second-order differential, standard variable transformation, multivariate scatter correction, centering, detrending, wavelet transform, etc.

[0053] Preferably, the waterproofing agent mixing uniformity detection device also includes an alarm 3, and the processor 2 is connected to the alarm 3; the processor 2 performs uniformity judgment based on the waterproofing agent addition detection value of each spectral probe corresponding to the raw material detection area, including: judging whether the waterproofing agent addition detection value of each spectral probe corresponding to the raw material detection area is between a preset upper limit value of the waterproofing agent addition and a preset lower limit value of the waterproofing agent addition; if so, the waterproofing agent mixing uniformity of the raw material detection area corresponding to the spectral probe is uniform; if it is greater than the upper limit value of the waterproofing agent addition, the waterproofing agent mixing uniformity of the raw material detection area corresponding to the spectral probe is uneven, the waterproofing agent addition is too high, and an alarm message is issued; if it is less than the lower limit value of the waterproofing agent addition, the waterproofing agent mixing uniformity of the raw material detection area corresponding to the spectral probe is uneven, the waterproofing agent addition is too low, and an alarm message is issued.

[0054] Preferably, the device for detecting the mixing uniformity of waterproofing agents in the production process of artificial boards provided by the present invention also includes a model trainer 4; the spectral detection module 1 detects the same type of artificial board raw materials with different amounts of waterproofing agents added in sequence, and outputs a spectral data set to the model trainer 4; the model trainer 4 divides the spectral data set of the same type of artificial board raw materials with different amounts of waterproofing agents added according to the received data, obtains a training set, a validation set and a test set, and uses an optimizer to train the multi-layer convolutional neural network model according to the training set, the validation set and the test set to obtain a trained multi-layer convolutional neural network model; the model trainer 4 is connected to the processor 2, and the model trainer 4 outputs the trained multi-layer convolutional neural network model to the processor 2 for model updating.

[0055] The waterproofing agent mixing uniformity detection device configured in this way can instantly train the corresponding multi-layer convolutional neural network model according to the different types of artificial boards prepared on the production line, and input the trained multi-layer convolutional neural network model into the processor 2 to detect the mixing uniformity of the artificial board waterproofing agent.

[0056] The differences in the types of artificial boards are mainly reflected in: the source of the raw materials, production time, size, moisture content, tree species conditions, etc.

[0057] Preferably, in the spectral data set of the same type of wood-based panel raw materials with different amounts of waterproofing agent added, the number of spectra with the same amount of waterproofing agent added is greater than 30.

[0058] It should be noted that the spectral data set of the same type of wood-based panel raw materials with different amounts of waterproofing agent added is divided to obtain a training set, a validation set and a test set, including: converting the spectral data set into a two-dimensional data set, and obtaining the training set, the validation set and the test set according to the division of the two-dimensional data set.

[0059] Preferably, the spectral data set of the same type of artificial board raw material with different water-repellent addition amounts is randomly divided into a training set, a validation set and a test set according to a common rule or ratio. For example, a five-fold cross-validation division or an 8:1:1 random division is used. The divided training set, validation set and test set require that each data set simultaneously contains the spectral data of the artificial board raw material under the conditions of different water-repellent addition amounts.

[0060] Preferably, the multi-layer convolutional neural network model is trained using an optimizer, and the training of the multi-layer convolutional neural network model is completed when the correlation coefficient between the predicted results and the actual results of the test set data is greater than 0.98.

[0061] Preferably, the device for detecting the mixing uniformity of waterproofing agents in the production process of artificial boards provided by the present invention also includes a model library 5, which pre-stores trained multi-layer convolutional neural network models of different types of artificial boards; the processor 2 is connected to the model library 5, and the processor 2 selects the corresponding multi-layer convolutional neural network model from the model library 5 for model update according to the type of artificial board to be tested.

[0062] Further preferably, the model trainer 4 is connected to the model library 5, and the model trainer 4 outputs the trained multi-layer convolutional neural network model to the model library 5 for storage.

[0063] The present invention also provides a device for controlling the mixing uniformity of a waterproofing agent during the production process of artificial boards, comprising a controller and the above-mentioned waterproofing agent mixing uniformity detection device, wherein the controller is connected to a processor 2, and the controller controls the mixing uniformity of the artificial board waterproofing agent according to the artificial board waterproofing agent mixing uniformity detection result output by the processor 2.

[0064] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0065] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0066] In the present invention, unless otherwise clearly specified and limited, when a first feature is “on” or “below” a second feature, it may be that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. Moreover, when a first feature is “above”, “above” or “above” a second feature, it may be that the first feature is directly above or obliquely above the second feature, or it may simply mean that the first feature is higher in level than the second feature. When a first feature is “below”, “below” or “below” a second feature, it may be that the first feature is directly below or obliquely below the second feature, or it may simply mean that the first feature is lower in level than the second feature.

[0067] In the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0068] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may alter, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A device for detecting the mixing uniformity of a waterproofing agent in the production process of artificial boards, characterized in that: It comprises a spectrum detection module (1) and a processor (2); The spectrum detection module (1) is arranged on the artificial board production line and is located after the step of adding a waterproofing agent. The detection area of ​​the spectrum detection module (1) covers the conveying width of the artificial board raw material. The spectrum detection module (1) comprises at least one spectrum probe. The spectrum detection module (1) detects the artificial board raw material and outputs a spectrum data set to the processor (2) in real time. The spectrum data set comprises spectrum data collected by each spectrum probe. The processor (2) receives a spectral data set, inputs the spectral data collected by each spectral probe into a pre-trained machine learning detection model or a multi-layer convolutional neural network model in real time, or converts the spectral data collected by each spectral probe into two-dimensional data and inputs it into a pre-trained multi-layer convolutional neural network model, obtains a waterproofing agent addition amount detection value of each spectral probe corresponding to a raw material detection area, performs uniformity judgment based on the waterproofing agent addition amount detection value of each spectral probe corresponding to the raw material detection area, and outputs a test result of the mixing uniformity of the artificial board waterproofing agent based on the judgment result; The spectrum probe is arranged on a conveying pipeline of the artificial board raw material, above the artificial board raw material after paving is completed, or above the artificial board raw material after pre-pressing is completed.

2. The device for detecting the mixing uniformity of waterproofing agent in the production process of artificial board according to claim 1, characterized in that: The multi-layer convolutional neural network model includes an input layer, a first convolutional layer, a first maximum pooling layer, a first residual connection module, a second residual connection module, a third residual connection module, a second convolutional layer, a second maximum pooling layer, a fully connected layer and an output layer connected in sequence; each residual connection module includes a first convolutional unit and a second convolutional unit arranged in parallel, the first convolutional unit includes two convolutional layers connected in sequence, the second convolutional unit includes a convolutional layer, each convolutional layer is connected to a batch normalization layer, and the output features of the first convolutional unit and the output features of the second convolutional unit are added as the input of the next layer.

3. The device for detecting the mixing uniformity of waterproofing agent in the production process of artificial board according to claim 1, characterized in that: The detection area of ​​the spectrum detection module (1) completely overlaps with the conveying width of the artificial board raw material.

4. The device for detecting the mixing uniformity of waterproofing agent in the production process of artificial board according to claim 1, characterized in that: It also includes an alarm (3), and the processor (2) is connected to the alarm (3); The processor (2) performs uniformity judgment based on the waterproofing agent addition amount detection value of each spectral probe corresponding to the raw material detection area, including: judging whether the waterproofing agent addition amount detection value of each spectral probe corresponding to the raw material detection area is between a preset upper limit value of the waterproofing agent addition amount and a preset lower limit value of the waterproofing agent addition amount, if so, the waterproofing agent mixing uniformity of the raw material detection area corresponding to the spectral probe is uniform; if greater than the upper limit value of the waterproofing agent addition amount, the waterproofing agent mixing uniformity of the raw material detection area corresponding to the spectral probe is uneven and the waterproofing agent addition amount is too high, and the alarm (3) is instructed to issue an alarm message; if less than the lower limit value of the waterproofing agent addition amount, the waterproofing agent mixing uniformity of the raw material detection area corresponding to the spectral probe is uneven and the waterproofing agent addition amount is too low, and the alarm (3) is instructed to issue an alarm message.

5. The device for detecting the mixing uniformity of waterproofing agent in the production process of artificial board according to claim 1, characterized in that: Also included is a model trainer (4); The spectral detection module (1) detects the same type of artificial board raw materials with different amounts of waterproofing agent added in sequence, and outputs the spectral data set to the model trainer (4); The model trainer (4) divides the received spectral data sets of the same type of artificial board raw materials with different amounts of waterproofing agent added to obtain a training set, a validation set and a test set, and uses an optimizer to train the machine learning detection model or the multi-layer convolutional neural network model based on the training set, the validation set and the test set to obtain a trained machine learning detection model or the multi-layer convolutional neural network model; The model trainer (4) is connected to the processor (2), and the model trainer (4) outputs the trained machine learning detection model or the multi-layer convolutional neural network model to the processor (2) for model updating.

6. The device for detecting the mixing uniformity of waterproofing agent in the production process of artificial board according to claim 1 or 5, characterized in that: It also includes a model library (5), in which the model library (5) pre-stores trained machine learning detection models or multi-layer convolutional neural network models of different types of artificial boards; The processor (2) is connected to the model library (5), and the processor (2) selects a corresponding machine learning detection model or a multi-layer convolutional neural network model from the model library (5) according to the type of the artificial board to be tested to perform model update.

7. The device for detecting the mixing uniformity of waterproofing agent in the production process of artificial board according to claim 6, characterized in that: The model trainer (4) is connected to the model library (5), and the model trainer (4) outputs the trained multi-layer convolutional neural network model to the model library (5) for storage.

8. The device for detecting the mixing uniformity of waterproofing agent in the production process of artificial board according to claim 1, characterized in that: The spectrum probe is statically set on the wood-based panel production line; or, The spectrum probes are movably arranged on the artificial board production line along the raw material conveying width direction. During the movement of the spectrum probes, the detection areas of all the spectrum probes always cover the conveying width of the artificial board raw materials.

9. The device for detecting the mixing uniformity of waterproofing agent in the production process of artificial board according to claim 1, characterized in that: The spectrum probe collects spectrum wavelengths including visible spectrum wavelengths, combined spectrum wavelengths of visible light and near-infrared light, near-infrared spectrum wavelengths, mid-infrared spectrum wavelengths, combined spectrum wavelengths of near-infrared light and mid-infrared light, or combined spectrum wavelengths of visible light, near-infrared light and mid-infrared light.

10. A device for controlling the mixing uniformity of a waterproofing agent in the production process of artificial boards, characterized in that: The invention comprises a controller and a waterproofing agent mixing uniformity detection device as claimed in any one of claims 1 to 9, wherein the controller is connected to a processor (2), and the controller controls the mixing uniformity of the artificial board waterproofing agent according to the artificial board waterproofing agent mixing uniformity detection result output by the processor (2).

Images