Method and apparatus for controlling the modification process of hygroscopic materials

Abstract

A method and apparatus (10) for controlling a reforming process of a hygroscopic material (15) includes the steps of measuring at least one process variable from the reforming process at least during the reforming, measuring at least one process variable from the hygroscopic material at least during the reforming, calculating at least one intermediate control parameter by a neural network by using at least the measured process variable as an input parameter for the neural network, and controlling the reforming process using a genetic algorithm and genetic programming based on the at least one intermediate control parameter determined by the neural network.

Description

[Technical Field]

[0001] The present invention relates to a method and apparatus for controlling a modification process, such as a thermomechanical modification process, of a hygroscopic material. [Background technology]

[0002] Hygroscopic materials are materials that have the ability to absorb and store moisture from the surrounding air. When the relative humidity changes, the difference in partial vapor pressure causes the material to reach equilibrium, resulting in the absorption or desorption of moisture. Depending on the properties of hygroscopic materials, they have many applications in different sectors of industry. Typical hygroscopic materials used in industry are wood, wood-plastic composites, plant-based materials, concrete, among others.

[0003] Increasing environmental awareness has led the market to seek more sustainable options. In particular, materials that are not refined using toxic chemicals can be produced more responsibly by companies and can be conveniently disposed of when necessary. Wood, a natural, virtually long-lasting, and moisture-absorbing material, offers an attractive and cost-effective solution to these environmental sustainability demands. However, in developing countries, 80% of all harvested wood is currently discarded or burned. In particular, less than 5% of tropical forests are managed sustainably.

[0004] Current wood processing methods for forming hardwood, such as traditional drying and pressing processes, do not sufficiently offer an economically attractive solution for investors to increase the proportion of harvested wood used in the production of sustainable wood products, such as building and furniture. In particular, some lower-grade tropical timbers have been considered unsuitable for such uses because they have not been able to reach quality levels that meet the demands regarding dimensional accuracy and surface quality as well as the mechanical properties of such products. Furthermore, when using modern drying and pressing processes, the process delivery cycle for such timbers, and for some higher-grade timber species, is often too long.

[0005] Thermomechanical modification processes, such as thermomechanical wood modification (TMTM), allow for greater utilization of harvested wood. Thermomechanical modification processes can modify properties of hygroscopic materials, such as compressive strength, stiffness, density, hardness, and dimensional stability, to suit specific uses of the material. One known thermomechanical modification process is disclosed in WO 2022 / 175585 A1.

[0006] However, effective and optimal control of a reforming process, such as the thermomechanical reforming process described above, is problematic due to the many factors that affect the process. Summary of the Invention

[0007] The objective of the present invention is to realize a new method for controlling the modification process of hygroscopic materials, which significantly shortens the process delivery cycle and improves the quality level of the modified hygroscopic materials. In particular, the objective of the present invention is to develop a method for controlling the modification process of hygroscopic materials, which significantly improves the quality parameters of the modified hygroscopic materials, such as material strength, surface hardness, and dimensional stability, while significantly shortening the process delivery cycle.

[0008] The object of the present invention is achieved by the method for treating hygroscopic materials according to the present invention, in that the modification process for modifying the hygroscopic material is controlled using genetic algorithms and genetic programming based on at least one intermediate parameter calculated by the neural network by using at least one measured process variable of the modification process and at least one measured process variable of the hygroscopic material as input parameters of the neural network, since the neural network is trained taking into account the case-by-case nonlinear dependencies of the various input parameters depending on the material and its initial state, and the desired properties to be achieved in the final product.

[0009] Measured process variables of the reformulation process include, but are not limited to, for example, air temperature, air velocity, air relative moisture content, air pressure, compression force, and compression speed.

[0010] Measuring process variables for hygroscopic materials include, but are not limited to, humidity, temperature, moisture gradient, microcrack development, size of modified pieces, compression (thickness), and the like.

[0011] Other suitable additional input parameters for the neural network may also be used.

[0012] Preferably, the modification process controlled by the method of the present invention is a thermomechanical modification process of a hygroscopic material. Furthermore, in the method of the present invention, the moisture gradient of the hygroscopic material is preferably measured by electrical impedance spectroscopy (EIS), and the occurrence of microcracks in the hygroscopic material is preferably monitored by acoustic emission (AE).

[0013] More precisely, the method according to the invention is characterized by what is stated in independent claim 1, and the device according to the invention is characterized by what is stated in independent claim 15. Dependent claims 2 to 14 show some advantageous embodiments of the method according to the invention.

[0014] The advantages of the present invention are that the method and apparatus can maintain influencing factors such as the moisture gradient value and the amount of microcracks in the hygroscopic material during modification within acceptable levels, and the modified hygroscopic material achieves an overall appropriate quality, significantly shortening the process delivery cycle. In particular, the hygroscopic material's ability to absorb moisture from the air and its strength are improved. Therefore, the devised new method and apparatus allows hygroscopic materials such as low-grade tropical wood to be used for more sustainable purposes, such as furniture and building products, instead of being burned or discarded, as is largely the case today.

[0015] Furthermore, genetic algorithms and genetic programming offer additional advantages in this type of application by solving many different problems in controlling the reforming process that cannot be solved by current control methods. These include: Drying stage monitoring techniques can only be used if the mutual and overall influence of changes in the observed process values ​​on changes in the properties of the material being dried is known. Hygroscopic materials such as wood are typically heterogeneous natural materials, with thousands of varieties, all with different physical properties. Each has its own unique characteristics that must be taken into account in order to achieve the desired properties in the final product. If the objective is to modify even approximately certain physical properties (surface hardness, strength, moisture content, torsional rigidity, ageing resistance without chemically or environmentally harmful substances), the task forms a theoretical optimization situation where genetic algorithms (GA) and genetic programming (GP) methods, together with appropriate online measurement techniques, provide more effective tools than traditional methods, where control programs based on a single, predetermined mathematical model are used. [Brief explanation of the drawings]

[0016] The invention will now be described in more detail with reference to the accompanying drawings. [Figure 1] FIG. 1 shows a schematic cross section of a reforming chamber of a reformer applied in one embodiment of the device according to the invention. [Figure 2] FIG. 2 illustrates one embodiment of an artificial neural network for use in the present invention. [Figure 3] FIG. 3 shows a flow chart of an embodiment of the method according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0017] The method of the present invention controls a modification process of a hygroscopic material, such as a thermomechanical modification process. In such a thermomechanical modification process, a hygroscopic material, such as wood, is typically treated in a modification apparatus having a modification chamber through which the hygroscopic material is transported. The hygroscopic material is modified by applying a process that includes multiple process steps, such as wetting, drying, heating, and compression. Therefore, atmospheric conditions in the modification chamber, such as air temperature, relative moisture content of the air, and compression force, are varied according to an applied control program. The procedure used depends, among other things, on the type and quality of the hygroscopic material before modification, the size of the pieces to be modified, and the desired appearance and mechanical properties to be achieved in the hygroscopic material upon modification. After the hygroscopic material has reached the desired properties, it is transferred out of the modification chamber for further processing, such as packaging, storage, and / or transportation to another internal manufacturing department or to the end user of the material.

[0018] An example of such a reforming apparatus 10 is shown in Figure 1. The apparatus includes a reforming chamber 11 into which a hygroscopic material 15 can be thermomechanically reformed. In this example, the hygroscopic material to be reformed is in the form of a batch, i.e., it includes multiple pieces of such material arranged adjacent to and / or on top of each other, with or without one or more intermediate pieces between each row and / or column of pieces of hygroscopic material. These intermediate pieces 16 are preferably cellular plates, plates, or stickers, having hollow spaces or channels through which air blown into the reforming chamber flows.

[0019] The hygroscopic material to be modified in the modification apparatus 10 shown in Figure 1 may be, for example, wood, a wood-plastic composite, or any plant-based material, preferably one that reacts similarly to wood when modified.

[0020] Within the reforming chamber 11 is a compression device 12, which includes a first compression member 13 and a second compression member 14, between which a batch of hygroscopic material 15 to be reformed is placed and thereby compressed during reforming. The first compression member 13 and the second compression member 14 are platform-like elements with flat compression surfaces between which the material to be reformed can be placed as it is reformed. In this embodiment, the compression device has only two compression members and is therefore applicable to compressing the material to be reformed in the thickness direction of the piece of hygroscopic material. However, in other embodiments of the method and apparatus, the compression device may include additional compression members for compressing the piece of hygroscopic material in other directions, i.e., widthwise and / or lengthwise.

[0021] The reforming chamber 11 of the reformer 10 shown in Figure 1 is also provided with heating means 18, air blowing means 19, and humidifying means (not shown). The heating means 18 may be, for example, an electric heater, an oil heater, or a suitable biofuel heating device. The air blowing means 19 is preferably an electric fan, and the humidifying means may consist of liquid spray and / or steam generating equipment and devices.

[0022] The reformer 10 shown in FIG. 1 also includes measurement means for measuring various properties of the hygroscopic material. In particular, when applying the method of the present invention, the moisture gradient of the hygroscopic material to be reformed is preferably measured by electrical impedance spectroscopy (EIS), and the amount of microcracks in the hygroscopic material is preferably measured by acoustic emission (AE). Other types of sensors and measurements can also be used to determine the moisture gradient and the amount of microcracks. These measurements are performed at least during the thermomechanical reforming, but can also be performed before and / or after the thermomechanical reforming. Online measurements can be arranged by suitable sensors connected to the control unit 20 of the reformer 10 wirelessly, for example, via a WLAN network or other suitable wireless data communication technology. Alternatively, a wired connection can also be used.

[0023] Additionally, several other process parameters and physical quantities, such as temperature, moisture content of the air and / or hygroscopic material being reformed, and weight of the hygroscopic material being reformed within the reformer 10, can be measured using appropriate measurement sensor technology. Online measurement devices / sensors and testing devices / instruments can be connected to the reformer 10. Separate laboratory measurements can also be applied to verify determined property values. These laboratory measurements are most preferably applied to properties such as moisture content, hardness, and strength of the hygroscopic material. The control unit 20, to which the online measurement means are connected and / or to which additional input data are supplied, is configured to control various devices of the reformer, such as the compressor 12, heating means 18, blower means 19, and humidifier, according to a control program executed by the control unit 20. Accordingly, the control unit 20 includes a computing means for executing the control program and a memory for storing the control program and processed data. The computing means is therefore capable of controlling these devices so as to implement the method according to the invention, i.e. it is equipped with a program for implementing a neural network as well as a genetic algorithm or genetic programming for carrying out data processing according to the method of the invention. The control unit also requires electronic circuits and components for controlling actuators such as the fan and actuator of the compressor, and measuring devices of the reformer for controlling them, so that the reforming process carried out by the reformer is fully automatically controlled by the control unit 20.

[0024] Therefore, the above control program in the control unit 20 controls the thermomechanical modification process of the hygroscopic material according to the method of the present invention. The method of this embodiment includes at least the following method steps: Measure the temperature and moisture in the reforming chamber; measuring the moisture gradient of the hygroscopic material, preferably by electrical impedance spectroscopy (EIS), at least during the modification; monitoring the amount of microcracks occurring in the hygroscopic material, preferably by acoustic emission (AE), at least during the modification; calculating at least one intermediate control parameter by the neural network using at least the measured variables as input parameters of the neural network; Based on the at least one intermediate control parameter determined by the neural network, a genetic algorithm and genetic programming are used to control the thermomechanical reforming process.

[0025] Electrical impedance spectroscopy (EIS) is the measurement of an object's impedance (AC resistance) at multiple frequencies. The result is a frequency spectrum that provides information about the object's structure and properties. Impedance spectroscopy has a variety of applications, one of the most important of which is the study of biological materials, for example in medicine. In this invention, EIS technology is used to monitor the structure and moisture distribution and gradients of hygroscopic materials in real time.

[0026] The electrical impedance spectroscopy measuring device may include suitable electrodes that are placed on the surface of the material to be measured and through which impedance measurements are taken during reforming. A separate measuring device with electrodes may be provided within the reforming chamber and may be wirelessly connected to the control unit 20 of the reformer.

[0027] Acoustic emission (AE) can be used to monitor microcracks that may occur during drying. When hygroscopic materials such as wood are exposed to cracking during drying, microcracks are the first phenomenon observed. The material emits ultrasonic frequency sounds during microcracks. By monitoring acoustic emissions and adjusting the modification process accordingly, macrocracks that negatively affect the quality of the final product can be eliminated.

[0028] The acoustic emission measuring device can consist, for example, of a piezoelectric sensor that measures ultrasonic waves formed by microcracks in the hygroscopic material being monitored. The sensor can be wirelessly connected to the control unit 20 and provide the measurement data accordingly.

[0029] Genetic algorithms are heuristic optimization methods that mimic the mechanisms of natural evolution. They are suitable for tasks where the solution space is very large (e.g., large-scale combinatorial problems) and where near-optimal solutions are sufficient. With the rapid growth in computing power, the potential applications of genetic algorithms have expanded significantly over the past decade.

[0030] Genetic algorithms (GAs) and genetic programming (GPs) are methods for tuning and controlling thermomechanical reforming processes to improve final product quality. This is because traditional mathematical analysis does not or cannot provide an analytical solution. GAs and GPs are adaptive methods, adapting to changes in the reforming process. The process improves without external intervention through real-time optimization or machine learning. Adaptation is continuous. Currently, the internal relationships between the relevant variables are not fully understood (or there is reason to suspect that the current understanding is incorrect). Finding the size and shape of the final solution to the problem is a critical part of the problem. Approximate solutions are acceptable (or are the only solution likely to be reached). Such tasks generate large amounts of computer-readable data that must be reviewed, sorted, and aggregated. Small improvements in performance are measured periodically (or are easily measurable) and are important to the overall result.

[0031] Furthermore, the added value of genetic algorithms and genetic programming in the measurement and control technology of thermomechanical modification processes is that when drying hygroscopic materials during the thermomechanical modification process, the priority is not to control the humidity and temperature of the air, but to ensure that the hygroscopic material being dried is not damaged as a result of drying, and that the final moisture content of the dried hygroscopic material is sufficiently uniform and the moisture gradient of the material is low, thereby achieving a processed hygroscopic material with an intact structure and with the moisture distribution, gradient, and desired moisture content kept as low as possible.

[0032] To achieve this and other objectives of this method, the EIS and AE monitoring methods described above allow for real-time detection of the time and circumstances under which damage to trees begins. Knowing the time and circumstances allows efforts to be made to prevent such situations. This allows for conditions to be created that prevent hygroscopic materials from cracking or becoming damaged.

[0033] An artificial neural network (hereafter "neural network"), an embodiment of which is shown in FIG. 2, is a nonlinear statistical data modeling or decision-making tool. It can be used to model complex relationships between inputs and outputs or to find patterns in data. It comprises a network of simple processing elements (artificial neurons) that can exhibit complex overall behavior determined by the relationships between the processing elements and element parameters. In the present invention, a neural network is applied to determine the relationship between the properties of a hygroscopic material and measured moisture gradients and microcracks, as well as other measured quantities of the hygroscopic material being modified.

[0034] In the embodiment of FIG. 2, measurable input data for Input Layer A include, but are not limited to, initial moisture content of the material, process temperature, relative humidity, steam consumption, moisture content gradient, material weight and / or density, microcracks, wood type, compaction pressure, compaction rate and / or velocity, air velocity and / or direction, and / or changes in input values.

[0035] In the neural network, the complex relationships are defined by the hidden layer BD, a deep learning stage between the input data and the output data from the output layer E, which may include, but are not limited to, color gamut, hardness, strength, modulus of elasticity (MOE), modulus of rupture (MOR), compression, density, dimensional stability, fire resistance, rot resistance, fungus resistance, termite resistance, temperature / humidity calibration and correction of process sensors, etc.

[0036] The intermediate control parameters may be moisture gradient and microcrack values ​​calculated using a neural network based on the initial state of the hygroscopic material being modified, or they may be other values ​​and / or combinations of values ​​calculated using a neural network and then used as inputs to a genetic algorithm used to control the modification device.

[0037] A flow chart of an embodiment of the method according to the present invention is shown in FIG.

[0038] In the embodiment of Figure 3, the modification process of the hygroscopic material is controlled based on available data, as shown in box 101. At the start of the modification process, this data may be based on or consist of measurable variables related to the treated material, such as moisture content, weight, density, wood type, etc., and the process itself, such as process temperature, air velocity, compaction pressure, etc.

[0039] During the reformulation process, data related to the process is collected by appropriate sensors and measurements, as shown in box 102 .

[0040] The acquired process data is used as input data for a neural network, as shown in box 103. The output data from the neural network is then used as input data for a genetic algorithm, as shown in box 104.

[0041] Improved control data for the reformulation process is obtained from the genetic algorithm, as shown in box 105, and is then used to control the actual reformulation process, as shown in box 101.

[0042] The process shown in FIG. 3 can be carried out multiple times until the modification process achieves the desired quality and properties of the modified hygroscopic material.

[0043] Embodiments of the method of the present invention may further include determining expected values ​​for the moisture gradient and the amount of microcracks, training a neural network, and controlling the modification process so that the measured values ​​of the moisture gradient and the amount of microcracks are as close as possible to the expected values ​​of the moisture gradient and the amount of microcracks (i.e., the values ​​calculated using the neural network).

[0044] One embodiment of the method of the present invention may further include determining the initial state of the hygroscopic material to be modified. Determining the initial state of the hygroscopic material means determining characteristics such as the initial moisture content, initial moisture gradient, and initial amount of microcracks before modification. These can be determined by appropriate investigations and / or measurements. Measurements to determine the initial state of the hygroscopic material can be performed as laboratory measurements before modification or using online measurements in the modification chamber before modification begins. Knowing the initial state of the hygroscopic material to be modified can improve and speed up the process and prevent situations in which deviations between the expected and actual initial states of the hygroscopic material cause the control program to fail to achieve the best possible results.

[0045] Embodiments of the method of the present invention can further include determining temporary control values ​​and parameters. In some cases, some hygroscopic materials may have parameter values ​​that may deviate from normal values. Some hygroscopic materials may also have properties or behaviors, so that their modification may require additional parameters to be used to control the process. In such cases, these parameters or values ​​can be determined before starting the process or during the process, for example, when individual values ​​that trigger normal parameters or combinations of multiple values ​​of normal parameters are obtained.

[0046] Method embodiments of the present invention may further include drying the hygroscopic material using a function of a problem, variables, and parameters. The problem function may be a function provided to mathematically calculate the relationship between measured or monitored parameters, such as moisture gradient and amount of microcracks, taking into account the atmospheric conditions or applied pressure of a compression device used in the reforming chamber.

[0047] An embodiment of the method of the present invention may further include evaluating the dried hygroscopic material, which may include, for example, detecting the surface quality of the hygroscopic material and dimensional accuracy (e.g., straightness) of the piece, measuring strength and / or hardness, or determining the moisture content of the dried hygroscopic material.

[0048] An embodiment of the method of the present invention may further include generating a new control program based on the data obtained from the previous step. In developing the new control program, the results of the neural network calculations may be used to generate a control program that responds to changes in the measured parameters according to certain desired properties of the particular hygroscopic material being modified, thereby controlling the modification process to achieve optimal properties for the hygroscopic material's intended use.

[0049] Embodiments of the inventive method may further include copying a best existing control program. Copying an existing optimal control program speeds up calculations in control unit 20. This is, of course, because a better chosen starting point reduces the amount of different steps and calculations required to reach an optimal result with given intermediate parameters.

[0050] One embodiment of the inventive method may further include creating new control programs through mutation, which involves replacing random parts of the program with other random parts. Thus, in this embodiment, an iterative process is used to test different combinations of permutations to find the best possible solution to the problem case.

[0051] In one embodiment of the method of the present invention, the method may further include generating a new control program by crossing. If the desired properties of the modified moisture-absorbing material fall between those of two or more existing control programs, an appropriate control program may be achieved by combining features of those existing control programs. Therefore, applying crossing in such cases can rapidly derive an appropriate control program with fewer calculations than starting from a single existing control program that is far from the final solution, rather than combining two or more existing programs that contain all or at least most of the desired functionality.

[0052] An embodiment of the method of the present invention may further include selecting the best control program that appears in any given population and using that control program to control the reformulation process. These method steps are performed in the application of genetic programming to find the optimal solution for each specific case. As can be seen, there are multiple achievable property combinations for any given hygroscopic material. Furthermore, there are several different types of hygroscopic materials, and typically many applications for such materials. Therefore, the number of optimal control parameter combinations for all these different goals is enormous. Therefore, it is necessary to evaluate the results achieved by different control programs, rank them by suitability, and select the one from the evaluated options that provides the best results for that particular product, using the selected hygroscopic material as the starting material.

[0053] In one embodiment of the method of the present invention, the amount of moisture in the hygroscopic material is determined, which may be determined using a microwave resonator or by measuring the initial weight of the hygroscopic material and its weight during the modification process, for example as described in Applicant's earlier published application WO 2022 / 175585 A1.

[0054] In one embodiment of the method of the present invention, the amount of moisture in the hygroscopic material is taken into account when calculating at least one intermediate control parameter using a neural network. The amount of moisture in the hygroscopic material affects the modification process at least through the moisture gradient measured during the modification process. Therefore, if the amount of moisture in the hygroscopic material is known, it becomes easier to predict the change in the moisture gradient and the amount of microcracks that occur in the hygroscopic material, which are key parameters that represent the state of the hygroscopic material to be optimized during the modification process in this method.

[0055] In other embodiments of the method of the present invention, quality parameters other than moisture content, moisture gradient, and amount of microcracks may also be present. For example, if compression is applied in the modification process, the density of the pieces can be determined by measuring the volume and weight of the pieces being processed. Controlling the modification process using these additional control parameters can further improve the process and make it more suitable for specific uses and applications.

[0056] The method and apparatus for controlling a modification process of the present invention are not limited to the above-described embodiments and may be modified within the scope of the claims. In further embodiments, one or more of the above-described embodiments may be appropriately combined to impart desired properties to the modified hygroscopic material.

Claims

1. A method for controlling the modification process of a hygroscopic material (15), comprising the steps of: measuring at least one process variable from the reformulation process at least during reformulation; measuring at least one process variable from the hygroscopic material at least during reformulation; calculating at least one intermediate control parameter with a neural network by using at least the measured process variables as input parameters for the neural network; controlling the reformulation process using a genetic algorithm and genetic programming based on the at least one intermediate control parameter determined by the neural network; A method comprising:

2. The modification process is a thermomechanical modification process, and / or the at least one measured process variable from the hygroscopic material (15) is moisture gradient and / or microcrack development. The method of claim 1.

3. determining an expected value of the measured process variable and training the neural network and controlling the reformulation process so that the measured value of the process variable is as close as possible to the expected value of the process variable.

3. The method according to claim 1 or 2.

4. determining the initial state of the hygroscopic material (15) to be modified. The method according to any one of claims 1 to 3.

5. determining temporary control values ​​and parameters The method according to any one of claims 1 to 4.

6. Drying a hygroscopic material (15) by using a function of the problem, variables, and parameters. The method according to any one of claims 1 to 5.

7. Includes evaluation of dry, hygroscopic materials (15) The method of claim 6.

8. This involves creating a new control program based on data obtained from the previous step. The method according to any one of claims 1 to 7.

9. This involves copying the best existing control program. The method according to any one of claims 1 to 8.

10. This involves creating new regulatory programs through mutation and / or crossing. The method according to any one of claims 1 to 9.

11. selecting the best control program that appears in any population and using that control program in controlling the reforming process. The method according to any one of claims 1 to 10.

12. determining the amount of moisture in said hygroscopic material (15). The method according to any one of claims 1 to 11.

13. The amount of moisture in the hygroscopic material (15) is taken into account when calculating at least one intermediate control parameter by a neural network. The method of claim 12.

14. The amount of moisture in the hygroscopic material (15) is measured by a microwave resonator.

14. The method according to claim 12 or 13.

15. 15. An apparatus (10) for controlling a modification process of a hygroscopic material (15), comprising a control unit (20) configured to control the modification process of said hygroscopic material (15) by the method of any one of claims 1 to 14.

Images