Method and device for controlling modification process of hygroscopic material

Abstract

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

Description

Technical Field

[0001] The present invention relates to a method and an apparatus for controlling a modification process of a moisture-absorbing material, such as a thermo-mechanical modification process. Background Art

[0002] A moisture-absorbing material is a material that has the ability to absorb and store moisture from the surrounding air. When the relative humidity changes, the difference in vapor pressure causes the material to absorb or desorb moisture to reach equilibrium. Due to its characteristics, moisture-absorbing materials have a wide range of applications in various industrial fields. Typical moisture-absorbing materials include wood, wood-plastic composites, certain plant-based materials, and materials such as concrete.

[0003] With the increasing environmental awareness, the market is looking for more sustainable options. In response to this demand, some materials refined without using toxic chemicals enable enterprises to build more responsibly and dispose of them conveniently when needed. As a natural and long-existing moisture-absorbing material, wood provides an attractive and cost-effective solution for these environmental sustainability requirements. However, currently in developing countries, approximately 80% of the harvested wood is wasted or burned. In particular, less than 5% of the wood in tropical forests is sustainably managed.

[0004] Current wood processing methods, such as traditional drying and compression processes, have not provided investors with sufficiently economically attractive solutions to increase the proportion of harvested wood used in the production of sustainable wooden products such as buildings and furniture. In particular, some low-grade tropical wood species are considered unsuitable for such applications because they cannot meet the quality levels required for mechanical properties, dimensional accuracy, and surface quality. In addition, many existing drying and compression processes have too long processing cycles, even for some high-grade wood species.

[0005] Thermo-mechanical modification processes, such as the thermo-mechanical wood modification (TMTM) process, allow for a greater degree of utilization of harvested wood. Through the thermo-mechanical modification process, the properties of the moisture-absorbing material, such as compressive strength, stiffness, density, hardness, and dimensional stability, can be modified to make it suitable for specific applications. A known thermo-mechanical modification process has been disclosed in the published patent WO2022 / 175585 A1.

[0006] However, due to the numerous factors affecting these processes, effectively and optimally controlling the modification process (such as the above-mentioned thermo-mechanical modification process) remains a problem. Summary of the Invention

[0007] The object of the present invention is to provide a new method for controlling the modification process of a hygroscopic material, significantly shortening the process cycle and improving the quality level of the modified material. Specifically, the present invention aims to form a method for controlling the modification process of a hygroscopic material, such that quality parameters such as the strength, surface hardness, and dimensional stability of the modified hygroscopic material are significantly improved, while significantly reducing the process cycle.

[0008] The above object is achieved by the following solution: According to the method of the present invention, the modification process of the hygroscopic material is controlled by using genetic algorithms and genetic programming, based on at least one intermediate parameter calculated by a neural network, the input of the intermediate parameter being at least one measured modification process variable and at least one measured hygroscopic material process variable; and the neural network, after learning, can take into account the non-linear dependencies between different input parameters, specifically, these dependencies depend on the material and its initial state as well as the nature of the final product to be achieved.

[0009] The measured process variables of the modification process include but are not limited to: air temperature, air velocity, relative air humidity, air pressure, compression force, and compression velocity, etc.

[0010] The measured process variables of the hygroscopic material include but are not limited to: humidity, temperature, humidity gradient, occurrence of microcracks, size of the block to be modified, compression (thickness), etc.

[0011] In addition, other suitable additional input parameters can also be used for the neural network.

[0012] Preferably, the modification process controlled by the present invention is a thermo-mechanical modification process of the hygroscopic material. In addition, in the method of the present invention, the humidity gradient of the hygroscopic material is preferably measured by electrical impedance spectroscopy (EIS), and the microcracks in the hygroscopic material are preferably monitored by acoustic emission (AE).

[0013] More precisely, the method of the present invention is characterized by what is described in independent claim 1, and the device of the present invention is characterized by what is described in independent claim 15. The dependent claims 2 - 14 introduce some advantageous embodiments of the method of the present invention.

[0014] The advantages of the method and device of the present invention are that, through the method and device of the present invention, the factors affected during the modification process of the hygroscopic material (such as the humidity gradient value and the number of microcracks) can be controlled within an acceptable level, so that even if the process delivery cycle is significantly shortened, an appropriate overall quality of the modified hygroscopic material can still be obtained. In particular, the ability of the hygroscopic material to absorb moisture from the air and the strength of the material are enhanced. Therefore, thanks to the new method and device of the present invention, hygroscopic materials (such as low-grade tropical wood) can be used for more sustainable applications, such as furniture or building products, instead of being burned or discarded as is currently the case.

[0015] In addition, genetic algorithms and genetic programming offer more advantages in this application by solving different problems in the control modification process that many existing control methods cannot solve. These problems include:

[0016] During the drying stage, the monitoring technology is only useful when the interaction and total effect of the known observed process values on the property changes of the material to be dried are known;

[0017] Hygroscopic materials, such as wood, are usually heterogeneous natural materials with thousands of different species, each with different physical properties. Therefore, during the processing of these materials, each species has its specific characteristics that need to be considered during processing to achieve a final product with the desired properties;

[0018] When the goal is to modify certain physical properties (such as surface hardness, strength, humidity, torsional stiffness, aging resistance, etc.), the task forms a theoretically optimized problem. Genetic algorithm (GA) and genetic programming (GP) methods, along with appropriate online measurement techniques, provide more effective tools than traditional methods (where certain single pre-determined mathematical model control programs are used). BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Next, the present invention will be described in more detail with reference to the accompanying drawings, where:

[0020] Figure 1 Schematically shows a cross-section of the modification chamber of the modification device applied in the device according to the present invention;

[0021] Figure 2 Shows an embodiment of the artificial neural network used in the present invention;

[0022] Figure 3 Shows a flowchart of an embodiment of the method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0023] According to the method of the present invention, the modification process of the moisture-absorbing material is controlled, such as the thermo-mechanical modification process. In such a thermo-mechanical process, a moisture-absorbing material such as wood is processed in a modification device, which usually has a modification chamber into which the moisture-absorbing material is transferred and modified by applying a process comprising a plurality of process stages, such as humidification, drying, heating, and compression. Accordingly, the atmospheric conditions in the modification chamber, such as air temperature, air relative humidity, and, for example, the compression force, are adjusted according to the applied control program. Thus, the program used depends on the type and quality of the moisture-absorbing material before modification, the size of the block to be modified, and the desired appearance and mechanical properties of the moisture-absorbing material to be achieved. After the desired properties of the moisture-absorbing material are reached, it is transferred out of the modification chamber for further processing, such as packaging, storage, and / or transportation to other in-house manufacturing departments or the end-users of the material.

[0024] Figure 1 An example of such a modification device 10 is shown. It includes a modification chamber 11 in which the moisture-absorbing material 15 can be thermo-mechanically modified. In this example, the moisture-absorbing material to be modified is present in batches, i.e., it comprises a plurality of material blocks placed adjacent to and / or stacked on top of each other, such that there may or may not be one or more intermediate blocks between these blocks. These intermediate blocks 16 are preferably honeycomb-like plates, sheets, or tabs having cavities or channels through which air can flow in the modification chamber.

[0025] Figure 1 The moisture-absorbing material to be modified in the shown modification device can be, for example, wood, wood-plastic composite, or certain plant-based materials. Preferably, the properties of these materials are similar to those of wood during the modification process.

[0026] In the modification chamber 11, there is a compression device 12 which comprises a first compression element 13 and a second compression element 14, between which the moisture-absorbing material 15 to be modified is placed, and by which the material can be compressed during the modification process. The first compression element 13 and the second compression element 14 are platform-like elements having flat compression surfaces on which the material to be modified can be placed during modification. In this embodiment, the compression device has only two compression elements and can thus be used to compress the material to be modified in the thickness direction of the moisture-absorbing material. However, in some other embodiments of the present method and device, the compression device can also have additional compression elements for compressing the blocks of the moisture-absorbing material in other directions, i.e., the width and / or length directions.

[0027] In Figure 1In the modification chamber 11 of the modification device 10 shown, there are also a heating device 18, a blowing device 19, and a humidifying device (not shown in the figure). The heating device 18 can be, for example, an electric heater, an oil heater, or a suitable biofuel heating device. The blowing device 19 is preferably an electric fan, and the humidifying device can include a spraying device and / or a steam device.

[0028] As Figure 1 The modification device 10 shown also includes a measuring device for measuring different properties of the hygroscopic material. In particular, when applying the method of the present invention, the humidity gradient of the hygroscopic material to be modified is usually measured by electrical impedance spectroscopy (EIS), while the number of microcracks in the hygroscopic material is usually measured by acoustic emission (AE). Other types of sensors and measuring means can also be used to determine the humidity gradient and the number of microcracks. These measurements can be carried out during the thermomechanical modification process, but can also be carried out before and / or after the thermomechanical modification. Online measurements can be carried out by providing the modification device with appropriate sensors and connecting them to the control unit 20 of the modification device 10, for example, through a WLAN network or other suitable wireless data communication technologies, or a wired connection can also be used.

[0029] In addition, appropriate measurement sensor technologies can also be used to measure some other process parameters and physical quantities, such as the temperature and humidity content of the air and / or the hygroscopic material to be modified, and the weight of the hygroscopic material to be modified. The modification device 10 may also be equipped with online measurement devices / sensors and test devices / instruments. The determined property values can also be verified through separate laboratory measurements. Laboratory measurements are most suitable for determining properties such as the humidity content, hardness, and strength of the hygroscopic material. The control unit 20 is connected to the online measurement device and / or receives additional input data for controlling different devices of the modification device, such as the compression device 12, the heating device 18, the blowing device 19, and the humidifying device, and executes the control according to the control program. Therefore, the control unit 20 includes a computing device for executing the control program and a memory for storing the control program and processing data. Therefore, the computing device is capable of controlling these devices in order to implement the method of the present invention, that is, it has programs for running neural networks, genetic algorithms, and genetic programming for data processing according to the method of the present invention. The control unit may also have the necessary electronic circuits and components for controlling actuators, such as the actuators of the fan and the compression device, and the measuring devices of the modification device, so that the modification process carried out by the modification device can be completely automatically controlled by the control unit 20.

[0030] 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 in this embodiment at least includes the following steps:

[0031] - Measuring the temperature and humidity in the modification chamber,

[0032] - At least during the modification process, measure the humidity gradient of the moisture-absorbing material by means of impedance spectroscopy (EIS);

[0033] - At least during the modification process, monitor the occurrence amount of microcracks in the moisture-absorbing material by acoustic emission (AE),

[0034] - Calculate at least one intermediate control parameter by using at least the measured variables as input parameters of a neural network,

[0035] - Based on the at least one intermediate control parameter determined by the neural network, control the thermomechanical modification process using genetic algorithms and genetic programming.

[0036] Impedance spectroscopy (EIS) refers to measuring the impedance (alternating current resistance) of an object at multiple frequencies. The result is a frequency spectrum that provides information about the structure and properties of the object. Impedance spectroscopy has been applied in a variety of applications, one of the most important being the study of biological substances, for example in medicine. In the present invention, EIS technology is used to monitor in real time the structure, humidity distribution and gradient of the moisture-absorbing material.

[0037] The impedance spectroscopy measurement device may include appropriate electrodes that are placed on the surface of the material to be measured, and impedance measurements are carried out through these electrodes during the modification process. The measurement device may be wirelessly connected to the control unit 20 of the modification device, and the electrodes are located in the modification chamber.

[0038] Acoustic emission (AE) can be used to monitor the microcracks that may occur during the drying process. If a moisture-absorbing material such as wood is exposed to cracks during drying, the first observable phenomenon is microcracks. When microcracks occur, the material emits sound at ultrasonic frequencies. By monitoring the acoustic emission and correspondingly adjusting the modification process, macroscopic cracks that are harmful to the quality of the final product can be eliminated.

[0039] The acoustic emission measurement device may include, for example, piezoelectric sensors for measuring the ultrasonic sound waves formed by the microcracks of the moisture-absorbing material to be monitored. The sensors can be wirelessly connected to the control unit 20 to provide measurement data accordingly.

[0040] Genetic algorithms are heuristic optimization methods that mimic the mechanisms of natural evolution. They are suitable for tasks with a very large solution space (for example, large-scale combinatorial problems), and even an approximate optimal solution is sufficient to solve the problem. With the rapid growth of computer computing power, the application possibilities of genetic algorithms have been greatly expanded in the past decade.

[0041] Genetic algorithms (GA) and genetic programming (GP), as methods for regulating and controlling the thermomechanical modification process, can improve the quality of the final product. This is because traditional mathematical analysis cannot or does not provide an analytical solution. GA and GP, as adaptive methods, can adapt to changes in the modification process. Through real-time optimization or machine learning, the process becomes better without external intervention. Adaptation is continuous. Currently, the internal relationships between relevant variables are not clear (or there is reason to suspect that the current understanding is wrong). Finding the size and shape of the final solution to the problem is an important part of problem-solving. An approximate solution is acceptable (or the only solution that can be achieved). In such tasks, a large amount of computer-readable data is encountered, which needs to be reviewed, classified, and aggregated. A small improvement in performance is usually measurable (or easily measurable) and is significant for the overall result.

[0042] In addition, the added value of genetic algorithms and genetic programming in the measurement and control techniques of the thermomechanical modification process is that when drying hygroscopic materials in the thermomechanical modification process, it is not the humidity and temperature of the air that are preferentially controlled, but rather to ensure that the hygroscopic material to be dried is not damaged by drying, and the final humidity of the dried hygroscopic material is sufficiently uniform, and the material has a low humidity gradient. The treated hygroscopic material has a complete structure, and the humidity distribution, gradient, and required humidity content reach the lowest possible values.

[0043] To achieve the above and other objectives of the method, the aforementioned electrochemical impedance spectroscopy (EIS) and acoustic emission (AE) monitoring methods can detect in real time the time and conditions at which wood damage begins. Once the time and conditions are known, measures can be taken to avoid such situations. This makes it possible to create conditions to avoid cracking or damage to hygroscopic materials.

[0044] An artificial neural network (hereinafter referred to as "neural network"), as implemented as Figure 2 shown, is a non-linear statistical data modeling or decision-making tool. It can be used to establish complex relationships between inputs and outputs, or to find patterns in data. It involves a network composed of simple processing units (artificial neurons), and these processing units can exhibit complex global behavior through mutual connection and different element parameters. In the present invention, the neural network is applied to determine the relationship between the properties of hygroscopic materials and the measured humidity gradient, microcracks, and other measured quantities.

[0045] In Figure 2 the embodiment of, the measurable input data of the input layer A can include, but are not limited to: the initial moisture content of the material; the process temperature; the relative humidity; the steam consumption; the humidity gradient; the weight and / or density of the material; microcracks; the type of wood; the compression pressure, rate, and / or speed; the air velocity and / or direction; and / or the change in the input value.

[0046] In a neural network, complex relationships are defined by hidden layers B - D (i.e., the deep learning stage), connecting the input data to the output data of output layer E. The output data can include, but is not limited to: color gamut; hardness; strength; modulus of elasticity (MOE); modulus of rupture (MOR); compression; density; dimensional stability; fire resistance; corrosion resistance; fungicidal resistance; termite resistance; temperature / humidity calibration and compensation of process sensors.

[0047] Intermediate control parameters can be the humidity gradient and the amount of microcracks calculated using a neural network, and these calculations are based on the initial state of the hygroscopic material to be modified. They may also be some other values and / or combinations of values that are calculated using a neural network and then used as input values for a genetic algorithm to control the modification device.

[0048] A flowchart of an embodiment of the method of the present invention is as Figure 3 shown.

[0049] In Figure 3 the embodiment, the modification process of the hygroscopic material is controlled based on available data, as shown in block 101. At the start of the modification process, this data may be based on or include measurable variables related to the material being processed, such as moisture content, weight, density, wood species, etc., and variables related to the process itself, such as process temperature, air flow rate, compression pressure, etc.

[0050] During the modification process, process - related data is collected through appropriate sensors and measurement devices, as shown in block 102.

[0051] The acquired process data is used as input data for the neural network, as shown in block 103. Then, the output data from the neural network is used as input data for the genetic algorithm, as shown in block 104.

[0052] Improved modification process control data is obtained through the genetic algorithm, as shown in block 105, and these data are subsequently used to control the actual modification process, as shown in block 101.

[0053] As Figure 3 shown, the process can be implemented multiple times until the desired quality and properties of the modified hygroscopic material are obtained through the modification process.

[0054] An embodiment of the method of the present invention may also include determining expected values of the humidity gradient and the amount of microcracks, and training the neural network to control the modification process such that the measured values of the humidity gradient and the amount of microcracks are as close as possible to the expected values (i.e., the values calculated using the neural network).

[0055] An embodiment of the method of the present invention may further include determining the initial state of the moisture-absorbing material to be modified. Determining the initial state of the moisture-absorbing material refers to determining its characteristics, such as the initial moisture content, the initial humidity gradient, and the number of microcracks before modification. These can be determined by appropriate investigations and / or measurements. To determine the initial state of the moisture-absorbing material, laboratory measurements can be carried out before modification or in-line measurements can be used in the modification chamber for pre-modification measurements. Understanding the initial state of the moisture-absorbing material to be modified will improve and accelerate the process, while preventing the situation where the control program fails to achieve the best results due to the deviation between the expected initial state and the actual initial state.

[0056] An embodiment of the method of the present invention may further include determining temporary control values and parameters. Depending on the specific situation, some moisture-absorbing materials may contain parameter values that may deviate from the conventional values. Some moisture-absorbing materials may also have characteristics or behaviors, so their modification requires the use of some additional parameters to control the process. In this case, these parameters or values can be determined before or during the process, for example, when a single trigger value of ordinary parameters or a combination of multiple ordinary parameter values is obtained.

[0057] An embodiment of the method of the present invention may further include drying the moisture-absorbing material by using problem functions, variables, and parameters. The problem function can be a function used for mathematical calculation to measure or monitor the relationship between parameters, such as the humidity gradient and the number of microcracks, taking into account the atmospheric conditions in the modification chamber or the applied pressure of the compression device.

[0058] An embodiment of the method of the present invention may further include evaluating the dried moisture-absorbing material. The evaluation of the dried moisture-absorbing material may include, for example: detecting the surface quality and dimensional accuracy (such as straightness) of the moisture-absorbing material fragments, measuring the strength and / or hardness, or determining the moisture content of the dried moisture-absorbing material.

[0059] An embodiment of the method of the present invention may further include creating a new control program based on the data obtained in the previous stage. During the development of the new control program, the calculation results of the neural network may be utilized so that the control program can respond in real time to the changes in the measured parameters according to the changes in the characteristics of the specific moisture-absorbing material desired. In this way, the modification process can be controlled to provide the moisture-absorbing material with the characteristics most suitable for its intended use.

[0060] An embodiment of the method of the present invention may further include replicating the existing best control program. Replicating the existing best control program can accelerate the calculation in the control unit 10 because if the starting point is selected more appropriately, the number of different stages and the amount of calculation required to achieve the best results will be reduced by using the predetermined intermediate parameters.

[0061] An embodiment of the method of the present invention may further include creating a new control program by mutation. Mutation involves replacing some random parts of the program with other random parts of the program. Thus, in such an embodiment, iterative testing of different replacement combinations is employed to find the best possible solution.

[0062] An embodiment of the method of the present invention may further include creating a new control program by crossover. When the characteristics of the target modified hygroscopic material are between two or more existing control programs, a suitable control program can be achieved by combining the characteristics of these existing control programs. Thus, applying crossover in such a case enables obtaining a suitable control program more quickly, and the computational amount is less than using a single existing control program (if it is far from the final solution).

[0063] An embodiment of the method of the present invention may further include selecting the best control program that appears in any population and using this control program to control the modification process. This step is used to find the best solution for each specific case in the application of genetic programming. As can be understood, for certain specific hygroscopic materials, there may be several different combinations of characteristics. In addition, there are various different types of hygroscopic materials and their common applications. Therefore, for all these different targets, there may be a large number of different combinations of control parameters that are most suitable. Therefore, it is necessary to evaluate the results obtained through different control programs, sort them according to their applicability, and then select which one of the evaluated alternatives gives the best result, using the selected hygroscopic material as the raw material.

[0064] In an embodiment of the method of the present invention, the moisture content in the hygroscopic material is determined. The moisture content in the hygroscopic material can be measured by a microwave resonator or by measuring the initial weight of the hygroscopic material and the weight that changes during the modification process, as described in the applicant's earlier application publication WO2022 / 175585A1.

[0065] In an embodiment of the method of the present invention, the moisture content in the hygroscopic material is considered when calculating at least one intermediate control parameter by a neural network. The moisture content in the hygroscopic material affects the modification process through a humidity gradient, and the humidity gradient is determined at least during the modification process. Thus, if the moisture content in the hygroscopic material is known, it is easier to predict the change in the humidity gradient and the occurrence of microcracks in the hygroscopic material, which are the main parameters describing the state of the hygroscopic material and need to be optimized during the modification process of this method.

[0066] In other embodiments of the method of the present invention, there may also be other quality parameters in addition to the moisture content, humidity gradient, and number of microcracks. For example, when pressure is applied during the modification process, the density of the processed fragment can be determined by measuring the volume and weight of the fragment. By controlling the modification process with these additional control parameters, the process can be further improved, especially to adapt it to more specific uses and applications.

[0067] The method and device for controlling the modification process of the present invention are not limited to the above embodiments and can be varied within the scope of the claims. In further embodiments, one or more of the above embodiments can be combined to obtain a suitable combination of characteristics required for modifying the hygroscopic material.

Claims

1. A method for controlling the modification process of a moisture-absorbing material (15), comprising the following method steps: - Measuring at least one process variable from the modification process, at least during the modification process; - Measuring at least one process variable from the moisture-absorbing material (15), at least during the modification process; - Calculating at least one intermediate control parameter by a neural network, using at least one of the measured process variables as input parameters of the neural network; - Controlling the modification process using a genetic algorithm and genetic programming based on the at least one intermediate control parameter determined by the neural network.

2. The method according to claim 1, characterized in that The modification process is a thermo-mechanical modification process, and / or at least one of the measured process variables from the moisture-absorbing material (15) is a humidity gradient and / or the occurrence of microcracks.

3. The method according to any one of the preceding claims, characterized in that, The method includes determining an expected value of the measured process variable, training the neural network, and controlling the modification process so that the measured value of the process variable is as close as possible to the expected value of the process variable.

4. The method according to any one of the preceding claims, characterized in that, The method includes determining the initial state of the moisture-absorbing material (15) to be modified.

5. The method according to any one of the preceding claims, characterized in that, The method includes determining temporary control values and parameters.

6. The method according to any one of the preceding claims, characterized in that, The method includes drying the moisture-absorbing material (15) by using problem functions, variables, and parameters.

7. The method according to claim 6, wherein The method includes evaluating the dried moisture-absorbing material (15).

8. The method according to any one of the preceding claims, characterized in that, The method includes creating a new control program based on data obtained from the foregoing stages.

9. The method according to any one of the preceding claims, characterized in that, The method includes copying the existing best control program.

10. The method according to any one of the preceding claims, characterized in that, The method includes creating a new control program by mutation and / or crossover.

11. The method according to any one of the preceding claims, characterized in that, The method includes selecting the best control program that appears in any population and using this control program to control the modification process.

12. The method according to any one of the preceding claims, characterized in that Determining the water content in the moisture-absorbing material (15).

13. The method according to claim 12, characterized in that, Considering the water content in the moisture-absorbing material (15) when calculating the at least one intermediate control parameter by the neural network.

14. The method according to claim 12 or 13, characterized in that, Measuring the water content in the moisture-absorbing material (15) by a microwave resonator.

15. A device (10) for controlling the modification process of a moisture-absorbing material, characterized in that, The device includes a control unit (20), which is configured to control the modification process of the moisture-absorbing material (15) by the method according to any one of claims 1-14.

Images