Intelligent board painting method and system based on Internet of Things

Abstract

The invention relates to the technical field of data processing, in particular to an intelligent board painting method and system based on the Internet of Things, and the method comprises the steps: collecting surface parameters of all path points on a preset painting path to determine first parameter values of control parameters of all the path points; if the variable quantity of the control parameter between the adjacent path points is greater than a first threshold value, taking each path point in the neighborhood range of the path points as a candidate path point, and defining the control parameter of each candidate path point as a particle; and second parameter values of all the control parameters of the candidate path points are obtained through a particle swarm optimization algorithm, and control over wood board intelligent painting is achieved according to the second parameter values of all the control parameters of the candidate path points and the first parameter values of all the control parameters of the non-candidate path points. By means of the technical scheme, the global optimal control scheme can be rapidly and accurately obtained, and accurate control over intelligent painting is achieved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a smart painting method and system for wood panels based on the Internet of Things. Background Technology

[0002] In the furniture production process, painting, glazing, or gluing operations are required on the surface of the wood boards used to produce furniture. Because the surface of the wood boards has texture information, the surface characteristics of different locations are different, and it is difficult to control the painting quality of different locations on the wood board surface by manual painting. Therefore, it is necessary to use painting equipment to achieve intelligent painting of the wood board surface according to the surface characteristics of each location.

[0003] Currently, patent application CN120595604A discloses a trajectory adaptive control method for furniture spraying processes. The method includes the following steps: establishing a parametric furniture geometric model, which is constructed by reading design data; obtaining the spraying requirements of the target furniture, dividing the spraying area into units based on the spraying requirements and the parametric furniture geometric model, and establishing the spraying area unit division results; extracting geometric feature parameters from the spraying area unit division results and establishing a geometric partition label set; extracting the main direction of the spraying area unit division results based on the geometric partition label set, and establishing a basic spraying trajectory based on the main direction extraction results; optimizing the spraying path after layering the basic spraying trajectory based on the spraying requirements and the geometric partition label set, and establishing a following spraying trajectory, wherein the following spraying trajectory is mapped with spraying parameter labels; when spraying is performed based on the following spraying trajectory, real-time monitoring feedback data of the spray gun is obtained, and adaptive following optimization is performed based on the monitoring feedback data and the following spraying trajectory to complete trajectory adaptive control.

[0004] The above method achieves intelligent spraying of furniture surface by following the spraying trajectory mapped with spraying parameter labels during the spraying process. However, while the above method improves the spraying accuracy of the furniture surface by adaptively optimizing the spraying trajectory, it ignores the changes in control parameters between adjacent positions on the spraying trajectory. If the change in control parameters between adjacent positions is large, it will cause trajectory oscillation, which in turn leads to a decrease in control accuracy. Summary of the Invention

[0005] To address the technical problem of decreased control precision, this application provides an IoT-based intelligent wood coating method and system that can quickly and accurately determine the optimal global control scheme, thereby achieving precise control of intelligent coating.

[0006] In a first aspect, this application provides an intelligent wood painting method based on the Internet of Things (IoT). The painting method includes: collecting surface parameters of each path point on a preset painting path, and determining a first parameter value for each control parameter of each path point based on the surface parameters, wherein the first parameter value maximizes the painting quality of each path point, and the painting quality is obtained by a preset mapping relationship; responding to a change in the first parameter value of any control parameter between any path point and the previous path point exceeding a first threshold, selecting each path point within the neighborhood of the path point as a candidate path point, and defining the control parameter of each candidate path point as a particle; iteratively updating the control parameters of each candidate path point using a particle swarm optimization algorithm until the change in the fitness function is less than a second threshold for multiple consecutive iterations, and using the optimal position of the swarm as the second parameter value of each control parameter of each candidate path point, wherein the fitness function is negatively correlated with the loss of painting quality of each candidate path point and the number of path points where the change in any control parameter between the path point and the previous path point exceeds the first threshold; and realizing intelligent wood painting control based on the second parameter value of each control parameter at the candidate path point and the first parameter value of each control parameter at non-candidate path points.

[0007] By introducing a particle swarm optimization algorithm, the neighborhood range is adaptively selected and the parameters of the particles are set at path points where control parameters undergo abrupt changes. This balances painting quality and path parameter smoothness while reducing the computational cost of the particle swarm optimization algorithm. Since the fitness function is negatively correlated with quality loss and the number of abrupt path points, it can effectively control the optimization convergence direction, ensuring the quality of regional painting while avoiding drastic changes in control parameters between adjacent path points, thus improving global control stability and quality consistency.

[0008] Preferably, collecting the surface parameters of each path point on the preset painting path includes: obtaining the painting area of ​​any path point in the surface image, obtaining the gray-level co-occurrence matrix of the painting area and the gradient direction of each pixel; and using the mean of the gradient direction, as well as the contrast and entropy of the gray-level co-occurrence matrix, as the surface parameters of the path point.

[0009] Preferably, determining the first parameter value of each control parameter at each path point based on surface parameters includes: inputting the initial value of each control parameter and the surface parameter of any path point into a preset mapping relationship to obtain the coating quality; adjusting the initial value of each control parameter, and using the initial value corresponding to the maximum coating quality as the first parameter value of each control parameter at the path point.

[0010] By inputting initial values ​​of control parameters and surface parameters at path points to enter a preset mapping relationship, the control parameter corresponding to the maximum coating quality is selected as the first parameter value. This can be regarded as obtaining the optimal control parameter at a local single point, forming a precise control starting point and providing a foundation for subsequent optimization.

[0011] Preferably, the input to the preset mapping relationship is the normalized surface parameters of any path point and the parameter values ​​of each control parameter, and the output is the probability value that the coating quality of the path point is qualified.

[0012] By using the probability value of a coating quality being qualified as the coating quality, the coating quality can be accurately quantified, providing a reliable quality evaluation indicator for subsequent optimization.

[0013] Preferably, the method for obtaining the neighborhood range of the path point includes: taking a control parameter whose change in the value of a first parameter in the path point is greater than a first threshold as a target parameter; calculating the neighborhood value of the target parameter based on the product of the quality sensitivity and the change in the target parameter, wherein the neighborhood value is positively correlated with the product; and taking a preset number of path points on both sides of the path point as the neighborhood range of the path point, wherein the preset number is the maximum value of the neighborhood value among the target parameters.

[0014] By using a composite index composed of quality sensitivity multiplied by parameter change as the basis for determining the neighborhood range of path points, a dynamic neighborhood adjustment mechanism oriented towards the influence of control disturbances is formed. Based on the actual impact of the mutation parameter on the coating quality, it is determined whether to expand the optimization range, thereby ensuring that the coating quality does not decrease due to drastic parameter changes while avoiding the introduction of too many path points, and providing a precise boundary for defining particles.

[0015] Preferably, the method for obtaining the quality sensitivity of the target parameter includes: inputting the first parameter value of each control parameter at any path point into a preset mapping relationship to obtain the initial quality; applying at least one random perturbation to the target parameter to obtain the perturbation quality output by the preset mapping relationship; calculating the difference between the initial quality and the perturbation quality, and using the ratio of the difference to the initial quality as the quality sensitivity of the target parameter.

[0016] The quality sensitivity is obtained by calculating the relative change rate of coating quality before and after the control parameter disturbance, thus achieving accurate quantification of the true impact of the control parameters on the coating quality.

[0017] Preferably, the target parameter neighborhood value for: ; Based on the neighborhood value, For target parameters Quality sensitivity, For target parameters Changes after standardization This is the scaling factor. This is the floor symbol.

[0018] Preferably, the fitness function Satisfying the relation: ; This represents the number of candidate path points in the particle. Candidate path points The coating quality when each control parameter is at its first value. Candidate path points after control parameter update The quality of the coating, For adjustment coefficients, For quality requirements; if , The value of is 1, if , The value of is 2; The goal is to determine the number of path points whose control parameters, after being adjusted to the values ​​in the particles, result in a change in any control parameter between the current path point and the previous path point that exceeds a first threshold.

[0019] The number of path points where the change in any control parameter between the path point and the previous path point is greater than the first threshold. The larger the value, the worse the control effect; the fitness function... The smaller; The amount of loss in coating quality can characterize the loss of candidate path points after adjusting the control parameters of the candidate path points to the parameter values ​​in the particles. The degree of degradation in coating quality; when candidate path points When the coating quality is less than the quality requirement, the candidate path points will be... adjustment coefficient Set to 2 to increase the number of candidate waypoints. fitness function The degree of influence makes the fitness function It can balance the smoothness of control parameters and the quality of coating.

[0020] Preferably, in response to the fact that the change in the first parameter value of any control parameter between any path point and the previous path point is not greater than the first threshold, the intelligent painting control of the wood board is realized based on the first parameter value of each control parameter at each path point.

[0021] In a second aspect, this application also provides an IoT-based intelligent wood painting system, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the IoT-based intelligent wood painting method according to the first aspect of this application is implemented.

[0022] The technical solution of this application has the following beneficial technical effects: Surface parameters of each path point on the preset painting path are collected. Then, with the goal of maximizing the painting quality at each path point, the first parameter value of each control parameter at each path point is obtained. Next, path points where the control parameters change abruptly are located. Path points within the neighborhood of these path points are selected as candidate path points, and the control parameters of each candidate path point are defined as particles. In the subsequent particle swarm optimization algorithm, only the control parameters of the candidate positions need to be updated, avoiding updating the control parameters of all path points on the preset painting path, reducing the computational load of the optimization process and improving optimization efficiency. Furthermore, a fitness function is constructed. The number of path points is negatively correlated with the loss of painting quality at each candidate path point and the number of path points where the change in any control parameter between a path point and the previous path point is greater than the first threshold. By using the particle swarm optimization algorithm to obtain the second parameter values ​​of each control parameter at each candidate path point that maximizes the fitness function, and controlling the wood painting process based on the second parameter values ​​of each control parameter at the candidate path point and the first parameter values ​​of each control parameter at the non-candidate path point, the global optimal control scheme can be obtained, ensuring that the painting quality at each position point meets the quality requirements, while avoiding abrupt changes in the control parameters of adjacent path points on the preset painting path, thus achieving precise control of intelligent painting. Attached Figure Description

[0023] Figure 1 This is a flowchart of an IoT-based intelligent painting method for wood panels according to an embodiment of this application.

[0024] Figure 2 This is a schematic diagram of a preset brushing path according to an embodiment of this application.

[0025] Figure 3 This is a structural block diagram of an IoT-based intelligent wood coating system according to an embodiment of this application. Detailed Implementation

[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] According to a first aspect of this application, this application provides an intelligent painting method for wood panels based on the Internet of Things. Figure 1 This is a flowchart of an IoT-based intelligent wood painting method according to an embodiment of this application. Figure 1 As shown, the IoT-based smart wood coating method includes steps S101 to S104, which are described in detail below.

[0028] S101, collect the surface parameters of each path point on the preset coating path, and determine the first parameter value of each control parameter of each path point based on the surface parameters. The first parameter value maximizes the coating quality of each path point. The coating quality is obtained by a preset mapping relationship.

[0029] In one embodiment, the painting process requires uniformly applying paint (oil or glaze, etc.) to the surface of the wood panel; the dimensions and surface image of the wood panel can be obtained in advance, and all points on the surface of the wood panel can be traversed along a preset painting path. See also... Figure 2 This is a schematic diagram of a preset coating path according to an embodiment of this application. The path width of the preset coating path is the coating width of a single coating operation by the coating device, which is a coating roller or a brush.

[0030] The acquisition of surface parameters for each path point along a preset painting path includes: obtaining the painting area of ​​any path point in the surface image, obtaining the gray-level co-occurrence matrix of the painting area and the gradient direction of each pixel; and using the mean of the gradient direction, as well as the contrast and entropy of the gray-level co-occurrence matrix, as the surface parameters of the path point. The mean of the gradient direction reflects the texture direction within the painting area, while the contrast, energy, and entropy of the gray-level co-occurrence matrix reflect the severity and complexity of texture changes within the painting area, thereby accurately assessing the surface roughness within the painting area.

[0031] The painting area is a rectangle centered on the path point, the width of the rectangle is the path width, and the height of the rectangle is... ,in, In this embodiment of the application, the number of adjacent path points on one side of a path point is used. The value of is 2.

[0032] Understandably, for a path point on the edge of the wooden board, the rectangle corresponding to the path point extends beyond the wooden board area. For the part that extends beyond the wooden board area, the pixel value of the pixel in that part is set to 0, and the painting area corresponding to the path point can be obtained.

[0033] In one embodiment, after obtaining the surface parameters of each path point, the first parameter value of each control parameter of each path point is determined according to a preset mapping relationship. The input of the preset mapping relationship is the normalized surface parameters of any path point and the parameter values ​​of each control parameter, and the output is the probability value that the coating quality of the path point is qualified. Specifically, the construction process of the preset mapping relationship includes: inputting multiple training samples into the preset mapping relationship to obtain the output result, wherein the training samples are the normalized surface parameters and control parameters of any path point; iteratively updating the preset mapping relationship according to the cross-entropy loss between the quality label of the training samples and the output result until the number of iterations is greater than a preset number, or the cross-entropy loss is less than a preset loss, thus completing the construction of the preset mapping relationship. The preset number of iterations is 60, and the preset loss is 0.01.

[0034] The control parameters include the pressure, angle, and speed of the coating equipment.

[0035] Understandably, the quality labels of training samples can be manually labeled, including qualified and unqualified. A qualified training sample has a quality label of 1, and an unqualified sample has a quality label of 0. The preset mapping relationship can be a binary classification network or a logical classification network; the output of the preset mapping relationship is a probability value between 0 and 1. The closer the output is to 1, the better the painting quality of the corresponding path point.

[0036] In one embodiment, determining the first parameter value of each control parameter at each path point based on surface parameters includes: inputting the initial value of each control parameter and the surface parameter of any path point into a preset mapping relationship to obtain the coating quality; adjusting the initial value of each control parameter, and using the initial value corresponding to the maximum coating quality as the first parameter value of each control parameter at the path point.

[0037] The first parameter value is the value of each control parameter when the coating quality reaches its maximum value. Based on the first parameter value of each path point, the coating quality of each path point can be maximized.

[0038] Thus, the first parameter value of each control parameter at each path point on the preset coating path is determined to maximize the coating quality.

[0039] S102, in response to the change in the value of the first control parameter between any path point and the previous path point being greater than the first threshold, each path point in the neighborhood of the path point is taken as a candidate path point, and the control parameter of each candidate path point is defined as a particle.

[0040] In one embodiment, after obtaining the first parameter value of each control parameter at each path point, the change in the first parameter value of each control parameter between any path point and the previous path point is calculated. If the change in the first parameter value of any control parameter between any path point and the previous path point is not greater than a first threshold, it indicates that no abrupt change in the control parameters occurred during the painting operation along the preset painting path, enabling a smooth transition of control parameters between adjacent path points. At this point, intelligent painting control of the wood panel is achieved based on the first parameter value of each control parameter at each path point. The first threshold can be adjusted according to production requirements; in this embodiment, the first threshold is set to 3.

[0041] If the change in the first parameter value of any control parameter between any path point and the previous path point is greater than the first threshold, it indicates that a sudden change in the control parameter has occurred at that path point, which will cause uneven coating or oscillation of the coating trajectory. Therefore, it is necessary to adjust the control parameters of multiple path points near that path point.

[0042] Specifically, the method for obtaining the neighborhood range of the path point includes: taking a control parameter whose change in the value of a first parameter in the path point is greater than a first threshold as a target parameter; calculating the neighborhood value of the target parameter based on the product of the quality sensitivity and the change in the target parameter, wherein the neighborhood value is positively correlated with the product; and taking a preset number of path points on both sides of the path point as the neighborhood range of the path point, wherein the preset number is the maximum value of the neighborhood value among the target parameters.

[0043] The quality sensitivity is used to reflect the degree of influence of the target parameter on the coating quality. Specifically, the method for obtaining the quality sensitivity of the target parameter includes: inputting the first parameter value of each control parameter at any path point into a preset mapping relationship to obtain the initial quality; applying at least one random perturbation to the target parameter to obtain the perturbed quality output by the preset mapping relationship; calculating the difference between the initial quality and the perturbed quality, and using the ratio of the difference to the initial quality as the quality sensitivity of the target parameter.

[0044] Target parameters quality sensitivity Satisfying the relation: ;in, For the initial mass, For the target parameters Conduct the first The perturbation quality after a random perturbation. This represents the number of random perturbations. It should be noted that the initial quality is the maximum paint quality achievable at the corresponding path point, and its value approaches 1. Therefore, Less than or equal to Therefore, the target parameter quality sensitivity A value between 0 and 1, quality sensitivity The closer a value is to 0, the better it represents the target parameter. The less impact on coating quality, the lower the quality sensitivity. The closer a value is to 1, the better it represents the target parameter. The greater the impact on the quality of the coating.

[0045] Among them, the target parameter neighborhood value for: ; Based on the neighborhood value, For target parameters Quality sensitivity, For target parameters Changes after standardization This is the scaling factor. The floor function is used for rounding up. The base neighborhood value. The value of is 1, meaning the minimum value in the neighborhood of any target parameter is 1, and the scaling factor is... The value is 5.

[0046] Understandably, for any path point where the change in the value of the first control parameter is greater than the first threshold, if the target parameter... The quality sensitivity is 0, i.e., the target parameter Changes in this parameter will not alter the coating quality; in this case, only the target parameters at that path point need to be adjusted. The parameter value can ensure that the change in the first parameter value of the path point does not exceed the first threshold, while maintaining high painting quality. In this case, the neighborhood of the path point is the path point itself; conversely, if the target parameter... The quality sensitivity is 1, which is the target parameter. Even minor changes can alter the coating quality; in this case, simply adjusting the target parameters at that path point will not solve the problem. While the parameter values ​​can ensure that the change in the first parameter value at a path point does not exceed the first threshold, they will severely affect the painting quality at that path point. Therefore, it is necessary to adjust the control parameters of multiple adjacent path points within the neighborhood of that path point to ensure a smooth transition in the control parameters of each path point while maintaining the painting quality at each path point that meets the quality requirements. In summary, the greater the quality sensitivity and the greater the change in the target parameter at a path point, the larger the neighborhood of that path point should be.

[0047] Thus, after determining the first parameter values ​​of each control parameter at each path point on the preset painting path to maximize painting quality, the path points where the control parameters exhibit abrupt changes are located. The neighborhood range of each path point is determined based on the abrupt change amount and quality sensitivity of each control parameter, thereby identifying candidate path points. By treating the control parameters of each candidate point as particles, the subsequent particle swarm optimization algorithm only needs to update the control parameters of the candidate points, avoiding the need to update the control parameters of all path points on the preset painting path. This reduces the computational load of the optimization process and improves optimization efficiency.

[0048] S103, the control parameters of each candidate path point are iteratively updated using the particle swarm optimization algorithm until the change in the fitness function is less than the second threshold for multiple consecutive times. The optimal position of the population is then used as the second parameter value of each control parameter of each candidate path point.

[0049] In one embodiment, multiple particles are randomly generated, each particle containing the values ​​of control parameters for all candidate path points. The number of particles can be preset; in this embodiment, the number of particles is 100. After determining the number of particles, the control parameters of each candidate path point in each particle are iteratively updated using a particle swarm optimization algorithm, and the fitness function of each particle is calculated. The fitness function is negatively correlated with the loss of coating quality at each candidate path point and the number of path points where the change in any control parameter between a path point and the previous path point is greater than a first threshold.

[0050] Specifically, the fitness function Satisfying the relation: ; This represents the number of candidate path points in the particle. Candidate path points The coating quality when each control parameter is at its first value. Candidate path points after control parameter update The quality of the coating, For adjustment coefficients, For quality requirements; if , The value of is 1, if , The value of is 2; This refers to the number of path points whose control parameters, after being adjusted to the values ​​in the particles, result in a change in any control parameter between the current path point and the previous path point exceeding a first threshold. Among these, the quality requirement... The value is 0.6.

[0051] Understandably, a particle contains control parameters for each candidate path point. Adjusting the control parameters of the candidate path points to the values ​​within the particle yields the control parameters for all path points on the preset painting path. On one hand, the change in any control parameter between any path point and the previous path point is calculated. If the change exceeds a first threshold, it indicates that the path point still experiences abrupt changes in control parameters, suggesting poor control of the intelligent wood painting process. Therefore, the number of path points where the change in any control parameter between a path point and the previous path point exceeds the first threshold is limited. The larger the value, the worse the control effect; the fitness function... The smaller. On the other hand, Candidate path points The maximum paint quality that can be achieved The amount of loss in coating quality can characterize the loss of candidate path points after adjusting the control parameters of the candidate path points to the parameter values ​​in the particles. The degree of degradation in coating quality; when candidate path points When the coating quality is less than the quality requirement, the candidate path points will be... adjustment coefficient Set to 2 to increase the number of candidate waypoints. fitness function The greater the impact, the greater the loss in the painting quality of each candidate path point, and the more significant the fitness function becomes. The smaller the value, the better.

[0052] After determining the fitness function, the control parameters of each candidate path point are iteratively updated until the change in the fitness function is less than the second threshold for multiple consecutive iterations. This indicates that the control parameters of each candidate path point have stopped being updated, and the optimal position of the population is used as the second parameter value of each candidate path point's control parameter. The second threshold is set to 0.1.

[0053] In this way, the second parameter value of each control parameter is determined. The second parameter value can ensure that the loss of painting quality at each candidate path point is small, while ensuring that the control parameters between adjacent path points on the preset painting path are smoothly transitioned.

[0054] S104, based on the second parameter value of each control parameter at the candidate path point and the first parameter value of each control parameter at the non-candidate path point, realizes the intelligent painting control of the wood board.

[0055] In one embodiment, each path point on the preset painting path can be divided into candidate path points and non-candidate path points. Based on the second parameter value of each control parameter at the candidate path point and the first parameter value of each control parameter at the non-candidate path point, the wood painting process is controlled to obtain the global optimal control scheme, ensuring that the painting quality at each position point meets the quality requirements, while avoiding abrupt changes in the control parameters of adjacent path points on the preset painting path, thus achieving precise control of intelligent painting.

[0056] According to a second aspect of this application, this application also provides an Internet of Things-based intelligent wood panel painting system. Figure 3 This is a structural block diagram of an IoT-based intelligent wood coating system according to an embodiment of this application. Figure 3 As shown, the system 50 includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the IoT-based intelligent wood painting method according to the first aspect of this application. The system also includes other components well-known to those skilled in the art, such as a communication bus and a communication interface. Their configurations and functions are known in the art and will not be described further here.

[0057] It should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of this application, and these all fall within the scope of protection of this application.

Claims

1. A smart painting method for wood panels based on the Internet of Things, characterized in that, The coating method includes: collecting surface parameters of each path point on a preset coating path, and determining a first parameter value of each control parameter of each path point based on the surface parameters. The first parameter value maximizes the coating quality of each path point, and the coating quality is obtained by a preset mapping relationship. In response to the change in the value of the first control parameter between any path point and the previous path point being greater than a first threshold, each path point in the neighborhood of the path point is taken as a candidate path point, and the control parameter of each candidate path point is defined as a particle. The control parameters of each candidate path point are iteratively updated using the particle swarm optimization algorithm until the change in the fitness function is less than the second threshold for multiple consecutive times. The optimal position of the swarm is used as the second parameter value of each control parameter of each candidate path point. The fitness function is negatively correlated with the loss of the painting quality of each candidate path point and the number of path points where the change in any control parameter between the path point and the previous path point is greater than the first threshold. Intelligent painting of wood panels is achieved by using the second parameter value of each control parameter at candidate path points and the first parameter value of each control parameter at non-candidate path points.

2. The intelligent wood coating method based on the Internet of Things according to claim 1, characterized in that, The surface parameters collected at each point along the preset painting path include: Obtain the painted area of ​​any path point in the surface image, obtain the gray-level co-occurrence matrix of the painted area and the gradient direction of each pixel; use the mean of the gradient direction, the contrast and entropy of the gray-level co-occurrence matrix as the surface parameters of the path point.

3. The intelligent wood coating method based on the Internet of Things according to claim 1, characterized in that, The first parameter value for determining each control parameter of each path point based on surface parameters includes: The initial values ​​of each control parameter and the surface parameters of any path point are input into a preset mapping relationship to obtain the coating quality; the initial values ​​of each control parameter are adjusted, and the initial value corresponding to the maximum coating quality is used as the first parameter value of each control parameter at the path point.

4. The intelligent wood coating method based on the Internet of Things according to claim 1, characterized in that, The input to the preset mapping relationship is the normalized surface parameters and the parameter values ​​of each control parameter of any path point, and the output is the probability value of the path point having qualified coating quality.

5. The intelligent wood coating method based on the Internet of Things according to claim 1, characterized in that, The method for obtaining the neighborhood range of the path point includes: The control parameter whose change in the first parameter value at the path point is greater than the first threshold is taken as the target parameter; The neighborhood value of the target parameter is calculated based on the product of the quality sensitivity of the target parameter and the amount of change, and the neighborhood value is positively correlated with the product; A predetermined number of path points on both sides of the path point are taken as the neighborhood range of the path point, and the predetermined number is the maximum value of the neighborhood value among the target parameters.

6. The intelligent wood coating method based on the Internet of Things according to claim 5, characterized in that, The method for obtaining the target parameter quality sensitivity includes: Input the first parameter value of each control parameter at any path point into a preset mapping relationship to obtain the initial quality; after applying at least one random perturbation to the target parameter, obtain the perturbation quality output by the preset mapping relationship; calculate the difference between the initial quality and the perturbation quality, and use the ratio of the difference to the initial quality as the quality sensitivity of the target parameter.

7. The intelligent wood coating method based on the Internet of Things according to claim 5, characterized in that, Target parameters neighborhood value for: ; Based on the neighborhood value, For target parameters Quality sensitivity, For target parameters Changes after standardization This is the scaling factor. The rounding up symbol.

8. The intelligent wood coating method based on the Internet of Things according to claim 1, characterized in that, fitness function Satisfying the relation: ; This represents the number of candidate path points in the particle. Candidate path points The coating quality when each control parameter is at its first value. Candidate path points after control parameter update The quality of the coating, For adjustment coefficients, For quality requirements; if , The value of is 1, if , The value of is 2; The goal is to determine the number of path points whose control parameters, after being adjusted to the values ​​in the particles, result in a change in any control parameter between the current path point and the previous path point that exceeds a first threshold.

9. The intelligent wood coating method based on the Internet of Things according to claim 1, characterized in that, In response to the fact that the change in the first parameter value of any control parameter between any path point and the previous path point is not greater than the first threshold, intelligent painting control of wood boards is realized based on the first parameter values ​​of each control parameter at each path point.

10. A smart wood coating system based on the Internet of Things, characterized in that, It includes a processor and a memory, the memory storing computer program instructions, which, when executed by the processor, implement a smart wood painting method based on the Internet of Things according to any one of claims 1 to 9.

Images