Coating line adaptive spray optimization system, method, computing device, and storage medium

Abstract

The application discloses a kind of coating line adaptive spraying optimization system and method, wherein, perception module is used to collect the geometric shape information and surface characteristic information of target workpiece and environmental information;Data processing module is used to extract workpiece characteristics based on geometric shape information and surface characteristic information, and generates spraying path and spraying parameter in combination with environmental information;Control module is used to generate spraying control instruction based on optimized spraying path and spraying parameter, to control execution unit to carry out spraying operation;Feedback module is used to monitor spraying quality data obtained by spraying operation in real time, and transmit it as feedback information to data processing module.By the application, the collective characteristics of workpiece are extracted by collecting the geometric shape and surface characteristic information of workpiece, and spraying path and parameter are generated in combination with environmental information, to adapt to diversified workpieces, greatly shorten the spraying time and optimize the paint distribution, and further optimize the spraying process through feedback information.

Description

Technical Field

[0001] This invention relates to the field of industrial coating automation, and in particular to an adaptive spraying optimization system, method, computing device and storage medium for coating lines. Background Technology

[0002] With technological advancements, an increasing number of industrial sectors require coatings for related equipment. These coatings enhance corrosion resistance, improve durability, and enhance aesthetics. Transformer radiators, as essential equipment in large power plants and energy companies, are widely used in the global energy industry. The inner walls of transformer radiators require painting to increase their corrosion resistance, extend their lifespan, and minimize replacement frequency and costs.

[0003] Currently, the method for spraying the inner wall of transformer radiators typically involves using a spraying system to perform the spraying operation based on a preset path and fixed spraying parameters. However, this method has extremely poor ability to plan the spraying path for complex or non-standard workpieces, only able to complete the operation within the preset path and unable to reasonably determine the spraying path. Furthermore, it cannot dynamically adjust the spraying parameters according to environmental changes (such as temperature and humidity), resulting in inconsistent spraying effects under different environments, affecting the compatibility of the workpiece during use, and causing significant differences in performance and durability between different batches of workpieces. In addition, due to the unreasonable spraying path and the inability to adjust the spraying parameters, the paint utilization rate is low, the waste rate is extremely high, and the overall operating cost is affected. Adjusting the parameters requires frequent manual intervention, increasing the complexity of the operation and raising labor costs. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides an adaptive spraying optimization system for coating lines, a corresponding adaptive spraying optimization method for coating lines, a computing device, and a computer storage medium.

[0005] According to one aspect of the present invention, an adaptive spraying optimization system for a coating line is provided, comprising: a sensing module, a data processing module, a control module, and a feedback module; wherein,

[0006] The sensing module is used to collect geometric shape information and surface characteristic information of the target workpiece, as well as environmental information;

[0007] The data processing module is used to extract workpiece features based on the geometric shape and surface characteristics of the target workpiece, and generate spraying paths and spraying parameters in combination with environmental information.

[0008] The control module is used to generate spray control commands based on the optimized spray path and spray parameters, so as to control the execution unit to perform spray operation.

[0009] The feedback module is used to monitor the spraying operation in real time to obtain spraying quality data, and transmit the spraying quality data as feedback information to the data processing module.

[0010] In the above scheme, the sensing module is further used for:

[0011] Using a 3D laser scanner and an RGB-D camera, the target workpiece is scanned and photographed to obtain its geometric shape information and surface characteristic information; wherein, the geometric shape information includes at least point cloud data; and the surface characteristic information includes at least the workpiece roughness and the workpiece material type.

[0012] Environmental information is acquired based on environmental sensors; wherein, the environmental information includes at least temperature and humidity.

[0013] In the above scheme, the data processing module is further used for:

[0014] Edge computing units are used to process geometric information based on the PointNet point cloud processing model to extract the workpiece features of the target workpiece;

[0015] The spraying parameters are predicted using the Transformer model based on workpiece features and environmental information.

[0016] Reinforcement learning techniques are used to determine the spraying path.

[0017] In the above scheme, the control module is further used for:

[0018] Generate corresponding spraying control commands based on spraying parameters and spraying path;

[0019] Send the spraying control command to the control execution unit;

[0020] The drive control execution unit performs the spraying operation according to the spraying control command.

[0021] In the above scheme, the feedback module is further used for:

[0022] The spraying monitoring unit is used to monitor the spraying effect of the target workpiece and obtain spraying quality data.

[0023] The spraying quality data is transmitted as feedback information to the data processing module.

[0024] In the above scheme, the step of using a spraying monitoring unit to monitor the spraying effect of the target workpiece and obtain spraying quality data further includes:

[0025] The surface quality data of the target workpiece is determined using a visual inspection camera; the surface quality data includes at least: surface defect feature data and coating physical property data;

[0026] The film thickness data of the target workpiece is detected by using a film thickness sensor and a vision inspection camera to determine the uniformity of the film thickness after spraying.

[0027] In the above scheme, the step of using reinforcement learning technology to determine the spraying path further includes:

[0028] Using the DQN algorithm, based on the input workpiece features and environmental information, at least one initial path is generated according to the ε-greedy strategy, and a reward function and basic reward are set.

[0029] The GAN-based generator converts the obtained initial path into the corresponding spraying simulation image;

[0030] The discriminator based on GAN obtains feedback information as a real quality inspection image, which is input into the discriminator along with the spraying simulation image, and outputs a path quality score.

[0031] Based on the reward function, DQN is triggered to replan the local path based on the path quality score, generating an optimized spraying path.

[0032] According to another aspect of the present invention, an adaptive spraying optimization method for a coating line is provided, comprising:

[0033] Collect geometric shape information, surface characteristic information, and environmental information of the target workpiece;

[0034] Workpiece features are extracted based on the geometric shape and surface characteristics of the target workpiece, and spraying paths and spraying parameters are generated by combining environmental information.

[0035] Based on the optimized spraying path and spraying parameters, spraying control commands are generated to control the execution unit to perform spraying operations.

[0036] Real-time monitoring of the spraying operation yields spraying quality data, which is then used as feedback information.

[0037] According to another aspect of the present invention, a computing device is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus;

[0038] The memory is used to store at least one executable instruction that causes the processor to perform operations as described above in an adaptive spraying optimization system for a coating line.

[0039] According to another aspect of the present invention, a computer storage medium is provided, the storage medium storing at least one executable instruction that causes a processor to perform operations as described above in an adaptive spraying optimization system for a coating line.

[0040] According to the technical solution provided by the present invention, an adaptive spraying optimization system for a coating line includes: a sensing module, a data processing module, a control module, and a feedback module; wherein, the sensing module is used to collect geometric shape information and surface characteristic information of the target workpiece, as well as environmental information; the data processing module is used to extract workpiece features based on the geometric shape information and surface characteristic information of the target workpiece, and generate a spraying path and spraying parameters in combination with environmental information; the control module is used to generate spraying control commands based on the optimized spraying path and spraying parameters to control the execution unit to perform spraying operations; the feedback module is used to monitor the spraying operation in real time to obtain spraying quality data, and transmit the spraying quality data as feedback information to the data processing module. The technical solution of this invention utilizes an adaptive spraying optimization system for coating lines. This system employs a 3D laser scanner and an RGB-D camera to acquire the geometric shape and surface characteristic information of the workpiece, improving the accuracy of workpiece scanning and obtaining more precise point cloud data, workpiece surface roughness, and workpiece material type. An edge computing unit runs a point cloud processing model to intelligently extract workpiece features from the point cloud data, improving the accuracy of feature identification. Furthermore, by combining workpiece features with environmental data, targeted and automatic spraying parameters are generated based on the current workpiece structural features, surface features, and the current environment. This allows for better spraying operations, improved spraying results, reduced film thickness errors after spraying, and enhanced spraying efficiency. The parameters meet environmental requirements, reducing paint waste. Simultaneously, by generating spray gun trajectories through reinforcement learning, feedback information can be used to further optimize the trajectory. Combined with GAN (Generative Adversarial Network), diverse path schemes are generated, ensuring coverage of complex workpiece surfaces. This effectively improves the intelligence and scientific rigor of path planning, meeting the spraying requirements of various complex workpiece structures, increasing spraying efficiency, further reducing paint waste, and saving material costs. Monitoring spraying quality generates feedback information that is fed back to the spraying parameter and path generation process. Based on the spraying quality, the current spraying parameters and paths are further optimized, perfecting the spraying process adjustment loop and better meeting the user's actual needs. Furthermore, through the collaboration of the DQN algorithm and Generative Adversarial Network (GAN), spraying paths are generated more scientifically and optimized based on feedback data, further improving the scientific rigor of the trajectory and adapting to complex workpiece surfaces. Therefore, this invention greatly improves the adaptability of the spraying system, meets the needs of handling complex workpiece shapes or non-standard workpieces, greatly improves the adaptability rate, and can better adapt to environmental changes through dynamic parameter adjustment and more precise path planning, improves the uniformity of coating, shortens the spraying time, and improves the spraying efficiency. At the same time, it reduces paint waste and defect rate, and reduces environmental pollution and operating costs.

[0041] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the corresponding accompanying drawings.

[0042] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0043] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0044] Figure 1 A structural block diagram of an adaptive spraying optimization system for a coating line according to an embodiment of the present invention is shown;

[0045] Figure 2 A flowchart illustrating a data processing method based on spraying baseline data and environmental data according to an embodiment of the present invention is shown.

[0046] Figure 3 A flowchart illustrating a spraying path determination method based on reinforcement learning technology according to an embodiment of the present invention is shown.

[0047] Figure 4 A flowchart illustrating a spraying effect feedback method according to an embodiment of the present invention is shown.

[0048] Figure 5 A flowchart illustrating an adaptive spraying optimization method for a coating line according to an embodiment of the present invention is shown.

[0049] Figure 6 A schematic diagram of the structure of a computing device according to an embodiment of the present invention is shown. Detailed Implementation

[0050] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0051] Figure 1 A structural block diagram of an adaptive spraying optimization system for a coating line according to an embodiment of the present invention is shown. The system includes: a sensing module 101, a data processing module 102, a control module 103, and a feedback module 104; wherein,

[0052] The sensing module 101 is used to collect geometric shape information and surface characteristic information of the target workpiece, as well as environmental information.

[0053] Preferably, the sensing module 101 is further used for:

[0054] Using a 3D laser scanner and an RGB-D camera, the target workpiece is scanned and photographed to obtain its geometric shape information and surface characteristic information; wherein, the geometric shape information includes at least point cloud data; and the surface characteristic information includes at least the workpiece roughness and the workpiece material type.

[0055] Environmental information is acquired based on environmental sensors; wherein, the environmental information includes at least temperature and humidity.

[0056] The data processing module 102 is used to extract workpiece features based on the geometric shape information and surface characteristic information of the target workpiece, and generate spraying path and spraying parameters in combination with environmental information.

[0057] Specifically, the data processing module 102 is further used for:

[0058] When generating spraying paths and spraying parameters, environmental information can be incorporated, or environmental information and feedback information can be incorporated.

[0059] Preferably, the spraying parameters may include at least the spray gun moving speed, spraying distance, and spraying amount; wherein, the spraying distance is the distance between the spray gun nozzle and the workpiece surface, and the spraying amount is the amount sprayed per second.

[0060] Preferably, the data processing module 102 communicates with the sensing module 101 through a high-speed interface and outputs the results to the control module 103.

[0061] The control module 103 is used to generate spraying control commands based on the optimized spraying path and spraying parameters, so as to control the execution unit to perform spraying operations.

[0062] Specifically, the control module 103 is further used for,

[0063] Generate corresponding spraying control commands based on spraying parameters and spraying path;

[0064] Send the spraying control command to the control execution unit;

[0065] The drive control execution unit performs the spraying operation according to the spraying control command.

[0066] Preferably, the spraying control instructions can be based on a standardized language of CNC programming, such as G-code;

[0067] The control execution unit may include a spray gun or a spraying robot, and a corresponding controller for each spray gun or spraying robot.

[0068] For example, if it is a painting robot, the ROS (Robot Operating System) is used as the controller to communicate with the painting robot and control the painting path, the speed of the spray gun, the painting distance and the amount of paint.

[0069] The feedback module 104 is used to monitor the spraying operation in real time to obtain spraying quality data, and transmit the spraying quality data as feedback information to the data processing module.

[0070] An adaptive spraying optimization system for a coating line, according to this embodiment, includes: a sensing module, a data processing module, a control module, and a feedback module. The sensing module collects geometric shape information and surface characteristic information of the target workpiece, as well as environmental information. The data processing module extracts workpiece features based on the geometric shape information and surface characteristic information of the target workpiece and generates a spraying path and spraying parameters in conjunction with the environmental information. The control module generates spraying control commands based on the optimized spraying path and spraying parameters to control the execution unit to perform spraying operations. The feedback module monitors the spraying operation in real time to obtain spraying quality data and transmits the spraying quality data as feedback information to the data processing module. This embodiment provides an adaptive spraying optimization system for coating lines. By using a 3D laser scanner and an RGB-D camera to collect the geometric shape and surface characteristic information of the workpiece, the system improves the accuracy of workpiece scanning, obtaining more precise point cloud data, workpiece surface roughness, and workpiece material type. It intelligently extracts workpiece features and, combined with workpiece features and environmental data, automatically generates targeted spraying parameters based on the current workpiece structure, surface features, and environment. This improves the spraying operation, enhances the spraying effect, ensures the spraying parameters meet environmental requirements, and reduces paint waste. Simultaneously, it generates spray gun trajectories through reinforcement learning, ensuring coverage of complex workpiece surfaces. This effectively improves the scientific nature of path planning, meets the spraying requirements of different complex workpiece structures, increases the efficiency of the spraying process, further reduces paint waste, and saves material costs. By monitoring the spraying quality, feedback information is generated and fed back to the process for generating spraying parameters and paths. Based on the spraying quality, the system further optimizes the current spraying parameters and paths, perfecting the closed-loop adjustment of the spraying process and better meeting the actual needs of users. Therefore, this invention greatly improves the adaptability of the spraying system, meets the needs of handling complex workpiece shapes or non-standard workpieces, greatly improves the adaptability rate, and can better adapt to environmental changes through dynamic parameter adjustment and more precise path planning, improves the uniformity of coating, shortens the spraying time, and improves the spraying efficiency. At the same time, it reduces paint waste and defect rate, and reduces environmental pollution and operating costs.

[0071] Figure 2 A flowchart illustrating a data processing method based on spraying baseline data and environmental data according to an embodiment of the present invention is shown.

[0072] like Figure 2 As shown, the method includes the following steps:

[0073] Step S201: The edge computing unit processes the geometric information based on the PointNet point cloud processing model to extract the workpiece features of the target workpiece.

[0074] Step S202: Using the Transformer model, spraying parameters are predicted based on workpiece features and environmental information.

[0075] Preferably, the spray gun moving speed, spraying distance and spraying amount can be determined based on the preset paint film thickness corresponding to different positions, so that the preset paint film thickness can be achieved at the position after spraying. The specific thickness value can be set by the user according to their needs and is not limited here.

[0076] For example, if environmental parameters are collected in real time and the temperature is 20-30℃ and the humidity is 40-60%, then the spraying volume can be dynamically adjusted to 0.1-0.5L / min and the spray gun moving speed to 0.2-0.5m / s.

[0077] Step S203: Determine the spraying path using reinforcement learning techniques.

[0078] Specifically, the spraying paths include diverse path schemes generated by combining Generative Adversarial Networks (GANs).

[0079] Furthermore, based on the above method, Figure 3 A flowchart illustrating a spraying path determination method based on reinforcement learning technology according to an embodiment of the present invention is shown.

[0080] like Figure 3 As shown, the method includes the following steps:

[0081] Step S301: Using the DQN (Deep Q Network) algorithm, based on the input workpiece features and environmental information, generate at least one initial path according to the ε-greedy strategy, and set the reward function and basic reward.

[0082] Preferably, the update rule in the DQN algorithm is as follows:

[0083] Q(s,a)←Q(s,a)+α[r+γmax a′ Q(s′,a′)-Q(s,a)]

[0084] Where s represents the current state; a represents the currently selectable action; Q(s,a) is the expected value of the long-term cumulative reward obtained by taking action a in state s; α is the learning rate; r is the immediate reward of the current action; γ is a discount factor to measure the current value of future rewards; s′ is the next state after taking action a; a′ is the selectable action in the next state s′; max a′ Q(s′,a′) is the maximum Q value in the next state s′.

[0085] Wherein, the reward function is

[0086] r new =r base +λ*D(G(path))

[0087] Where, r new The new reward after the update; base λ is the base reward; G(path) is the synergy coefficient; D(G(path)) is the selected feature in the path; and D(G(path)) is the reward component corresponding to the selected feature.

[0088] Furthermore, the target Q-value is calculated through the established target network, i.e.

[0089] y = r + γmax a′ Q(s′,a′;θ - )

[0090] Where, θ - The network parameters θ of the current network are copied and updated at preset time intervals.

[0091] Preferably, in order to make the target Q value relatively stable, θ - It does not update frequently; the preset time interval can be set by the user to a longer period, such as every 100, 200, or 1000 steps.

[0092] In step S302, the GAN-based generator converts the obtained initial path into the corresponding spraying simulation image.

[0093] Step S303: The GAN-based discriminator obtains feedback information as a real quality inspection image, and inputs it into the discriminator along with the spraying simulation image, and outputs a path quality score.

[0094] Step S304: Based on the reward function and the path quality score, DQN is triggered to replan the local path and generate an optimized spraying path.

[0095] Preferably, DQN updates the neural network parameters by minimizing the mean square error (MSE) between the predicted Q-value and the target Q-value using gradient descent.

[0096]

[0097] Where L(θ) is the mean square error between the predicted and target values ​​of Q; N is the number of empirical data points for mini-batch training; y i Let Q(s) be the target Q value for the i-th empirical data. i ,a i ;θ) is the predicted Q-value for the i-th empirical data; s i a i Let θ represent the state and its executable action corresponding to the i-th empirical data point; θ represents the network parameters of the current network.

[0098] Based on the above method, the point cloud processing model, running through edge computing units, intelligently extracts workpiece features from point cloud data, improving the accuracy of feature identification. Furthermore, by combining workpiece features with environmental data, targeted and dynamic spraying parameters are generated based on the current workpiece structure, surface features, and environment. This improves the spraying operation, enhances the spraying effect, keeps film thickness error within ±5%, and ensures the spraying parameters meet environmental requirements, reducing paint waste. Simultaneously, the spray gun trajectory is generated using reinforcement learning, and feedback information can be used to further optimize the trajectory. Combined with GAN, diverse path schemes are generated to ensure coverage of complex workpiece surfaces, effectively improving the intelligence and scientific rigor of path planning. This meets the spraying requirements of various complex workpiece structures, increases spraying efficiency, further reduces paint waste, and saves material costs. Moreover, through the collaboration of the DQN algorithm and Generative Adversarial Network (GAN), spraying paths are generated more scientifically and optimized based on feedback data, further improving the scientific rigor of the trajectory and adapting to complex workpiece surfaces.

[0099] Figure 4 A flowchart illustrating a spraying effect feedback method according to an embodiment of the present invention is shown.

[0100] like Figure 4 As shown, the method includes the following steps:

[0101] Step S401: Use the spraying monitoring unit to monitor the spraying effect of the target workpiece spraying operation and obtain spraying quality data.

[0102] Preferably, the coating quality data includes at least: film thickness data and surface coating quality.

[0103] Specifically, a visual inspection camera is used to determine the surface quality data of the target workpiece; the surface quality data includes at least: surface defect feature data and coating physical property data.

[0104] Preferably, the surface defect feature data includes defect type identification and the number and area ratio of each type of defect; the coating physical property data includes at least color difference value and gloss.

[0105] Specifically, the film thickness data of the target workpiece is detected by using a film thickness sensor and a vision inspection camera to determine the uniformity of the film thickness after spraying.

[0106] Preferably, the film thickness uniformity is calculated using a film thickness sensor combined with the color difference value determined by a visual inspection camera.

[0107] Step S402: The spraying quality data is transmitted to the data processing module as feedback information.

[0108] Based on the above method, by monitoring the spraying quality, feedback information is generated and fed back to the process of generating spraying parameters and spraying paths. According to the spraying quality, the current spraying parameters and spraying paths are further optimized in a targeted manner, improving the closed loop of spraying process adjustment and better meeting the actual needs of users.

[0109] Figure 5 A flowchart illustrating an adaptive spraying optimization method for a coating line according to an embodiment of the present invention is shown.

[0110] like Figure 5 As shown, the method includes the following steps:

[0111] Step S501: Collect the geometric shape information and surface characteristic information of the target workpiece, as well as environmental information.

[0112] Specifically, the acquisition of geometric shape information and surface characteristic information of the target workpiece, as well as environmental information, further includes,

[0113] Using a 3D laser scanner and an RGB-D camera, the target workpiece is scanned and photographed to obtain its geometric shape information and surface characteristic information; wherein, the geometric shape information includes at least point cloud data; and the surface characteristic information includes at least the workpiece roughness and the workpiece material type.

[0114] Environmental information is acquired based on environmental sensors; wherein, the environmental information includes at least temperature and humidity.

[0115] Step S502: Extract workpiece features based on the geometric shape and surface characteristics of the target workpiece, and generate spraying path and spraying parameters by combining environmental information.

[0116] Specifically, the step of extracting workpiece features based on the geometric shape and surface characteristic information of the target workpiece, and generating a spraying path and spraying parameters by combining environmental information, further includes:

[0117] Edge computing units are used to process geometric information based on the PointNet point cloud processing model to extract the workpiece features of the target workpiece;

[0118] The spraying parameters are predicted using the Transformer model based on workpiece features and environmental information.

[0119] Reinforcement learning techniques are used to determine the spraying path.

[0120] Specifically, using the DQN algorithm, based on the input workpiece features and environmental information, at least one initial path is generated according to the ε-greedy strategy, and a reward function and basic reward are set;

[0121] The GAN-based generator converts the obtained initial path into the corresponding spraying simulation image;

[0122] The discriminator based on GAN obtains feedback information as a real quality inspection image, which is input into the discriminator along with the spraying simulation image, and outputs a path quality score.

[0123] Based on the reward function, DQN is triggered to replan the local path based on the path quality score, generating an optimized spraying path.

[0124] Preferably, the update rule in the DQN algorithm is as follows:

[0125] Q(s,a)←Q(s,a)+α[r+γmax a′ Q(s′,a′)-Q(s,a)]

[0126] Where s represents the current state; a represents the currently selectable action; Q(s,a) is the expected value of the long-term cumulative reward obtained by taking action a in state s; α is the learning rate; r is the immediate reward of the current action; γ is a discount factor to measure the current value of future rewards; s′ is the next state after taking action a; a′ is the selectable action in the next state s′; max a′ Q(s′,a′) is the maximum Q value in the next state s′.

[0127] Wherein, the reward function is

[0128] r new =r base +λ*D(G(path))

[0129] Where, r new The new reward after the update; base λ is the base reward; G(path) is the synergy coefficient; D(G(path)) is the selected feature in the path; and D(G(path)) is the reward component corresponding to the selected feature.

[0130] Furthermore, the target Q-value is calculated through the established target network, i.e.

[0131] y = r + γmax a′ Q(s′,a′;θ - )

[0132] Where, θ - The network parameters θ of the current network are copied and updated according to a preset time interval. Preferably, the preset time interval can be every 100, 200, or 1000 steps.

[0133] DQN uses gradient descent to minimize the mean squared error (MSE) between the predicted Q-value and the target Q-value, and updates the neural network parameters accordingly.

[0134]

[0135] Where L(θ) is the mean square error between the predicted and target values ​​of Q; N is the number of empirical data points for mini-batch training; y i Let Q(s) be the target Q value for the i-th empirical data. i ,a i ;θ) is the predicted Q-value for the i-th empirical data; s i a i Let θ represent the state and its executable action corresponding to the i-th empirical data point; θ represents the network parameters of the current network.

[0136] Step S503: Based on the optimized spraying path and spraying parameters, generate spraying control commands to control the execution unit to perform spraying operations.

[0137] Specifically, the step of generating spray control commands based on optimized spraying paths and spraying parameters to control the execution unit to perform spraying operations further includes:

[0138] Generate corresponding spraying control commands based on spraying parameters and spraying path;

[0139] Send the spraying control command to the control execution unit;

[0140] The drive control execution unit performs the spraying operation according to the spraying control command.

[0141] Step S504: Real-time monitoring of the spraying operation to obtain spraying quality data, and using the spraying quality data as feedback information.

[0142] Specifically, the real-time monitoring of the spraying operation obtains spraying quality data, and the spraying quality data is used as feedback information, further including:

[0143] The spraying monitoring unit is used to monitor the spraying effect of the target workpiece and obtain spraying quality data.

[0144] The coating quality data will be used as feedback information.

[0145] Preferably, a visual inspection camera is used to determine the surface quality data of the target workpiece; wherein, the surface quality data includes at least: surface defect feature data and coating physical property data;

[0146] The film thickness data of the target workpiece is detected by using a film thickness sensor and a vision inspection camera to determine the uniformity of the film thickness after spraying.

[0147] According to the adaptive spraying optimization method for coating lines provided in this embodiment, the geometric shape information and surface characteristic information of the target workpiece, as well as environmental information, are collected; workpiece features are extracted based on the geometric shape information and surface characteristic information of the target workpiece, and spraying path and spraying parameters are generated in combination with environmental information; based on the optimized spraying path and spraying parameters, spraying control commands are generated to control the execution unit to perform spraying operations; spraying operations are monitored in real time to obtain spraying quality data, and the spraying quality data is used as feedback information. This embodiment provides an adaptive spraying optimization method for coating lines. By using a 3D laser scanner and an RGB-D camera to acquire the geometric shape and surface characteristic information of the workpiece, the accuracy of workpiece scanning is improved, resulting in more precise point cloud data, workpiece surface roughness, and workpiece material type. An edge computing unit runs a point cloud processing model to intelligently extract workpiece features from the point cloud data, improving the accuracy of feature identification. Furthermore, by combining workpiece features with environmental data, targeted and automatic spraying parameters are generated based on the current workpiece structural features, surface features, and the current environment. This allows for better spraying operations, improved spraying results, reduced film thickness errors after spraying, and optimized spraying parameters. Meeting environmental requirements and reducing paint waste, this system generates spray gun trajectories using reinforcement learning. Feedback information further optimizes these trajectories, and combined with GAN (Generative Adversarial Network) to generate diverse path solutions, ensuring coverage of complex workpiece surfaces. This significantly improves the intelligence and scientific rigor of path planning, meeting the spraying requirements of various complex workpiece structures, increasing spraying efficiency, reducing paint waste, and saving material costs. Monitoring spraying quality generates feedback information that is fed back to the spraying parameters and path generation process. Based on the spraying quality, the current parameters and paths are further optimized, creating a closed-loop adjustment system that better meets user needs. Furthermore, the collaboration between the DQN algorithm and Generative Adversarial Network (GAN) generates spraying paths more scientifically and optimizes them based on feedback data, further enhancing the scientific rigor of the trajectory and adapting to complex workpiece surfaces. Therefore, this invention greatly improves the adaptability of the spraying system, meets the needs of handling complex workpiece shapes or non-standard workpieces, greatly improves the adaptability rate, and can better adapt to environmental changes through dynamic parameter adjustment and more precise path planning, improves the uniformity of coating, shortens the spraying time, and improves the spraying efficiency. At the same time, it reduces paint waste and defect rate, and reduces environmental pollution and operating costs.

[0148] The present invention also provides a non-volatile computer storage medium storing at least one executable instruction that can perform the operations that can be achieved by an adaptive spraying optimization system for a coating line in any of the above system embodiments.

[0149] Figure 6 The diagram illustrates the structure of a computing device according to an embodiment of the present invention. The specific embodiments of the present invention do not limit the specific implementation of the computing device.

[0150] like Figure 6 As shown, the computing device may include: a processor 602, a communications interface 604, a memory 606, and a communications bus 608.

[0151] in:

[0152] The processor 602, communication interface 604, and memory 606 communicate with each other via communication bus 608.

[0153] Communication interface 604 is used to communicate with other network elements such as clients or other servers.

[0154] The processor 602 is used to execute program 610, specifically to perform the operations that can be achieved by the adaptive spraying optimization system for a coating line in the above embodiment of the adaptive spraying optimization system for a coating line.

[0155] Specifically, program 610 may include program code, which includes computer operation instructions.

[0156] Processor 602 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The computing device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0157] Memory 606 is used to store program 610. Memory 606 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0158] Specifically, program 610 can be used to cause processor 602 to perform operations achievable by an adaptive spraying optimization system for a coating line in any of the above system embodiments. The specific implementation of each step in program 610 can be found in the corresponding descriptions of the steps and units in the above embodiments of an adaptive spraying optimization system for a coating line, and will not be repeated here. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices and modules described above can be referred to the corresponding process descriptions in the foregoing method embodiments, and will not be repeated here.

[0159] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0160] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0161] Similarly, it should be understood that, in order to streamline this disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof. However, this method of disclosure should not be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the claims, inventive aspects lie in fewer than all features of a single foregoing disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0162] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0163] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, in the claims, any of the claimed embodiments can be used in any combination.

[0164] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components according to the embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0165] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An adaptive spraying optimization system for a coating line, comprising: The system comprises a sensing module, a data processing module, a control module, and a feedback module; among which, The sensing module is used to collect geometric shape information and surface characteristic information of the target workpiece, as well as environmental information; The data processing module is used to extract workpiece features based on the geometric shape and surface characteristics of the target workpiece, and generate spraying paths and spraying parameters in combination with environmental information. The control module is used to generate spray control commands based on the optimized spray path and spray parameters, so as to control the execution unit to perform spray operation. The feedback module is used to monitor the spraying operation in real time to obtain spraying quality data, and transmit the spraying quality data as feedback information to the data processing module. The data processing module is further used for: Edge computing units are used to process geometric information based on the PointNet point cloud processing model to extract the workpiece features of the target workpiece; The spraying parameters are predicted using the Transformer model based on workpiece features and environmental information. Reinforcement learning techniques are used to determine the spraying path; The method of determining the spraying path using reinforcement learning technology further includes: Using the DQN algorithm, based on the input workpiece features and environmental information, at least one initial path is generated according to the ε-greedy strategy, and a reward function and basic reward are set. The GAN-based generator converts the obtained initial path into the corresponding spraying simulation image; The discriminator based on GAN obtains feedback information as a real quality inspection image, which is input into the discriminator along with the spraying simulation image, and outputs a path quality score. Based on the reward function, DQN is triggered to replan the local path based on the path quality score, generating an optimized spraying path.

2. The adaptive spraying optimization system for a coating line according to claim 1, characterized in that, The sensing module is further used for: Using a 3D laser scanner and an RGB-D camera, the target workpiece is scanned and photographed to obtain its geometric shape information and surface characteristic information; wherein, the geometric shape information includes at least point cloud data; and the surface characteristic information includes at least the workpiece roughness and the workpiece material type. Environmental information is acquired based on environmental sensors; wherein, the environmental information includes at least temperature and humidity.

3. The adaptive spraying optimization system for a coating line according to claim 1, characterized in that, The control module is further used for: Generate corresponding spraying control commands based on spraying parameters and spraying path; Send the spraying control command to the control execution unit; The drive control execution unit performs the spraying operation according to the spraying control command.

4. The adaptive spraying optimization system for a coating line according to claim 1, characterized in that, The feedback module is further used for: The spraying monitoring unit is used to monitor the spraying effect of the target workpiece and obtain spraying quality data. The spraying quality data is transmitted as feedback information to the data processing module.

5. The adaptive spraying optimization system for a coating line according to claim 4, characterized in that, The method of using a spraying monitoring unit to monitor the spraying effect of the target workpiece and obtain spraying quality data further includes: The surface quality data of the target workpiece is determined using a visual inspection camera; the surface quality data includes at least: surface defect feature data and coating physical property data; The film thickness data of the target workpiece is detected by using a film thickness sensor and a vision inspection camera to determine the uniformity of the film thickness after spraying.

6. An adaptive spraying optimization method for a coating line, comprising: Collect geometric shape information, surface characteristic information, and environmental information of the target workpiece; Workpiece features are extracted based on the geometric shape and surface characteristics of the target workpiece, and spraying paths and spraying parameters are generated by combining environmental information. Based on the optimized spraying path and spraying parameters, spraying control commands are generated to control the execution unit to perform spraying operations. Real-time monitoring of the spraying operation yields spraying quality data, which is then used as feedback information. Edge computing units are used to process geometric information based on the PointNet point cloud processing model to extract the workpiece features of the target workpiece; The spraying parameters are predicted using the Transformer model based on workpiece features and environmental information. Reinforcement learning techniques are used to determine the spraying path; The method of determining the spraying path using reinforcement learning technology further includes: Using the DQN algorithm, based on the input workpiece features and environmental information, at least one initial path is generated according to the ε-greedy strategy, and a reward function and basic reward are set. The GAN-based generator converts the obtained initial path into the corresponding spraying simulation image; The discriminator based on GAN obtains feedback information as a real quality inspection image, which is input into the discriminator along with the spraying simulation image, and outputs a path quality score. Based on the reward function, DQN is triggered to replan the local path based on the path quality score, generating an optimized spraying path.

7. A computing device, comprising: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform the operations implemented by the coating line adaptive spraying optimization system as described in any one of claims 1-5.

8. A computer storage medium storing at least one executable instruction that causes a processor to perform the operations implemented by a coating line adaptive spraying optimization system as claimed in any one of claims 1-5.

Images