Automobile part spraying control method and system

Abstract

The invention provides an automobile part spraying control method and system, and relates to the technical field of automobile manufacturing, and the method comprises the steps: collecting spraying environment parameters in real time, generating a comprehensive digital model containing the environment parameters and part characteristics, and carrying out the intelligent partitioning of the surface of an automobile part through a clustering partitioning algorithm, and dividing the surface into a plurality of sub-regions; on the basis of each sub-region, combining with the current environment parameters, generating an optimal spraying parameter for each sub-region by utilizing an intelligent agent trained by a reinforcement learning algorithm, and on the basis of the spatial relationship between the optimal spraying parameters and the sub-regions, optimizing a global spraying sequence by adopting an ant colony path planning algorithm, and generating an optimal spraying path; the spraying robot sprays, a real-time quality evaluation result of spraying is generated, a pre-constructed fuzzy logic controller is adopted, spraying parameters are adjusted, sprayed parts are obtained, a full-period comprehensive data set is integrated, and spraying control parameters are iteratively updated.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile manufacturing, and in particular to a method and system for controlling the spraying of automobile parts. Background Art

[0002] The spraying quality of automobile parts directly affects the appearance and anti-corrosion performance of the car, and is a key link in the automobile manufacturing process. The traditional control method of automobile parts spraying mainly relies on manual experience and fixed parameter settings, which is prone to problems.

[0003] In the existing technology, the spraying parameter settings lack flexibility, resulting in unstable spraying quality; manual inspection methods are highly subjective and the quality control accuracy is insufficient; the trial and error method is usually adopted, and the optimization process is inefficient; the shapes of automotive parts are complex and diverse, and traditional methods are difficult to cope with complex working conditions; there is a lack of data-driven continuous optimization capabilities, making it difficult to take into account multiple quality indicators.

[0004] In summary, it is necessary to apply artificial intelligence technology, combined with intelligent control technology, to the spraying process of automobile parts to develop a more intelligent and adaptive spraying control method. The present invention can solve the problems in the prior art. Summary of the invention

[0005] The embodiments of the present invention provide a method and system for controlling the spraying of automobile parts, which can solve the problems in the prior art.

[0006] According to a first aspect of the embodiments of the present invention, Provided is a method for controlling spraying of automobile parts, comprising: The spraying environment parameters are collected in real time through temperature and humidity sensors, air pressure sensors and air quality sensors; the surface spectrum information and three-dimensional structure data of automobile parts are obtained by using multi-spectral cameras and three-dimensional structured light scanners; the spraying environment parameters, surface spectrum information and three-dimensional structure data are input into the pre-trained multimodal fusion neural network to generate a comprehensive digital model containing environmental parameters and parts characteristics. Based on the comprehensive digital model, the surface of automobile parts is intelligently partitioned into multiple sub-areas using a clustering partitioning algorithm; Based on each sub-area and combined with the current environmental parameters, an intelligent agent trained by a reinforcement learning algorithm generates optimal spraying parameters for each sub-area. Based on the optimal spraying parameters and the spatial relationship between the sub-areas, an ant colony path planning algorithm is used to optimize the global spraying sequence and generate the optimal spraying path. The optimal spraying parameters and the optimal spraying path are sent to a spraying robot with an adaptive control function. During the spraying process, the spraying status data is collected in real time through a high-speed camera and a coating thickness sensor. The spraying status data is input into a pre-trained convolutional neural network to generate real-time quality evaluation results for the spraying. A pre-built fuzzy logic controller is used to adjust the spraying parameters according to the deviation between the real-time quality assessment result and the preset quality target until the spraying of the current automobile parts is completed, and the sprayed parts are obtained. The sprayed parts are scanned and inspected to obtain the final spraying quality data. The final spraying quality data, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set. Based on the full-cycle comprehensive data set, the gradient boosting decision tree algorithm is used to extract the spraying quality influencing variables. Based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network and the algorithm parameters of the reinforcement learning algorithm are updated.

[0007] In an optional embodiment, The spraying environment parameters, surface spectrum information and three-dimensional structure data are input into the pre-trained multimodal fusion neural network to generate a comprehensive digital model containing environmental parameters and component characteristics. Based on the comprehensive digital model, the surface of the automotive parts is intelligently partitioned using a clustering partitioning algorithm and divided into multiple sub-areas, including: Collect spraying environment parameters, including temperature data, humidity data, air pressure data, and air particle concentration data; collect hyperspectral information on the surface of automobile parts, including spectral reflectance data of multiple bands; collect three-dimensional structure data of automobile parts, including point cloud data; Constructing a multimodal fusion neural network, wherein the multimodal fusion neural network includes an environmental parameter branch, a spectral information branch, and a three-dimensional structure branch; The spraying environment parameters are input into the environment parameter branch, and the time series environment data is processed using a one-dimensional convolutional neural network to obtain an environment feature vector; the hyperspectral information is input into the spectral information branch, and the spatial-spectral features are extracted using a three-dimensional convolutional neural network to obtain a spectral feature vector; the three-dimensional structure data is input into the three-dimensional structure branch, and the point cloud data is processed using a point cloud convolutional network to obtain a structure feature vector; The environmental feature vector, the spectral feature vector and the structural feature vector are fused by using a multi-head self-attention module to obtain a fused feature; the fused feature is input into a multi-layer perceptron, and a feature vector of a fixed dimension is output as a comprehensive digital model; Based on the comprehensive digital model, a clustering partitioning algorithm is used to intelligently partition the surface of the automobile parts to determine a plurality of sub-areas.

[0008] In an optional embodiment, The cluster partitioning algorithm includes: Constructing a cluster partitioning algorithm based on K-means clustering; initializing cluster centers and using feature vectors in the comprehensive digital model as initial data points; Calculate the distance from each data point to each cluster center, introduce spatial constraints when calculating the distance, add spatial coordinate terms to the distance calculation formula, and assign adjustable weight coefficients; assign each data point to the nearest cluster center; calculate the average value of all data points in each cluster, determine the new cluster center, and update it; calculate the local curvature of each data point, and use the local curvature as the cluster feature dimension; Repeat the iteration until the position change of the cluster center is less than the preset cluster center change threshold, and the optimal partitioning scheme is obtained; Calculating the silhouette coefficient of the optimal partitioning scheme by a similarity measurement method; Within the preset range of partition numbers, the initialization and iterative calculation process is repeated to obtain the silhouette coefficients under different numbers of partitions, and the partition scheme with the largest silhouette coefficient is selected as the final intelligent partitioning result.

[0009] In an optional embodiment, Based on each sub-area, combined with the current environmental parameters, the intelligent agent trained by the reinforcement learning algorithm generates the optimal spraying parameters for each sub-area, including: Building a main network based on a deep Q network, the main network includes an input layer, two hidden layers and an output layer, wherein the number of nodes in the input layer is the same as the dimension of the state vector, the number of nodes in the output layer is equal to the size of the action space, the hidden layer uses a ReLU activation function, and the output layer uses a linear activation function; Based on the geometric features, material properties and environmental parameters of the sub-area, the state space is determined; based on the spraying parameters, the action space is determined; based on the preset coating quality indicators, the weighted sum of the coating quality indicators in the sub-area is calculated to construct a reward function; Based on the state space, action space and reward function, a simulation environment is constructed to simulate the spraying process of sub-areas on the surface of automobile parts; Construct an experience pool to store the experience data obtained by the interaction between the intelligent agent and the simulation environment, initialize the experience pool, create a target network based on the structure of the main network, and calculate the target Q value; The ε-greedy method is used to set the exploration strategy. The ε value is initially set to 0.9 and linearly decays to 0.1 as the number of training rounds increases. The intelligent agent selects spraying parameters according to the current sub-region state and exploration strategy, interacts with the simulation environment, obtains rewards and the next sub-region state, and calculates the temporal difference error based on the current sub-region state, the selected spraying parameters, the rewards obtained, the next sub-region state, and the target Q value calculated by the target network; The current sub-region state, the selected spraying parameters, the reward obtained, the next sub-region state and the time series difference error are taken as one piece of experience data, stored in and updated in the experience pool; a sampling probability is assigned to each piece of experience data based on the time series difference error in the experience pool, batch data is selected based on the sampling probability, and the loss function is calculated based on the batch data, the main network and the target network; Minimize the loss function through the gradient descent method, update the main network parameters, assign the weighted average of the main network parameters to the corresponding parameters of the target network, update the target network parameters, add Gaussian noise to the main network parameters and the target network parameters, and continue to iterate until the preset training rounds are reached. Determine the main network that has been trained, input the current state of each sub-area into the trained main network, and input the optimal spraying parameter combination of the sub-area.

[0010] In an optional embodiment, Based on the optimal spraying parameters and the spatial relationship between sub-areas, the ant colony path planning algorithm is used to optimize the global spraying sequence and generate the optimal spraying path including: A node is generated for each sub-region. The edges between nodes are determined based on the spatial relationship between sub-regions. The graph structure is determined based on the nodes and edges. Based on the spatial distance between the sub-regions, the optimal spraying parameters corresponding to each sub-region and the surface curvature change of the sub-region, the weight of the edge is calculated, wherein the spatial distance is determined based on the Euclidean distance between adjacent sub-regions, the similarity between the corresponding vectors of the optimal spraying parameters of the two sub-regions is calculated by cosine similarity, the spraying parameter similarity is determined, the curvature between adjacent sub-regions is calculated, the curvature difference is determined, and the surface curvature change is obtained; According to the preset ant population size, the ant population and the corresponding pheromone matrix are initialized, and each ant randomly selects a sub-region as the starting sub-region; Based on the weight of the edge, determine the heuristic information, based on the pheromone matrix, determine the pheromone concentration, calculate the state transition probability according to the heuristic information and the pheromone concentration, determine the probability of each ant selecting the next sub-area, select the sub-area corresponding to the maximum probability, and move the ant to the corresponding sub-area; After the ant moves, the local pheromone of the edge passed by the ant during the movement is updated, and the local pheromone is updated based on the corresponding pheromone concentration in the pheromone matrix, the preset pheromone volatility coefficient and the preset initial pheromone concentration; Repeat the steps of sub-region selection and local pheromone update until all ants complete a complete path. Determine the path length of the ants based on the spatial distance between sub-regions and the time distance of the spray gun adjustment. Determine the corresponding spray quality score of the ants based on the similarity of spraying parameters and curvature changes between adjacent sub-regions. Update the global pheromone based on the path length and spray quality score. Repeat until the preset maximum number of iterations is reached, select the spray path corresponding to the maximum spray quality score, and determine the optimal spray path.

[0011] In an optional embodiment, The path length is given by the following formula: ; Where L represents the total path length, m represents the number of sub-regions contained in the path, i represents the index of the sub-region in the path, r(·) represents the sub-region with the corresponding index in the path, d(·,·) represents the spatial distance between sub-regions, T represents the time coefficient required to adjust the spraying parameters, P(·) represents the optimal spraying parameter vector, and ||·|| represents the calculation of the Euclidean distance; The spraying quality score is as follows: ; Where Q represents, λ represents the weight coefficient, sim(·) represents the calculation of spraying parameter similarity, c(·) represents the calculation of average curvature, and |·| represents the absolute value of curvature difference.

[0012] In an optional embodiment, A pre-built fuzzy logic controller is used to adjust the spraying parameters according to the deviation between the real-time quality evaluation result and the preset quality target until the spraying of the current automobile parts is completed, and the sprayed parts are obtained. The sprayed parts are scanned and tested to obtain the final spraying quality data. The final spraying quality data, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set. Based on the full-cycle comprehensive data set, the gradient boosting decision tree algorithm is used to extract the spraying quality influencing variables. Based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network and the algorithm parameters of the reinforcement learning algorithm are updated, including: According to the deviation between the real-time quality assessment results and the preset quality targets, the spraying parameters are adjusted through the pre-built fuzzy logic controller until the current automotive parts spraying is completed, including: A dynamic fuzzy neural network is introduced, and the input and output variable membership function parameters of the fuzzy logic controller are adjusted according to the real-time spraying effect feedback, and the loss function is calculated, wherein the loss function includes the mean square error between the predicted quality and the actual quality, and the L2 regularization term of the difference before and after the membership function parameter adjustment; based on the loss function, the gradient is calculated by the back propagation algorithm, and the input and output variable membership function parameters of the fuzzy logic controller are updated in combination with the stochastic gradient descent method; the loss function calculation and parameter updating steps are repeated until the preset number of iterations is reached to obtain the sprayed parts; Use a high-precision three-dimensional laser scanner to perform an all-round scan on the spray-painted parts to generate a high-precision three-dimensional model; Based on the high-precision three-dimensional model, a multi-scale spraying quality assessment method is used to calculate a comprehensive quality score, the comprehensive quality score, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set, and the full-cycle comprehensive data set is stored in a time series database; Based on the full-cycle comprehensive data set, the LightGBM algorithm is used to extract the variables affecting the spraying quality, and the SHAP value is used to explain the LightGBM model and quantify the importance of each feature to obtain the ranking of key factors affecting the spraying quality; Based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network are updated, wherein the multimodal fusion neural network includes a one-dimensional convolutional neural network for processing time series spraying parameter data, a two-dimensional convolutional neural network for processing component geometric feature images, a long short-term memory network for processing environmental parameter time series, and a multi-layer perceptron sub-network for processing static features, and the outputs of each sub-network are fused through a Transformer-based attention layer; Based on the spraying quality influencing variables, the algorithm parameters of the reinforcement learning algorithm are updated, and the updated multimodal fusion neural network and the reinforcement learning algorithm are applied to the spraying process of the next batch of automobile parts to continuously adjust the spraying quality.

[0013] According to a second aspect of the embodiments of the present invention, Provided is an automobile parts spraying control system, comprising: The first unit is used to collect spraying environment parameters in real time through temperature and humidity sensors, air pressure sensors and air quality sensors; use multispectral cameras and three-dimensional structured light scanners to obtain surface spectral information and three-dimensional structural data of automobile parts; input spraying environment parameters, surface spectral information and three-dimensional structural data into a pre-trained multimodal fusion neural network to generate a comprehensive digital model containing environmental parameters and component characteristics, and based on the comprehensive digital model, use a clustering partitioning algorithm to intelligently partition the surface of automobile parts into multiple sub-areas; The second unit is used to generate the optimal spraying parameters for each sub-area based on each sub-area and in combination with the current environmental parameters, using an intelligent agent trained by a reinforcement learning algorithm, and based on the optimal spraying parameters and the spatial relationship between the sub-areas, using an ant colony path planning algorithm to optimize the global spraying sequence and generate the optimal spraying path; the optimal spraying parameters and the optimal spraying path are sent to a spraying robot with an adaptive control function, and during the spraying process, the spraying state data is collected in real time through a high-speed camera and a coating thickness sensor, and the spraying state data is input into a pre-trained convolutional neural network to generate a real-time quality evaluation result of the spraying; The third unit is used to use a pre-built fuzzy logic controller to adjust the spraying parameters according to the deviation between the real-time quality assessment result and the preset quality target until the spraying of the current automobile parts is completed, the sprayed parts are obtained, the sprayed parts are scanned and inspected, and the final spraying quality data is obtained. The final spraying quality data, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set, based on the full-cycle comprehensive data set, the gradient boosting decision tree algorithm is used to extract the spraying quality influencing variables, and based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network and the algorithm parameters of the reinforcement learning algorithm are updated.

[0014] According to a third aspect of the embodiments of the present invention, An electronic device is provided, comprising: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0015] A fourth aspect of the embodiments of the present invention is: A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0016] The present invention uses multiple sensors to collect environmental parameters in real time, combines multi-spectral cameras and three-dimensional structured light scanners to obtain component surface information, and achieves comprehensive perception of the spraying environment and the object to be sprayed. A multimodal fusion neural network is used to generate a comprehensive digital model, and a clustering partitioning algorithm is used for intelligent partitioning to formulate personalized spraying strategies for different areas. This refined spraying method significantly improves the quality and accuracy of spraying, can adapt to automotive parts of different shapes and materials, and achieve a uniform, high-quality coating effect.

[0017] The present invention uses a reinforcement learning algorithm to generate optimal spraying parameters for each sub-area, and uses an ant colony path planning algorithm to optimize the global spraying sequence. This intelligent parameter setting and path planning method greatly improves the spraying efficiency and reduces unnecessary spraying time and paint waste. At the same time, through real-time quality evaluation and dynamic adjustment of the fuzzy logic controller, the stability and consistency of the spraying process are ensured, resource utilization is further optimized, and production costs are reduced.

[0018] The present invention constructs a closed-loop adaptive optimization system. By collecting spraying status data in real time and combining it with the final spraying quality data, a full-cycle comprehensive data set is generated. The gradient boosting decision tree algorithm is used to extract the variables affecting the spraying quality, and the parameters of the multimodal fusion neural network and the reinforcement learning algorithm are updated accordingly. This mechanism enables the system to continuously learn and optimize, adapt to different spraying environments and component characteristics, and achieve continuous improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of a method for controlling spraying of automobile parts according to an embodiment of the present invention; Figure 2 The figure is a schematic diagram of the structure of the automobile parts spraying control system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0022] Figure 1 FIG. 1 is a flow chart of a method for controlling the spraying of automobile parts according to an embodiment of the present invention. Figure 1 As shown, the method includes: S101. Collect spraying environment parameters in real time through temperature and humidity sensors, air pressure sensors and air quality sensors; obtain surface spectrum information and three-dimensional structure data of automobile parts by using multispectral cameras and three-dimensional structured light scanners; input spraying environment parameters, surface spectrum information and three-dimensional structure data into pre-trained multimodal fusion neural networks to generate a comprehensive digital model including environmental parameters and component characteristics; based on the comprehensive digital model, use clustering partitioning algorithms to intelligently partition the surface of automobile parts into multiple sub-areas; S102. Based on each sub-area and in combination with the current environmental parameters, an intelligent agent trained by a reinforcement learning algorithm is used to generate optimal spraying parameters for each sub-area. Based on the optimal spraying parameters and the spatial relationship between the sub-areas, an ant colony path planning algorithm is used to optimize the global spraying sequence and generate an optimal spraying path. The optimal spraying parameters and the optimal spraying path are sent to a spraying robot with an adaptive control function. During the spraying process, spraying status data is collected in real time by a high-speed camera and a coating thickness sensor. The spraying status data is input into a pre-trained convolutional neural network to generate a real-time quality evaluation result of the spraying. S103. Use a pre-built fuzzy logic controller to adjust the spraying parameters according to the deviation between the real-time quality assessment result and the preset quality target until the spraying of the current automobile parts is completed, obtain the sprayed parts, scan and detect the sprayed parts, obtain the final spraying quality data, integrate the final spraying quality data, environmental parameters, component characteristics and spraying parameters into a full-cycle comprehensive data set, based on the full-cycle comprehensive data set, extract the spraying quality influencing variables through the gradient boosting decision tree algorithm, and based on the spraying quality influencing variables, update the network parameters of the multimodal fusion neural network and the algorithm parameters of the reinforcement learning algorithm.

[0023] The specific implementation is as follows: The spraying control method for automobile parts first collects spraying environmental parameters in real time through a variety of sensors. Specifically, the DHT22 temperature and humidity sensor is used to collect ambient temperature and relative humidity with a sampling frequency of 1Hz; the BMP280 air pressure sensor is used to collect atmospheric pressure with a sampling frequency of 0.1Hz; the PMS5003 air quality sensor is used to collect PM2.5 and PM10 concentrations with a sampling frequency of 0.2Hz. These environmental parameters have an important impact on the spraying quality and need to be monitored in real time.

[0024] At the same time, multispectral cameras and 3D structured light scanners are used to obtain surface feature information of automotive parts. The multispectral camera uses the FLIR Blackfly S BFS-U3-51S5C-C model, which can simultaneously capture images in the visible light, near infrared and short-wave infrared bands, with a resolution of 2448x2048 pixels and a frame rate of 30fps. The 3D structured light scanner uses the Artec Leo model, with a scanning accuracy of up to 0.1mm and a scanning speed of up to 80 frames per second. These two devices can fully obtain the spectral information and 3D structural data of the surface of the parts.

[0025] The collected environmental parameters, surface spectral information and three-dimensional structure data are input into the pre-trained multimodal fusion neural network. The network uses ResNet50 as the backbone network and realizes the effective fusion of multimodal features through the attention mechanism. The network outputs a 300-dimensional feature vector as a comprehensive digital model of the component. Based on this model, the DBSCAN clustering algorithm is used to intelligently partition the surface of the component. The eps parameter of the DBSCAN algorithm is set to 2mm, and the minPts parameter is set to 10. By adjusting these two parameters, partition results of different granularities can be obtained. Generally, a medium-sized automotive part will be divided into 20-30 sub-areas.

[0026] For each sub-area, the optimal spraying parameters are generated by an intelligent agent trained using the Deep Deterministic Policy Gradient (DDPG) algorithm in combination with the current environmental parameters. The Actor network and Critic network of the DDPG algorithm both use a 3-layer fully connected network structure, with 256 and 128 hidden layer neurons, respectively. The input is the sub-area characteristics and environmental parameters, and the output is key parameters such as spraying distance, spraying speed, spraying angle, and atomization pressure. The reward function is designed as the weighted sum of spraying quality indicators and energy consumption. Through repeated training, the intelligent agent can generate optimal spraying parameters for sub-areas with different characteristics.

[0027] After obtaining the optimal spraying parameters for each sub-area, the global spraying sequence needs to be further optimized. Here, the ant colony algorithm is used for spraying path planning. The number of ant colonies is set to 50, the pheromone importance factor α=1, the heuristic factor β=2, and the pheromone volatility coefficient ρ=0.1. Through iterative optimization, an optimal spraying path with the shortest total path length and the highest spraying efficiency can be obtained.

[0028] The optimal spraying parameters and optimal spraying path are sent to the ABB IRB 5500 spraying robot with adaptive control function. The robot has 6 degrees of freedom, repeated positioning accuracy of ±0.15mm, and a maximum load of 13kg, which is very suitable for spraying automotive parts. During the spraying process, the spraying status data is collected in real time through the Phantom v2640 high-speed camera (maximum frame rate up to 25,030fps) and the Micro-Epsilon optoCONTROL 2520 coating thickness sensor (measuring range 46-650μm, resolution 1μm).

[0029] The collected spray status data is input into the pre-trained ResNet18 convolutional neural network to generate real-time spray quality assessment results. The network is pre-trained on a large amount of historical spray data and can quickly and accurately assess the current spray quality. The assessment results include multiple indicators such as coating uniformity, adhesion, glossiness, etc., expressed as scores from 0 to 100.

[0030] A pre-built fuzzy logic controller is used to adjust the spraying parameters according to the deviation between the real-time quality assessment results and the preset quality target. The input variables of the fuzzy controller are the quality deviation and the deviation change rate, and the output variables are the spraying distance adjustment amount, spraying speed adjustment amount, etc. The fuzzy rule base contains 25 IF-THEN rules. Through fuzzy reasoning, the appropriate parameter adjustment amount can be obtained to achieve closed-loop control of the spraying process.

[0031] After the current automotive parts are sprayed, the Zeiss COMET L3D 2 3D scanner is used to scan and inspect the sprayed parts to obtain the final spray quality data. The scanner has a measurement range of 45-500mm and an accuracy of up to 2μm. Through scanning, detailed quality data such as coating thickness distribution and surface roughness can be obtained.

[0032] The final spraying quality data, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set. Based on this data set, the XGBoost gradient boosting decision tree algorithm is used to extract the variables affecting the spraying quality. The main parameters of the XGBoost algorithm are set as follows: the maximum tree depth is 6, the minimum number of leaf node samples is 5, the learning rate is 0.1, and the subsampling ratio is 0.8. Through feature importance analysis, the key variables affecting the spraying quality and their importance can be obtained.

[0033] Finally, based on the extracted variables affecting spraying quality, the network parameters of the multimodal fusion neural network and the algorithm parameters of the reinforcement learning algorithm are updated. For the neural network, the Adam optimizer is used for fine-tuning, and the learning rate is set to 0.0001; for the reinforcement learning algorithm, the exploration rate ε and the discount factor γ are adjusted. Through continuous model updates, the intelligence level and control accuracy of the system can be continuously improved.

[0034] A specific data case is as follows: For a car bumper, the ambient temperature is 25°C, the relative humidity is 60%, and the atmospheric pressure is 101.3kPa. Through multimodal feature extraction, a 300-dimensional digital model is obtained. After DBSCAN clustering, the bumper is divided into 25 sub-areas. For the central area, the optimal spraying parameters generated by the DDPG algorithm are: spraying distance 200mm, spraying speed 300mm / s, spraying angle 75°, and atomization pressure 0.3MPa. The total length of the optimal path planned by the ant colony algorithm is 15.6m. During the spraying process, the real-time quality assessment score is 92 points. The final scanning test results show that the average coating thickness is 52μm, the thickness uniformity is ±3μm, and the surface roughness Ra is 0.8μm. XGBoost algorithm analysis shows that spraying distance and atomization pressure are the two most critical factors affecting quality.

[0035] In an optional implementation, spraying environment parameters, surface spectrum information and three-dimensional structure data are input into a pre-trained multimodal fusion neural network to generate a comprehensive digital model including environmental parameters and component characteristics. Based on the comprehensive digital model, a clustering partitioning algorithm is used to intelligently partition the surface of the automotive parts into multiple sub-areas including: Collect spraying environment parameters, including temperature data, humidity data, air pressure data, and air particle concentration data; collect hyperspectral information on the surface of automobile parts, including spectral reflectance data of multiple bands; collect three-dimensional structure data of automobile parts, including point cloud data; Constructing a multimodal fusion neural network, wherein the multimodal fusion neural network includes an environmental parameter branch, a spectral information branch, and a three-dimensional structure branch; The spraying environment parameters are input into the environment parameter branch, and the time series environment data is processed using a one-dimensional convolutional neural network to obtain an environment feature vector; the hyperspectral information is input into the spectral information branch, and the spatial-spectral features are extracted using a three-dimensional convolutional neural network to obtain a spectral feature vector; the three-dimensional structure data is input into the three-dimensional structure branch, and the point cloud data is processed using a point cloud convolutional network to obtain a structure feature vector; The environmental feature vector, the spectral feature vector and the structural feature vector are fused by using a multi-head self-attention module to obtain a fused feature; the fused feature is input into a multi-layer perceptron, and a feature vector of a fixed dimension is output as a comprehensive digital model; Based on the comprehensive digital model, a clustering partitioning algorithm is used to intelligently partition the surface of the automobile parts to determine a plurality of sub-areas.

[0036] In order to realize the intelligent spraying method of automobile parts based on multimodal fusion neural network and clustering partitioning, the following steps can be taken: First, collect spraying environment parameters. Use temperature and humidity sensors to collect temperature and humidity data, with a sampling frequency of 1Hz, and collect data continuously for 24 hours to obtain 86,400 temperature data points and 86,400 humidity data points. Use pressure sensors to collect pressure data, with a sampling frequency of 0.1Hz, and collect data continuously for 24 hours to obtain 8,640 pressure data points. Use a particle concentration detector to collect PM2.5 and PM10 concentration data, with a sampling frequency of 0.01Hz, and collect data continuously for 24 hours to obtain 864 PM2.5 concentration data points and 864 PM10 concentration data points.

[0037] Secondly, collect hyperspectral information on the surface of automotive parts. Use a hyperspectral imager to scan the surface of the parts and obtain reflectance data of 210 bands in the wavelength range of 400-2500nm. The scanning resolution is 1mm×1mm. For a 1m×1m part surface, a 1000×1000×210 hyperspectral data cube can be obtained.

[0038] Then, the 3D structure data of the auto parts is collected. The parts are scanned in all directions using a 3D laser scanner to obtain point cloud data. The scanning accuracy is 0.1mm. For a 1m×1m×0.5m part, about 5 million points of point cloud data can be obtained, and each point contains spatial coordinate (x, y, z) information.

[0039] Next, a multimodal fusion neural network is constructed. The environmental parameter branch uses 5 layers of one-dimensional convolutional layers, with a convolution kernel size of 3, a step size of 1, and the number of channels is 32, 64, 128, 256, and 512, respectively. The spectral information branch uses 4 layers of three-dimensional convolutional layers, with a convolution kernel size of 3×3×3, a step size of 1, and the number of channels is 32, 64, 128, and 256, respectively. The three-dimensional structure branch uses the PointNet++ network, which contains 4 set abstraction layers and 2 feature propagation layers. The multi-head self-attention module uses 8 attention heads and the hidden layer dimension is 512. The multi-layer perceptron contains 3 fully connected layers, and the hidden layer dimensions are 1024, 512, and 256, respectively.

[0040] The collected data is input into the constructed neural network for processing. The environmental parameter branch inputs an 86400×5 environmental parameter matrix, and after one-dimensional convolution processing, a 512-dimensional environmental feature vector is obtained. The spectral information branch inputs a 1000×1000×210 hyperspectral data cube, and after three-dimensional convolution processing, a 256-dimensional spectral feature vector is obtained. The three-dimensional structure branch inputs 5 million points of point cloud data, and after PointNet++ network processing, a 128-dimensional structural feature vector is obtained.

[0041] Then, the three feature vectors are fused using a multi-head self-attention module. First, the three feature vectors are concatenated into an 896-dimensional vector, and then input into eight parallel self-attention submodules. Each submodule generates a query vector, a key vector, and a value vector through linear transformation, calculates the attention weights, and weighted sums them to obtain a 64-dimensional output. The outputs of the eight submodules are concatenated to obtain a 512-dimensional fused feature.

[0042] Finally, the fused features are input into a three-layer fully connected network to obtain a 256-dimensional comprehensive digital model feature vector, which contains comprehensive information on environmental parameters, surface spectra, and three-dimensional structures.

[0043] Based on the generated comprehensive digital model, the K-means clustering algorithm is used to intelligently partition the surface of the component. First, the 256-dimensional feature vector is reduced to 3 dimensions, and then the number of cluster centers K=5 is set to cluster the reduced feature points. The clustering result is 5 sub-areas on the surface of the component, and different areas have different environmental adaptability and surface characteristics.

[0044] Through the above steps, the intelligent spraying partitioning method of automobile parts based on multimodal fusion and clustering partitioning is realized. This method comprehensively considers environmental factors, surface characteristics and structural features, and can partition the surface of parts more accurately, providing a basis for subsequent intelligent spraying.

[0045] In an optional implementation, the cluster partitioning algorithm includes: Constructing a cluster partitioning algorithm based on K-means clustering; initializing cluster centers and using feature vectors in the comprehensive digital model as initial data points; Calculate the distance from each data point to each cluster center, introduce spatial constraints when calculating the distance, add spatial coordinate terms to the distance calculation formula, and assign adjustable weight coefficients; assign each data point to the nearest cluster center; calculate the average value of all data points in each cluster, determine the new cluster center, and update it; calculate the local curvature of each data point, and use the local curvature as the cluster feature dimension; Repeat the iteration until the position change of the cluster center is less than the preset cluster center change threshold, and the optimal partitioning scheme is obtained; Calculating the silhouette coefficient of the optimal partitioning scheme by a similarity measurement method; Within the preset range of partition numbers, the initialization and iterative calculation process is repeated to obtain the silhouette coefficients under different numbers of partitions, and the partition scheme with the largest silhouette coefficient is selected as the final intelligent partitioning result.

[0046] In a specific implementation, a cluster partitioning algorithm is constructed based on K-means clustering to intelligently partition the comprehensive digital model. First, the cluster center is initialized, and the feature vector in the comprehensive digital model is used as the initial data point. The feature vector contains information such as geometric features and material properties.

[0047] Next, calculate the distance from each data point to each cluster center. Introduce spatial constraints when calculating the distance, add spatial coordinate terms to the distance calculation formula, and assign adjustable weight coefficients. For example, you can use the Euclidean distance weighted formula, where the weight of the spatial coordinate term can be set to 0.3-0.5. This ensures that points with similar spatial locations are more easily classified into the same cluster.

[0048] Each data point is then assigned to the cluster center that is closest to it. The average value of all data points in each cluster is calculated, a new cluster center is determined, and the cluster center position is updated.

[0049] During the iteration process, the local curvature of each data point is also calculated and used as a characteristic dimension for clustering. The local curvature can be calculated by fitting a quadratic surface around the data points. This helps to identify characteristic areas in the model, such as edges, corners, etc.

[0050] The process of iteratively calculating and updating the cluster center is repeated until the position change of the cluster center is less than a preset threshold, such as 0.001. In this way, the optimal partitioning scheme under the current number of partitions is obtained.

[0051] In order to evaluate the partitioning effect, the silhouette coefficient of the optimal partitioning scheme is calculated. The silhouette coefficient measures the compactness and separation of clusters, and its value ranges from -1 to 1. The closer it is to 1, the better the clustering effect.

[0052] In the preset range of partition numbers, such as 2 to 10, the initialization and iterative calculation process is repeated to obtain the silhouette coefficients under different numbers of partitions. The partition scheme with the largest silhouette coefficient is selected as the final intelligent partition result.

[0053] For example, for a car front bumper model, the maximum silhouette coefficient of 0.72 may be obtained when the number of partitions is 5. This shows that it is most appropriate to divide the model into 5 regions, which not only ensures the similarity within each region, but also ensures the differences between regions.

[0054] The final intelligent partitioning results are visualized, with different areas marked with different colors. This provides a basis for subsequent finite element analysis, 3D printing, etc. By adjusting the weight coefficient in the distance calculation, changing the clustering characteristics, etc., the partitioning effect can be further optimized to make it more in line with engineering application requirements.

[0055] In an optional embodiment, based on each sub-area and in combination with current environmental parameters, an intelligent agent trained by a reinforcement learning algorithm generates optimal spraying parameters for each sub-area, including: Building a main network based on a deep Q network, the main network includes an input layer, two hidden layers and an output layer, wherein the number of nodes in the input layer is the same as the dimension of the state vector, the number of nodes in the output layer is equal to the size of the action space, the hidden layer uses a ReLU activation function, and the output layer uses a linear activation function; Based on the geometric features, material properties and environmental parameters of the sub-area, the state space is determined; based on the spraying parameters, the action space is determined; based on the preset coating quality indicators, the weighted sum of the coating quality indicators in the sub-area is calculated to construct a reward function; Based on the state space, action space and reward function, a simulation environment is constructed to simulate the spraying process of sub-areas on the surface of automobile parts; Construct an experience pool to store the experience data obtained by the interaction between the intelligent agent and the simulation environment, initialize the experience pool, create a target network based on the structure of the main network, and calculate the target Q value; The ε-greedy method is used to set the exploration strategy. The ε value is initially set to 0.9 and linearly decays to 0.1 as the number of training rounds increases. The intelligent agent selects spraying parameters according to the current sub-region state and exploration strategy, interacts with the simulation environment, obtains rewards and the next sub-region state, and calculates the temporal difference error based on the current sub-region state, the selected spraying parameters, the rewards obtained, the next sub-region state, and the target Q value calculated by the target network; The current sub-region state, the selected spraying parameters, the reward obtained, the next sub-region state and the time series difference error are taken as one piece of experience data, stored in and updated in the experience pool; a sampling probability is assigned to each piece of experience data based on the time series difference error in the experience pool, batch data is selected based on the sampling probability, and the loss function is calculated based on the batch data, the main network and the target network; Minimize the loss function through the gradient descent method, update the main network parameters, assign the weighted average of the main network parameters to the corresponding parameters of the target network, update the target network parameters, add Gaussian noise to the main network parameters and the target network parameters, and continue to iterate until the preset training rounds are reached. Determine the main network that has been trained, input the current state of each sub-area into the trained main network, and input the optimal spraying parameter combination of the sub-area.

[0056] In order to achieve the optimization of surface spraying parameters of automotive parts based on deep reinforcement learning, it is first necessary to build a deep Q network as the main network. The main network consists of an input layer, two hidden layers and an output layer. The number of nodes in the input layer is the same as the dimension of the state vector. For example, it can be set to 10 nodes, corresponding to the geometric characteristics of the sub-region (such as area, curvature, etc.), material properties (such as surface roughness, hardness, etc.) and environmental parameters (such as temperature, humidity, etc.). The two hidden layers are set with 64 and 32 nodes respectively, and the ReLU activation function is used. The number of nodes in the output layer is equal to the size of the action space. For example, it can be set to 5 nodes, corresponding to spraying parameters such as spray gun distance, spray angle, spray speed, paint flow rate and atomization pressure. The output layer uses a linear activation function.

[0057] Next, the state space is determined based on the characteristics of the sub-area. For example, the area, curvature, surface roughness, hardness, ambient temperature, etc. of the sub-area can be used as state variables to form a 10-dimensional state vector. At the same time, a 5-dimensional action space is determined based on the spraying parameters, including the 5 aforementioned spraying parameters. Then, based on the preset coating quality indicators, such as coating thickness uniformity, adhesion, etc., the weighted sum of these indicators in the sub-area is calculated to construct a reward function. For example, the weight of coating thickness uniformity can be set to 0.6, the weight of adhesion can be set to 0.4, and the weighted sum can be calculated as the reward value.

[0058] Based on the state space, action space and reward function defined above, a simulation environment is constructed to simulate the spraying process of the sub-area of ​​the surface of automobile parts. The simulation environment can simulate the spraying process and calculate the corresponding reward value according to the input state and action.

[0059] Then, an experience pool is constructed to store the experience data obtained by the intelligent agent interacting with the simulation environment. The capacity of the experience pool can be set to 10,000 and randomly initialized. Based on the structure of the main network, a target network with the same structure is created to calculate the target Q value.

[0060] During the training process, the ε-greedy method is used to set the exploration strategy. The ε value is initially set to 0.9, and as the number of training rounds increases, the ε value decays linearly to 0.1. For example, this decay process can be achieved in 10,000 rounds of training.

[0061] The intelligent agent selects the spraying parameters according to the current sub-region state and the exploration strategy. Specifically, a random number between 0 and 1 is generated. If it is less than the current ε value, an action is randomly selected; otherwise, the action with the largest Q value is selected. The intelligent agent interacts with the simulation environment to obtain rewards and the next sub-region state. The temporal difference error is calculated based on the current sub-region state, the selected spraying parameters, the rewards obtained, the next sub-region state, and the target Q value calculated by the target network.

[0062] The current sub-area state, the selected spraying parameters, the reward obtained, the next sub-area state, and the time difference error in the above interaction process are taken as an experience data, stored in and updated in the experience pool. A sampling probability is assigned to each experience data based on the time difference error in the experience pool. The larger the time difference error, the higher the sampling probability. For example, the time difference error can be normalized and used as the sampling probability. Based on these sampling probabilities, a batch of data is selected from the experience pool, and the batch size can be set to 64. The loss function is calculated using this batch of data, the main network, and the target network.

[0063] Minimize the loss function by gradient descent and update the main network parameters. You can use the Adam optimizer with a learning rate of 0.001. Assign the weighted average of the updated main network parameters to the corresponding parameters of the target network to achieve a soft update of the target network. The update frequency can be set to update once every 10 rounds of training, with an update coefficient of 0.01. In order to enhance the robustness of the model, you can add Gaussian noise with a mean of 0 and a standard deviation of 0.01 to the main network parameters and the target network parameters.

[0064] Continue to iterate the above process until the preset training rounds, such as 10,000 rounds, are reached. After determining the trained main network, input the current state of each sub-area into the network to obtain the optimal spray parameter combination for the sub-area. For example, for a sub-area with an area of ​​100 square centimeters, a curvature of 0.05, a surface roughness of 3μm, a hardness of 60HRC, and an ambient temperature of 25°C, the trained network may output the optimal spray parameters as follows: spray gun distance 20cm, spray angle 75°, spray speed 30cm / s, paint flow 100ml / min, and atomization pressure 0.4MPa.

[0065] In this way, customized optimal spraying parameters can be generated for each sub-area of ​​the automotive component surface, thereby optimizing the overall spraying effect.

[0066] In an optional implementation, based on the optimal spraying parameters and the spatial relationship between the sub-areas, an ant colony path planning algorithm is used to optimize the global spraying sequence, and generating the optimal spraying path includes: A node is generated for each sub-region. The edges between nodes are determined based on the spatial relationship between sub-regions. The graph structure is determined based on the nodes and edges. Based on the spatial distance between the sub-regions, the optimal spraying parameters corresponding to each sub-region and the surface curvature change of the sub-region, the weight of the edge is calculated, wherein the spatial distance is determined based on the Euclidean distance between adjacent sub-regions, the similarity between the corresponding vectors of the optimal spraying parameters of the two sub-regions is calculated by cosine similarity, the spraying parameter similarity is determined, the curvature between adjacent sub-regions is calculated, the curvature difference is determined, and the surface curvature change is obtained; According to the preset ant population size, the ant population and the corresponding pheromone matrix are initialized, and each ant randomly selects a sub-region as the starting sub-region; Based on the weight of the edge, determine the heuristic information, based on the pheromone matrix, determine the pheromone concentration, calculate the state transition probability according to the heuristic information and the pheromone concentration, determine the probability of each ant selecting the next sub-area, select the sub-area corresponding to the maximum probability, and move the ant to the corresponding sub-area; After the ant moves, the local pheromone of the edge passed by the ant during the movement is updated, and the local pheromone is updated based on the corresponding pheromone concentration in the pheromone matrix, the preset pheromone volatility coefficient and the preset initial pheromone concentration; Repeat the steps of sub-region selection and local pheromone update until all ants complete a complete path. Determine the path length of the ants based on the spatial distance between sub-regions and the time distance of the spray gun adjustment. Determine the corresponding spray quality score of the ants based on the similarity of spraying parameters and curvature changes between adjacent sub-regions. Update the global pheromone based on the path length and spray quality score. Repeat until the preset maximum number of iterations is reached, select the spray path corresponding to the maximum spray quality score, and determine the optimal spray path.

[0067] To achieve the optimal spraying path planning based on the ant colony algorithm, the spraying area needs to be divided into multiple sub-areas. For each sub-area, a corresponding node is generated, and the connecting edges between nodes are determined based on the spatial relationship between the sub-areas, thereby constructing a graph structure. This graph structure will serve as the basis for ant movement.

[0068] Next, we need to calculate the weight of each edge in the graph. The weight of the edge is determined by three factors: spatial distance, spray parameter similarity, and surface curvature change. The spatial distance is obtained by calculating the Euclidean distance between the center points of adjacent sub-regions. The spray parameter similarity is obtained by calculating the cosine similarity between the optimal spray parameter vectors of two sub-regions. The higher the similarity, the smaller the weight. The surface curvature change is represented by calculating the difference in the average curvature of adjacent sub-regions. The greater the curvature difference, the greater the weight. After these three factors are normalized, they are weighted and summed according to a certain ratio to obtain the final edge weight.

[0069] When initializing the ant colony, you need to set the number of ants, such as 50. Each ant randomly selects a starting sub-region. At the same time, the pheromone matrix is ​​initialized. The matrix size is the square of the number of nodes, and the initial value can be set to a small constant, such as 0.1.

[0070] During the movement of ants, each ant calculates the probability of selecting the adjacent node to which it may move next based on the current node. The selection probability consists of two parts: heuristic information and pheromone concentration. Heuristic information reflects the attractiveness of the edge and can be taken as the inverse of the edge weight. The pheromone concentration is obtained from the pheromone matrix. These two parts of information are weighted according to a certain proportion (such as 50% each) to obtain the normalized selection probability. The ant randomly selects the next node to move based on this probability distribution.

[0071] Every time an ant completes a move, it needs to update the local pheromone of the edge it passes through. The update formula is: new pheromone = (1-ρ)*old pheromone + ρ*initial pheromone. ρ is the pheromone volatility coefficient, which can be 0.1. This local update is conducive to increasing the diversity of paths.

[0072] When all ants have completed a complete path, the path quality of each ant needs to be evaluated and the global pheromone needs to be updated. The path quality consists of two parts: path length and spray quality score. The path length takes into account the spatial distance between sub-areas and the spray gun adjustment time. The spray quality score is based on the similarity of spray parameters and curvature changes between adjacent sub-areas on the path. The shorter the path and the higher the spray quality score, the better the path quality. The global pheromone is updated according to the path quality. The better the path quality, the greater the increase in pheromone.

[0073] The above process is repeated until the preset maximum number of iterations is reached, such as 1000. Finally, the path with the highest spraying quality score in all iterations is selected as the optimal spraying path.

[0074] The following is a specific case to illustrate the implementation process of this method: Assume that the outer surface of a car body is divided into 20 sub-regions. First, a graph structure with 20 nodes is constructed, and there are edge connections between adjacent sub-regions. In the calculated edge weight matrix, for example, the edge weight from node 1 to node 2 is 0.8, the edge weight from node 1 to node 3 is 1.2, and so on.

[0075] Initialize 50 ants and randomly distribute them on 20 nodes. The pheromone matrix is ​​initialized to 0.1.

[0076] In the first iteration, assume that the movement path of an ant is: 1->2->5->8->12->15->18->20->17->14->11->7->4->3->6->9->10->13->16->19.

[0077] The local pheromone is continuously updated during the movement. For example, after the ant moves from node 1 to node 2, the pheromone of the edge (1, 2) is updated to: 0.9*0.1+0.1*0.1=0.09.

[0078] After one complete iteration, the total length of the ant path is calculated to be 100 (unit: cm), and the spray quality score is 85 (out of 100). Based on this, the global pheromone is updated, such as adding 0.05 pheromones to all edges on the path.

[0079] After 1000 iterations, the optimal path is: 1->4->7->10->13->16->19->20->17->14->11->8->5->2->3->6->9->12->15->18. The total length of the path is 90 cm, and the spraying quality score is 92, which is the best result among all iterations.

[0080] This ant colony algorithm-based method can effectively balance the spray path length and spray quality, and obtain the optimal spray path that takes into account both efficiency and effect, thereby improving the overall performance of the automatic spray system.

[0081] In an optional implementation, the path length is expressed as follows: ; Where L represents the total path length, m represents the number of sub-regions contained in the path, i represents the index of the sub-region in the path, r(·) represents the sub-region with the corresponding index in the path, d(·,·) represents the spatial distance between sub-regions, T represents the time coefficient required to adjust the spraying parameters, P(·) represents the optimal spraying parameter vector, and ||·|| represents the calculation of the Euclidean distance; The spraying quality score is as follows: ; Where Q represents, λ represents the weight coefficient, sim(·) represents the calculation of spraying parameter similarity, c(·) represents the calculation of average curvature, and |·| represents the absolute value of curvature difference.

[0082] In practical applications, this technical solution can be implemented in the following ways: First, the 3D model data of the workpiece to be sprayed needs to be obtained and divided into multiple sub-areas. For each sub-area, its geometric features such as surface curvature are calculated. Then, according to the process requirements, the optimal spraying parameters are set for each sub-area, including spray gun distance, spray angle, spray speed, etc.

[0083] Next, determine the spray path. Starting from the starting sub-area, traverse all sub-areas in turn and calculate the spatial distance between adjacent sub-areas. At the same time, consider the time required to adjust the spray parameters to obtain the total path length. Specifically, for the i-th and i+1-th sub-areas in the path, calculate the distance between their center points, plus the time to adjust the spray parameters multiplied by the degree of parameter difference. Add up this "distance" between all adjacent sub-areas to get the total path length.

[0084] While calculating the path length, the spraying quality also needs to be evaluated. Two factors are mainly considered: one is the similarity of the spraying parameters of adjacent sub-areas, and the other is the difference in average curvature between sub-areas. The higher the similarity of the spraying parameters and the smaller the difference in curvature, the better the spraying quality. These two indicators can be calculated separately and weighted summed according to certain weights to obtain the final spraying quality score.

[0085] For example, suppose the workpiece to be sprayed is divided into 5 sub-areas. The optimal spraying parameters of the first sub-area are (10, 30, 50), and the second sub-area is (12, 35, 55). The distance between their center points is 100mm, and it takes 2 seconds to adjust the parameters. Then the path length between the two sub-areas can be calculated as: 100 + 2 * ((12-10)^2 + (35-30)^2 +(55-50)^2)^0.5 ≈ 111.4mm. Similarly, calculate other adjacent sub-areas and add them up to get the total path length.

[0086] For the spraying quality score, assume that the average curvatures of the first and second sub-regions are 0.01 and 0.015 respectively, and the parameter similarity is 0.9. Take the weight coefficient λ=0.6, then the spraying quality score between these two sub-regions is: 0.6 * 0.9 + 0.4 *(1 - |0.01-0.015|) = 0.938. Calculate the other adjacent sub-regions in the same way, and take the average value as the overall score.

[0087] The above method can be used to comprehensively evaluate the performance of a given spray path. In practical applications, multiple candidate paths can be generated by adjusting the sub-area division, changing the traversal order, etc., and the optimal path can be selected using this evaluation system, thereby realizing intelligent spray path planning.

[0088] In an optional implementation, a pre-built fuzzy logic controller is used to adjust the spraying parameters according to the deviation between the real-time quality evaluation result and the preset quality target until the current automobile parts are sprayed, the sprayed parts are obtained, the sprayed parts are scanned and tested, and the final spraying quality data is obtained. The final spraying quality data, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set. Based on the full-cycle comprehensive data set, the gradient boosting decision tree algorithm is used to extract the spraying quality influencing variables. Based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network and the algorithm parameters of the reinforcement learning algorithm are updated, including: According to the deviation between the real-time quality assessment results and the preset quality targets, the spraying parameters are adjusted through the pre-built fuzzy logic controller until the current automotive parts spraying is completed, including: A dynamic fuzzy neural network is introduced, and the input and output variable membership function parameters of the fuzzy logic controller are adjusted according to the real-time spraying effect feedback, and the loss function is calculated, wherein the loss function includes the mean square error between the predicted quality and the actual quality, and the L2 regularization term of the difference before and after the membership function parameter adjustment; based on the loss function, the gradient is calculated by the back propagation algorithm, and the input and output variable membership function parameters of the fuzzy logic controller are updated in combination with the stochastic gradient descent method; the loss function calculation and parameter updating steps are repeated until the preset number of iterations is reached to obtain the sprayed parts; Use a high-precision three-dimensional laser scanner to perform an all-round scan on the spray-painted parts to generate a high-precision three-dimensional model; Based on the high-precision three-dimensional model, a multi-scale spraying quality assessment method is used to calculate a comprehensive quality score, the comprehensive quality score, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set, and the full-cycle comprehensive data set is stored in a time series database; Based on the full-cycle comprehensive data set, the LightGBM algorithm is used to extract the variables affecting the spraying quality, and the SHAP value is used to explain the LightGBM model and quantify the importance of each feature to obtain the ranking of key factors affecting the spraying quality; Based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network are updated, wherein the multimodal fusion neural network includes a one-dimensional convolutional neural network for processing time series spraying parameter data, a two-dimensional convolutional neural network for processing component geometric feature images, a long short-term memory network for processing environmental parameter time series, and a multi-layer perceptron sub-network for processing static features, and the outputs of each sub-network are fused through a Transformer-based attention layer; Based on the spraying quality influencing variables, the algorithm parameters of the reinforcement learning algorithm are updated, and the updated multimodal fusion neural network and the reinforcement learning algorithm are applied to the spraying process of the next batch of automobile parts to continuously adjust the spraying quality.

[0089] In practical applications, the specific implementation methods of the method of the present invention are as follows: First, based on the pre-built fuzzy logic controller, the spraying parameters are adjusted according to the deviation between the real-time quality assessment results and the preset quality targets. Specifically, a dynamic fuzzy neural network is introduced to dynamically adjust the input and output variable membership function parameters of the fuzzy logic controller according to the real-time spraying effect feedback. The loss function is calculated, including the mean square error between the predicted quality and the actual quality, and the L2 regularization term of the difference before and after the membership function parameter adjustment. For example, the weight of the mean square error between the predicted quality and the actual quality can be set to 0.7, and the weight of the L2 regularization term can be set to 0.3. Based on the loss function, the gradient is calculated by the back propagation algorithm, and the input and output variable membership function parameters of the fuzzy logic controller are updated in combination with the stochastic gradient descent method. Repeat the loss function calculation and parameter update steps until the preset number of iterations (such as 100 times) is reached to obtain the sprayed parts.

[0090] Next, a high-precision 3D laser scanner is used to perform an all-round scan of the sprayed parts to generate a high-precision 3D model. The scanning accuracy can reach 0.1mm, and the scanning frequency is 200Hz. Based on the generated high-precision 3D model, a multi-scale spraying quality evaluation method is used to calculate the comprehensive quality score. The multi-scale evaluation method includes three levels: macro appearance evaluation, meso surface morphology evaluation, and micro coating structure evaluation. The macro appearance evaluation mainly considers indicators such as color difference and gloss; the meso surface morphology evaluation mainly considers indicators such as surface roughness and waviness; the micro coating structure evaluation mainly considers indicators such as coating thickness uniformity and porosity. The evaluation results of each level are weighted according to a certain weight (such as 0.4:0.3:0.3) to obtain the final comprehensive quality score.

[0091] The comprehensive quality score, environmental parameters (such as temperature, humidity, dust concentration, etc.), component characteristics (such as material, shape complexity, etc.) and spraying parameters (such as spray gun movement speed, spraying distance, paint flow, etc.) are integrated into a full-cycle comprehensive data set and stored in a time series database. The time series database uses InfluxDB, which can efficiently store and query time series data.

[0092] Based on the stored full-cycle comprehensive data set, the variables affecting spraying quality are extracted by the LightGBM algorithm. LightGBM is an ensemble learning algorithm based on a decision tree, which has the advantages of fast training speed and low memory usage. In this embodiment, the main parameters of LightGBM are set as follows: the maximum depth of the tree is 6, the minimum amount of data for leaf nodes is 20, and the minimum gain for feature segmentation is 0.1. The SHAP (SHapley Additive exPlanations) value is used to interpret the LightGBM model and quantify the importance of each feature to obtain the ranking of key factors affecting spraying quality. For example, the ranking results that may be obtained are: spraying distance> paint flow> spray gun moving speed> ambient temperature> part surface roughness, etc.

[0093] According to the extracted variables affecting the spraying quality, the network parameters of the multimodal fusion neural network are updated. The multimodal fusion neural network includes multiple subnetworks: a one-dimensional convolutional neural network (1D-CNN) for processing time-series spraying parameter data, a two-dimensional convolutional neural network (2D-CNN) for processing component geometric feature images, a long short-term memory network (LSTM) for processing environmental parameter time series, and a multi-layer perceptron (MLP) subnetwork for processing static features. 1D-CNN adopts a 3-layer convolution structure with convolution kernel sizes of 3, 5, and 7, respectively, and each layer is followed by a maximum pooling layer; 2D-CNN adopts the ResNet-18 architecture; LSTM adopts a 2-layer structure with 128 hidden units in each layer; MLP adopts a 3-layer structure with 64, 32, and 16 neurons in each layer. The outputs of each subnetwork are fused through the Transformer-based attention layer. Transformer adopts a 4-head attention mechanism with a hidden layer dimension of 256.

[0094] At the same time, based on the extracted variables affecting spraying quality, the algorithm parameters of the reinforcement learning algorithm are updated. The reinforcement learning algorithm uses a deep Q network (DQN), including 3 fully connected layers, with 128, 64, and 32 neurons in each layer. The parameters of the epsilon-greedy strategy are updated, with the initial epsilon value set to 0.9, decaying by 0.995 per round, and the minimum value being 0.01. The experience replay buffer size is updated to 10,000, and the batch size of each sampling is 64.

[0095] Finally, the updated multimodal fusion neural network and reinforcement learning algorithm are applied to the spraying process of the next batch of automotive parts to continuously adjust the spraying quality. Specifically, the multimodal fusion neural network is used to predict the spraying quality in the current state, and the reinforcement learning algorithm selects the optimal spraying parameter adjustment strategy based on the prediction results. Through continuous iterative optimization, continuous improvement of spraying quality is achieved.

[0096] Through the above technical solution, the method of the present invention can effectively improve the spraying quality of automobile parts, reduce the defective product rate, and improve production efficiency. The method has strong adaptability and robustness and can cope with different types of parts and changing production environments.

[0097] Figure 2 FIG. 1 is a schematic diagram of the structure of a spray control system for automobile parts according to an embodiment of the present invention. Figure 2 As shown, the system comprises: The first unit is used to collect spraying environment parameters in real time through temperature and humidity sensors, air pressure sensors and air quality sensors; use multispectral cameras and three-dimensional structured light scanners to obtain surface spectral information and three-dimensional structural data of automobile parts; input spraying environment parameters, surface spectral information and three-dimensional structural data into a pre-trained multimodal fusion neural network to generate a comprehensive digital model containing environmental parameters and component characteristics, and based on the comprehensive digital model, use a clustering partitioning algorithm to intelligently partition the surface of automobile parts into multiple sub-areas; The second unit is used to generate the optimal spraying parameters for each sub-area based on each sub-area and in combination with the current environmental parameters, using an intelligent agent trained by a reinforcement learning algorithm, and based on the optimal spraying parameters and the spatial relationship between the sub-areas, using an ant colony path planning algorithm to optimize the global spraying sequence and generate the optimal spraying path; the optimal spraying parameters and the optimal spraying path are sent to a spraying robot with an adaptive control function, and during the spraying process, the spraying state data is collected in real time through a high-speed camera and a coating thickness sensor, and the spraying state data is input into a pre-trained convolutional neural network to generate a real-time quality evaluation result of the spraying; The third unit is used to use a pre-built fuzzy logic controller to adjust the spraying parameters according to the deviation between the real-time quality assessment result and the preset quality target until the spraying of the current automobile parts is completed, the sprayed parts are obtained, the sprayed parts are scanned and inspected, and the final spraying quality data is obtained. The final spraying quality data, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set, based on the full-cycle comprehensive data set, the gradient boosting decision tree algorithm is used to extract the spraying quality influencing variables, and based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network and the algorithm parameters of the reinforcement learning algorithm are updated.

[0098] According to a third aspect of the embodiments of the present invention, An electronic device is provided, comprising: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0099] A fourth aspect of the embodiments of the present invention is: A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0100] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling the spraying of automobile parts, characterized in that: include: The spraying environment parameters are collected in real time through temperature and humidity sensors, air pressure sensors and air quality sensors; the surface spectrum information and three-dimensional structure data of automobile parts are obtained by using multi-spectral cameras and three-dimensional structured light scanners; the spraying environment parameters, surface spectrum information and three-dimensional structure data are input into the pre-trained multimodal fusion neural network to generate a comprehensive digital model containing environmental parameters and parts characteristics. Based on the comprehensive digital model, the surface of automobile parts is intelligently partitioned into multiple sub-areas using a clustering partitioning algorithm; Based on each sub-area and combined with the current environmental parameters, an intelligent agent trained by a reinforcement learning algorithm generates optimal spraying parameters for each sub-area. Based on the optimal spraying parameters and the spatial relationship between the sub-areas, an ant colony path planning algorithm is used to optimize the global spraying sequence and generate the optimal spraying path. The optimal spraying parameters and the optimal spraying path are sent to a spraying robot with an adaptive control function. During the spraying process, the spraying status data is collected in real time through a high-speed camera and a coating thickness sensor. The spraying status data is input into a pre-trained convolutional neural network to generate real-time quality evaluation results for the spraying. A pre-built fuzzy logic controller is used to adjust the spraying parameters according to the deviation between the real-time quality assessment result and the preset quality target until the spraying of the current automobile parts is completed, and the sprayed parts are obtained. The sprayed parts are scanned and inspected to obtain the final spraying quality data. The final spraying quality data, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set. Based on the full-cycle comprehensive data set, the gradient boosting decision tree algorithm is used to extract the spraying quality influencing variables. Based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network and the algorithm parameters of the reinforcement learning algorithm are updated.

2. The method according to claim 1, characterized in that The spraying environment parameters, surface spectrum information and three-dimensional structure data are input into the pre-trained multimodal fusion neural network to generate a comprehensive digital model containing environmental parameters and component characteristics. Based on the comprehensive digital model, the surface of the automotive parts is intelligently partitioned using a clustering partitioning algorithm and divided into multiple sub-areas, including: Collect spraying environment parameters, including temperature data, humidity data, air pressure data, and air particle concentration data; collect hyperspectral information on the surface of automobile parts, including spectral reflectance data of multiple bands; collect three-dimensional structure data of automobile parts, including point cloud data; Constructing a multimodal fusion neural network, wherein the multimodal fusion neural network includes an environmental parameter branch, a spectral information branch, and a three-dimensional structure branch; The spraying environment parameters are input into the environment parameter branch, and the time series environment data is processed using a one-dimensional convolutional neural network to obtain an environment feature vector; the hyperspectral information is input into the spectral information branch, and the spatial-spectral features are extracted using a three-dimensional convolutional neural network to obtain a spectral feature vector; the three-dimensional structure data is input into the three-dimensional structure branch, and the point cloud data is processed using a point cloud convolutional network to obtain a structure feature vector; The environmental feature vector, the spectral feature vector and the structural feature vector are fused by using a multi-head self-attention module to obtain a fused feature; the fused feature is input into a multi-layer perceptron, and a feature vector of a fixed dimension is output as a comprehensive digital model; Based on the comprehensive digital model, a clustering partitioning algorithm is used to intelligently partition the surface of the automobile parts to determine a plurality of sub-areas.

3. The method according to claim 2, characterized in that The cluster partitioning algorithm includes: Constructing a cluster partitioning algorithm based on K-means clustering; initializing cluster centers and using feature vectors in the comprehensive digital model as initial data points; Calculate the distance from each data point to each cluster center, introduce spatial constraints when calculating the distance, add spatial coordinate terms to the distance calculation formula, and assign adjustable weight coefficients; assign each data point to the nearest cluster center; calculate the average value of all data points in each cluster, determine the new cluster center, and update it; calculate the local curvature of each data point, and use the local curvature as the cluster feature dimension; Repeat the iteration until the position change of the cluster center is less than the preset cluster center change threshold, and the optimal partitioning scheme is obtained; Calculating the silhouette coefficient of the optimal partitioning scheme by a similarity measurement method; Within the preset range of partition numbers, the initialization and iterative calculation process is repeated to obtain the silhouette coefficients under different numbers of partitions, and the partition scheme with the largest silhouette coefficient is selected as the final intelligent partitioning result.

4. The method according to claim 1, characterized in that: Based on each sub-area and combined with the current environmental parameters, the intelligent agent trained by the reinforcement learning algorithm generates the optimal spraying parameters for each sub-area, including: Building a main network based on a deep Q network, the main network includes an input layer, two hidden layers and an output layer, wherein the number of nodes in the input layer is the same as the dimension of the state vector, the number of nodes in the output layer is equal to the size of the action space, the hidden layer uses a ReLU activation function, and the output layer uses a linear activation function; Based on the geometric features, material properties and environmental parameters of the sub-area, the state space is determined; based on the spraying parameters, the action space is determined; based on the preset coating quality indicators, the weighted sum of the coating quality indicators in the sub-area is calculated to construct a reward function; Based on the state space, action space and reward function, a simulation environment is constructed to simulate the spraying process of sub-areas on the surface of automobile parts; Construct an experience pool to store the experience data obtained by the interaction between the intelligent agent and the simulation environment, initialize the experience pool, create a target network based on the structure of the main network, and calculate the target Q value; The ε-greedy method is used to set the exploration strategy. The ε value is initially set to 0.9 and linearly decays to 0.1 as the number of training rounds increases. The intelligent agent selects spraying parameters according to the current sub-region state and exploration strategy, interacts with the simulation environment, obtains rewards and the next sub-region state, and calculates the temporal difference error based on the current sub-region state, the selected spraying parameters, the rewards obtained, the next sub-region state, and the target Q value calculated by the target network; The current sub-region state, the selected spraying parameters, the reward obtained, the next sub-region state and the time series difference error are taken as one piece of experience data, stored in and updated in the experience pool; a sampling probability is assigned to each piece of experience data based on the time series difference error in the experience pool, batch data is selected based on the sampling probability, and the loss function is calculated based on the batch data, the main network and the target network; Minimize the loss function through the gradient descent method, update the main network parameters, assign the weighted average of the main network parameters to the corresponding parameters of the target network, update the target network parameters, add Gaussian noise to the main network parameters and the target network parameters, and continue to iterate until the preset training rounds are reached. Determine the main network that has been trained, input the current state of each sub-area into the trained main network, and input the optimal spraying parameter combination of the sub-area.

5. The method according to claim 4, characterized in that Based on the optimal spraying parameters and the spatial relationship between sub-areas, the ant colony path planning algorithm is used to optimize the global spraying sequence and generate the optimal spraying path including: A node is generated for each sub-region. The edges between nodes are determined based on the spatial relationship between sub-regions. The graph structure is determined based on the nodes and edges. Based on the spatial distance between the sub-regions, the optimal spraying parameters corresponding to each sub-region and the surface curvature change of the sub-region, the weight of the edge is calculated, wherein the spatial distance is determined based on the Euclidean distance between adjacent sub-regions, the similarity between the corresponding vectors of the optimal spraying parameters of the two sub-regions is calculated by cosine similarity, the spraying parameter similarity is determined, the curvature between adjacent sub-regions is calculated, the curvature difference is determined, and the surface curvature change is obtained; According to the preset ant population size, the ant population and the corresponding pheromone matrix are initialized, and each ant randomly selects a sub-region as the starting sub-region; Based on the weight of the edge, determine the heuristic information, based on the pheromone matrix, determine the pheromone concentration, calculate the state transition probability according to the heuristic information and the pheromone concentration, determine the probability of each ant selecting the next sub-area, select the sub-area corresponding to the maximum probability, and move the ant to the corresponding sub-area; After the ant moves, the local pheromone of the edge passed by the ant during the movement is updated, and the local pheromone is updated based on the corresponding pheromone concentration in the pheromone matrix, the preset pheromone volatility coefficient and the preset initial pheromone concentration; Repeat the steps of sub-region selection and local pheromone update until all ants complete a complete path. Determine the path length of the ants based on the spatial distance between sub-regions and the time distance of the spray gun adjustment. Determine the corresponding spray quality score of the ants based on the similarity of spraying parameters and curvature changes between adjacent sub-regions. Update the global pheromone based on the path length and spray quality score. Repeat until the preset maximum number of iterations is reached, select the spray path corresponding to the maximum spray quality score, and determine the optimal spray path.

6. The method according to claim 5, characterized in that The path length is given by the following formula: ； Where L represents the total path length, m represents the number of sub-regions contained in the path, i represents the index of the sub-region in the path, r(·) represents the sub-region with the corresponding index in the path, d(·,·) represents the spatial distance between sub-regions, T represents the time coefficient required to adjust the spraying parameters, P(·) represents the optimal spraying parameter vector, and ||·|| represents the calculation of the Euclidean distance; The spraying quality score is as follows: ； Where Q represents, λ represents the weight coefficient, sim(·) represents the calculation of spraying parameter similarity, c(·) represents the calculation of average curvature, and |·| represents the absolute value of curvature difference.

7. The method according to claim 1, characterized in that A pre-built fuzzy logic controller is used to adjust the spraying parameters according to the deviation between the real-time quality evaluation result and the preset quality target until the spraying of the current automobile parts is completed, and the sprayed parts are obtained. The sprayed parts are scanned and tested to obtain the final spraying quality data. The final spraying quality data, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set. Based on the full-cycle comprehensive data set, the gradient boosting decision tree algorithm is used to extract the spraying quality influencing variables. Based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network and the algorithm parameters of the reinforcement learning algorithm are updated, including: According to the deviation between the real-time quality assessment results and the preset quality targets, the spraying parameters are adjusted through the pre-built fuzzy logic controller until the current automotive parts spraying is completed, including: A dynamic fuzzy neural network is introduced, and the input and output variable membership function parameters of the fuzzy logic controller are adjusted according to the real-time spraying effect feedback, and the loss function is calculated, wherein the loss function includes the mean square error between the predicted quality and the actual quality, and the L2 regularization term of the difference before and after the membership function parameter adjustment; based on the loss function, the gradient is calculated by the back propagation algorithm, and the input and output variable membership function parameters of the fuzzy logic controller are updated in combination with the stochastic gradient descent method; the loss function calculation and parameter updating steps are repeated until the preset number of iterations is reached to obtain the sprayed parts; Use a high-precision three-dimensional laser scanner to perform an all-round scan on the spray-painted parts to generate a high-precision three-dimensional model; Based on the high-precision three-dimensional model, a multi-scale spraying quality assessment method is used to calculate a comprehensive quality score, the comprehensive quality score, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set, and the full-cycle comprehensive data set is stored in a time series database; Based on the full-cycle comprehensive data set, the LightGBM algorithm is used to extract the variables affecting the spraying quality, and the SHAP value is used to explain the LightGBM model and quantify the importance of each feature to obtain the ranking of key factors affecting the spraying quality; Based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network are updated, wherein the multimodal fusion neural network includes a one-dimensional convolutional neural network for processing time series spraying parameter data, a two-dimensional convolutional neural network for processing component geometric feature images, a long short-term memory network for processing environmental parameter time series, and a multi-layer perceptron sub-network for processing static features, and the outputs of each sub-network are fused through a Transformer-based attention layer; Based on the spraying quality influencing variables, the algorithm parameters of the reinforcement learning algorithm are updated, and the updated multimodal fusion neural network and the reinforcement learning algorithm are applied to the spraying process of the next batch of automobile parts to continuously adjust the spraying quality.

8. An automobile parts spraying control system, used to implement the method according to any one of claims 1 to 7, characterized in that: include: The first unit is used to collect spraying environment parameters in real time through temperature and humidity sensors, air pressure sensors and air quality sensors; use multispectral cameras and three-dimensional structured light scanners to obtain surface spectral information and three-dimensional structural data of automobile parts; input spraying environment parameters, surface spectral information and three-dimensional structural data into a pre-trained multimodal fusion neural network to generate a comprehensive digital model containing environmental parameters and component characteristics, and based on the comprehensive digital model, use a clustering partitioning algorithm to intelligently partition the surface of automobile parts into multiple sub-areas; The second unit is used to generate the optimal spraying parameters for each sub-area based on each sub-area and in combination with the current environmental parameters, using an intelligent agent trained by a reinforcement learning algorithm, and based on the optimal spraying parameters and the spatial relationship between the sub-areas, using an ant colony path planning algorithm to optimize the global spraying sequence and generate the optimal spraying path; the optimal spraying parameters and the optimal spraying path are sent to a spraying robot with an adaptive control function, and during the spraying process, the spraying state data is collected in real time through a high-speed camera and a coating thickness sensor, and the spraying state data is input into a pre-trained convolutional neural network to generate a real-time quality evaluation result of the spraying; The third unit is used to use a pre-built fuzzy logic controller to adjust the spraying parameters according to the deviation between the real-time quality assessment result and the preset quality target until the spraying of the current automobile parts is completed, the sprayed parts are obtained, the sprayed parts are scanned and inspected, and the final spraying quality data is obtained. The final spraying quality data, environmental parameters, component characteristics and spraying parameters are integrated into a full-cycle comprehensive data set, based on the full-cycle comprehensive data set, the gradient boosting decision tree algorithm is used to extract the spraying quality influencing variables, and based on the spraying quality influencing variables, the network parameters of the multimodal fusion neural network and the algorithm parameters of the reinforcement learning algorithm are updated.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images