A spraying control method, device, electronic device and storage medium

Abstract

The present application provides a spray control method, device, electronic device and storage medium, which are applied to the field of die-casting spray technology. Through real-time and accurate monitoring of the characterization parameters of the die-casting, a dynamic mapping relationship between the characterization parameters and the spraying parameters is established, and adaptive adjustment is made according to the actual spraying effect on the surface of the die-casting to ensure the consistency and reliability of the coating quality. It also has an effective feedback mechanism and self-learning ability, and can continuously optimize the control strategy according to the deviation of the actual spraying effect to cope with the long-term evolution of the surface characteristics of the die-casting.

Description

Technical Field

[0001] The present application relates to the technical field of die casting spraying, and in particular to a spraying control method, device, electronic equipment and storage medium. Background Art

[0002] In existing die casting production lines, the die casting spraying process faces a series of technical challenges. Since die castings need to be recycled and repeatedly sprayed, their surfaces will produce uneven wear, which gradually deepens with the increase in the number of uses. In the same batch of die castings, the degree and form of wear at different locations vary, especially in areas with high frequency of use, the surface roughness will change dynamically. This change in surface characteristics directly affects the adhesion of the coating and the uniformity of the coating thickness.

[0003] Existing spray systems usually operate with a multi-nozzle system with fixed parameters. Although this method can meet basic spraying needs, it cannot effectively cope with the dynamic changes of the die-casting surface. Some systems use manual detection or simple automatic detection to adjust the spraying parameters, but this method is difficult to achieve accurate real-time control, resulting in unstable coating quality. In addition, after multiple spray-use cycles, the residual paint from the previous spraying will gradually peel off during the wear process, changing the local surface characteristics and further affecting the next spraying effect.

[0004] These problems lead to instability in coating quality, which is manifested in uneven coating thickness, insufficient adhesion, and local coating defects. This not only affects the appearance quality of the product, but may also reduce the protective function of the coating, thereby affecting the service life and performance of the parts. At the same time, the inability to accurately control the spraying parameters also causes waste of paint and environmental pollution.

[0005] It is difficult for the existing technology to achieve real-time and accurate monitoring of the surface characteristics of die castings, and it is impossible to establish a dynamic mapping relationship between surface characteristics and optimal spraying parameters. This results in the spraying system being unable to adaptively adjust according to the actual conditions of the die casting surface, thereby failing to ensure the consistency and reliability of the coating quality. In addition, the existing technology lacks an effective feedback mechanism and self-learning capability. In the actual production process, even if an automatic detection and parameter adjustment system is used, it is difficult to continuously optimize the control strategy based on the deviation of the spraying effect. This makes it difficult for the spraying system to cope with changes in the production environment and the long-term evolution of the surface characteristics of die castings.

[0006] In view of the above problems, the existing technology needs to be improved urgently. Summary of the invention

[0007] In view of the above-mentioned shortcomings of the prior art, a spray control method, device, electronic device and storage medium provided in the present application are applied to the field of die-casting spray technology. It can realize real-time and accurate monitoring of the characterization parameters of die-castings, establish a dynamic mapping relationship between the characterization parameters and the spraying parameters, and adaptively adjust according to the actual spraying effect on the surface of the die-casting to ensure the consistency and reliability of the coating quality. It also has an effective feedback mechanism and self-learning ability, and can continuously optimize the control strategy according to the deviation of the actual spraying effect to cope with the long-term evolution of the surface characteristics of the die-castings.

[0008] In a first aspect, the present application provides a spraying control method, the method comprising the steps of:

[0009] S1: Acquire current parameter data of the surface of the die casting, wherein the current parameter data at least includes a current roughness parameter, a current wear degree parameter and a current coating residue parameter;

[0010] S2: establishing a mapping relationship between surface features and spraying parameters according to the current roughness parameter, the current wear degree parameter and the current coating residue parameter according to a preset weight coefficient;

[0011] S3: calculating the spraying parameters of the surface of the die casting according to the mapping relationship;

[0012] S4: Control the pressure, flow rate and angle of the nozzle according to the spraying parameters, obtain the deviation between the actual spraying effect and the expected spraying effect, and update the correction coefficient of the mapping relationship according to the deviation, so as to update the mapping relationship, calculate the updated spraying parameters of the die-casting surface according to the updated mapping relationship, and spray according to the updated spraying parameters of the die-casting surface.

[0013] The present application provides a spray control method, which obtains the current parameter data of the die casting surface and establishes a mapping relationship between the surface features and the spray parameters according to the current parameter data. The optimal spray parameter control nozzle pressure, flow rate and angle can be calculated from the mapping relationship to achieve precise spray control. The actual spraying effect is obtained by visual inspection or other means and compared with the expected effect. If there is a deviation, the system uses the deviation to update the correction coefficient, thereby updating the mapping relationship. This dynamic correction mechanism ensures that the system can continuously adapt to the long-term change trend of the die casting surface and maintain the stability of the spraying effect. This control method based on the mapping relationship can cope with the complex nonlinear relationship between the surface features of the die casting and the optimal spray parameters. By establishing a dynamic mapping relationship, the system can more flexibly cope with various surface changes and improve the quality and efficiency of spraying.

[0014] Further, step S2 includes:

[0015] S21: Acquire historical parameter data of the surface of the die casting, wherein the historical parameter data at least includes a historical roughness parameter, a historical wear degree parameter, and a historical coating residue parameter;

[0016] S22: comparing the historical parameter data with the current parameter data, and establishing a time series model for representing parameter change trends of die castings according to the comparison result;

[0017] S23: dynamically adjusting the preset weight coefficient according to the time series model;

[0018] S24: According to the current roughness parameter, the current wear degree parameter and the current coating residue parameter, a mapping relationship between surface features and spraying parameters is established according to the dynamically adjusted preset weight coefficient.

[0019] The present application provides a spraying control method, which accurately captures and predicts the changing trend of the characterization parameters by introducing historical parameter data analysis to establish a time series model. By dynamically adjusting the preset weight coefficient, the mapping relationship can better adapt to the dynamic changes of the characterization parameters of the die casting, thereby improving the calculation accuracy and control effect of the spraying parameters.

[0020] Further, step S22 includes:

[0021] S221: Acquire a sampling time interval of the die casting characterization parameter, wherein the die casting characterization parameter includes historical parameter data and the current parameter data;

[0022] S222: Arranging the comparison results in chronological order according to the sampling time interval;

[0023] S223: Calculating the parameter change rate of the die casting characterization parameters collected at adjacent sampling time points;

[0024] S224: establishing a differential equation according to the parameter change rate, substituting the differential equation into a preset time series basis function, and obtaining a time series model for characterizing the parameter change trend of the die casting by least squares fitting calculation.

[0025] The present application provides a spraying control method, which obtains the sampling time interval, sorts the comparison results of historical parameter data and current parameter data in chronological order, and calculates the parameter change rate of the die casting characterization parameters at adjacent sampling time points, and establishes a time series model based on the parameter change rate. The time series model has strong adaptability and prediction ability. It can accurately capture the change law of the surface characteristic parameters of the die casting and provide reliable data support for the subsequent optimization of spraying parameters.

[0026] Furthermore, the step of updating the correction coefficient of the mapping relationship according to the deviation in step S4 includes:

[0027] S41: Obtaining deviations between the actual spraying effect and the expected spraying effect at multiple preset time points;

[0028] S42: Calculate the deviation change trend of the plurality of preset time points, and when the deviation change trend exceeds a preset range, use an adaptive learning strategy and the deviation at the current moment to update the correction coefficient of the mapping relationship.

[0029] The present application provides a spraying control method, which realizes intelligent control of the spraying process by introducing mechanisms such as multi-time point deviation analysis, change trend calculation and adaptive learning strategy. This method can not only quickly respond to instantaneous changes in surface features, but also capture long-term change trends, so as to make more accurate adjustments. Especially when dealing with recycled die castings, this solution can effectively cope with the cumulative changes in surface features and ensure the quality stability of each spraying.

[0030] Further, step S42 includes:

[0031] S421: Obtaining the fluctuation frequency of the deviation change trend;

[0032] S422: when the fluctuation frequency is greater than a first preset threshold, reducing the learning rate of the adaptive learning strategy according to a first preset ratio; when the fluctuation frequency is less than a second preset threshold, increasing the learning rate of the adaptive learning strategy according to the first preset ratio; the first preset threshold is greater than the second preset threshold;

[0033] S423: updating the correction coefficient of the mapping relationship according to the updated adaptive learning strategy and the deviation at the current moment.

[0034] Further, step S423 includes:

[0035] S4231: Obtaining the change rate of the deviation change trend;

[0036] S4232: Calculating a stability index of the deviation according to the change rate;

[0037] S4233: When the stability index is greater than a third preset value, the updated adaptive learning strategy is multiplied by the stability index to obtain a revised adaptive learning strategy, and the correction coefficient of the mapping relationship is updated according to the revised adaptive learning strategy and the deviation at the current moment.

[0038] Further, the mapping relationship adopts the following mathematical model: P(x,y) = K[α·R(x,y) + β·W(x,y) + γ·C(x,y)], wherein P(x,y) is the spraying parameter at the position (x,y), R(x,y) is the current roughness parameter at the position (x,y), W(x,y) is the current wear degree parameter at the position (x,y), C(x,y) is the current coating residual parameter at the position (x,y), K is the correction coefficient, α, β, γ are preset weight coefficients; the correction coefficient is updated using the following mathematical model: , where K(t) is the correction coefficient at time t, η is the learning rate of the adaptive learning strategy, and E(t) is the deviation between the actual effect and the expected effect. Represents the gradient of the deviation E(t) to the correction coefficient K(t).

[0039] In a second aspect, a spray control device is provided, the device comprising:

[0040] Acquisition module: acquiring current parameter data of the die casting surface, wherein the current parameter data at least includes current roughness parameter, current wear degree parameter and current coating residue parameter;

[0041] Mapping module: establishing a mapping relationship between surface features and spraying parameters according to the current roughness parameter, the current wear degree parameter and the current coating residue parameter according to a preset weight coefficient;

[0042] Calculation module: calculates the spraying parameters of the die casting surface according to the mapping relationship;

[0043] Correction module: Control the pressure, flow rate and angle of the nozzle according to the spraying parameters, obtain the deviation between the actual spraying effect and the expected spraying effect, and update the correction coefficient of the mapping relationship according to the deviation, so as to update the mapping relationship, calculate the updated spraying parameters of the die-casting surface according to the updated mapping relationship, and spray according to the updated spraying parameters of the die-casting surface.

[0044] In a third aspect, the present application provides an electronic device, including a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in any of the above methods are executed.

[0045] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in any of the above methods are executed.

[0046] Beneficial effects: The present application provides a spray control method, device, electronic device and storage medium, which obtains the current parameter data of the die-casting surface and establishes a mapping relationship between the surface features and the spray parameters based on the current parameter data. The optimal spray parameter control nozzle pressure, flow rate and angle can be calculated from the mapping relationship to achieve precise spray control. The actual spraying effect is obtained through visual inspection or other means and compared with the expected effect. If there is a deviation, the system will use the deviation to update the correction coefficient, thereby updating the mapping relationship. This dynamic correction mechanism ensures that the system can continuously adapt to the long-term change trend of the die-casting surface and maintain the stability of the spraying effect. This control method based on the mapping relationship can cope with the complex nonlinear relationship between the surface features of the die-casting and the optimal spray parameters. By establishing a dynamic mapping relationship, the system can more flexibly respond to various surface changes and improve the quality and efficiency of spraying. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flow chart of a spraying control method proposed in this application.

[0048] Figure 2 This is a structural diagram of a spray control device proposed in this application.

[0049] Figure 3 A schematic diagram of the structure of the electronic device provided in this application.

[0050] Description of reference numerals: 201, acquisition module; 202, mapping module; 203, calculation module; 204, correction module; 301, processor; 302, memory; 303, communication bus; 3, electronic device. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application usually described and marked in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0052] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first, second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0053] The following disclosure provides many different implementations or examples for achieving the purpose of the present invention, which solves the problem that the existing spraying equipment may not spray enough in the areas of severe wear of the die casting, resulting in insufficient protection; while in the areas with less wear, it may over-spray, resulting in material waste and subsequent processing difficulties. In addition, due to the lack of real-time surface property mapping capabilities, the spraying equipment cannot identify and respond to the impact of the previous coating residue, which may lead to poor interface bonding between the new and old coatings, increasing the risk of coating peeling. In order to solve this problem, the present application proposes a spraying control method, device, electronic device and storage medium.

[0054] Please refer to Figure 1 In a first aspect, the present application provides a spraying control method, the method comprising the steps of:

[0055] S1: obtaining current parameter data of the die casting surface, the current parameter data at least including current roughness parameter, current wear degree parameter and current coating residue parameter;

[0056] S2: According to the current roughness parameter, the current wear degree parameter and the current coating residue parameter, a mapping relationship between the surface characteristics and the spraying parameters is established according to the preset weight coefficient;

[0057] S3: Calculate the spraying parameters on the surface of the die casting according to the mapping relationship;

[0058] S4: Control the pressure, flow rate and angle of the nozzle according to the spraying parameters, obtain the deviation between the actual spraying effect and the expected spraying effect, and update the correction coefficient of the mapping relationship according to the deviation, so as to update the mapping relationship, calculate the updated spraying parameters of the die-casting surface according to the updated mapping relationship, and spray according to the updated spraying parameters of the die-casting surface.

[0059] Among them, the existing spraying systems usually adopt a multi-nozzle configuration with fixed parameters, which cannot effectively cope with the dynamic changes of the die-casting surface. Although some systems introduce manual detection or simple automatic detection mechanisms to adjust the spraying parameters, these methods are difficult to achieve precise real-time control. Among them, the dynamic change speed of the surface characteristics of die-castings is much faster than the response speed of traditional detection and adjustment mechanisms, resulting in significant time lag and accuracy deviation between the spraying parameters and actual needs. Specifically, when the roughness of the die-casting surface suddenly increases, the fixed parameter spraying system cannot increase the paint flow or adjust the nozzle angle in time, resulting in insufficient coating thickness or uneven coverage. At the same time, areas with increased local wear may require higher spraying pressure, and the system cannot adjust the parameters of a single nozzle in a targeted manner. In addition, the irregular peeling of the residual paint from the previous spraying during the wear process will cause uneven adhesion between the new coating and the substrate, further exacerbating the instability of the coating quality.

[0060] For example, in automotive parts, die-castings in the area around the combustion chamber of the cylinder may experience rapid localized wear, while other areas may wear more slowly. Traditional fixed parameter spray systems cannot guarantee the consistency of coating quality when faced with such complex surface changes. Spraying equipment may under-spray in areas with severe wear, resulting in inadequate protection, while over-spraying in areas with less wear may cause material waste and subsequent processing difficulties. In addition, due to the lack of real-time surface property mapping capabilities, the system cannot identify and respond to the impact of previous coating residues, which may lead to poor interface bonding between the new and old coatings, increasing the risk of coating peeling.

[0061] Therefore, a spraying control method proposed in the present application is a spraying control method based on die-casting surface feature mapping, wherein the roughness parameter refers to a numerical value characterizing the microscopic unevenness of the die-casting surface, which can be specifically represented by the arithmetic mean roughness Ra or the maximum height roughness Rz. The wear degree parameter refers to the degree of material loss on the die-casting surface due to use, which can be specifically quantified by the wear depth or volume loss. The coating residual parameter refers to the amount of coating retained on the die-casting surface during the previous spraying, which can be specifically characterized by the residual coating thickness or the coverage area ratio. It can be understood that the current roughness parameter, the current wear degree parameter, and the current coating residual parameter all indicate the roughness parameter, the wear degree parameter, and the coating residual parameter detected by the sensor at the current moment.

[0062] The mapping relationship refers to a mathematical model that converts the characterization parameters into spraying parameters, which can be established by using methods such as multivariate linear regression or neural network. Spraying parameters refer to key variables that control the spraying process, which can include nozzle pressure, flow rate and angle.

[0063] The present application captures the changes of the surface parameter data of the die casting in real time by establishing a mapping relationship between the characterization surface features and the spraying parameters, and converts these changes into accurate spraying control instructions. By introducing weight coefficients and correction coefficients, the system can continuously optimize the mapping relationship according to the actual spraying effect, and realizes the adaptive control of the spraying process. Specifically, in some embodiments, the present application can obtain the current roughness parameter, the current wear degree parameter and the current coating residual parameter of the die casting surface through a special sensor system, for example, a laser scattering sensor can be used to measure the current roughness parameter of the die casting surface, an eddy current probe can be used to detect the current wear degree parameter of the die casting surface, and a spectrum analyzer can be equipped to identify the current coating residual parameter of the die casting surface. These parameters are transmitted to the central processing unit in real time through a data acquisition module. The central processing unit inputs these current parameter data into the mapping relationship model according to a pre-set weight coefficient. The mapping relationship model may be a multi-layer neural network, which can convert the current parameter data of the input die casting surface into corresponding spraying parameters.

[0064] The mapping relationship is established based on a large amount of historical data and is trained by machine learning algorithms. In actual applications, the system calculates the optimal spraying parameters from the mapping relationship according to the current specific situation of the die casting. These parameters are then sent to the spray control module, which is responsible for adjusting the pressure, flow rate and angle of each nozzle to achieve precise spray control. After the spraying process is completed, the system obtains the actual spraying effect through visual inspection or other means and compares it with the expected effect. If there is a deviation, the system uses the deviation to update the correction coefficient, thereby updating the mapping relationship. This dynamic correction mechanism ensures that the system can continuously adapt to the long-term change trend of the die casting surface and maintain the stability of the spraying effect. This method updates the mapping relationship by the deviation between the actual spraying effect and the expected effect, thereby updating the mapping relationship formula, so that the updated spraying parameters calculated by the updated mapping relationship formula are more accurate. It solves the problem that the spraying system with fixed spraying parameters in the traditional technology cannot increase the paint flow rate or adjust the nozzle angle in time, resulting in insufficient coating thickness or uneven coverage, so that the spraying effect of the die casting surface is better.

[0065] Among them, in some preferred embodiments, after data collection is completed, the central control system will input the current roughness parameters (such as Ra value), current wear degree parameters (such as maximum wear depth) and current coating residue parameters (such as residual area ratio) obtained into the pre-trained neural network model. The neural network model adopts a three-layer structure, the input layer contains the above three parameters, the hidden layer has 20 neurons, and the output layer corresponds to the three parameters of nozzle pressure, flow rate and angle. The preset weight coefficient of the neural network model is obtained through training with a large amount of historical data, and the initial value can be set as: current roughness parameter weight 0.4, current wear degree parameter weight 0.4, current coating residue parameter weight 0.2. The spraying parameters output by the neural network model will be sent to the spraying control system. The system includes multiple independently controllable nozzles, each of which is equipped with a precise pressure regulating valve, flow controller and angle adjustment mechanism. For example, for the die casting surface located in the peripheral area of ​​the combustion chamber, the system may set the nozzle pressure to 0.5 MPa, the flow rate to 2 L / min, and the spray angle to 60° to cope with the more severe wear that may exist in this area. After spraying is completed, a high-speed camera system will perform real-time inspection of the spraying effect and measure the coating thickness and uniformity. If the coating thickness deviation in certain areas is detected to exceed ±5%, the system uses the deviation to update the correction factor. For example, if the coating is found to be generally thin, the correction factor may be updated to 1.05 to appropriately increase the spraying amount during the next spraying. This implementation method can effectively respond to various changes in the surface of die-casting parts through real-time surface feature mapping and dynamic parameter adjustment, ensuring that each spraying can achieve the best effect, thereby improving product quality and production efficiency.

[0066] Further, step S2 includes:

[0067] S21: Acquire historical parameter data of the surface of the die casting, where the historical parameter data at least includes a historical roughness parameter, a historical wear degree parameter, and a historical coating residue parameter;

[0068] S22: Compare historical parameter data with current parameter data, and establish a time series model for representing parameter change trends of die castings based on the comparison results;

[0069] S23: dynamically adjusting the preset weight coefficient according to the time series model;

[0070] S24: According to the current roughness parameter, the current wear degree parameter and the current coating residue parameter, a mapping relationship between the surface feature and the spraying parameter is established according to the dynamically adjusted preset weight coefficient.

[0071] The solution accurately captures and predicts the changing trend of surface characteristic parameters by introducing historical data analysis and time series models. And by dynamically adjusting the preset weight coefficients, the mapping relationship can better adapt to the dynamic changes of the die casting characterization parameters, thereby improving the calculation accuracy and control effect of the spraying parameters. The die casting characterization parameters include historical parameter data and current parameter data.

[0072] Specifically, in some embodiments, the historical parameter data of the die casting surface is collected from the data collected during the previous spraying process, including surface characteristic parameters at different time points and different positions. The acquisition of the historical parameter data can be achieved by querying a database or reading from a data storage device.

[0073] The purpose of comparing historical data with currently collected data is to identify the change pattern and trend of the characterization parameters. A variety of statistical methods can be used for comparative analysis, such as calculating the mean, variance, correlation coefficient, etc. of the parameters. Through comparative analysis, the change pattern of the parameters over time and the relationship between different parameters can be found. Among them, the time series model can take a variety of forms, such as autoregressive model, moving average model or autoregressive moving average model. The selection of the time series model depends on the characteristics and change pattern of the data. For example, for characterization parameters with periodic changes, a seasonal time series model can be considered. The model establishment process includes characterization parameter estimation and model verification to ensure that the time series model can accurately reflect the change trend of the characterization parameters. Among them, the adjustment of the preset weight coefficient is to adapt to the dynamic change of the characterization parameter. The adjustment method can be based on the future parameter data predicted by the time series model, or it can be based on the change rate of the characterization parameter. For example, for a characterization parameter with a faster change rate, its corresponding preset weight coefficient can be increased to increase the influence of the characterization parameter in the calculation of the spraying parameter. Among them, the adjusted preset weight coefficient is substituted into the mapping relationship to obtain the updated mapping relationship between the surface feature and the spraying parameter. This step realizes the dynamic update of the mapping relationship, so that the calculation of spraying parameters can reflect the changes of surface characteristics in time.

[0074] For example, in some preferred embodiments, the system obtains historical parameter data of the die casting surface in the past 30 days, including measurement results at different time points every day. These historical data are compared with the current parameter data to calculate the rate of change and correlation of each parameter data. For example, it is found that the roughness parameter of the die casting surface increases by 0.1 μm on average every day, the wear degree parameter increases by 0.05 mm every week, and the coating residue parameter shows periodic changes. Based on these analysis results, an ARIMA model containing trend terms and seasonal terms is established to describe the parameter change trend. The parameters of the ARIMA model are estimated by the least squares method, and the model is selected using the AIC criterion. Using the established ARIMA model, the change trend of each parameter data in the next 24 hours is predicted. According to the prediction results, the preset weight coefficient is dynamically adjusted. For example, if it is predicted that the roughness parameter will increase rapidly, the preset weight coefficient of the current roughness parameter is increased from 0.4 to 0.5, and the weight of other parameter data is correspondingly reduced. Finally, the adjusted preset weight coefficient is substituted into the mapping relationship formula: P(x,y) = K[α·R(x,y) + β·W(x,y) + γ·C(x,y)], where α, β, and γ are the adjusted preset weight coefficients, K is the correction coefficient, and R(x,y), W(x,y), and C(x,y) are the current roughness parameter, current wear degree parameter, and current coating residue parameter at the position (x,y). Through this updated mapping relationship, the optimal spraying parameters at each position on the die casting surface are calculated.

[0075] Further, step S22 includes:

[0076] S221: Acquire sampling time intervals of die casting characterization parameters, where the die casting characterization parameters include historical parameter data and current parameter data;

[0077] S222: Arrange the comparison results in chronological order according to the sampling time interval;

[0078] S223: Calculating the parameter change rate of the die casting characterization parameters collected at adjacent sampling time points;

[0079] S224: establishing a differential equation according to the parameter change rate, substituting the differential equation into a preset time series basis function, and obtaining a time series model for characterizing the parameter change trend of the die casting by least squares fitting calculation.

[0080] Among them, by obtaining the sampling time interval of the die casting characterization parameters, the time accuracy of data acquisition is ensured, laying the foundation for the subsequent time series analysis. The selection of the sampling time interval needs to consider the speed of change of the die casting characterization parameters and the actual operation of the production line. Generally, every hour or every shift can be selected as the sampling time interval, depending on the production process and quality control requirements. In some preferred embodiments, when obtaining the sampling time interval, an adaptive sampling strategy can be considered. When the parameters change rapidly, the sampling time interval is automatically shortened; when the parameters change slowly, the sampling time interval is appropriately extended. This strategy can optimize the use of system resources while ensuring data accuracy.

[0081] Among them, by arranging the comparison results in chronological order according to the sampling time interval, the discrete parameter data is arranged in an orderly time series, which provides a structured data basis for subsequent trend analysis. The time series can be arranged in the form of timestamps or serial numbers to ensure the continuity and traceability of the parameter data. Among them, the calculation of the parameter change rate can adopt a simple difference method, or a more complex sliding average or exponential smoothing technology to reduce the impact of noise. The calculation method is selected according to the needs of the actual application. For example, the raw data is preprocessed using the moving average method or the exponential smoothing method to eliminate the impact of short-term fluctuations and highlight the long-term change trend. The size of the smoothing window can be determined according to the actual production cycle, and 3-5 sampling points can usually be selected as a smoothing window.

[0082] The differential equation is established according to the parameter change rate. Specifically, when calculating the parameter change rate, a smoothing process technology can be introduced. The specific steps of establishing the differential equation by the smoothing process technology can be:

[0083] When establishing a first-order differential equation, the least squares method can be used to fit the data points to obtain the general form of the differential equation: y(t+1) - y(t) = a·y(t) + b, where a and b are unknown coefficients. By solving the normal equations, the optimal a and b values ​​can be obtained, thereby determining the specific form of the differential equation. y(t) is the die casting characterization parameter data at time t.

[0084] The calculation of the coefficient matrix of the unknown coefficients is a key step in the process of establishing the difference equation. For a first-order difference equation, the coefficient matrix is ​​usually a 2×2 matrix whose elements are determined by the coefficients of the difference equation. For example, for the equation y(t+1) = (1+a)y(t) + b, its coefficient matrix can be expressed as: [1+ab][0 1].

[0085] Substituting this coefficient matrix into the difference equation, we can obtain the discrete expression of parameter change: y(t) = (1+a)^t·y(0) + b·[(1+a)^t - 1] / a, where y(0) is the initial value and t is the discrete time point.

[0086] By removing abnormal fluctuation points through smoothing, the reliability of data is improved. When implementing it specifically, the following steps can be adopted:

[0087] 1. Data preprocessing: Use the 5-point moving average method to smooth the parameter change rate data. Suppose the original data sequence is {x1, x2, ..., xn}, then the smoothed data is {y3, y4, ..., yn-2}, where yi = (xi-2+ xi-1 + xi + xi+1 + xi+2) / 5.

[0088] 2. Establish the difference equation: Use the least squares method to fit the smoothed data points and obtain the first-order difference equation y(t+1) - y(t) = a·y(t) + b. Assume that the coefficients to be determined are a = 0.05 and b = 0.01.

[0089] 3. Calculate the coefficient matrix: According to the difference equation y(t+1) = (1+a)y(t) + b, the coefficient matrix of the unknown coefficients is: [1.05 0.01][0 1 ].

[0090] 4. Obtain the discrete expression: Substitute the coefficient matrix into the difference equation to obtain the discrete expression of the parameter change: y(t) = 1.05^t·y(0) + 0.01·(1.05^t - 1) / 0.05.

[0091] Through this method, the differential equation reflecting the parameter change trend can be effectively established, providing a reliable mathematical model for the subsequent spraying parameter optimization.

[0092] Among them, the preset time series basis function can select a suitable basis function in advance according to the characteristics of the parameter change, such as a linear function, an exponential function or a periodic function. Among them, the least squares method can minimize the sum of squares of the errors between the predicted value and the actual value of the time series model predicting the characterization parameters of the die casting, thereby obtaining the optimal time series model. In practical applications, matrix operations or iterative algorithms can be used to improve the computational efficiency in the least squares fitting process. When selecting the time series basis function, polynomial basis functions, Fourier basis functions or wavelet basis functions can be used. For parameter changes with obvious periodicity, the Fourier basis function may be more suitable; for nonlinear change trends, spline functions or radial basis functions (RBF) can be considered. The selection of basis functions can determine the best solution by cross-validation. In some preferred embodiments, in the process of least squares fitting, a regularization term can be introduced to prevent overfitting. For example, using ridge regression or LASSO regression, the model complexity is controlled by adding a penalty term. The selection of regularization parameters can determine the optimal value by grid search combined with cross-validation.

[0093] The time series model established by this method has strong adaptability and predictive ability. It can accurately capture the changing rules of the parameters characterizing the die castings and provide reliable data support for the subsequent optimization of the spraying parameters. For example, the model can predict the changing trend of the surface roughness in the next few hours, so as to adjust the spraying parameters in advance and achieve preventive maintenance.

[0094] Furthermore, the step of updating the correction coefficient of the mapping relationship according to the deviation in step S4 includes:

[0095] S41: Obtaining deviations between actual spraying effects and expected spraying effects at multiple preset time points;

[0096] S42: Calculate the deviation change trend at multiple preset time points, and when the deviation change trend exceeds a preset range, use an adaptive learning strategy and the deviation at the current moment to update the correction coefficient of the mapping relationship.

[0097] Among them, by dynamically adjusting the correction coefficient, real-time optimization of the spraying parameters is achieved. Specifically, by obtaining the deviations at multiple preset time points, a comprehensive understanding of the changes in the spraying effect over time can be obtained. By calculating the changing trend of these deviation data, it is possible to promptly detect whether there are abnormal fluctuations in the spraying effect. When the changing trend exceeds the preset range, it means that the current mapping relationship can no longer adapt well to the changes in the die casting characterization parameters, and the correction coefficient needs to be adjusted.

[0098] Among them, when the adaptive learning strategy and the deviation at the current moment are used to calculate the new correction coefficient, the correction amplitude is determined by the learning rate of the adaptive learning strategy. Too large a learning rate may cause system instability, while too small a learning rate may cause the system to react slowly. Specifically, the learning rate of the adaptive learning strategy can be set to 0.01, and then dynamically adjusted according to the response of the system. If the deviation is still large after multiple consecutive corrections, the learning rate can be appropriately increased; if the deviation fluctuates violently, the learning rate can be reduced to improve system stability.

[0099] The new correction coefficient can be calculated using the gradient descent method. Assuming the current correction coefficient is K, the learning rate is η, and the deviation at the current moment is E, the new correction coefficient K' can be expressed as:

[0100] ,in, It represents the gradient of the deviation E to the correction coefficient K. This method can adjust the correction coefficient in the direction of reducing the deviation.

[0101] When updating the new correction coefficients to the mapping relationship, the importance of different parameters needs to be taken into account. Different preset weight coefficients can be assigned to different characterization parameters (such as the current roughness parameter, the current wear degree parameter, and the current coating residue parameter) to make the correction process more refined. For example, if it is found that the current wear degree parameter has the greatest impact on the spraying effect, it can be given a higher preset weight coefficient.

[0102] Further, step S42 includes:

[0103] S421: Obtain the fluctuation frequency of the deviation change trend;

[0104] S422: when the fluctuation frequency is greater than a first preset threshold, reducing the learning rate of the adaptive learning strategy according to a first preset ratio; when the fluctuation frequency is less than a second preset threshold, increasing the learning rate of the adaptive learning strategy according to the first preset ratio; the first preset threshold is greater than the second preset threshold;

[0105] S423: Update the correction coefficient of the mapping relationship according to the updated adaptive learning strategy and the deviation at the current moment.

[0106] Among them, the fluctuation frequency of the deviation change trend is obtained. The fluctuation frequency here can be calculated by performing Fourier transform or wavelet analysis on the deviation data within a certain time window. Then, the learning rate is dynamically adjusted according to the size of the fluctuation frequency. When the fluctuation frequency is greater than the first preset threshold, the learning rate is reduced according to the first preset ratio to avoid excessive sensitivity of the system; when the fluctuation frequency is less than the second preset threshold, the learning rate is increased according to the second preset ratio to speed up the system response speed. Finally, the new correction coefficient is calculated using the updated learning rate and the deviation data at the current moment. When the fluctuation frequency is between the second preset threshold and the first preset threshold, it remains unchanged.

[0107] This method of dynamically adjusting the learning rate is closely combined with the aforementioned mapping relationship establishment and spray parameter calculation steps to form a closed-loop adaptive control system. By adjusting the learning rate in real time, the system can better adapt to the dynamic changes in the surface characteristics of the die-casting and improve the stability and uniformity of the spraying effect. For example, in actual applications, the first preset threshold can be set to 0.5Hz and the second preset threshold can be set to 0.1Hz. When the fluctuation frequency is detected to be greater than 0.5Hz, the learning rate can be reduced to 80% of the original; when the fluctuation frequency is less than 0.1Hz, the learning rate can be increased to 120% of the original. Such a setting can slow down the adjustment speed when the system fluctuates violently, and speed up the adjustment speed when the system changes slowly, thereby achieving a smoother and more stable control effect.

[0108] Further, step S423 includes:

[0109] S4231: Obtain the change rate of the deviation change trend;

[0110] S4232: Calculate the stability index of the deviation based on the rate of change;

[0111] S4233: When the stability index is greater than a third preset value, the updated adaptive learning strategy is multiplied by the stability index to obtain a revised adaptive learning strategy, and the correction coefficient of the mapping relationship is updated according to the revised adaptive learning strategy and the deviation at the current moment.

[0112] The change rate of the deviation change trend is obtained. This can be achieved by calculating the difference between the deviation data at two adjacent sampling moments and dividing it by the sampling time interval. For example, the following formula can be used: change rate = (E(t) - E(t-1)) / Δt, where E(t) represents the deviation data at the current moment, E(t-1) represents the deviation data at the previous moment, and Δt is the sampling time interval.

[0113] In calculating the stability index of the deviation based on the rate of change, the stability index can be defined in a variety of ways, such as using the inverse of the rate of change, or using an exponential function to map the rate of change of the deviation. One possible calculation method is as follows:

[0114] Stability index = exp(-|rate of change|). Under this definition, when the deviation change rate is close to zero, the stability index is close to 1, indicating that the system is in a stable state; when the change rate is large, the stability index is close to 0, indicating that the system fluctuates greatly.

[0115] Among them, the third preset value refers to the standard stability index set in advance by the technicians. When the calculated stability index is greater than the third preset value, it is considered that the system is in a relatively stable state. At this time, the learning rate can be appropriately increased to speed up the correction process. Updating the adaptive learning strategy refers to updating the learning rate of the adaptive learning strategy. The specific operation is to multiply the updated learning rate by the stability index to obtain the corrected learning rate. The updated learning rate here refers to the learning rate after the fluctuation frequency is adjusted. The calculation formula of the corrected learning rate can be expressed as:

[0116] Corrected learning rate = updated learning rate × stability index. Use the corrected learning rate and the deviation data at the current moment to calculate the new correction coefficient. The correction coefficient can be updated using the gradient descent method. The specific formula is as follows: New correction coefficient = old correction coefficient - corrected learning rate × gradient of deviation data. Among them, the gradient of the deviation can be approximately calculated by the numerical differentiation method.

[0117] Furthermore, the mapping relationship adopts the following mathematical model: P(x,y) = K[α·R(x,y) + β·W(x,y) +γ·C(x,y)], where P(x,y) is the spraying parameter at the position (x,y), R(x,y) is the current roughness parameter at the position (x,y), W(x,y) is the current wear degree parameter at the position (x,y), C(x,y) is the current coating residual parameter at the position (x,y), K is the correction coefficient, α, β, γ are preset weight coefficients; the correction coefficient is updated using the following mathematical model: , where K(t) is the correction coefficient at time t, η is the learning rate of the adaptive learning strategy, and E(t) is the deviation between the actual effect and the expected effect. Represents the gradient of the deviation E(t) to the correction coefficient K(t).

[0118] Among them, this model describes the dynamic update process of the correction coefficient. By introducing the learning rate η, the update speed of the correction coefficient can be controlled. When the actual spraying effect deviates from the expected, the model can automatically adjust the correction coefficient so that the mapping relationship is gradually optimized. In some specific embodiments, the time evolution characteristics of the current wear degree parameter adopt the following mathematical model: dW(x,y,t) / dt = f[W(x,y,t), σ(x,y,t)], where σ(x,y,t) is the stress distribution function, which describes the stress distribution at the position (x,y) at time t, f[W(x,y,t), σ(x,y,t)] is the wear evolution function, W(x,y,t) is the current wear degree parameter, which describes the wear degree at the position (x,y) after time t. By introducing the stress distribution function σ(x,y,t), the wear evolution process at different positions can be predicted more accurately. The wear evolution function f[W(x,y,t), σ(x,y,t)] can be fitted and optimized based on the actual observed data. dW(x,y,t) / dt represents the rate of change of the wear degree parameter W(x,y,t) over time.

[0119] In a second aspect, the present application provides a spray control device, the device comprising:

[0120] Acquisition module 201: Acquisition of current parameter data of the die casting surface, the current parameter data at least including current roughness parameter, current wear degree parameter and current coating residue parameter;

[0121] Mapping module 202: establishing a mapping relationship between surface features and spraying parameters according to a current roughness parameter, a current wear degree parameter and a current coating residue parameter according to a preset weight coefficient;

[0122] Calculation module 203: Calculate the spraying parameters of the die casting surface according to the mapping relationship;

[0123] Correction module 204: Controls the pressure, flow rate and angle of the nozzle according to the spraying parameters, obtains the deviation between the actual spraying effect and the expected spraying effect, and updates the correction coefficient of the mapping relationship according to the deviation, thereby updating the mapping relationship, so as to calculate the updated spraying parameters of the die-casting surface according to the updated mapping relationship, and sprays according to the updated spraying parameters of the die-casting surface.

[0124] The acquisition module 201 can use a variety of sensor technologies to acquire current parameter data of the die casting surface. For example, a laser scattering sensor can be used to measure the current roughness parameter of the surface, an eddy current probe can be used to detect the current wear degree parameter, and a spectrum analyzer can be used to identify the current coating residue parameter. These sensors can be installed at the feed end of the spraying equipment to achieve real-time detection of each die casting.

[0125] The mapping module 202 uses a preset weight coefficient to establish a mapping relationship between surface features and spraying parameters. This mapping relationship can be expressed as a mathematical model, for example: P(x,y) = K[α·R(x,y) + β·W(x,y) + γ·C(x,y)], where P(x,y) is the spraying parameter at the position (x,y), R(x,y), W(x,y) and C(x,y) are respectively the current roughness parameter, current wear degree parameter and current coating residue parameter of the die casting surface at the position (x,y), K is the correction coefficient, and α, β, γ are the preset weight coefficients.

[0126] The calculation module 203 calculates specific spraying parameters for each position on the die casting surface based on the mapping relationship established by the mapping module 202. These parameters may include spraying pressure, flow rate, nozzle angle, etc. The calculation process may use parallel processing technology to improve calculation efficiency and ensure real-time performance.

[0127] Wherein, correction module 204 can adopt precise servo control system to adjust the pressure, flow and angle of nozzle.For example, pressure proportional valve can be used to control spraying pressure, flow meter and control valve can adjust flow, and stepper motor can accurately adjust nozzle angle.These control elements can be connected with central controller through industrial bus to achieve fast response and accurate control.Correction module 204 can also adopt machine vision system to evaluate spraying effect.The surface image after spraying can be captured by high-speed camera, combined with image processing algorithm, coating thickness, uniformity and other indicators can be analyzed in real time, and compared with preset standard, deviation can be obtained, and mapping relationship can be updated according to the deviation, and updated spraying parameters can be calculated by updated mapping relationship at last, and the surface of die casting can be sprayed according to updated spraying parameters, so that spraying effect can be better.

[0128] Please refer to Figure 3 , Figure 3This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The present application provides an electronic device 3, including: a processor 301 and a memory 302. The processor 301 and the memory 302 are interconnected and communicate with each other through a communication bus 303 and / or other forms of connection mechanisms (not shown). The memory 302 stores computer-readable instructions executable by the processor 301. When the electronic device is running, the processor 301 executes the computer-readable instructions to execute the method in any optional implementation of the above embodiment to achieve the following functions: obtaining current parameter data of the surface of the die casting, the current parameter data at least including when The method comprises the following steps: calculating the previous roughness parameter, the current wear degree parameter and the current coating residual parameter; establishing a mapping relationship between surface features and spraying parameters according to the preset weight coefficients according to the current roughness parameter, the current wear degree parameter and the current coating residual parameter; calculating the spraying parameters of the die-casting surface according to the mapping relationship; controlling the pressure, flow rate and angle of the nozzle according to the spraying parameters, obtaining the deviation between the actual spraying effect and the expected spraying effect, and updating the correction coefficient of the mapping relationship according to the deviation, thereby updating the mapping relationship, calculating the updated spraying parameters of the die-casting surface according to the updated mapping relationship, and performing spraying according to the updated spraying parameters of the die-casting surface.

[0129] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method in any optional implementation of the above-mentioned embodiment is executed to achieve the following functions: obtaining current parameter data of the surface of a die-casting, the current parameter data at least including a current roughness parameter, a current wear degree parameter and a current coating residue parameter; establishing a mapping relationship between surface features and spraying parameters according to a preset weight coefficient based on the current roughness parameter, the current wear degree parameter and the current coating residue parameter; calculating the spraying parameters of the surface of the die-casting according to the mapping relationship; controlling the pressure, flow rate and angle of the nozzle according to the spraying parameters to obtain the deviation between the actual spraying effect and the expected spraying effect, and updating the correction coefficient of the mapping relationship according to the deviation, thereby updating the mapping relationship, so as to calculate the updated spraying parameters of the surface of the die-casting according to the updated mapping relationship, and spraying according to the updated spraying parameters of the surface of the die-casting.

[0130] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0131] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0132] Furthermore, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0133] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0134] The above description is only an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the scope of protection of the present application.

Claims

1. A spraying control method, characterized in that: The method comprises the steps of: S1: Acquire current parameter data of the surface of the die casting, wherein the current parameter data at least includes a current roughness parameter, a current wear degree parameter and a current coating residue parameter; S2: According to the current roughness parameter, the current wear degree parameter and the current coating residual parameter, a mapping relationship between surface features and spraying parameters is established according to preset weight coefficients; the mapping relationship adopts the following mathematical model: P(x, y) = K[α·R(x, y) + β·W(x, y) + γ·C(x, y)], wherein P(x, y) is the spraying parameter at the position (x, y), R(x, y) is the current roughness parameter at the position (x, y), W(x, y) is the current wear degree parameter at the position (x, y), C(x, y) is the current coating residual parameter at the position (x, y), K is the correction coefficient, and α, β, and γ are preset weight coefficients; the correction coefficient is updated using the following mathematical model: , where K(t) is the correction coefficient at time t, η is the learning rate of the adaptive learning strategy, and E(t) is the deviation between the actual effect and the expected effect. Represents the gradient of the deviation E(t) to the correction coefficient K(t); S3: calculating the spraying parameters of the surface of the die casting according to the mapping relationship; S4: Control the pressure, flow rate and angle of the nozzle according to the spraying parameters, obtain the deviation between the actual spraying effect and the expected spraying effect, and update the correction coefficient of the mapping relationship according to the deviation, so as to update the mapping relationship, calculate the updated spraying parameters of the die-casting surface according to the updated mapping relationship, and spray according to the updated spraying parameters of the die-casting surface.

2. A spraying control method according to claim 1, characterized in that: Step S2 includes: S21: Acquire historical parameter data of the surface of the die casting, wherein the historical parameter data at least includes a historical roughness parameter, a historical wear degree parameter, and a historical coating residue parameter; S22: comparing the historical parameter data with the current parameter data, and establishing a time series model for representing parameter change trends of die castings according to the comparison result; S23: dynamically adjusting the preset weight coefficient according to the time series model; S24: According to the current roughness parameter, the current wear degree parameter and the current coating residue parameter, a mapping relationship between surface features and spraying parameters is established according to the dynamically adjusted preset weight coefficient.

3. A spraying control method according to claim 2, characterized in that: Step S22 includes: S221: Acquire a sampling time interval of the die casting characterization parameter, wherein the die casting characterization parameter includes historical parameter data and the current parameter data; S222: Arranging the comparison results in chronological order according to the sampling time interval; S223: Calculating the parameter change rate of the die casting characterization parameters collected at adjacent sampling time points; S224: establishing a differential equation according to the parameter change rate, substituting the differential equation into a preset time series basis function, and obtaining a time series model for characterizing the parameter change trend of the die casting by least squares fitting calculation.

4. A spraying control method according to claim 1, characterized in that: The step of updating the correction coefficient of the mapping relationship according to the deviation in step S4 includes: S41: Obtaining deviations between the actual spraying effect and the expected spraying effect at multiple preset time points; S42: Calculate the deviation change trend of the plurality of preset time points, and when the deviation change trend exceeds a preset range, use an adaptive learning strategy and the deviation at the current moment to update the correction coefficient of the mapping relationship.

5. A spraying control method according to claim 4, characterized in that: Step S42 includes: S421: Obtaining the fluctuation frequency of the deviation change trend; S422: when the fluctuation frequency is greater than a first preset threshold, reducing the learning rate of the adaptive learning strategy according to a first preset ratio; when the fluctuation frequency is less than a second preset threshold, increasing the learning rate of the adaptive learning strategy according to the first preset ratio; the first preset threshold is greater than the second preset threshold; S423: updating the correction coefficient of the mapping relationship according to the updated adaptive learning strategy and the deviation at the current moment.

6. A spraying control method according to claim 5, characterized in that: Step S423 includes: S4231: Obtaining the change rate of the deviation change trend; S4232: Calculating a stability index of the deviation according to the change rate; S4233: When the stability index is greater than a third preset value, the updated adaptive learning strategy is multiplied by the stability index to obtain a revised adaptive learning strategy, and the correction coefficient of the mapping relationship is updated according to the revised adaptive learning strategy and the deviation at the current moment.

7. A spraying control device, characterized in that: According to any one of the steps of the method described in claims 1 to 6, the device comprises: Acquisition module: acquiring current parameter data of the die casting surface, wherein the current parameter data at least includes current roughness parameter, current wear degree parameter and current coating residue parameter; Mapping module: According to the current roughness parameter, the current wear degree parameter and the current coating residual parameter, a mapping relationship between surface features and spraying parameters is established according to a preset weight coefficient; the mapping relationship adopts the following mathematical model: P(x, y) = K[α·R(x, y) + β·W(x, y) + γ·C(x, y)], wherein P(x, y) is the spraying parameter at the position (x, y), R(x, y) is the current roughness parameter at the position (x, y), W(x, y) is the current wear degree parameter at the position (x, y), C(x, y) is the current coating residual parameter at the position (x, y), K is the correction coefficient, and α, β, and γ are preset weight coefficients; the correction coefficient is updated using the following mathematical model: , where K(t) is the correction coefficient at time t, η is the learning rate of the adaptive learning strategy, and E(t) is the deviation between the actual effect and the expected effect. Represents the gradient of the deviation E(t) to the correction coefficient K(t); Calculation module: calculates the spraying parameters of the die casting surface according to the mapping relationship; Correction module: Control the pressure, flow rate and angle of the nozzle according to the spraying parameters, obtain the deviation between the actual spraying effect and the expected spraying effect, and update the correction coefficient of the mapping relationship according to the deviation, so as to update the mapping relationship, calculate the updated spraying parameters of the die-casting surface according to the updated mapping relationship, and spray according to the updated spraying parameters of the die-casting surface.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the method according to any one of claims 1 to 6 are executed.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are executed.

Images