Horizontal rotary coating control method and system based on neural network

Abstract

The invention discloses a horizontal rotary coating control method and system based on a neural network, and relates to the technical field of data processing, and the method comprises the following steps: obtaining surface point cloud data of a part at a spraying station inlet; obtaining the current curvature of the part according to the surface point cloud data; paint viscosity and environment temperature are obtained; inputting the current curvature, the coating viscosity and the environment temperature into a rotating speed prediction model, and outputting a predicted rotating speed; acquiring the current curvature, the coating viscosity and the environment temperature at a preset time interval, and updating the predicted rotating speed; and controlling the rotation speed of the part according to the predicted rotation speed. According to the method, the part data and the environment data on the horizontal rotary coating line are obtained in real time, the rotating speed of the part is dynamically adjusted through the rotating speed prediction model, paint film forming of the part at the proper rotating speed is facilitated, and the quality of the paint film is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a horizontal rotary coating control method and system based on a neural network. BACKGROUND

[0002] A horizontal rotary coating line is a special coating line. The parts to be sprayed are hung on a trolley, and the trolley transports the parts to the spraying station. When spraying work is performed at the spraying station, the parts are rotated by a motor. The parts rotate continuously during the spraying process, and the paint film flows in different directions periodically under the action of gravity, improving the quality of the paint film.

[0003] In the process of horizontal rotary coating, the same rotary speed is used to rotate the parts, and the rotary speed is not automatically adjusted. However, different rotary speeds are required for different parts and different station environments. If the rotary speed is slow, the uniform flow of the paint film will be slow, and if the rotary speed is fast, it will easily cause paint splashing, resulting in surface defects of the paint film. Using a fixed and unchanged rotary speed to rotate the parts is not conducive to forming a high-quality paint film. Therefore, a control method is needed to dynamically control the rotary speed of the parts. SUMMARY

[0004] The purpose of the present application is to provide a horizontal rotary coating control method and system based on a neural network. The present application dynamically adjusts the rotary speed of the parts by real-time acquisition of part data and environmental data on the horizontal rotary coating line, using a rotational speed prediction model. This is conducive to the formation of a paint film at an appropriate rotary speed, and improves the quality of the paint film.

[0005] The purpose of the present application is achieved by the following technical means:

[0006] In a first aspect, the present application provides a horizontal rotary coating control method based on a neural network, comprising the following steps:

[0007] Acquire surface point cloud data of the parts at the entrance of the spraying station;

[0008] According to the surface point cloud data, acquire the current curvature of the parts;

[0009] Acquire the paint viscosity and the environmental temperature;

[0010] Input the current curvature, the paint viscosity and the environmental temperature into a rotational speed prediction model, and output a predicted rotational speed;

[0011] Acquire the current curvature, the paint viscosity and the environmental temperature at a preset time interval, and update the predicted rotational speed;

[0012] Control the rotary speed of the parts according to the predicted rotational speed.

[0013] Preferably, the step of inputting the current curvature, the paint viscosity, and the ambient temperature into a speed prediction model and outputting the predicted speed further includes: training the speed prediction model, specifically including the following steps:

[0014] Obtaining a large number of the current curvature, the paint viscosity, the ambient temperature, the paint film grade, and the rotation speed as a training set;

[0015] The training set is input into the MLP regression model for training to obtain the speed prediction model.

[0016] Preferably, the obtaining of a large number of the current curvature, the paint viscosity, the ambient temperature, the paint film grade and the rotation speed as a training set comprises the following steps:

[0017] Acquire the current curvature, the paint viscosity, and the ambient temperature at the start of spraying;

[0018] The parts are sprayed with paint at the rotational speed of the parts;

[0019] When the working time of the spraying work is equal to the preset time interval, the spraying work is stopped and the paint film thickness is obtained;

[0020] Calculating the paint film grade according to the paint film thickness;

[0021] A plurality of the current curvatures, the paint viscosities, the ambient temperatures, the paint film grades, and the rotation speeds are collected as a training set.

[0022] Preferably, calculating the paint film grade according to the paint film thickness comprises the following steps:

[0023] Calculate the average thickness and thickness standard deviation based on the paint film thickness;

[0024] Calculating a thickness score based on the average thickness and the target thickness;

[0025] Calculate the uniformity score based on the thickness standard deviation;

[0026] Performing a weighted summation of the thickness score and the uniformity score to obtain a paint film quality score;

[0027] The paint film quality score is divided into grades to obtain the paint film grade.

[0028] Preferably, the calculation formula of the thickness score is expressed as follows:

[0029] , , ,

[0030] wherein, is a thickness score, and is a weight coefficient, is an average thickness, is a target thickness, is an allowable error, is a number of thickness points, is a thickness of a first point, is a minimum allowable thickness, a maximum allowable thickness;

[0031] The calculation formula of the uniformity score is shown as follows:

[0032] ,

[0033] wherein, is a uniformity score, is a penalty coefficient, is a thickness standard deviation;

[0034] The calculation formula of the paint film quality score is shown as follows:

[0035] ,

[0036] wherein, is a paint film quality score, is a thickness weight coefficient, is a thickness score, is a uniformity weight coefficient, is a uniformity score.

[0037] Preferably, the step of controlling the rotation speed of the part according to the predicted rotation speed comprises the following steps:

[0038] obtaining a maximum rotation speed and a minimum rotation speed to determine a speed interval;

[0039] when the predicted rotation speed is within the speed interval, adjusting the rotation speed to the predicted rotation speed;

[0040] when the predicted rotation speed is outside the speed interval, reducing a preset paint film grade, and reacquiring the predicted rotation speed according to the rotation speed prediction model.

[0041] Preferably, after the step of controlling the rotation speed of the part according to the predicted rotation speed, the method further comprises the step of correcting the rotation speed prediction model, specifically comprising the following steps:

[0042] acquiring an actual paint film thickness of the part at an exit of a spraying station;

[0043] According to the actual paint film thickness, an actual paint film grade is calculated;

[0044] The actual paint film grade is compared with a preset paint film grade, and the rotational speed prediction model is corrected according to a comparison result.

[0045] In a second aspect, the present application provides a horizontal rotary coating control system based on a neural network, which applies the above-mentioned horizontal rotary coating control method based on a neural network, and comprises a point cloud data acquisition module, a current curvature acquisition module, an environment parameter acquisition module, a predicted rotational speed output module, a predicted rotational speed updating module and a rotary speed control module.

[0046] The point cloud data acquisition module is configured to acquire surface point cloud data of a part at an entrance of a spraying station.

[0047] The current curvature acquisition module is configured to acquire a current curvature of the part according to the surface point cloud data.

[0048] The environment parameter acquisition module is configured to acquire a paint viscosity and an environment temperature.

[0049] The predicted rotational speed output module is configured to input the current curvature, the paint viscosity and the environment temperature into a rotational speed prediction model and output a predicted rotational speed.

[0050] The predicted rotational speed updating module is configured to acquire the current curvature, the paint viscosity and the environment temperature at a preset time interval and update the predicted rotational speed.

[0051] The rotary speed control module is configured to control a rotary speed of the part according to the predicted rotational speed.

[0052] In a third aspect, the present application provides an electronic device comprising a processor and a memory, wherein the memory is configured to store computer program code, and the computer program code comprises computer instructions, and when the processor executes the computer instructions, the electronic device executes the above-mentioned horizontal rotary coating control method based on a neural network.

[0053] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program comprises program instructions, and when the program instructions are executed by a processor of an electronic device, the processor executes the above-mentioned horizontal rotary coating control method based on a neural network.

[0054] Compared with the prior art, the present application has the following beneficial effects:

[0055] The application obtains the current curvature of the part on the horizontal rotary coating line, the coating viscosity and the environmental temperature in real time, dynamically adjusts the rotary speed of the part using the rotary speed prediction model under the preset paint film grade and time interval, is favorable for the part to form the paint film under the suitable rotary speed, and improves the quality of the paint film;

[0056] The application obtains the paint film thickness of each point of the paint film, calculates the thickness score and the uniformity score according to the paint film thickness, obtains the paint film quality score and divides the paint film grade, is favorable for the control of the paint film quality requirement in the spraying process, filters the training set data according to the paint film quality score and the paint film grade, reduces the training time of the rotary speed prediction model, and improves the quality of the paint film;

[0057] When the predicted rotary speed is located outside the speed interval, the application reduces the preset paint film grade in the rotary speed prediction model, is favorable for generating the paint film with higher quality and locating the rotary speed of the part in the control range of the motor, and improves the stability and accuracy of the rotary speed control. BRIEF DESCRIPTION OF DRAWINGS

[0058] The drawings incorporated into the specification and constituting a part of the specification show embodiments consistent with the application and, together with the specification, serve to explain the principles of the application.

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced as follows. Obviously, for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0060] Figure 1 A flowchart of a horizontal rotary coating control method based on a neural network is provided for the embodiment;

[0061] Figure 2 A flowchart of training a rotary speed prediction model is provided for step S39 of the embodiment;

[0062] Figure 3 A flowchart of obtaining a large number of current curvatures, coating viscosities, environmental temperatures, paint film grades and rotary speeds as training sets is provided for step S391 of the embodiment;

[0063] Figure 4 A flowchart of calculating the paint film grade according to the paint film thickness is provided for step S3914 of the embodiment;

[0064] Figure 5 A flowchart of controlling the rotary speed of the part according to the predicted rotary speed is provided for step S60 of the embodiment;

[0065] Figure 6 The step S70 provided for the embodiment corrects the rotational speed prediction model, and a flowchart of the step S70 is shown in FIG. 7.

[0066] Figure 7 A structure diagram of a horizontal rotary coating control system based on a neural network provided for the embodiment is shown in FIG. 8.

[0067] Figure 8 A structure diagram of an electronic device provided for the embodiment is shown in FIG. 9. DETAILED DESCRIPTION

[0068] The technical solutions in the embodiments of the present application will be clearly and completely described 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 of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the present application.

[0069] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present application are only used to explain the relative positional relationship, movement condition, etc. between components in a certain posture, and if the certain posture changes, the directional indications also change accordingly.

[0070] In addition, the description of “first”, “second” and the like in the present application is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by “first”, “second” can explicitly or implicitly include at least one of the features. In addition, the technical solutions of each embodiment can be combined with each other, but it must be based on the realization of a person of ordinary skill in the art, and when the combination of technical solutions contradicts each other or cannot be realized, it should be considered that the combination of technical solutions does not exist, and is not within the protection scope required by the present application.

[0071] Embodiment one

[0072] The embodiment provides a horizontal rotary coating control method based on a neural network, as shown in FIG. 8, which comprises the following steps: Figure 1

[0073] S10, acquiring surface point cloud data of a part at an entrance of a spraying station;

[0074] S20, acquiring current curvature of the part according to the surface point cloud data;

[0075] S30, acquiring coating viscosity and environmental temperature;

[0076] ​S40, inputting the current curvature, paint viscosity and ambient temperature into a speed prediction model and outputting a predicted speed;

[0077] S50, obtaining the current curvature, paint viscosity and ambient temperature at preset time intervals, and updating the predicted rotation speed;

[0078] S60: Control the rotation speed of the component according to the predicted rotation speed.

[0079] It should be noted that the spraying workflow of the horizontal rotary coating line is to transport the parts that need to be sprayed to the spraying station by a trolley. Before the trolley enters the spraying station, the rotation of the parts is limited by a limit assembly. After the trolley enters the spraying station, the limit assembly is separated from the trolley, and the parts are rotated by a motor control. The parts rotate continuously during the spraying and surface film formation process, and the paint film flows periodically in different directions under the action of gravity, thereby improving the quality of the paint film. The rotation speed of the parts in this embodiment plays an important role. If the rotation speed is too slow, the paint film cannot achieve the characteristic of uniform flow. If the rotation speed is too fast, the paint will splash, which will cause the paint film to fail to reach the ideal state, and surface defects or even leakage will occur. Therefore, it is necessary to control the rotation speed of the parts in real time so that they are in a suitable speed range to form a high-quality paint film.

[0080] Surface point cloud data can be obtained by placing a laser sensor above the spray station entrance to obtain three-dimensional point cloud data of the component. By analyzing and processing the surface point cloud data, the curvature of the component's current curve or surface can be obtained. This embodiment takes into account the different shapes of different components and the different shapes of different parts of the same component. Therefore, the current curvature of the component is used to measure the stability of the component during rotation.

[0081] Paint viscosity and ambient temperature are environmental data that affect the rotation speed in addition to the component's own data. Paint viscosity and ambient temperature can be directly measured by the paint and temperature stored in the spray station. The paint viscosity and ambient temperature can be used to measure the leveling speed and curing speed of the paint film.

[0082] The speed prediction model is a neural network model. This pre-trained model predicts rotational speed based on the current curvature, paint viscosity, ambient temperature, and a preset paint film grade. The paint film grade can be classified based on the quality of the paint film, with higher film grades indicating better film quality. Typically, the paint film grade is preset at the highest level and does not need to be modified to achieve higher-quality films. The current curvature, paint viscosity, and ambient temperature are repeatedly acquired at preset intervals, and an updated predicted speed is output, enabling real-time control of the component's rotational speed.

[0083] In the embodiment, by acquiring the current curvature, the coating viscosity and the ambient temperature of the part on the horizontal rotary coating line in real time, the rotary speed of the part is dynamically adjusted using the rotary speed prediction model under the preset paint film grade and time interval, which is beneficial to the formation of the paint film at a suitable rotary speed of the part and improves the quality of the paint film.

[0084] In some embodiments, before the step S40 of inputting the current curvature, the coating viscosity and the ambient temperature into the rotary speed prediction model and outputting the predicted rotary speed, the step S39 of training the rotary speed prediction model is further included, as shown in the following. Figure 2 The step S39 includes the following steps:

[0085] S391, a large number of current curvatures, coating viscosities, ambient temperatures, paint film grades and rotary speeds are acquired as a training set;

[0086] S392, the training set is input into the MLP regression model for training to obtain the rotary speed prediction model.

[0087] It should be noted that a large number of data during spraying work are acquired, and the data cover different current curvatures, coating viscosities, ambient temperatures, paint film grades and rotary speeds. The collected original data are preprocessed, and the processed data set is used to train the model to obtain the change rule of the curvature, the viscosity, the temperature, the quality and the rotary speed. In the embodiment, the MLP regression model can be used for training, and the structure thereof includes an input layer receiving four feature vectors of the current curvature, the coating viscosity, the ambient temperature and the paint film grade, is connected through three fully connected layers, fully iteratively learns the nonlinear relationship between the four parameters, and finally performs accurate prediction of the rotary speed in the output layer and outputs. The setting of the fully connected layer, such as the connection of the step-by-step dimension reduction and the application of the activation function, and the strategy of the training process can adopt the prior art, and will not be described here.

[0088] In some embodiments, the step S391 of acquiring a large number of current curvatures, coating viscosities, ambient temperatures, paint film grades and rotary speeds as a training set includes the following steps, as shown in the following. Figure 3 The step S391 includes the following steps:

[0089] S3911, the current curvature, the coating viscosity and the ambient temperature at the start of the spraying work are acquired;

[0090] S3912, the spraying work is performed on the part with the rotary speed of the part being zero;

[0091] S3913, the spraying work is stopped when the working time of the spraying work is equal to the preset time interval, and the paint film thickness is acquired;

[0092] S3914, the paint film grade is calculated according to the paint film thickness;

[0093] S3915, collect the current curvature, paint viscosity, ambient temperature, paint film grade and rotation speed as the training set.

[0094] It should be noted that when the spraying work is in progress, the rotation speed of the entire part will change when the rotation speed changes, which includes the part that has been sprayed and the part that has not been sprayed. By properly selecting the preset time interval, it can be considered that the sprayed part is less affected by the subsequent rotation speed change, and the data is obtained on this basis. The part affected by the data is corrected by the data obtained when the rotation speed prediction model is trained and put into production. In the data acquisition process, the current curvature, paint viscosity and ambient temperature at the start and the paint film thickness at the stop are obtained by presetting the rotation speed and time interval, and the paint film grade is calculated according to the paint film thickness. Repeat the above data acquisition steps to obtain a large amount of data as the training set of the rotation speed prediction model, and improve the prediction accuracy of the rotation speed prediction model.

[0095] In some embodiments, step S3914, according to the paint film thickness, calculate the paint film grade, as Figure 4 shown, comprising the following steps:

[0096] S39141, according to the paint film thickness, calculate the average thickness and thickness standard deviation;

[0097] S39142, according to the average thickness and target thickness, calculate the thickness score;

[0098] S39143, according to the thickness standard deviation, calculate the uniformity score;

[0099] S39144, the thickness score and uniformity score are weighted and summed to obtain the paint film quality score;

[0100] S39145, according to the paint film quality score, divide the grade to obtain the paint film grade.

[0101] It should be noted that the paint film quality needs to be quantified into a vector for the rotation speed prediction model to train the model to output the rotation speed that forms a higher paint film quality. By rotating the part during spraying, the thickness of the paint film is mainly improved, and the rotation speed is controlled to prevent phenomena such as sagging, surface defects and missed spraying of the paint film, which is reflected in the uniformity of the paint film. Therefore, the present embodiment obtains the thickness of each point on the paint film, calculates the score on the thickness and uniformity according to the paint film thickness, and obtains the paint film quality score. In the thickness score, whether the average thickness meets the target thickness and its allowable error range, and whether the thickness of each point on the paint film meets the target thickness and its allowable error range; In the uniformity score, whether the thickness of each point on the paint film is uniform.

[0102] Specifically, the calculation formula of the thickness score is as follows:

[0103] , , ,

[0104] in, Score thickness, and is the weight coefficient, is the average thickness, is the target thickness, To allow error, is the number of thickness points, For the The paint film thickness at the point is the minimum allowable thickness, Maximum allowable thickness;

[0105] The calculation formula of the uniformity score is as follows:

[0106] ,

[0107] in, Score for uniformity, is the penalty coefficient, is the thickness standard deviation;

[0108] The calculation formula for the paint film quality score is as follows:

[0109] ,

[0110] in, Score the paint film quality. is the thickness weight coefficient, Score thickness, is the uniformity weight coefficient, Score for evenness.

[0111] After the paint film quality score is calculated, the paint film quality score is divided into grades, such as grades 1, 2, 3, 4, and 5 corresponding to paint film quality scores [0, 20], (20, 40], (40, 60], (60, 80], and (80, 100], respectively. In this regard, when a plurality of current curvatures, paint viscosities, ambient temperatures, paint film grades, and rotation speeds are collected as a training set at step S3915, a case where the current curvatures, paint viscosities, ambient temperatures, and paint film grades are the same, but the rotation speeds are different, can occur. The reason is that the paint film quality scores are different due to different rotation speeds, but are divided into the same grade when grading. At this time, the collected data is filtered. Specifically, the paint film quality scores corresponding to the paint film grades are compared, and the data with a lower paint film quality score is filtered out. By filtering the data, the redundancy of the data in the training set is reduced, and the training time of the rotation speed prediction model is reduced.

[0112] In the present embodiment, by obtaining the paint film thickness of each point of the paint film, calculating the thickness score and the uniformity score according to the paint film thickness, obtaining the paint film quality score and dividing the paint film grade, the control of the paint film quality requirement in the spraying process is facilitated. The training set data is filtered according to the paint film quality score and the paint film grade, the training time of the rotation speed prediction model is reduced, and the quality of the paint film is improved.

[0113] In some embodiments, at step S60, the rotation speed of the part is controlled according to the predicted rotation speed, as shown in FIG. 6, including the following steps: Figure 5

[0114] S601, obtaining the maximum rotation speed and the minimum rotation speed, and determining the speed interval;

[0115] S602, when the predicted rotation speed is within the speed interval, adjusting the rotation speed to the predicted rotation speed;

[0116] S603, when the predicted rotation speed is outside the speed interval, reducing the preset paint film grade, and reacquiring the predicted rotation speed according to the rotation speed prediction model.

[0117] It should be noted that the rotation speed of the motor control has a maximum rotation speed and a minimum rotation speed. When the speed is predicted by the rotation speed prediction model, the output predicted rotation speed can be greater than the maximum rotation speed or less than the minimum rotation speed, i.e., the predicted rotation speed is outside the speed interval. Specifically, when the predicted rotation speed is greater than the minimum rotation speed and less than the maximum rotation speed, or the predicted rotation speed is equal to the minimum rotation speed or the maximum rotation speed, the rotation speed is adjusted to the predicted rotation speed by the motor. When the predicted rotation speed is greater than the maximum rotation speed or less than the minimum rotation speed, the rotation speed cannot be simply adjusted to the maximum rotation speed or the minimum rotation speed. At this time, the preset paint film grade in the rotation speed prediction model is reduced, and a new predicted rotation speed is obtained under the input of the new paint film grade. ​

[0118] In the embodiment, when the predicted rotating speed is located outside the speed interval, the preset paint film grade in the rotating speed prediction model is reduced, which is beneficial to generate paint film with higher quality and the rotating speed of the part is located in the control range of the motor, thereby improving the stability and accuracy of the rotating speed control.

[0119] In some embodiments, after the step S60 of controlling the rotating speed of the part according to the predicted rotating speed, the method further comprises a step S70 of correcting the rotating speed prediction model, as shown in the following. Figure 6 The step S70 specifically comprises the following steps:

[0120] S701, obtaining the actual paint film thickness of the part at the exit of the spraying station;

[0121] S702, calculating the actual paint film grade according to the actual paint film thickness;

[0122] S703, comparing the actual paint film grade with the preset paint film grade, and correcting the rotating speed prediction model according to the comparison result.

[0123] It should be noted that, in the process of obtaining the training set of the rotating speed prediction model, the part is only sprayed and rotated in the preset time interval, and the spraying work is stopped when the working time reaches the preset time interval. In actual production, the rotating speed controlled in the second time interval will affect the quality of the paint film formed in the first time interval, the rotating speed controlled in the third time interval will affect the quality of the paint film formed in the first time interval and the second time interval, and so on. Therefore, the actual paint film thickness is obtained at the exit of the spraying station, i.e. the paint film thickness after the spraying work is completed, and the actual paint film thickness includes the paint film thicknesses of multiple continuous areas on the part, which are the paint film thicknesses formed in the part in each time interval. The actual paint film grade is calculated according to the actual paint film thickness, and the corresponding part data and environmental data are obtained to correct the rotating speed prediction model, thereby improving the prediction accuracy of the rotating speed prediction model.

[0124] Embodiment Two

[0125] The embodiment provides a horizontal rotary coating control system based on a neural network, which applies the above-mentioned horizontal rotary coating control method based on a neural network, as shown in the following. Figure 7 The horizontal rotary coating control system based on a neural network comprises a point cloud data acquisition module, a current curvature acquisition module, an environmental parameter acquisition module, a predicted rotating speed output module, a predicted rotating speed updating module and a rotating speed control module.

[0126] The point cloud data acquisition module is configured to obtain the surface point cloud data of the part at the entrance of the spraying station.

[0127] The current curvature acquisition module is used to obtain the current curvature of the component based on the surface point cloud data;

[0128] Environmental parameter acquisition module, used to obtain paint viscosity and ambient temperature;

[0129] The predicted speed output module is used to input the current curvature, paint viscosity and ambient temperature into the speed prediction model and output the predicted speed;

[0130] A predicted speed update module is used to obtain the current curvature, paint viscosity and ambient temperature at preset time intervals and update the predicted speed;

[0131] The rotation speed control module is used to control the rotation speed of the component according to the predicted rotation speed.

[0132] In this embodiment, by acquiring the current curvature, paint viscosity and ambient temperature of the parts on the horizontal rotary coating line in real time, the rotation speed of the parts is dynamically adjusted using a speed prediction model under a preset paint film level and time interval, which is conducive to the formation of a paint film on the parts at an appropriate rotation speed and improves the quality of the paint film.

[0133] It should be understood that the disclosed system can be implemented in other ways. For example, the system embodiment described above is merely illustrative. For example, the division of the above modules is merely a logical functional division. In actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, each functional module may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0134] Example 3

[0135] This embodiment provides an electronic device 2, such as Figure 8 As shown, it includes a processor 21 and a memory 22. The memory 22 is used to store computer program codes. The computer program codes include computer instructions. When the processor 22 executes the computer instructions, the electronic device 2 executes the above-mentioned horizontal rotary coating control method based on neural network.

[0136] The electronic device 2 includes a processor 21, a memory 22, an output device 23, and an input device 24. The processor 21, the memory 22, the output device 23, and the input device 24 are coupled by a connector, which includes various interfaces, transmission lines, buses, and the like, and embodiments of the present application are not limited in this regard. It will be understood that, in various embodiments of the present application, coupled means connected by some means, either directly or indirectly, such as by various interfaces, transmission lines, buses, and the like.

[0137] The processor 21 can be one or more graphics processing units (GPUs), and in the case where the processor 21 is a GPU, the GPU can be a single-core GPU or a multi-core GPU. Alternatively, the processor 21 can be a processor group composed of a plurality of GPUs, and the plurality of processors are coupled to each other by one or more buses. Alternatively, the processor 21 can be another type of processor, and embodiments of the present application are not limited in this regard.

[0138] The memory 22 can be used to store computer program instructions, and various types of computer program codes for executing the present application. Alternatively, the memory 22 includes, but is not limited to, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), or a compact disc read-only memory (CD-ROM), and the memory 22 is used for related instructions and data.

[0139] The input device 24 is used to input data and / or signals, and the output device 23 is used to output data and / or signals. The output device 23 and the input device 24 can be independent devices, or can be an integral device.

[0140] The embodiment provides a computer readable storage medium, and the computer readable storage medium stores a computer program. The computer program includes program instructions, and the program instructions, when executed by the processor 21 of the electronic device 2, cause the processor 21 to execute the horizontal rotary coating control method based on a neural network.

[0141] The foregoing is considered as illustrative only of the principles of the application. Numerous modifications and changes will readily occur to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Therefore, the scope of the application is indicated by the appended claims rather than by the foregoing description, and all changes that come within the meaning and range of equivalents are intended to be embraced therein.

Claims

1. A horizontal rotary coating control method based on neural network, characterized in that: The following steps are involved: Obtain surface point cloud data of parts at the entrance of the spraying station; Obtaining a current curvature of a component according to the surface point cloud data; Obtain coating viscosity and ambient temperature; Inputting the current curvature, the coating viscosity and the ambient temperature into a rotation speed prediction model and outputting a predicted rotation speed; acquiring the current curvature, the paint viscosity, and the ambient temperature at preset time intervals, and updating the predicted rotation speed; The rotation speed of the component is controlled based on the predicted rotation speed.

2. A neural network-based horizontal rotary coating control method according to claim 1, characterized in that: Before inputting the current curvature, the paint viscosity, and the ambient temperature into the speed prediction model and outputting the predicted speed, the method further includes: training the speed prediction model, specifically including the following steps: Obtaining a large number of the current curvature, the paint viscosity, the ambient temperature, the paint film grade, and the rotation speed as a training set; The training set is input into the MLP regression model for training to obtain the speed prediction model.

3. A neural network-based horizontal rotary coating control method according to claim 2, characterized in that: The step of obtaining a large amount of the current curvature, the paint viscosity, the ambient temperature, the paint film grade, and the rotation speed as a training set includes the following steps: Acquire the current curvature, the paint viscosity, and the ambient temperature at the start of spraying; The parts are sprayed with paint at the rotational speed of the parts; When the working time of the spraying work is equal to the preset time interval, the spraying work is stopped and the paint film thickness is obtained; Calculating the paint film grade according to the paint film thickness; A plurality of the current curvatures, the paint viscosities, the ambient temperatures, the paint film grades, and the rotation speeds are collected as a training set.

4. A neural network-based horizontal rotary coating control method according to claim 3, characterized in that: Calculating the paint film grade according to the paint film thickness comprises the following steps: Calculate the average thickness and thickness standard deviation based on the paint film thickness; Calculating a thickness score based on the average thickness and the target thickness; Calculate the uniformity score based on the thickness standard deviation; Performing a weighted summation of the thickness score and the uniformity score to obtain a paint film quality score; The paint film quality score is divided into grades to obtain the paint film grade.

5. A neural network-based horizontal rotary coating control method according to claim 4, characterized in that: The calculation formula of the thickness score is as follows: ， ， ， in, Score thickness, and is the weight coefficient, is the average thickness, is the target thickness, To allow error, is the number of thickness points, For the The paint film thickness at the point is the minimum allowable thickness, Maximum allowable thickness; The calculation formula of the uniformity score is as follows: ， in, Score for uniformity, is the penalty coefficient, is the thickness standard deviation; The calculation formula of the paint film quality score is as follows: ， in, Score the paint film quality. is the thickness weight coefficient, Score thickness, is the uniformity weight coefficient, Score for evenness.

6. The neural network-based horizontal rotary coating control method according to claim 1, characterized in that: The step of controlling the rotation speed of the component according to the predicted rotation speed comprises the following steps: Get the maximum rotation speed and the minimum rotation speed and determine the speed range; When the predicted rotational speed is within the speed range, adjusting the rotational speed to the predicted rotational speed; When the predicted speed is outside the speed range, the preset paint film level is lowered and the predicted speed is re-obtained according to the speed prediction model.

7. The neural network-based horizontal rotary coating control method according to claim 1, characterized in that: After controlling the rotation speed of the component according to the predicted rotation speed, the method further includes: correcting the rotation speed prediction model, specifically including the following steps: Obtain the actual paint film thickness of parts at the spray station exit; Calculating the actual paint film grade according to the actual paint film thickness; The actual paint film grade is compared with a preset paint film grade, and the speed prediction model is corrected according to the comparison result.

8. A neural network-based horizontal rotary coating control system, applying the neural network-based horizontal rotary coating control method according to any one of claims 1 to 7, characterized in that: include: Point cloud data acquisition module, current curvature acquisition module, environmental parameter acquisition module, predicted speed output module, predicted speed update module and rotation speed control module; The point cloud data acquisition module is used to acquire surface point cloud data of parts at the entrance of the spraying station; The current curvature acquisition module is used to acquire the current curvature of the component according to the surface point cloud data; The environmental parameter acquisition module is used to obtain the coating viscosity and the ambient temperature; The predicted speed output module is configured to input the current curvature, the coating viscosity, and the ambient temperature into a speed prediction model and output a predicted speed; The predicted speed updating module is configured to obtain the current curvature, the paint viscosity, and the ambient temperature at preset time intervals and update the predicted speed; The rotation speed control module is used to control the rotation speed of the component according to the predicted rotation speed.

9. An electronic device, characterized in that: It includes a processor and a memory, the memory is used to store computer program code, the computer program code includes computer instructions, and when the processor executes the computer instructions, the electronic device executes a horizontal rotary coating control method based on neural network as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes a neural network-based horizontal rotary coating control method as described in any one of claims 1 to 7.

Images