Control optimization method for hardware spraying equipment

Abstract

The present invention discloses a control optimization method for a hardware spraying device, which relates to the field of spraying control technology. The method comprises: extracting the structural data and spraying requirements of the hardware in the target spraying task; adjusting the parameters of the spraying device according to the initial control scheme to obtain a configured initial spraying device; using the initial spraying device to spray K hardware pieces in sequence, performing spraying quality inspection on M finished hardware pieces, and generating M spraying quality inspection results; performing quality situation analysis on the M spraying quality inspection results to generate abnormal quality analysis results; optimizing the control scheme of the initial spraying device to obtain an adjustment control scheme, and adjusting the parameters of the initial spraying device to obtain an optimized target spraying device. The method solves the technical problem in the prior art that the operating loss of the hardware spraying device leads to reduced spraying quality, and achieves the technical effect of improving the spraying quality by performing parameter compensation on the spraying device.

Description

Technical Field

[0001] The present invention relates to the technical field of spraying control, and in particular to a control optimization method for a hardware spraying device. Background Art

[0002] With the rapid development of the manufacturing industry and intensified market competition, the requirements for the spray coating quality of hardware parts are becoming increasingly higher. As an indispensable part of industrial production and daily life, the surface spray treatment of hardware parts plays a vital role in improving product performance, extending service life, and beautifying appearance. However, during the hardware spraying process, the spraying device often faces problems such as performance degradation and spray parameter deviation after long-term operation, which directly affects the stability and consistency of the spraying quality. Traditional spraying control methods usually rely on manual experience and regular maintenance, which have many limitations. First, manual experience makes it difficult to accurately judge subtle changes in spraying parameters, resulting in large fluctuations in spraying quality. Second, although regular maintenance can delay the performance degradation of the spraying device to a certain extent, it cannot respond to changes in spraying parameters in real time, and the maintenance cost is high. Summary of the Invention

[0003] The embodiments of the present application provide a control optimization method for a hardware spraying device, which solves the technical problem in the prior art that the operating loss of the hardware spraying device causes a decrease in spraying quality.

[0004] In view of the above problems, an embodiment of the present application provides a control optimization method for a hardware spraying device.

[0005] The present application provides a control optimization method for a hardware spraying device, the method comprising:

[0006] Extract the structural data and spraying requirements of the hardware in the target spraying task; execute the initial control scheme configuration of the spraying device, and adjust the parameters of the spraying device according to the initial control scheme to obtain the configured initial spraying device, wherein the initial control scheme is obtained by searching in the control scheme library based on the structural data and the spraying requirements; respectively perform traceability identification on the K hardware in the target spraying task to generate K traceability identifications, wherein the K traceability identifications are used to record the time when the K hardware perform different spraying processes; use the initial spraying device to spray the K hardware in turn, and extract the traceability information within the first preset detection window. M finished hardware parts that have been sprayed are subjected to spraying quality inspection to generate M spraying quality inspection results; based on the M traceability identifications of the M finished hardware parts, a quality situation analysis is performed on the M spraying quality inspection results to generate abnormal quality analysis results, wherein the abnormal quality analysis results include Q abnormal quality deviation coefficients and Q abnormal surface defect characteristics; based on the Q abnormal quality deviation coefficients and Q abnormal surface defect characteristics, a control scheme of the initial spraying device is optimized to obtain an adjusted control scheme, and the initial spraying device is adjusted according to the adjusted control scheme to obtain an optimized target spraying device.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] First, the structural data and spraying requirements of the hardware components in the target spraying task are extracted. Next, the initial control scheme configuration of the spraying device is executed, and the parameters of the spraying device are adjusted according to the initial control scheme to obtain a configured initial spraying device. The initial control scheme is obtained by searching the control scheme library based on the structural data and spraying requirements. Furthermore, the K hardware components in the target spraying task are individually traceable, generating K traceability identifiers. These K traceability identifiers are used to record the time it takes for the K hardware components to perform different spraying processes. Then, the initial spraying device is used to sequentially spray the K hardware components, extracting M finished hardware components that have been sprayed within a first preset inspection window. These M finished hardware components are then subjected to spraying quality inspection, generating M spraying quality inspection results. Based on the M traceability identifiers of the M finished hardware components, a quality status analysis is performed on the M spraying quality inspection results to generate abnormal quality analysis results. The abnormal quality analysis results include Q abnormal quality deviation coefficients and Q abnormal surface defect characteristics. Finally, the control scheme for the initial spraying device was optimized based on Q abnormal mass deviation coefficients and Q abnormal surface defect characteristics, resulting in an adjusted control scheme. The parameters of the initial spraying device were then adjusted according to the adjusted control scheme to obtain the optimized target spraying device. This solved the technical problem of reduced spraying quality caused by operational losses in hardware spraying devices in the prior art. By compensating the parameters of the spraying device, the technical effect of improving spraying quality was achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A flow chart of a control optimization method for a hardware spraying device provided in an embodiment of the present application;

[0011] Figure 2 A schematic flow chart of spray quality detection in a control optimization method for a hardware spraying device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] The embodiments of the present application solve the technical problem in the prior art that the operating loss of the hardware spraying device leads to reduced spraying quality by providing a control optimization method for the hardware spraying device.

[0013] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0014] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0015] Examples, such as Figure 1 As shown, the embodiment of the present application provides a control optimization method for a hardware spraying device, wherein the method includes:

[0016] Extract the structural data and spraying requirements of the hardware in the target spraying task.

[0017] The structural data and spraying requirements of the hardware are extracted from the target spraying task, where the structural data includes the size and shape of the hardware, and the spraying requirements include the spraying color, coating thickness, etc.

[0018] Execute the initial control scheme configuration of the spray device, and adjust the parameters of the spray device according to the initial control scheme to obtain a configured initial spray device, wherein the initial control scheme is obtained by searching in a control scheme library based on the structural data and the spraying requirements.

[0019] Based on the hardware's structural data and coating requirements, a search is performed within a pre-set control solution library, which contains preset control solutions tailored to different hardware structures and coating requirements. By comparing the hardware's structural data (e.g., size, shape, material, etc.) with the coating requirements (e.g., color, coating thickness, and coating speed), the control solution that best matches or is similar to the current task is selected as the initial control solution. This initial control solution is then loaded into the coating device and configured according to the parameters set in the initial control solution. This includes adjusting parameters such as the spray gun's position, angle, spray pressure, flow rate, and speed. Once configured, the initial coating device is obtained.

[0020] Further, including:

[0021] Extract historical spraying record data of the spraying device within a historical time period, wherein the historical spraying record data includes historical structure data, historical spraying requirements, and historical spraying control schemes; perform data similarity authentication on the historical spraying record data to obtain historical similarity; if the historical similarity is greater than or equal to a preset similarity threshold, generate a cloud supplementary instruction; according to the cloud supplementary instruction, retrieve the cloud spraying record data, and store the cloud spraying record data and the historical spraying record data in the control scheme library.

[0022] Extract historical spraying record data within a historical period from the storage system of the spraying device, the historical spraying record data including historical structural data (such as size, shape, material, etc.), historical spraying requirements (such as color, coating thickness, spraying speed, etc.) and historical spraying control schemes (i.e., specific parameter settings used to perform spraying tasks in the past); perform data similarity authentication on the historical spraying record data, and obtain historical similarity by calculating the similarity between multiple spraying records in the historical spraying record data; if the historical similarity is greater than or equal to a preset similarity threshold, it means that the spraying record data is too single and needs to be supplemented from the cloud database. At this time, a cloud supplement instruction can be generated; according to the cloud supplement instruction, retrieve the cloud spraying record data from the cloud database, and store the cloud spraying record data together with the historical spraying record data in the control scheme library to optimize the control scheme library.

[0023] Further, including:

[0024] The historical spraying record data are subjected to pairwise similarity calculation using a cosine similarity calculation formula to generate a plurality of historical cosine similarities; the plurality of historical cosine similarities are subjected to variance calculation, and the inverse of the calculation result is used as the historical similarity.

[0025] When evaluating the similarity between historical spraying record data and the target spraying task, the cosine similarity calculation formula can be used to calculate the similarity between each pair of historical spraying record data. For example, there are two spraying records in the historical spraying record data, namely A and B, where A and B are both feature vectors (a combination of structural data and spraying requirements). Cosine similarity calculations are performed between each pair of historical records to generate multiple historical cosine similarities. To obtain an overall historical similarity, multiple historical cosine similarities can be aggregated by calculating the average of all historical cosine similarities, then calculating the variance of all historical cosine similarities and the mean, and taking the inverse of the variance as the historical similarity. The larger the variance, the smaller the similarity, indicating that the data quality is better and includes a variety of application scenarios; the smaller the variance, the higher the similarity, indicating that the data is too single and that additional data from the cloud is needed.

[0026] The K hardware pieces in the target spraying task are respectively traced to generate K traceability marks, wherein the K traceability marks are used to record the time when the K hardware pieces perform different spraying processes.

[0027] During the spraying process, hardware components are traced to track their status, time, and any potential quality issues throughout the entire spraying process. Specifically, each of the K hardware components in the target spraying task is assigned a unique traceability identifier, such as a barcode, QR code, or RFID tag. As the hardware components pass through different spraying steps, a recording system (such as a scanning device, RFID reader / writer, etc.) reads each hardware component's traceability identifier and records the start and end time of each step. By generating traceability identifiers for the K hardware components in the target spraying task and recording their time information as they perform different spraying steps, if a quality issue with a hardware component occurs, the process in which the problem occurred can be quickly located.

[0028] The K hardware parts are sprayed in sequence using the initial spraying device, and M finished hardware parts sprayed within a first preset detection window are extracted, and spraying quality inspection is performed on the M finished hardware parts to generate M spraying quality inspection results.

[0029] Using an initial spraying device, K hardware components are sequentially sprayed. After a period of time (e.g., waiting for the coating to dry and cure), M finished hardware components that have been sprayed within a first preset inspection window are extracted. The first preset inspection window is a time range used to determine which hardware components have been sprayed and are ready for quality inspection. The M extracted hardware components are then inspected for coating quality, including coating appearance (e.g., color, gloss, uniformity), coating thickness, coating hardness, and corrosion resistance. M spray quality inspection results are generated.

[0030] Furthermore, if Figure 2 As shown, the spraying quality test is performed on the M finished hardware parts to generate M spraying quality test results, including:

[0031] A surface acquisition device is used to acquire images of the M finished hardware parts to generate M surface image sets; a surface defect identifier is used to identify defect features of the M surface image sets to generate M surface defect feature value sets and M surface defect feature sets, wherein surface defect features correspond one-to-one to surface defect feature values; and the M surface defect feature sets and the M surface defect feature value sets are used as the M spraying quality inspection results.

[0032] Preferably, a surface acquisition device (such as a high-resolution camera, image sensor, etc.) is used to capture multi-angle surface images of M finished hardware products to obtain M surface image sets. A surface defect identifier is a deep learning model trained based on historical defect feature data, used to detect defects in images and extract feature information of the defects. The M surface image sets are input into the surface defect identifier, which then performs defect feature recognition on the surface image of each finished hardware product, outputting M surface defect feature value sets and M surface defect feature sets, where the surface defect features correspond one-to-one to the surface defect feature values, i.e., each defect feature has a corresponding feature value to describe its specific situation. The M surface defect feature sets and M surface defect feature value sets are used as M spraying quality inspection results. The surface acquisition device and surface defect identifier are used to capture surface images and recognize defect features of the finished hardware products after spraying, thereby more accurately evaluating the spraying quality.

[0033] Based on the M traceability identifications of the M hardware finished products, a quality situation analysis is performed on the M spraying quality inspection results to generate abnormal quality analysis results, wherein the abnormal quality analysis results include Q abnormal quality deviation coefficients and Q abnormal surface defect characteristics.

[0034] Hardware spray coating quality issues typically include paint shrinkage, puckering, sagging, pinholes, contamination and blistering, moisture-induced blistering, and uneven color. Paint shrinkage refers to wrinkles or shrinkage on the paint surface during the painting process or the drying process. Puckering refers to bulges formed in certain areas due to expansion of the paint surface during the painting process or the drying process. Sagging refers to localized thickening of the coating, resulting in sags due to gravity. Pinholes are pinholes that appear as sunken, pinhole-like pores on the paint film surface. Contamination and blistering refer to irregular blisters on the paint surface. Moisture-induced blistering refers to evenly distributed, pit-like bubbles of varying sizes, which are more likely to occur in very humid and hot conditions. Different hardware spray coating quality issues are related to different process steps. By linking the M traceability marks of M finished hardware products with their respective spray coating quality test results, we can trace back to the specific production processes and conditions of each hardware piece. A quality situation analysis is performed on M spraying quality inspection results, that is, the quality problems are traced back to the specific process flow to determine the quality problems, and then the abnormal quality analysis results are obtained. The abnormal quality analysis results include Q abnormal quality deviation coefficients and Q abnormal surface defect characteristics.

[0035] Further, including:

[0036] The M surface defect feature sets within the M spraying quality inspection results are aggregated across finished products to generate L aggregated surface defect feature sets, each aggregated surface defect feature corresponding to a hardware finished product, L being the total number of types of surface defect features in the M surface defect feature sets, and L being an integer greater than or equal to 1; based on the one-to-one correspondence between the surface defect features and the surface defect feature values, feature value matching is performed on the L aggregated surface defect feature sets to generate L aggregated surface defect feature value sets; quality situation analysis is performed based on the M traceability identifiers and the L aggregated surface defect feature value sets to generate the abnormal quality analysis result.

[0037] Preferably, the M surface defect feature sets of M hardware finished products are clustered across finished products, that is, the hardware finished products with the same or similar surface defect features are classified into one category, and by comparing the surface defect features of different hardware finished products, the same or similar features are grouped together to form L clustered surface defect feature sets, wherein L is the total number of types of surface defect features in the M surface defect feature sets, and is an integer greater than or equal to 1; based on the one-to-one correspondence between surface defect features and surface defect feature values, for each clustered surface defect feature set, the surface defect feature values ​​of all hardware finished products belonging to the set are searched, and these feature values ​​are clustered together to form corresponding L clustered surface defect feature value sets; quality situation analysis is performed based on the M traceability identifications and the L clustered surface defect feature value sets, each clustered surface defect feature value set is analyzed, and the severity and distribution of its corresponding clustered surface defect features are evaluated, and based on the traceability identification of the hardware finished products, the process flows in which quality defects occur are further identified, and abnormal quality analysis results are generated.

[0038] Further, including:

[0039] Performing abnormal screening on the L aggregated surface defect feature value sets using the L surface defect tolerance values ​​to generate L abnormal aggregated surface defect feature value sets;

[0040] Serializing the L abnormal aggregated surface defect feature value sets according to the M traceability identifiers, and generating L abnormal feature value curves based on the processing results, wherein the abscissa of the abnormal feature value curve is time, the ordinate is the abnormal aggregated surface defect feature value, and the L abnormal feature value curves include L abnormal coordinate point sets;

[0041] L center points of the L abnormal coordinate point sets are extracted, and quality situation analysis is performed based on the L center points and the L abnormal coordinate point sets to generate the abnormal quality analysis result.

[0042] Preferably, L surface defect tolerance values ​​are set according to historical data and industry standards, and these tolerance values ​​correspond to L different surface defect characteristics. Each value in the L aggregated surface defect feature value sets is compared with the corresponding surface defect tolerance value. If a feature value exceeds its corresponding tolerance value, the feature value is considered abnormal. The above comparison is performed on each aggregated surface defect feature value set, and abnormal feature values ​​are screened out to generate L abnormal aggregated surface defect feature value sets; the L abnormal aggregated surface defect feature value sets are serialized according to M traceability identifiers, that is, each abnormal aggregated surface defect feature value set is arranged in chronological order (production process flow), and for each abnormal aggregated surface defect feature value set, an abnormal feature value curve is generated, the horizontal axis of this curve represents time, and the vertical axis represents the abnormal aggregated surface defect feature value, and then L abnormal feature value curves are generated, each curve contains a series of abnormal coordinate point sets, and each coordinate point corresponds to an abnormal aggregated surface defect feature value; L center points (such as the average value of each set) are extracted from the L abnormal coordinate point sets, and quality situation analysis is performed based on the L center points and the L abnormal coordinate point sets to generate abnormal quality analysis results.

[0043] Further, including:

[0044] Determine L first straight lines passing through the L center points, move the L first straight lines according to a preset movement angle to generate L second straight lines; judge whether the L first clustering densities of the L first straight lines are greater than or equal to the L second clustering densities of the L second straight lines, if not, move the L second straight lines according to the preset movement angle to generate L third straight lines; iteratively update the L second straight lines based on the L third straight lines until a preset number of iterations is met, and use the straight line corresponding to the maximum clustering density as the L target straight lines; use the slopes of the L target straight lines as L mass deviation coefficients; judge whether the L mass deviation coefficients are greater than or equal to the preset deviation coefficient, if so, add an abnormal mass deviation coefficient set, wherein the abnormal mass deviation coefficient set includes Q abnormal mass deviation coefficients, Q is a positive integer less than or equal to L; use the surface defect features corresponding to the Q abnormal mass deviation coefficients as Q abnormal surface defect features.

[0045] Preferably, L first straight lines passing through L center points are determined, which can be obtained by a linear fitting method (such as the least squares method), and they respectively represent the trend of the L sets of characteristic values ​​of abnormal aggregated surface defects changing over time; the L first straight lines are moved according to a preset movement angle (such as a fixed angle increment, such as 1°, 5°, etc.) to generate L second straight lines; the L first aggregation densities of the L first straight lines are calculated, that is, the density of abnormal coordinate points on each first straight line. Similarly, the L second aggregation densities of the L second straight lines are also calculated. If any one of the L first aggregation densities is not less than the corresponding second aggregation density, then the linear movement is stopped, that is, the first straight line has provided a better fit. If the aggregation density of the first straight line does not meet the conditions, then the L second straight lines are moved according to the preset movement angle to generate L third straight lines; repeat the above steps of judging the aggregation density, compare the aggregation density of the L second straight lines and the L third straight lines, and select the one with the larger aggregation density as the new reference straight line; perform iterative updates until the preset number of iterations is met. After the iteration is completed, the straight line corresponding to the maximum value of the clustering density is used as L target straight lines; the slopes of the L target straight lines are calculated, and these slopes are used as L quality deviation coefficients, which represent the deviation trend of the spraying quality over time; a preset deviation coefficient is set as a threshold to determine whether the L quality deviation coefficients are greater than or equal to the preset deviation coefficient. If a certain quality deviation coefficient is greater than or equal to the preset deviation coefficient, it is added to the abnormal quality deviation coefficient set, and the abnormal quality deviation coefficient set will contain Q abnormal quality deviation coefficients, where Q is a positive integer less than or equal to L; the surface defect features corresponding to the Q abnormal quality deviation coefficients are used as Q abnormal surface defect features, which are the main reasons for the abnormal deviation of the spraying quality; the Q abnormal quality deviation coefficients and the Q abnormal surface defect features are used as abnormal quality analysis results.

[0046] The control scheme of the initial spraying device is optimized based on the Q abnormal mass deviation coefficients and the Q abnormal surface defect characteristics to obtain an adjustment control scheme, and the parameters of the initial spraying device are adjusted according to the adjustment control scheme to obtain an optimized target spraying device.

[0047] Based on Q abnormal quality deviation coefficients, the trend and degree of spray quality deviation are obtained. At the same time, based on the characteristics of Q abnormal surface defects, the specific impact of these defects on spray quality is determined. Based on the analysis results, the control scheme of the initial spray device is optimized to obtain an adjusted control scheme. According to the adjusted control scheme, the parameters of the initial spray device are adjusted to obtain the optimized target spray device. In short, based on the abnormal quality deviation coefficients and abnormal surface defect characteristics, the control scheme of the initial spray device is optimized and the parameters are adjusted to obtain the optimized target spray device, thereby improving spray quality and production efficiency.

[0048] Further, including:

[0049] A control scheme optimizer is constructed, and the control scheme optimizer is used to identify the Q abnormal quality deviation coefficients, the Q abnormal surface defect features and the initial control scheme to generate the adjusted control scheme.

[0050] Preferably, in order to effectively optimize the control scheme of the initial spraying device, a control scheme optimizer can be constructed based on optimization algorithms such as genetic algorithms, particle swarm optimization algorithms, etc. The control scheme optimizer will use Q abnormal mass deviation coefficients, Q abnormal surface defect characteristics and the initial control scheme as inputs, and find the optimal control scheme through iterative search. In each iteration, the pros and cons of the current control scheme are evaluated according to the fitness function, and a new control scheme candidate set is generated, wherein the fitness function can reflect the requirements of the optimization target, such as the weighted sum of the mass deviation coefficient and the surface defect characteristics. After multiple iterations, until the preset number of iterations is reached, the adjusted control scheme is output.

[0051] In summary, the embodiments of the present application have at least the following technical effects:

[0052] First, the structural data and spraying requirements of the hardware components in the target spraying task are extracted. Next, the initial control scheme configuration of the spraying device is executed, and the parameters of the spraying device are adjusted according to the initial control scheme to obtain a configured initial spraying device. The initial control scheme is obtained by searching the control scheme library based on the structural data and spraying requirements. Furthermore, the K hardware components in the target spraying task are individually traceable, generating K traceability identifiers. These K traceability identifiers are used to record the time it takes for the K hardware components to perform different spraying processes. Then, the initial spraying device is used to sequentially spray the K hardware components, extracting M finished hardware components that have been sprayed within a first preset inspection window. These M finished hardware components are then subjected to spraying quality inspection, generating M spraying quality inspection results. Based on the M traceability identifiers of the M finished hardware components, a quality status analysis is performed on the M spraying quality inspection results to generate abnormal quality analysis results. The abnormal quality analysis results include Q abnormal quality deviation coefficients and Q abnormal surface defect characteristics. Finally, the control scheme for the initial spraying device was optimized based on Q abnormal mass deviation coefficients and Q abnormal surface defect characteristics, resulting in an adjusted control scheme. The parameters of the initial spraying device were then adjusted according to the adjusted control scheme to obtain the optimized target spraying device. This solved the technical problem of reduced spraying quality caused by operational losses in hardware spraying devices in the prior art. By compensating the parameters of the spraying device, the technical effect of improving spraying quality was achieved.

[0053] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0054] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0055] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A control optimization method for a hardware spraying device, characterized in that: The method comprises: Extract the structural data and spraying requirements of the hardware in the target spraying task; Executing an initial control scheme configuration for the spraying device, and adjusting parameters of the spraying device according to the initial control scheme to obtain a configured initial spraying device, wherein the initial control scheme is obtained by searching a control scheme library based on the structural data and the spraying requirements; Perform traceability identification on each of the K hardware pieces in the target spraying task to generate K traceability identifications, wherein the K traceability identifications are used to record the time when the K hardware pieces perform different spraying processes; Using the initial spraying device to spray the K hardware parts in sequence, extracting M finished hardware parts that have been sprayed within a first preset inspection window, performing spraying quality inspection on the M finished hardware parts, and generating M spraying quality inspection results; Based on the M traceability identifications of the M hardware finished products, a quality situation analysis is performed on the M spraying quality inspection results to generate abnormal quality analysis results, wherein the abnormal quality analysis results include Q abnormal quality deviation coefficients and Q abnormal surface defect characteristics; Optimizing a control scheme of an initial spraying device based on the Q abnormal mass deviation coefficients and the Q abnormal surface defect characteristics to obtain an adjusted control scheme, and adjusting parameters of the initial spraying device according to the adjusted control scheme to obtain an optimized target spraying device; Using a surface acquisition device to acquire images of the M finished hardware parts to generate M surface image sets; Using a surface defect identifier to perform defect feature recognition on the M surface image sets, generating M surface defect feature value sets and M surface defect feature sets, wherein the surface defect features correspond one to one to the surface defect feature values; The M surface defect feature sets and the M surface defect feature value sets are used as the M spraying quality detection results; Aggregating the surface defect features across finished products for the M surface defect feature sets in the M spraying quality inspection results to generate L aggregated surface defect feature sets, each aggregated surface defect feature corresponding to a hardware finished product, where L is the total number of types of surface defect features in the M surface defect feature sets, and L is an integer greater than or equal to 1; Based on a one-to-one correspondence between the surface defect features and the surface defect feature values, performing feature value matching on the L aggregated surface defect feature sets to generate L aggregated surface defect feature value sets; Performing a quality situation analysis based on the M traceability identifiers and the L aggregated surface defect feature value sets to generate the abnormal quality analysis result; Performing abnormal screening on the L aggregated surface defect feature value sets using the L surface defect tolerance values ​​to generate L abnormal aggregated surface defect feature value sets; Serializing the L abnormal aggregated surface defect feature value sets according to the M traceability identifiers, and generating L abnormal feature value curves based on the processing results, wherein the abscissa of the abnormal feature value curve is time, the ordinate is the abnormal aggregated surface defect feature value, and the L abnormal feature value curves include L abnormal coordinate point sets; Extracting L center points of the L abnormal coordinate point sets, performing quality situation analysis based on the L center points and the L abnormal coordinate point sets, and generating the abnormal quality analysis result; Determine L first straight lines passing through the L center points, and move the L first straight lines according to a preset movement angle to generate L second straight lines; determining whether the L first clustering densities of the L first straight lines are greater than or equal to the L second clustering densities of the L second straight lines; if not, moving the L second straight lines according to the preset movement angle to generate L third straight lines; Iteratively updating the L second straight lines based on the L third straight lines until a preset number of iterations is met, and taking the straight line corresponding to the maximum value of the clustering density as the L target straight lines; The slopes of the L target straight lines are taken as L quality deviation coefficients; Determine whether the L mass deviation coefficients are greater than or equal to a preset deviation coefficient, and if so, add an abnormal mass deviation coefficient set, wherein the abnormal mass deviation coefficient set includes Q abnormal mass deviation coefficients, where Q is a positive integer less than or equal to L; The surface defect features corresponding to the Q abnormal mass deviation coefficients are used as Q abnormal surface defect features.

2. The control optimization method for a hardware spraying device according to claim 1, characterized in that: include: Extracting historical spraying record data of the spraying device within a historical time, wherein the historical spraying record data includes historical structure data, historical spraying requirements, and historical spraying control schemes; Performing data similarity authentication on the historical spraying record data to obtain historical similarity; If the historical similarity is greater than or equal to a preset similarity threshold, generating a cloud supplementary instruction; According to the cloud supplementary instruction, the cloud spray record data is retrieved, and the cloud spray record data and the historical spray record data are stored in the control solution library.

3. The control optimization method for a hardware spraying device according to claim 1, characterized in that: include: A control scheme optimizer is constructed, and the control scheme optimizer is used to identify the Q abnormal quality deviation coefficients, the Q abnormal surface defect features and the initial control scheme to generate the adjusted control scheme.

4. The control optimization method for a hardware spraying device according to claim 2, characterized in that: include: Performing pairwise similarity calculation on the historical spraying record data using a cosine similarity calculation formula to generate multiple historical cosine similarities; Variance calculation is performed on the multiple historical cosine similarities, and the inverse of the calculation result is used as the historical similarity.

Images