Spraying parameter recommendation and track automatic planning method based on finished product effect driving
By building a process library and using machine learning, the system automatically plans spraying parameters and trajectories, solving the problems of low efficiency and poor accuracy in traditional spraying parameter settings. This enables fast and accurate recommendation of spraying parameters and trajectory optimization, thereby improving spraying quality and production efficiency.
Patent Information
- Application Number
- CN202511003700.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-21
- Publication Date
- 2025-11-21
AI Technical Summary
Traditional spraying parameter settings rely on manual experience, which is inefficient, makes it difficult to achieve global optimization, and results in poor trajectory planning accuracy and real-time performance, making it unable to quickly adapt to the spraying needs of complex workpieces.
By building a process library and acquiring spraying parameters from multiple fields, and utilizing machine learning and simulation technologies, spraying parameters and trajectories are automatically planned. Based on a finished product effect-driven approach, intelligent adjustment and optimization of parameters are achieved.
It enables rapid and accurate recommendation of optimal spraying parameters, reduces trial and error costs, improves parameter matching efficiency, dynamically adapts to workpiece characteristics, and improves spraying quality and production efficiency.
Smart Images
Figure CN120995660A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of spray printing, and particularly relates to a spray parameter recommendation and trajectory automatic planning method based on finished product effect driving. BACKGROUND
[0002] Currently, the setting of spray parameters on the market is mostly artificial according to experience and historical data: referring to mature parameters of similar workpieces and coatings (such as a distance of 200 mm and a flow of 50 ml / min commonly used for furniture spraying), or industry general standards (such as the experience value of "voltage 85, current 70" for electrostatic powder spraying) has great subjectivity. Trial and error iteration: through the "test-detect-adjust" cycle optimization, the efficiency is low (1-2 weeks are needed for complex workpieces), and the traditional method is difficult to find the global optimum due to the synergistic effect of pressure, distance and speed. Environmental and equipment adaptation: the temperature and humidity and equipment state (such as pressure fluctuation caused by spray gun wear) need to be monitored by artificial, which is poor in real-time performance.
[0003] After obtaining the corresponding parameters, the traditional trajectory simulation verification is artificial programming of the spray trajectory by using the robot controller. For complex workpieces, the manual teaching accuracy of the trajectory is poor, the efficiency is low, and the cost is high. In testing whether the corresponding parameters meet the processing conditions, the traditional test needs to test the coverage and film thickness in the real environment by artificial.
[0004] The traditional spray parameter determination is a process of "experience + experiment + iteration", which explores the parameter range by artificial, optimizes gradually by combining the coating characteristics, equipment capacity and environmental conditions, and finally forms a standardized process. Its shortcomings are low efficiency, dependence on artificial and poor dynamic adaptability (as described in the foregoing).
[0005] Therefore, there is an urgent need for a spray parameter recommendation and trajectory automatic planning method based on finished product effect driving. SUMMARY
[0006] Therefore, it is necessary to provide a spray parameter recommendation and trajectory automatic planning method based on finished product effect driving in view of the above technical problems.
[0007] In a first aspect, the application provides a spray parameter recommendation and trajectory automatic planning method based on finished product effect driving, comprising:
[0008] Obtaining a plurality of field spray parameters to form a process library;
[0009] Obtaining a target workpiece spraying effect;
[0010] According to the target workpiece spraying effect, searching in the process library to determine an initial spraying parameter;
[0011] According to the initial spraying parameter, simulating spraying to obtain a simulated workpiece spraying effect;
[0012] According to the difference between the simulation workpiece spraying effect and the target workpiece spraying effect, the initial spraying parameter is adjusted to obtain a target spraying parameter.
[0013] The target workpiece is sprayed by using the target spraying parameter to obtain the target workpiece spraying effect.
[0014] In some modes that can be implemented, the step of obtaining a plurality of field spraying parameters to form a process library comprises:
[0015] Spraying parameters related to each field of surface spraying are obtained and normalized to obtain normalized structured spraying data.
[0016] The normalized structured spraying data is used to form a process library.
[0017] In some modes that can be implemented, the step of obtaining a target workpiece spraying effect comprises:
[0018] The target workpiece spraying effect is parameterized to obtain a plurality of target workpiece parameters.
[0019] In some modes that can be implemented, the step of traversing the process library according to the target workpiece spraying effect to determine an initial spraying parameter comprises:
[0020] A plurality of target workpiece parameters are feature extracted and relation mapped, and normalized to obtain a structured feature vector parameter and mapping of the target workpiece.
[0021] According to the structured feature vector parameter and mapping of the target workpiece, the process library is traversed to determine the closest parameter to obtain an initial spraying parameter.
[0022] In some modes that can be implemented, the step of simulating spraying according to the initial spraying parameter to obtain a simulation workpiece spraying effect comprises:
[0023] A target workpiece model is constructed.
[0024] The target workpiece model is simulated to be sprayed according to the initial spraying parameter to obtain a surface deposition simulation of the target workpiece model.
[0025] A simulation workpiece spraying effect is obtained according to the surface deposition simulation.
[0026] In some modes that can be implemented, the step of adjusting the initial spraying parameter according to the difference between the simulation workpiece spraying effect and the target workpiece spraying effect to obtain a target spraying parameter comprises:
[0027] The difference between the simulated workpiece spraying effect and the target workpiece spraying effect is calculated to obtain a difference report;
[0028] Based on the difference report, the initial spraying parameters are adjusted to obtain the target spraying parameters;
[0029] Based on the target spraying parameters, simulated spraying is performed on the target workpiece model to obtain the spraying result;
[0030] If the spraying result is greater than or equal to the set threshold, then the target spraying parameters are taken as the final result;
[0031] Otherwise, adjust the target spraying parameters again.
[0032] In some feasible methods, after the step of spraying the target workpiece using the target spraying parameters to obtain the spraying effect on the target workpiece, the method includes:
[0033] Based on the coating effect of the target workpiece, obtain the actual coating information of the target workpiece after coating, and construct a correction coefficient based on the actual coating information and the target coating parameters;
[0034] The correction coefficient is stored in the process library.
[0035] Secondly, this application provides a spraying parameter recommendation and trajectory automatic planning system based on finished product effect, applied to the aforementioned spraying parameter recommendation and trajectory automatic planning method based on finished product effect. The system includes:
[0036] The acquisition unit is used to acquire spraying parameters from multiple fields and form a process library.
[0037] The acquisition unit is also used to acquire the coating effect of the target workpiece;
[0038] An initialization unit is used to traverse the process library and determine initial spraying parameters based on the spraying effect of the target workpiece.
[0039] The simulation unit is used to simulate spraying based on the initial spraying parameters to obtain a simulated workpiece spraying effect;
[0040] The target unit is used to adjust the initial spraying parameters based on the difference between the simulated workpiece spraying effect and the target workpiece spraying effect to obtain the target spraying parameters;
[0041] The result unit is used to spray the target workpiece using the target spraying parameters to obtain the spraying effect of the target workpiece.
[0042] In a third aspect, the present application provides a computer storage medium, having stored thereon a computer program, which, when executed by a processor, implements the steps of the method described above.
[0043] In a fourth aspect, the present application provides a computer program, characterized by, when executed by a processor, implementing the steps of the method described above.
[0044] Beneficial effects: The present application provides a spraying parameter recommendation and trajectory automatic planning method based on finished product effect driving, comprising: obtaining a plurality of field spraying parameters to form a process library; obtaining a target workpiece spraying effect; according to the target workpiece spraying effect, traversing in the process library to determine initial spraying parameters; according to the initial spraying parameters, simulating spraying to obtain a simulated workpiece spraying effect; according to the difference between the simulated workpiece spraying effect and the target workpiece spraying effect, adjusting the initial spraying parameters to obtain target spraying parameters; and using the target spraying parameters to spray the target workpiece to obtain the target workpiece spraying effect. Through the above method, the process library integrates the spraying parameters of multiple fields, and realizes deep mining and intelligent analysis of historical data. Compared with traditional manual experience derivation, the optimal process parameters (such as distance, flow, speed, etc.) can be quickly and accurately recommended, the parameter trial and error cost is greatly reduced, and the parameter matching efficiency is improved. BRIEF DESCRIPTION OF DRAWINGS
[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.
[0046] Figure 1 A flowchart of a spraying parameter recommendation and trajectory automatic planning method based on finished product effect driving in an embodiment.
[0047] Figure 2 A logic diagram of a spraying parameter recommendation and trajectory automatic planning method based on finished product effect driving in an embodiment. DETAILED DESCRIPTION
[0048] In order to facilitate the understanding of the present application, the present application will be described more fully below with reference to the related drawings. The drawings show embodiments of the present application. However, the present application can be implemented in many different forms, and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.
[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description herein is for describing particular embodiments only and is not intended to be limiting of the application. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.
[0050] It is to be understood that the terms "first", "second", and the like, used herein do not connote any hierarchy or order, but are used to distinguish one element from another.
[0051] As shown in Figure 1 and Figure 2 In a first aspect, the application provides a spray parameter recommendation and trajectory automatic planning method based on finished product effect driving, the method comprising:
[0052] S100, obtaining a plurality of field spraying parameters to form a process library.
[0053] Specifically, the process library can include the following steps:
[0054] S101, obtaining spraying parameters related to surface spraying in each field, and normalizing to obtain normalized structured spraying data.
[0055] Specifically, the surface spraying fields can include furniture manufacturing (wood paint), automobile painting (metal / plastic parts), aerospace (composite materials), industrial equipment (steel structure corrosion protection), electronic appliances (housing spraying), building decoration (curtain wall / profile), etc.
[0056] Spraying parameters can include distance, pressure, flow rate, and speed, etc.
[0057] After obtaining the spraying parameters of each field, due to the different data structures, the spraying parameters are first normalized to form normalized structured data.
[0058] S102, using the normalized structured spraying data to form a process library.
[0059] Specifically, the structured spraying data can be matched with the substrate, process, paint viscosity, environment, etc. to form a process library.
[0060] Exemplarily, for the spraying parameters of each field, an entity type and relationship framework are constructed to form a knowledge graph.
[0061] The entity types can include substrate, paint, trajectory planning, equipment, film thickness target, and environment, etc. After obtaining these entity types, next, a relationship framework is constructed between the entity types, for example, between the substrate and the paint, there is an adsorption coefficient; between the substrate and the trajectory planning, there is a surface curvature; between the paint and the equipment and the film thickness target, there are corresponding viscosity threshold and curing temperature; between the equipment and the environment, there is a spray width parameter; between the environment and the substrate, there are temperature and humidity restrictions. Thus, the relationship framework between the entity types is formed. It can be understood that the foregoing entities and relationships are only illustratively described, and there are more contents involved in the spraying parameters, and the specific relationship between the entities and the spraying parameters is not limited in the application.
[0062] It should be noted that the construction of the knowledge graph can determine the correlation effect between the parameters, break the data island, realize the cross-field knowledge fusion, and convert the spraying parameters, process rules, and failure cases scattered in various industries (automobiles, furniture, aviation, etc.) into a unified knowledge network, solving the process development problems of small and medium-sized enterprises caused by data deficiency. For example, the high-precision parameters of automobile metal parts are mapped to household plastic parts through the "surface tension-viscosity" physical rule. For another example, the explicit correlation of "substrate-paint-equipment-environment" is established to avoid ignoring key constraints (such as the implicit influence of humidity on electrostatic spraying) in traditional manual trial and error.
[0063] It should also be noted that for each actual spraying result, the knowledge graph is fed back. Specifically, sensor data (such as film thickness, temperature and humidity, spray gun state, etc.) and quality inspection results (such as coverage, defect type) in the spraying process are collected to obtain the actual result; the actual result is compared with the expected effect in the knowledge graph, if the deviation is greater than the threshold, the rule engine framework is triggered to analyze the potential reasons (such as abnormal paint viscosity or unreasonable trajectory planning), and the thrust result is formed; the optimization suggestion is generated according to the thrust result, and the verification result is obtained by simulation; if the deviation of the verification result is less than the threshold, the new rule framework is recorded; according to the new rule framework, the original rule framework is adjusted, for example, the weight of the new rule framework is different from the weight of the original rule framework, the weight of the original rule framework is adjusted by using the weight of the new rule framework, and is stored in the process library. Thus, an autonomous optimization cycle is formed, gradually reducing the need for manual intervention.
[0064] S200, obtaining a target workpiece spraying effect.
[0065] Specifically, obtaining the target workpiece spraying effect can include the following steps:
[0066] The rational effect of the workpiece spraying proposed by the process personnel is taken as the target workpiece spraying effect, and a plurality of target workpiece parameters are obtained by parameter disassembly.
[0067] According to the ideal effect, the corresponding process parameters required to achieve this effect are deduced. For example, the spraying parameters of a certain workpiece are: distance 120 mm, fan pressure mpa, flow 50, speed 300 mm / s, etc.
[0068] In the subsequent step, the target workpiece spraying effect can be reproduced according to a plurality of target workpiece parameters.
[0069] S300, according to the target workpiece spraying effect, traversing the process library to determine the initial spraying parameters.
[0070] The natural language description or sensor data of the target workpiece is converted into a standardized feature vector that the process library can understand, and the physical correlation between the features is established.
[0071] Specifically, determining the initial spraying parameters can include the following steps:
[0072] S301, a plurality of target workpiece parameters are extracted and mapped, and normalized to obtain the structured feature vector parameters and mapping of the target workpiece.
[0073] Exemplarily, the target workpiece parameters are received, such as text expression or form data.
[0074] The key features of the target workpiece parameters are extracted, for example, substrate type: type (metal / plastic / wood), surface morphology (plane / curve); coating requirements: film thickness range, special performance; environmental conditions: temperature range, humidity range; equipment restrictions: spray gun model, nozzle size. Normalization is performed, which is a conventional method and will not be described here.
[0075] Next, the physical correlation mapping between features is performed. Exemplarily, the substrate type affects the coating adsorption characteristics, the surface morphology determines the spray gun movement path, the environmental humidity limits the maximum walkable speed, and the equipment model restricts the pressure adjustable range. According to this correlation, the structured feature vector parameters and mapping of the target workpiece are formed.
[0076] Further exemplarily, the workpiece parameters are as follows (distance 120 mm, fan pressure 0.3 MPa, flow 50 ml / min, speed 300 mm / s), and the features are extracted:
[0077] The first motion control feature group includes the relative motion parameters of the nozzle and the workpiece;
[0078] The second fluid control feature group includes coating delivery control parameters;
[0079] The second environmental constraint feature group includes workshop conditions such as temperature and humidity.
[0080] S302, according to the structured feature vector parameters of the target workpiece and the mapping, in the process library, traversal is performed to determine the closest parameters to obtain initial spraying parameters.
[0081] Exemplarily, according to step S301, process library feature matching retrieval is performed:
[0082] The first level is equipment matching, matching the same spray gun model, if there is no same spray gun model, then according to the parameters of the spray gun, the spray gun with the closest parameters is determined;
[0083] The second level is substrate matching, matching the same substrate, if there is no same substrate, then according to the parameter properties of the substrate, the substrate with the closest parameters is determined;
[0084] The third level is environment compatibility, matching the same substrate, if there is no same environment, then according to the temperature and humidity of the environment, the closest temperature and humidity is determined;
[0085] The fourth level is effect matching, matching the same film thickness, if there is no same film thickness, then according to the parameters of the film thickness, the film thickness with the closest parameters is determined.
[0086] Next, matching evaluation is performed, a score is given to the matching degree of each level to obtain a total score, if the total score is greater than a set threshold, such as the matching degree is greater than 95%, then it is directly used, waiting for the next step processing.
[0087] Exemplarily, with a workpiece spraying parameter of distance 120mm, fan pressure mpa, flow 50, speed 300mm / s.
[0088] The first stage is main parameter screening: screening records in the process library that meet the basic conditions;
[0089] The second stage is core feature matching degree calculation, which quantitatively evaluates the pre-screened records, which can include distance fit degree and flow coordination degree, etc.
[0090] The third stage is environment parameter correction, which implements environment compensation on the optimized records, such as humidity difference compensation. The first stage to the third stage are assigned weights, which can be adjusted according to actual conditions, if the total matching degree of the three stages is greater than or equal to 90%, it can be directly used, if the matching degree is less than 90%, then it waits for the next step processing.
[0091] When the matching degree is less than the set threshold, the parameters are adjusted at a preset pace.
[0092] It should be noted that the process library is loaded into the offline programming software according to the intelligent acquisition of the parameters accumulated in the history.
[0093] S400, according to the initial spraying parameters, simulating spraying to obtain the simulated workpiece spraying effect.
[0094] Specifically, obtaining the simulated workpiece spraying effect can include the following steps:
[0095] S401, constructing a target workpiece model.
[0096] Specifically, the physical characteristics of the target workpiece are converted into a calculable three-dimensional model. The geometric constraints in the initial spraying parameters (distance, speed, etc.) determined in the above step serve as the basis for modeling, without the need for complex algorithms.
[0097] It should be noted that constructing a three-dimensional model is a conventional three-dimensional modeling software. For the target workpiece model, the initial spraying parameters can be used to perform shape constraints as needed.
[0098] After obtaining the target workpiece model, the target workpiece model is meshed, the surface of the target workpiece model is divided into a plurality of grid units of a predetermined size, and each grid unit is marked with position coordinates, surface normal angle, and material characteristics for subsequent verification of the spraying effect.
[0099] S402, according to the initial spraying parameters, simulating spraying on the target workpiece model to obtain a surface deposition simulation of the target workpiece model.
[0100] The coating deposition process is simulated based on the basic principles of fluid dynamics.
[0101] S403, obtaining the simulated workpiece spraying effect according to the surface deposition simulation.
[0102] By predicting coating defects through physical effects, the simulated workpiece spraying effect is obtained, and the spraying effect of each grid unit is obtained according to the simulated workpiece spraying effect.
[0103] It should be noted that for the simulated spraying, the software used is existing simulation software (such as Ansys Fluent, Flow-3D, or Zhongwang Simulation, etc.), and the software is not improved in this application.
[0104] S500, according to the difference between the simulated workpiece spraying effect and the target workpiece spraying effect, adjusting the initial spraying parameters to obtain target spraying parameters.
[0105] Specifically, obtaining the target spraying parameters can include the following steps:
[0106] S501, calculating the difference between the simulated workpiece spraying effect and the target workpiece spraying effect to obtain a difference report.
[0107] Specifically, the spraying effect of each region of the target workpiece model is compared with the spraying effect to be achieved by the target workpiece.
[0108] Exemplarily, film thickness difference analysis: compare the simulated film thickness and the target film thickness, mark the positions exceeding the tolerance range (such as the top R corner film thickness is only 18 μm, lower than the lower limit 20 μm);
[0109] Coverage check: identify the area not sprayed (such as coverage is only 90%), calculate the area ratio.
[0110] Defect positioning: through image comparison, mark the positions of defects such as bubbles and sagging (such as defects are concentrated in the arc surface with a curvature > 0.7).
[0111] According to the above differences, a difference report is formed, which at least includes the position, numerical value and other differences.
[0112] S502, according to the difference report, adjusting the initial spraying parameters to obtain target spraying parameters.
[0113] Specifically, according to the difference report, the cause of the difference is determined. Then, combined with the cause, the initial spraying parameters are adjusted.
[0114] Exemplarily, according to the association between the spraying difference and the cause of forming the difference in the historical data, and the association between the spraying parameter adjustment strategy, a "difference-cause-parameter adjustment scheme" logical chain is formed to form training data, and the training data is used to train the difference model to obtain the target difference model.
[0115] The current difference report is input into the target difference model. The target difference model will first give the cause of the difference. For example, if the film thickness is insufficient in the difference report, the target difference model will give the possible reasons: too fast speed, insufficient flow, too far spraying distance. Next, the parameter adjustment scheme is suggested.
[0116] It should be noted that for spraying, to achieve the same effect, for example, film thickness, the speed can be increased, the spray amount can be increased, or the speed can be reduced, the spray amount can be reduced, so that the film thickness of the same grid unit is the same. That is, the same spraying effect can have multiple adjustment methods, but these many methods have different effects on the running time, cost, etc. of the equipment. Therefore, in the parameter adjustment scheme given by the target difference model, the parameter adjustment scheme with the greatest impact on the problem and the smallest side effect is given priority (such as when the film thickness is insufficient, the flow is increased instead of reducing the speed to avoid affecting the efficiency). That is, the preset conditions for solving the difference, such as the most cost-effective, the fastest, etc. can be used as the preset conditions to sort the parameter adjustment schemes, and the parameter adjustment scheme with the highest matching degree is presented first.
[0117] S503, according to the target spraying parameters, simulating spraying on the target workpiece model to obtain the spraying result;
[0118] If the spraying result is greater than or equal to the set threshold value, the target spraying parameter is taken as the final result.
[0119] Otherwise, the target spraying parameter is adjusted again.
[0120] The recommended parameter is used to define the brush tool, simulate the coverage simulation effect and film thickness simulation effect according to the trajectory planning, and finally export the spraying data analysis, including the spraying coverage of different positions of the workpiece. In the case of no error, the post-processing is carried out, such as on-machine spraying processing; in the case of error, the parameter value is adjusted.
[0121] Specifically, after obtaining the target spraying parameter, verification is also needed, that is, the target workpiece model is simulated again for spraying to form a spraying result. This result can be understood as a difference report. If each index in the difference report is greater than or equal to the set threshold value, the target spraying parameter is taken as the final result, and the target workpiece is sprayed in actual work. Otherwise, the target spraying parameter will be adjusted again. This adjustment can be carried out for multiple rounds, and after multiple rounds of adjustment, the spraying parameter with each index greater than or equal to the set threshold value is obtained.
[0122] Further, after obtaining the values and positions in the foregoing difference report, the grid unit is determined according to the position information. The target spraying parameter is used to spray the grid unit in the difference report to form a grid unit spraying result; and the accuracy of the target spraying parameter is determined according to the grid unit spraying result. In this way, the problem of wasting time caused by the need to spray simulate the complete target workpiece model after updating the target spraying parameter is avoided.
[0123] It should be noted that the target spraying parameter greater than or equal to the set threshold value is loaded into the process library, and the rule relationship of the knowledge graph is adjusted to complete the counter-benefiting of the knowledge graph. That is, the process library is continuously optimized.
[0124] Exemplarily, a factory uses a spraying robot to process an aluminum alloy automobile hub, and the initial spraying parameter is:
[0125] Walking speed: 3500 mm / s, flow rate: 180 ml / min, and overlap rate: 50%. Problem: The simulation result shows that the film thickness of the hub edge is insufficient (only 18 pm, target 25±5 pm).
[0126] Optimization process:
[0127] 1. Adjust the parameters:
[0128] According to the difference model suggestion, the flow rate is increased from 180 ml / min to 200 ml / min (+11%).
[0129] After re-simulation, the edge film thickness meets the standard (26 pm), and there is no new defect.
[0130] 2. Knowledge feedback:
[0131] New rule: update parameter relationship weight: "flow rate to aluminum alloy edge film thickness influence weight" from 0.5 to 0.7 (historical data proves that flow rate is the key factor).
[0132] After feedback, the next time the same type of hub is sprayed, the system directly recommends a flow rate of ≥200 ml / min to avoid repeated trial and error. Through continuous feedback, the process library gradually forms precise rules in the subdivided field.
[0133] S600, using the target spraying parameter, spraying the target workpiece to obtain the target workpiece spraying effect.
[0134] Specifically, after obtaining the target workpiece spraying effect, it further includes:
[0135] S601, according to the target workpiece spraying effect, obtaining the actual spraying information of the target workpiece after spraying, and constructing a correction coefficient according to the actual spraying information and the target spraying parameter.
[0136] S602, store the correction coefficient in the process library.
[0137] Specifically, after spraying the target workpiece according to the target spraying parameter, the paint surface of the target workpiece is detected by using a sensor and other related existing devices to obtain the actual spraying information. Next, according to the actual spraying information and the target spraying parameter, the difference is determined. Then, according to the difference, a correction coefficient is constructed, that is, the difference is converted into a parameter adjustment coefficient. This adjustment coefficient may represent the actual situation of the device, such as the spray gun. Illustratively, when the target spraying parameter has met the requirements of the spraying process, but due to the wear, delay, etc. of the actual device, there is a difference with the ideal spraying effect, but this difference can be quantified by comparing with the target spraying parameter, thereby forming a correction coefficient. And store it in the process library, when related spraying is carried out, the target spraying parameter can be multiplied by the correction coefficient before actual spraying, so as to achieve the target workpiece spraying effect.
[0138] In one embodiment, although S600 can realize the compensation of the subsequent process, it belongs to post-compensation and cannot realize real-time compensation, and therefore further includes the following steps to realize real-time compensation.
[0139] Obtaining original data collected by sensors in a target workpiece spraying process to form an original sensor data set;
[0140] Among them, for the spraying process, the environment temperature and humidity, the spray gun pressure / flow reading, and the coating image stream shot by the industrial camera, the multi-source sensor data timestamps are synchronized through the clock protocol to perform time alignment to form the original sensor data set.
[0141] For example, the change of temperature and humidity in the environment can be collected by a temperature and humidity sensor, the fluctuation amplitude of the spray gun pressure can be collected by a spray gun pressure / flow sensor, and the coating visual features can be collected by an industrial camera.
[0142] Feature extraction is performed on the data in the original sensor data set to obtain a structured feature vector;
[0143] Among them, the data in the original sensor data set is processed as follows:
[0144] Calculate the change rate of the environmental parameters (such as the minute-level fluctuation value of temperature and humidity);
[0145] Analyze the stability of the device state (such as the fluctuation amplitude of the spray gun pressure);
[0146] Analyze the coating visual features (obtain the flow leveling degree and film thickness uniformity through machine vision);
[0147] Through the above processing, a structured feature vector (including environmental, device, and coating features) is obtained, wherein the change rate, stability, and visual features can be calculated through conventional algorithms, and the algorithm is not limited by the present application. For example, the change rate of the environmental parameters can be calculated by time series and difference method, and the difference value between adjacent sampling points of the sensor data with continuous time stamps is calculated to reflect the instantaneous change trend of the parameters; the stability of the device state can be calculated by sliding window statistical process control, and the standard deviation and range of the parameters are calculated in the dynamic time window to quantify the fluctuation intensity of the device; the coating visual features can be analyzed by flow leveling degree analysis algorithm: gray level co-occurrence matrix texture entropy calculation, film thickness uniformity algorithm: sub-pixel edge detection combined with optical interference measurement. All the above algorithms are industrial standard methods, which can be directly realized by calling OpenCV, SciPy, etc.
[0148] Using the structured feature vector, the theoretical feature vector corresponding to the target spraying parameter is compared to obtain the differentiated data;
[0149] According to the differentiated data, the current target spraying parameter is compensated to obtain the real-time compensation of the target spraying parameter.
[0150] That is, according to the target spraying parameter at the previous moment and the differentiated data, the target spraying parameter at the next moment is compensated.
[0151] In summary, the application provides a spraying parameter recommendation and trajectory automatic planning method based on finished product effect driving, which has the following beneficial effects:
[0152] 1. Multi-field parameter accumulation and machine learning fusion
[0153] The process library integrates multi-field spraying parameters, and realizes deep mining and intelligent analysis of historical data by introducing a machine learning model. Compared with traditional manual experience derivation, it can quickly and accurately recommend optimal process parameters (such as distance, flow, speed, etc.), greatly reducing parameter trial and error costs and improving parameter matching efficiency. For example, for different workpiece spraying needs, based on historical cases and learning models, the parameter combination that meets the ideal effects of "uniform coverage, film thickness standard" can be directly output, significantly shortening the process preparation period.
[0154] 2. Dynamic optimization and adaptive adjustment
[0155] Using historical parameter accumulation and intelligent recommendation, the process library can dynamically adjust parameters according to workpiece characteristics (such as shape, material). When simulation verification finds that the parameters are incorrect, the system supports rapid iterative optimization (such as parameter adjustment in the second verification), forming a closed loop of "recommendation-simulation-verification-optimization", ensuring that the parameters always adapt to the actual spraying scene and improving process stability.
[0156] In addition, through coverage simulation and film thickness value simulation, the actual effect of the spraying trajectory in the virtual environment is simulated. For example, for the spraying trajectory planning of a conical workpiece, the film thickness uniformity and coverage completeness can be verified in advance to avoid problems such as missed spraying and thick spraying caused by unreasonable trajectory in actual processing. The data analysis derived from the simulation results (such as the spraying coverage of different positions of the workpiece) provides a quantitative basis for parameter adjustment, realizes "virtual verification first, actual processing later", and reduces material waste and equipment loss.
[0157] Closed-loop verification improves processing quality, and simulation verification and post-processing form a closed loop with actual processing:
[0158] Simulation stage: Through coverage and film thickness simulation, trajectory planning defects (such as whether the film thickness meets the standard under parameters such as "distance 120mm, flow 30, speed 3000mm / s") are found in advance.
[0159] Verification stage: If the data analysis is incorrect, immediately adjust the parameters and re-simulate (second verification) to ensure the matching degree of trajectory planning and parameters.
[0160] Processing stage: The parameters and trajectories verified by simulation are directly used for actual processing, which greatly improves the first-time processing yield, reduces the rework rate, and improves the production efficiency.
[0161] In a second aspect, the application provides a spraying parameter recommendation and trajectory automatic planning system based on product effect driving, which is applied to the spraying parameter recommendation and trajectory automatic planning method based on product effect driving. The system comprises:
[0162] An acquisition unit is configured to acquire spraying parameters in multiple fields to form a process library.
[0163] The acquisition unit is further configured to acquire a target workpiece spraying effect.
[0164] An initialization unit is configured to determine initial spraying parameters by traversing the process library according to the target workpiece spraying effect.
[0165] A simulation unit is configured to simulate spraying according to the initial spraying parameters to obtain a simulated workpiece spraying effect.
[0166] A target unit is configured to adjust the initial spraying parameters according to the difference between the simulated workpiece spraying effect and the target workpiece spraying effect to obtain target spraying parameters.
[0167] A result unit is configured to spray a target workpiece by using the target spraying parameters to obtain the target workpiece spraying effect.
[0168] In a third aspect, the application provides a computer storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the steps of the method.
[0169] In a fourth aspect, the application provides a computer program, wherein the computer program is executed by a processor to implement the steps of the method.
[0170] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0171] The various embodiments in the present disclosure are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments.
[0172] The scope of protection of the present disclosure is not limited to the above-mentioned embodiments. Obviously, those skilled in the art can make various modifications and changes to the present disclosure without departing from the scope and spirit of the present disclosure. If these modifications and changes belong to the scope of the claims of the present disclosure and its equivalent technologies, the present disclosure also includes these modifications and changes.
Claims
1. A method for recommending spraying parameters and automatically planning the trajectory based on the finished product effect, characterized in that the method... include: Acquire spraying parameters from multiple fields and compile them into a process library; Obtain the coating effect on the target workpiece; Based on the target workpiece coating effect, the initial coating parameters are determined by traversing the process library. Based on the initial spraying parameters, simulated spraying is performed to obtain the simulated workpiece spraying effect; Based on the difference between the simulated workpiece spraying effect and the target workpiece spraying effect, the initial spraying parameters are adjusted to obtain the target spraying parameters; Using the target spraying parameters, the target workpiece is sprayed to obtain the target workpiece spraying effect.
2. The method for recommending spraying parameters and automatically planning trajectories based on finished product effect as described in claim 1, characterized in that, The step of acquiring spraying parameters from multiple fields and forming a process library includes: Obtain spraying parameters from various fields involving surface spraying, and normalize them to obtain normalized structured spraying data; A process library is constructed using normalized structured spraying data.
3. The method for recommending spraying parameters and automatically planning trajectories based on finished product effect as described in claim 1, characterized in that, The step of obtaining the coating effect of the target workpiece includes: The coating effect on the target workpiece is analyzed by parameter decomposition to obtain several target workpiece parameters.
4. The method for recommending spraying parameters and automatically planning trajectories based on finished product effect as described in claim 3, characterized in that, The step of determining initial spraying parameters by traversing the process library based on the target workpiece spraying effect includes: For several target workpiece parameters, feature extraction and relation mapping are performed, and normalization is performed to obtain the structured feature vector parameters and mapping of the target workpiece; Based on the structured feature vector parameters and mapping of the target workpiece, the process library is traversed to determine the closest parameters and obtain the initial spraying parameters.
5. The method for recommending spraying parameters and automatically planning trajectories based on finished product effect as described in claim 1, characterized in that, The step of simulating spraying based on the initial spraying parameters to obtain a simulated workpiece spraying effect includes: Construct the target workpiece model; Based on the initial spraying parameters, the target workpiece model is simulated for spraying to obtain a surface deposition simulation of the target workpiece model; Based on the surface deposition simulation, the simulated workpiece coating effect is obtained.
6. The method for recommending spraying parameters and automatically planning trajectories based on finished product effect as described in claim 1, characterized in that, The step of adjusting the initial spraying parameters based on the difference between the simulated workpiece spraying effect and the target workpiece spraying effect to obtain the target spraying parameters includes: The difference between the simulated workpiece spraying effect and the target workpiece spraying effect is calculated to obtain a difference report; Based on the difference report, the initial spraying parameters are adjusted to obtain the target spraying parameters; Based on the target spraying parameters, simulated spraying is performed on the target workpiece model to obtain the spraying result; If the spraying result is greater than or equal to the set threshold, then the target spraying parameters are taken as the final result; Otherwise, adjust the target spraying parameters again.
7. The method for recommending spraying parameters and automatically planning trajectories based on finished product effect as described in claim 1, characterized in that, After the step of spraying the target workpiece using the target spraying parameters to obtain the spraying effect of the target workpiece, the following steps are included: Based on the coating effect of the target workpiece, obtain the actual coating information of the target workpiece after coating, and construct a correction coefficient based on the actual coating information and the target coating parameters; The correction coefficient is stored in the process library.
8. A spraying parameter recommendation and trajectory automatic planning system based on finished product effect-driven approach, characterized in that, The system, applied to the finished product effect-driven spraying parameter recommendation and trajectory automatic planning method according to any one of claims 1-7, comprises: The acquisition unit is used to acquire spraying parameters from multiple fields and form a process library. The acquisition unit is also used to acquire the coating effect of the target workpiece; An initialization unit is used to traverse the process library and determine initial spraying parameters based on the spraying effect of the target workpiece. The simulation unit is used to simulate spraying based on the initial spraying parameters to obtain a simulated workpiece spraying effect; The target unit is used to adjust the initial spraying parameters based on the difference between the simulated workpiece spraying effect and the target workpiece spraying effect to obtain the target spraying parameters; The result unit is used to spray the target workpiece using the target spraying parameters to obtain the spraying effect of the target workpiece.
9. A computer storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.
10. A computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.