A powder spraying detection system based on intelligent contour scanning

Abstract

The present invention discloses a powder spraying detection system based on intelligent contour scanning, which relates to the field of intelligent detection technology. The automated detection and evaluation functions of the system greatly improve production efficiency and reliability. The data detection module monitors the coating surface in real time, uses the intelligent contour scanning device to obtain high-quality image data, and accurately identifies defects after being processed by the deep learning model. The feature acquisition module summarizes the defects and environmental data in real time, and obtains the comprehensive defect coefficient Qxxs and the environmental impact coefficient Yxxs through calculation, providing a basis for subsequent evaluation. The defect evaluation module compares and evaluates the preset defect evaluation threshold Q with the comprehensive defect coefficient Qxxs, and starts the secondary evaluation mechanism according to the comparative evaluation results. If an abnormality is found, the second evaluation mechanism is automatically triggered to provide improvement suggestions. The system not only efficiently and automatically detects the coating, but also combines environmental data for comprehensive evaluation, helping producers to solve problems in a timely manner and improve product quality and production efficiency.

Description

Technical Field

[0001] The invention relates to the technical field of intelligent detection, and in particular to a powder spraying detection system based on intelligent contour scanning. Background Art

[0002] The intelligent contour scanning powder spraying inspection system is part of the field of intelligent manufacturing. By combining intelligent perception and deep learning technology, it innovates the inspection and evaluation of coating quality in the powder spraying process. Powder spraying, as a common coating process, is widely used in the fields of automobile manufacturing, aerospace and metal products. However, the traditional powder spray coating quality inspection method has problems such as reliance on manual judgment, low efficiency and strong subjectivity, which makes it difficult to meet the requirements of efficient, accurate and automated inspection.

[0003] At present, defects in powder spray coatings are often caused by a variety of factors, including abnormalities in spraying equipment, the influence of environmental conditions, and the adjustment of spraying process parameters. Traditional detection methods are often not easy to capture these defects comprehensively and accurately. The intelligent contour scanning powder spray detection system can monitor subtle changes in the coating surface in all directions and discover potential quality problems in advance. At the same time, traditional detection systems can only monitor environmental parameters in a single way, lacking a comprehensive analysis of the relationship between different environmental factors, resulting in an incomplete and in-depth understanding of environmental impacts. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the deficiencies in the prior art, the present invention provides a powder spraying detection system based on intelligent contour scanning, which solves the problems mentioned in the background technology.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a powder spraying detection system based on intelligent contour scanning, including a data detection module, a deep learning module, a feature acquisition module, a defect calculation module and a defect evaluation module;

[0008] The data detection module is used to scan the powder spray coating using an intelligent contour scanning device to obtain image data of the coating surface, and at the same time, a number of sensors are set to perform real-time detection of the powder spray environmental impact data;

[0009] The deep learning module is used to build a deep learning model, perform deep learning on the detected coating surface image data, and crawl a large amount of powder spraying defect image data from the Internet of Things for deep learning, accurately identify the comprehensive defect data of powder spraying, and set the sensor API application interface to receive the spraying environment data in real time;

[0010] The feature acquisition module is used to collect the comprehensive defect data and environmental data identified by the deep learning model in real time, and classify and summarize the collected comprehensive defect data and environmental data to generate a comprehensive defect data set and an environmental impact data set;

[0011] The defect calculation module is used to generate a comprehensive defect coefficient Qxxs and an environmental impact coefficient Yxxs after dimensionless processing based on the collected comprehensive defect data set and environmental impact data set, and to correlate the comprehensive defect coefficient Qxxs and the environmental impact coefficient Yxxs to obtain a comprehensive analysis coefficient Fxxs;

[0012] The defect assessment module is used to perform preliminary comparative assessment on a preset defect assessment threshold Q and the associated comprehensive analysis coefficient Fxxs, and generate an assessment result. The assessment result then triggers a second assessment mechanism. The second assessment mechanism performs secondary comparative assessment on a preset environmental assessment threshold Y and the obtained environmental impact coefficient Yxxs, and generates an assessment result.

[0013] Preferably, the data detection module includes an intelligent contour scanning unit and a sensor detection unit;

[0014] The intelligent contour scanning unit scans the powder spray coating by using an intelligent contour scanning device. Before starting the scanning, a light source is used to illuminate the coating surface to obtain data on the powder spray coating surface. The intelligent contour scanning device includes a laser scanner and an optical sensor. The laser scanner is turned on to emit a laser beam, and three-dimensional data of the coating surface is obtained by reflection and scattering of the coating surface. The optical sensor then uses a camera to capture image data of the coating surface.

[0015] The sensor detection unit is used to deploy an integrated sensor group in a powder spraying workshop to detect environmental impact data during the spraying process in real time. The integrated sensor group includes a temperature sensor, a humidity sensor, a flow sensor, a particle sensor and an electrostatic sensor.

[0016] Preferably, the deep learning module includes a model building unit and a feature recognition unit;

[0017] The model building unit crawls an image dataset containing powder spray coating defects from the Internet of Things through a crawler tool, and the images include crack defects, peeling defects, bubble defects, uneven defects, hole defects and surface uniformity defects, and then preprocesses the data, including resizing, standardization, enhancement and defect feature labeling, and selects a convolutional neural network and an attention mechanism as a deep learning model architecture, divides the prepared image dataset into a training set, a validation set and a test set, uses the training set to train the deep learning model, and verifies it on the validation set, and uses the test set to evaluate the trained deep learning model, and the evaluation indicators include accuracy, precision, recall rate and F1 score, and adjusts the model architecture, optimizes hyperparameters and data preprocessing according to the evaluation results, and after the model training is completed, deploys it in practical applications to monitor the quality of the powder spray coating in real time;

[0018] The feature recognition unit is used to set the API application interface of the intelligent contour scanning device and the API application interface of the sensor, connect to the deep learning model, transmit the acquired real-time image data to the deep learning model through wireless transmission, perform feature recognition on the comprehensive defect data in the coating image scanned by the deep learning model, and denoise, clean and normalize the environmental impact data detected by the sensor.

[0019] Preferably, the feature acquisition module includes a first acquisition unit and a second acquisition unit;

[0020] The first acquisition unit is used to classify and summarize the identified comprehensive defect data to generate a comprehensive defect data set, wherein the comprehensive defect data includes a surface defect data set, a coating quality defect data set, and a serious defect data set;

[0021] The surface defect data set includes spray surface curvature Bql, surface roughness Bcc, coating coverage Bfg, coating thickness Bhd and coating bonding strength Bjh;

[0022] The coating quality defect data set includes bubble diameter Tzj, bubble number Tgs, hole area Tkd, particle uniformity Tjy and pit area Tmd;

[0023] The severe defect data set includes crack length Ycd, crack number Ysl, spalling area Ymj, spalling number Ybs and total spraying area Yzm;

[0024] The second acquisition unit is used to classify and summarize the environmental impact data detected by the sensor group to generate an environmental impact data set, and the environmental impact data set includes spraying temperature Hwd, spraying humidity Hsd, air flow rate Hld, air particle concentration Hkl and static charge quantity Hjd.

[0025] Preferably, the defect calculation module includes a defect calculation unit, an environment calculation unit and a comprehensive calculation unit;

[0026] The defect calculation unit includes a surface defect calculation unit, a coating quality calculation unit and a serious defect calculation unit;

[0027] The surface defect calculation unit is used to obtain the surface defect coefficient Bmxs by analyzing and calculating the surface defect data set in the comprehensive defect data set after dimensionless processing;

[0028] The coating quality calculation unit is used to obtain the coating quality coefficient Tcxs by analyzing and calculating the coating quality defect data set in the comprehensive defect data set after dimensionless processing;

[0029] The severe defect calculation unit is used to perform analysis and calculation to obtain the severe defect coefficient Yzxs based on the severe defect data set in the comprehensive defect data set after dimensionless processing;

[0030] The surface defect coefficient Bmxs, coating quality coefficient Tcxs and severe defect coefficient Yzxs are obtained by the following formulas:

[0031]

[0032]

[0033]

[0034] Preferably, the environmental calculation unit is used to perform dimensionless processing on the environmental impact data set, summarize the calculation processing, and obtain the environmental impact coefficient Yxxs;

[0035] The environmental impact coefficient Yxxs is obtained by the following formula:

[0036]

[0037] Preferably, the comprehensive calculation unit includes a summarizing unit and an associating unit;

[0038] The summary unit is used to perform dimensionless processing on the obtained surface defect coefficient Bmxs, coating quality coefficient Tcxs and severe defect coefficient Yzxs, and then perform summary calculation to obtain the comprehensive defect coefficient Qxxs;

[0039] The comprehensive defect coefficient Qxxs is obtained by the following formula:

[0040]

[0041] Wherein, a1, a2 and a3 are the surface defect coefficient Bmxs, coating quality coefficient Tcxs and serious defect coefficient Yzxs respectively, and a1+a2+a3≠0, 0≤a1≤0.64, 0≤a2≤0.72, 0≤a3≤0.51, and the specific values ​​are adjusted and set by the user, and A is the first correction constant.

[0042] Preferably, the correlation unit is used to perform dimensionless processing on the comprehensive defect coefficient Qxxs obtained by the summary unit and the environmental impact coefficient Yxxs obtained by the environmental calculation unit, and then perform comprehensive correlation calculation to obtain the comprehensive analysis coefficient Fxxs;

[0043] The comprehensive analysis coefficient Fxxs is obtained by the following formula:

[0044]

[0045] Wherein, b1 and b2 represent the preset proportional coefficients of the comprehensive defect coefficient Qxxs and the environmental impact coefficient Yxxs respectively, and b1≠b2, b2≠0, 0≤b1≤0.64, 0<b2≤0.36, and the specific value is adjusted and set by the user, and B is the second correction constant.

[0046] Preferably, the defect assessment module comprises a first assessment unit and a second assessment unit;

[0047] The first evaluation unit is used to preset a defect evaluation threshold Q, and conduct a preliminary comparative evaluation with the obtained comprehensive analysis coefficient Fxxs. The specific evaluation results are as follows;

[0048] When the comprehensive analysis coefficient Fxxs>defect assessment threshold Q, it means that the powder coating currently obtained by intelligent profile scanning has defects. At this time, it is necessary to start the second assessment mechanism to analyze the cause of the current defect;

[0049] When the comprehensive analysis coefficient Fxxs ≤ the defect assessment threshold Q, it indicates that the powder spraying currently obtained by the intelligent profile scanning is in a normal state.

[0050] Preferably, the second evaluation unit is used to perform a secondary comparative evaluation on the preset environmental evaluation threshold Y and the obtained environmental impact coefficient Yxxs, and start the second evaluation mechanism, which is triggered by the first evaluation unit and generates a corresponding evaluation result. The specific evaluation scheme is as follows;

[0051] When the environmental impact coefficient Yxxs> environmental assessment threshold Y, it means that the current powder spraying is abnormal due to environmental factors. At this time, choose powder coatings with good quality and suitable for specific environments, including the weather resistance and anti-pollution ability of the coating. At the same time, use an air purifier to improve the spraying environment, and use appropriate surface treatment technology before spraying;

[0052] When the environmental impact coefficient Yxxs ≤ the environmental assessment threshold Y, it means that there is an abnormality in the current powder spraying and the environment is in a normal state. This means that there is an abnormality in the spraying equipment, which leads to coating defects in the spraying process. At this time, it is necessary to check and adjust the parameters of the spraying equipment, including spray gun pressure, nozzle cleanliness and powder flow rate, use high-quality spraying materials and accessories, and optimize the occurrence of defects.

[0053] (III) Beneficial effects

[0054] The present invention provides a powder spraying detection system based on intelligent contour scanning. It has the following beneficial effects:

[0055] (1) The automated detection and evaluation functions of the system make the production process more efficient and reliable. First, the data detection module is responsible for real-time monitoring of the coating surface and uses intelligent contour scanning equipment to obtain high-quality image data. These data are processed by the deep learning module. After a lot of training and verification, an accurate deep learning model for defect recognition is established. The deep learning model can not only identify common surface defects, but also detect subtle defects and other problems. At the same time, the feature acquisition module collects and summarizes the comprehensive defect data and environmental impact data of the coating surface in real time. These data are processed by the defect calculation module and the environmental calculation unit to obtain the comprehensive defect coefficient Qxxs and environmental impact data. The impact coefficient Yxxs reflects the coating quality and spraying environment, and provides an important basis for subsequent evaluation. The comprehensive evaluation process is completed by the defect evaluation module, which compares and evaluates the preset defect evaluation threshold Q and environmental evaluation threshold Y with the comprehensive analysis coefficient Fxxs and the environmental impact coefficient Yxxs. Once defects or abnormalities are detected in the coating, the system will automatically trigger the second evaluation mechanism. The second evaluation unit provides corresponding improvement suggestions based on the actual situation. The system not only realizes efficient and automated detection of powder spray coatings, but also can conduct comprehensive evaluation in combination with environmental data, helping producers to discover and solve problems in a timely manner, and ensure continuous improvement of product quality and production efficiency.

[0056] (2) The system also has the function of abnormal handling and improvement. When defects are detected in the coating, the system can automatically trigger the second evaluation mechanism based on the comprehensive analysis coefficient and the environmental impact coefficient to analyze and judge the cause of the defect. Based on the evaluation results provided by the second evaluation unit, the producer can take corresponding measures according to different situations, such as adjusting the parameters of the spraying equipment, improving the spraying environment conditions or optimizing the selection of spraying materials, so as to solve the problem in time and improve production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1A schematic diagram of a powder spraying detection system based on intelligent contour scanning according to the present invention; DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] Example 1

[0060] See also Figure 1 , the present invention provides a powder spraying detection system based on intelligent contour scanning, including a data detection module, a deep learning module, a feature acquisition module, a defect calculation module and a defect evaluation module;

[0061] The data detection module is used to scan the powder spray coating using an intelligent contour scanning device to obtain image data of the coating surface, and at the same time, a number of sensors are set to perform real-time detection of the powder spray environmental impact data;

[0062] The deep learning module is used to build a deep learning model, perform deep learning on the detected coating surface image data, and crawl a large amount of powder spraying defect image data from the Internet of Things for deep learning, accurately identify the comprehensive defect data of powder spraying, and set the sensor API application interface to receive the spraying environment data in real time;

[0063] The feature acquisition module is used to collect the comprehensive defect data and environmental data identified by the deep learning model in real time, and classify and summarize the collected comprehensive defect data and environmental data to generate a comprehensive defect data set and an environmental impact data set;

[0064] The defect calculation module is used to generate a comprehensive defect coefficient Qxxs and an environmental impact coefficient Yxxs after dimensionless processing based on the collected comprehensive defect data set and environmental impact data set, and to correlate the comprehensive defect coefficient Qxxs and the environmental impact coefficient Yxxs to obtain a comprehensive analysis coefficient Fxxs;

[0065] The defect assessment module is used to perform preliminary comparative assessment on a preset defect assessment threshold Q and the associated comprehensive analysis coefficient Fxxs, and generate an assessment result. The assessment result then triggers a second assessment mechanism. The second assessment mechanism performs secondary comparative assessment on a preset environmental assessment threshold Y and the obtained environmental impact coefficient Yxxs, and generates an assessment result.

[0066] In this embodiment, based on the five modules, this system provides a comprehensive solution for powder spray coating quality monitoring and evaluation. First, the data detection module uses intelligent contour scanning equipment and sensors to obtain coating surface image data and environmental impact data in real time, which provides a basis for subsequent processing. Secondly, the deep learning module realizes the accurate identification of comprehensive defects of powder spray coating by constructing a deep learning model and crawling a large amount of defect image data from the Internet of Things. Compared with traditional methods, the recognition accuracy and efficiency are greatly improved. The feature acquisition module collects and classifies the identified defect data and environmental data in real time, providing comprehensive data support for subsequent defect evaluation. The defect calculation module effectively combines the comprehensive defect coefficient and the environmental impact coefficient through dimensionless processing and comprehensive analysis, which improves the accuracy and reliability of the evaluation results. Finally, the defect evaluation module performs a preliminary evaluation based on a preset evaluation threshold, and combines the second evaluation mechanism for a secondary comparative evaluation, making the evaluation results more comprehensive and credible. Overall, the system realizes all-round monitoring and evaluation of the quality of powder spray coating, improves the evaluation accuracy and efficiency, and provides reliable quality assurance for industrial production. Compared with current technical means, it significantly improves the accuracy and reliability, and further promotes the optimization and improvement of powder spraying process.

[0067] Example 2

[0068] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the data detection module includes an intelligent contour scanning unit and a sensor detection unit;

[0069] The intelligent contour scanning unit scans the powder spray coating by using an intelligent contour scanning device. Before starting the scanning, a light source is used to illuminate the coating surface to obtain data on the powder spray coating surface. The intelligent contour scanning device includes a laser scanner and an optical sensor. The laser scanner is turned on to emit a laser beam, and three-dimensional data of the coating surface is obtained by reflection and scattering of the coating surface. The optical sensor then uses a camera to capture image data of the coating surface.

[0070] The sensor detection unit is used to deploy an integrated sensor group in a powder spraying workshop to detect environmental impact data during the spraying process in real time. The integrated sensor group includes a temperature sensor, a humidity sensor, a flow sensor, a particle sensor and an electrostatic sensor.

[0071] In this embodiment, the data detection module realizes comprehensive monitoring of the powder spray coating and its surrounding environment through the combination of the intelligent contour scanning unit and the sensor detection unit. The intelligent contour scanning unit uses a laser scanner and an optical sensor to obtain the three-dimensional data and image data of the coating surface, effectively capturing the subtle changes in the coating surface, and providing a rich data basis for the subsequent deep learning model. At the same time, the sensor detection unit deploys an integrated sensor group in the spraying workshop to detect the environmental impact data during the spraying process in real time, so as to fully grasp the changes in the spraying environment. This comprehensive monitoring method can not only detect defects on the coating surface in a timely manner, but also effectively evaluate the impact of various environmental factors on the coating quality during the spraying process. Therefore, the beneficial effect of the data detection module is to improve the comprehensiveness and accuracy of monitoring, provide reliable data support for subsequent deep learning models and defect assessments, and help improve the quality control level of the powder spraying process.

[0072] Example 3

[0073] This embodiment is explained in Example 2. Please refer to Figure 1 ,Specifically: the deep learning module includes a model building unit and a feature recognition unit;

[0074] The model building unit crawls an image dataset containing powder spray coating defects from the Internet of Things through a crawler tool, and the images include crack defects, peeling defects, bubble defects, uneven defects, hole defects and surface uniformity defects, and then preprocesses the data, including resizing, standardization, enhancement and defect feature labeling, and selects a convolutional neural network and an attention mechanism as a deep learning model architecture, divides the prepared image dataset into a training set, a validation set and a test set, uses the training set to train the deep learning model, and verifies it on the validation set, and uses the test set to evaluate the trained deep learning model, and the evaluation indicators include accuracy, precision, recall rate and F1 score, and adjusts the model architecture, optimizes hyperparameters and data preprocessing according to the evaluation results, and after the model training is completed, deploys it in practical applications to monitor the quality of the powder spray coating in real time;

[0075] The feature recognition unit is used to set the API application interface of the intelligent contour scanning device and the API application interface of the sensor, connect to the deep learning model, transmit the acquired real-time image data to the deep learning model through wireless transmission, perform feature recognition on the comprehensive defect data in the coating image scanned by the deep learning model, and denoise, clean and normalize the environmental impact data detected by the sensor.

[0076] In this embodiment, the model building unit and feature recognition unit of the deep learning module are combined to provide powerful analysis and recognition capabilities for powder spray coating quality monitoring. First, the model building unit obtains a large amount of powder spray coating defect image data from the Internet of Things through a crawler tool, and performs preprocessing and feature marking, and uses deep learning architectures such as convolutional neural networks and attention mechanisms to build an efficient deep learning model. After training, verification and testing, these models have achieved good accuracy, precision, recall and F1 scores in evaluation indicators, providing a reliable model basis for real-time monitoring. Secondly, the feature recognition unit transmits the coating image data and environmental impact data acquired in real time to the deep learning model through the API application interface of the intelligent contour scanning device and sensor. The deep learning model realizes the comprehensive analysis and recognition of coating quality by identifying the comprehensive defect data in the coating image and combining the environmental impact data detected by the sensor. At the same time, the environmental impact data detected by the sensor is denoised, cleaned and normalized to improve the quality and reliability of the data. Therefore, the beneficial effect of the deep learning module is to improve the accurate recognition ability of powder spray coating defects and the comprehensive analysis ability of environmental impacts, providing an efficient and reliable solution for real-time monitoring. Compared with traditional methods, the introduction of this module makes the monitoring process more intelligent and precise, and greatly improves the real-time and efficiency of monitoring, thus bringing more reliable quality assurance to industrial production.

[0077] Example 4

[0078] This embodiment is explained in Example 3, please refer to Figure 1 ,Specifically: the feature acquisition module includes a first acquisition unit and a second acquisition unit;

[0079] The first acquisition unit is used to classify and summarize the identified comprehensive defect data to generate a comprehensive defect data set, wherein the comprehensive defect data includes a surface defect data set, a coating quality defect data set, and a serious defect data set;

[0080] The surface defect data set includes spray surface curvature Bql, surface roughness Bcc, coating coverage Bfg, coating thickness Bhd and coating bonding strength Bjh;

[0081] The coating quality defect data set includes bubble diameter Tzj, bubble number Tgs, hole area Tkd, particle uniformity Tjy and pit area Tmd;

[0082] The severe defect data set includes crack length Ycd, crack number Ysl, spalling area Ymj, spalling number Ybs and total spraying area Yzm;

[0083] The second acquisition unit is used to classify and summarize the environmental impact data detected by the sensor group to generate an environmental impact data set, and the environmental impact data set includes spraying temperature Hwd, spraying humidity Hsd, air flow rate Hld, air particle concentration Hkl and static charge quantity Hjd.

[0084] In this embodiment, the beneficial effect of the two acquisition units of the feature acquisition module is that they provide a comprehensive and detailed data basis for subsequent defect calculation and evaluation. By classifying and summarizing the comprehensive coating defects and environmental impact data, the system can more accurately analyze coating quality problems and make corresponding adjustments and improvements in a timely manner, thereby improving the stability of the production process and quality controllability.

[0085] Example 5

[0086] This embodiment is explained in Example 4. Please refer to Figure 1 ,Specifically: the defect calculation module includes a defect calculation unit, an environment calculation unit and a comprehensive calculation unit;

[0087] The defect calculation unit includes a surface defect calculation unit, a coating quality calculation unit and a serious defect calculation unit;

[0088] The surface defect calculation unit is used to obtain the surface defect coefficient Bmxs by analyzing and calculating the surface defect data set in the comprehensive defect data set after dimensionless processing;

[0089] The coating quality calculation unit is used to obtain the coating quality coefficient Tcxs by analyzing and calculating the coating quality defect data set in the comprehensive defect data set after dimensionless processing;

[0090] The severe defect calculation unit is used to perform analysis and calculation to obtain the severe defect coefficient Yzxs based on the severe defect data set in the comprehensive defect data set after dimensionless processing;

[0091] The surface defect coefficient Bmxs, coating quality coefficient Tcxs and severe defect coefficient Yzxs are obtained by the following formulas:

[0092]

[0093]

[0094]

[0095] The environmental calculation unit is used to perform dimensionless processing on the environmental impact data set, summarize the calculation process, and obtain the environmental impact coefficient Yxxs;

[0096] The environmental impact coefficient Yxxs is obtained by the following formula:

[0097]

[0098] In this embodiment, the defect calculation module realizes the comprehensive analysis and evaluation of powder spray coating defects and environmental impacts through the combination of defect calculation unit, environmental calculation unit and comprehensive calculation unit. First, the defect calculation unit includes a surface defect calculation unit, a coating quality calculation unit and a severe defect calculation unit, which respectively calculate the coating surface, coating quality and severe defects. By calculating various defect data in the comprehensive defect data set, the surface defect coefficient Bmxs, coating quality coefficient Tcxs and severe defect coefficient Yzxs are obtained respectively. These coefficients reflect the degree and severity of coating defects and provide important indicators for subsequent comprehensive evaluation. Secondly, the environmental calculation unit calculates and processes according to the environmental impact data set to obtain the environmental impact coefficient Yxxs. This coefficient reflects the degree of influence of environmental conditions on coating quality by comprehensively analyzing and processing the key parameters in the environmental impact data set. This comprehensive calculation method can more comprehensively consider the influence of environmental factors on coating quality and provide an important basis for subsequent defect evaluation. The comprehensive calculation unit correlates the various coefficients obtained by the above calculation to obtain the comprehensive analysis coefficient Fxxs. This coefficient reflects the comprehensive relationship between coating defects and environmental impacts and comprehensively evaluates the coating quality. Therefore, the beneficial effect of the defect calculation module is that it realizes the comprehensive evaluation of coating defects and environmental factors, providing a comprehensive and reliable quality control method for industrial production. Through the comprehensive analysis of defects and environmental factors, problems can be discovered in time and corresponding measures can be taken, thus improving the stability and quality level of the production process.

[0099] Example 6

[0100] This embodiment is explained in Example 5. Please refer to Figure 1 , specifically: the comprehensive calculation unit includes a summary unit and an associated unit;

[0101] The summary unit is used to perform dimensionless processing on the obtained surface defect coefficient Bmxs, coating quality coefficient Tcxs and severe defect coefficient Yzxs, and then perform summary calculation to obtain the comprehensive defect coefficient Qxxs;

[0102] The comprehensive defect coefficient Qxxs is obtained by the following formula:

[0103]

[0104] Wherein, a1, a2 and a3 are the surface defect coefficient Bmxs, coating quality coefficient Tcxs and serious defect coefficient Yzxs respectively, and a1+a2+a3≠0, 0≤a1≤0.64, 0≤a2≤0.72, 0≤a3≤0.51, and the specific values ​​are adjusted and set by the user, and A is the first correction constant.

[0105] The correlation unit is used to perform dimensionless processing on the comprehensive defect coefficient Qxxs obtained by the summary unit and the environmental impact coefficient Yxxs obtained by the environmental calculation unit, and then perform comprehensive correlation calculation to obtain the comprehensive analysis coefficient Fxxs;

[0106] The comprehensive analysis coefficient Fxxs is obtained by the following formula:

[0107]

[0108] Wherein, b1 and b2 represent the preset proportional coefficients of the comprehensive defect coefficient Qxxs and the environmental impact coefficient Yxxs respectively, and b1≠b2, b2≠0, 0≤b1≤0.64, 0<b2≤0.36, and the specific value is adjusted and set by the user, and B is the second correction constant.

[0109] In this embodiment, the comprehensive calculation method can fully consider the situation that the coating quality is affected by defects and environmental factors, making the evaluation result more accurate and reliable. At the same time, the user can adjust the correction constant according to the specific situation and flexibly control the evaluation weight so that the evaluation result is more in line with the actual production needs. Therefore, the beneficial effect of the comprehensive calculation unit is to improve the comprehensive evaluation ability of the coating quality, provide a reliable basis for the quality control of the production process, and further improve the stability and quality level of industrial production.

[0110] Example 7

[0111] This embodiment is explained in Example 6. Please refer to Figure 1 ,Specifically: the defect assessment module includes a first assessment unit and a second assessment unit;

[0112] The first evaluation unit is used to preset a defect evaluation threshold Q, and conduct a preliminary comparative evaluation with the obtained comprehensive analysis coefficient Fxxs. The specific evaluation results are as follows;

[0113] When the comprehensive analysis coefficient Fxxs>defect assessment threshold Q, it means that the powder coating currently obtained by intelligent profile scanning has defects. At this time, it is necessary to start the second assessment mechanism to analyze the cause of the current defect;

[0114] When the comprehensive analysis coefficient Fxxs ≤ the defect assessment threshold Q, it indicates that the powder spraying currently obtained by the intelligent profile scanning is in a normal state.

[0115] The second evaluation unit is used to perform a secondary comparative evaluation between the preset environmental evaluation threshold value Y and the obtained environmental impact coefficient Yxxs, and start the second evaluation mechanism. The second evaluation mechanism is triggered by the first evaluation unit and generates a corresponding evaluation result. The specific evaluation scheme is as follows;

[0116] When the environmental impact coefficient Yxxs> environmental assessment threshold Y, it means that the current powder spraying is abnormal due to environmental factors. At this time, choose powder coatings with good quality and suitable for specific environments, including the weather resistance and anti-pollution ability of the coating. At the same time, use an air purifier to improve the spraying environment, and use appropriate surface treatment technology before spraying;

[0117] When the environmental impact coefficient Yxxs ≤ the environmental assessment threshold Y, it means that there is an abnormality in the current powder spraying and the environment is in a normal state. This means that there is an abnormality in the spraying equipment, which leads to coating defects in the spraying process. At this time, it is necessary to check and adjust the parameters of the spraying equipment, including spray gun pressure, nozzle cleanliness and powder flow rate, use high-quality spraying materials and accessories, and optimize the occurrence of defects.

[0118] In this embodiment, this dual evaluation mechanism effectively combines the comprehensive influence of coating quality and environmental factors, providing an important reference for accurately identifying and solving quality problems in the spraying process. Through the comprehensive evaluation of coating quality and environmental factors, it is possible to more effectively locate and solve abnormal situations that may occur in the spraying process, thereby improving production efficiency and coating quality stability. Therefore, the beneficial effect of this evaluation module is that it provides a comprehensive and reliable quality evaluation mechanism, which provides important support for the optimization and improvement of the powder spraying process.

[0119] Specific example:

[0120] Data collection:

[0121] Spraying surface curvature Bql: 3.74, surface roughness Bcc: 8.21, coating coverage Bfg: 5.69, coating thickness Bhd: 2.56, coating bonding strength Bjh: 9.83;

[0122] Bubble diameter Tzj: 6.47, bubble number Tgs: 4.12, hole area Tkd: 7.98, particle uniformity Tjy: 5.27, pit area Tmd: 2.08;

[0123] Crack length Ycd: 3.15, crack number Ysl: 7.42, peeling area Ymj: 4.89, peeling number Ybs: 6.58, total spraying area Yzm: 1.99;

[0124] Spraying temperature Hwd: 8.03, spraying humidity Hsd: 3.61, air flow rate Hld: 5.87, air particle concentration Hkl: 2.17, static charge quantity Hjd: 6.74;

[0125] a1=0.04, a2=0.27, a3=0.02, b1=0.12, b2=0.34, first correction constant A: 0.05, second correction constant B: 0.07;

[0126] Surface defect coefficient Bmxs:

[0127] Coating quality factor Tcxs:

[0128] Severe defect coefficient Yzxs:

[0129] Environmental impact coefficient Yxxs:

[0130] Comprehensive defect coefficient Qxxs:

[0131] Comprehensive analysis coefficient Fxxs:

[0132] All data values ​​are retained to two decimal places;

[0133] When the defect evaluation threshold Q is set to 1 and the comprehensive analysis coefficient Fxxs ≤ the defect evaluation threshold Q, it indicates that the powder coating currently obtained by the intelligent profile scanning is in a normal state.

[0134] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A powder spraying detection system based on intelligent contour scanning, characterized in that: It includes data detection module, deep learning module, feature acquisition module, defect calculation module and defect assessment module; The data detection module is used to scan the powder spray coating using an intelligent contour scanning device to obtain image data of the coating surface, and at the same time, a number of sensors are set to perform real-time detection of the powder spray environmental impact data; The deep learning module is used to build a deep learning model, perform deep learning on the detected coating surface image data, and crawl a large amount of powder spraying defect image data from the Internet of Things for deep learning, accurately identify the comprehensive defect data of powder spraying, and set the sensor API application interface to receive the spraying environment data in real time; The feature acquisition module is used to collect the comprehensive defect data and environmental data identified by the deep learning model in real time, and classify and summarize the collected comprehensive defect data and environmental data to generate a comprehensive defect data set and an environmental impact data set; The feature acquisition module includes a first acquisition unit and a second acquisition unit; The first acquisition unit is used to classify and summarize the identified comprehensive defect data to generate a comprehensive defect data set, wherein the comprehensive defect data includes a surface defect data set, a coating quality defect data set, and a serious defect data set; The surface defect data set includes spray surface curvature Bql, surface roughness Bcc, coating coverage Bfg, coating thickness Bhd and coating bonding strength Bjh; The coating quality defect data set includes bubble diameter Tzj, bubble number Tgs, hole area Tkd, particle uniformity Tjy and pit area Tmd; The severe defect data set includes crack length Ycd, crack number Ysl, spalling area Ymj, spalling number Ybs and total spraying area Yzm; The second acquisition unit is used to classify and summarize the environmental impact data detected by the sensor group to generate an environmental impact data set, wherein the environmental impact data set includes spraying temperature Hwd, spraying humidity Hsd, air flow rate Hld, air particle concentration Hkl and static charge quantity Hjd; The defect calculation module is used to generate a comprehensive defect coefficient Qxxs and an environmental impact coefficient Yxxs after dimensionless processing based on the collected comprehensive defect data set and environmental impact data set, and to correlate the comprehensive defect coefficient Qxxs and the environmental impact coefficient Yxxs to obtain a comprehensive analysis coefficient Fxxs; The defect calculation module includes a defect calculation unit, an environment calculation unit and a comprehensive calculation unit; The defect calculation unit includes a surface defect calculation unit, a coating quality calculation unit and a serious defect calculation unit; The surface defect calculation unit is used to obtain the surface defect coefficient Bmxs by analyzing and calculating the surface defect data set in the comprehensive defect data set after dimensionless processing; The coating quality calculation unit is used to obtain the coating quality coefficient Tcxs by analyzing and calculating the coating quality defect data set in the comprehensive defect data set after dimensionless processing; The severe defect calculation unit is used to perform analysis and calculation to obtain the severe defect coefficient Yzxs based on the severe defect data set in the comprehensive defect data set after dimensionless processing; The surface defect coefficient Bmxs, coating quality coefficient Tcxs and severe defect coefficient Yzxs are obtained by the following formulas: The environmental calculation unit is used to perform dimensionless processing on the environmental impact data set, summarize the calculation process, and obtain the environmental impact coefficient Yxxs; The environmental impact coefficient Yxxs is obtained by the following formula: The comprehensive calculation unit includes a summarizing unit and an associating unit; The summary unit is used to perform dimensionless processing on the obtained surface defect coefficient Bmxs, coating quality coefficient Tcxs and severe defect coefficient Yzxs, and then perform summary calculation to obtain the comprehensive defect coefficient Qxxs; The comprehensive defect coefficient Qxxs is obtained by the following formula: Wherein, a1, a2 and a3 represent the preset proportional coefficients of the surface defect coefficient Bmxs, the coating quality coefficient Tcxs and the serious defect coefficient Yzxs respectively, and their specific values ​​are adjusted and set by the user, and A is the first correction constant; The correlation unit is used to perform dimensionless processing on the comprehensive defect coefficient Qxxs obtained by the summary unit and the environmental impact coefficient Yxxs obtained by the environmental calculation unit, and then perform comprehensive correlation calculation to obtain the comprehensive analysis coefficient Fxxs; The comprehensive analysis coefficient Fxxs is obtained by the following formula: Wherein, b1 and b2 represent the preset proportional coefficients of the comprehensive defect coefficient Qxxs and the environmental impact coefficient Yxxs respectively, and their specific values ​​are adjusted and set by the user, and B is the second correction constant; The defect assessment module is used to perform preliminary comparative assessment on a preset defect assessment threshold Q and the associated comprehensive analysis coefficient Fxxs, and generate an assessment result. The assessment result then triggers a second assessment mechanism. The second assessment mechanism performs secondary comparative assessment on a preset environmental assessment threshold Y and the obtained environmental impact coefficient Yxxs, and generates an assessment result.

2. The powder spraying detection system based on intelligent contour scanning according to claim 1 is characterized in that: The data detection module includes an intelligent contour scanning unit and a sensor detection unit; The intelligent contour scanning unit scans the powder spray coating by using an intelligent contour scanning device. Before starting the scanning, a light source is used to illuminate the coating surface to obtain data on the powder spray coating surface. The intelligent contour scanning device includes a laser scanner and an optical sensor. The laser scanner is turned on to emit a laser beam, and three-dimensional data of the coating surface is obtained by reflection and scattering of the coating surface. The optical sensor then uses a camera to capture image data of the coating surface. The sensor detection unit is used to deploy an integrated sensor group in a powder spraying workshop to detect environmental impact data during the spraying process in real time. The integrated sensor group includes a temperature sensor, a humidity sensor, a flow sensor, a particle sensor and an electrostatic sensor.

3. The powder spraying detection system based on intelligent contour scanning according to claim 2 is characterized in that: The deep learning module includes a model building unit and a feature recognition unit; The model building unit crawls an image dataset containing powder spray coating defects from the Internet of Things through a crawler tool, and the images include crack defects, peeling defects, bubble defects, uneven defects, hole defects and surface uniformity defects, and then preprocesses the data, including resizing, standardization, enhancement and defect feature labeling, and selects a convolutional neural network and an attention mechanism as a deep learning model architecture, divides the prepared image dataset into a training set, a validation set and a test set, uses the training set to train the deep learning model, and verifies it on the validation set, and uses the test set to evaluate the trained deep learning model, and the evaluation indicators include accuracy, precision, recall rate and F1 score, and adjusts the model architecture, optimizes hyperparameters and data preprocessing according to the evaluation results, and after the model training is completed, deploys it in practical applications to monitor the quality of the powder spray coating in real time; The feature recognition unit is used to set the API application interface of the intelligent contour scanning device and the API application interface of the sensor, connect to the deep learning model, transmit the acquired real-time image data to the deep learning model through wireless transmission, perform feature recognition on the comprehensive defect data in the coating image scanned by the deep learning model, and denoise, clean and normalize the environmental impact data detected by the sensor.

4. The powder spraying detection system based on intelligent contour scanning according to claim 1 is characterized in that: The defect assessment module includes a first assessment unit and a second assessment unit; The first evaluation unit is used to preset a defect evaluation threshold Q, and conduct a preliminary comparative evaluation with the obtained comprehensive analysis coefficient Fxxs. The specific evaluation results are as follows; When the comprehensive analysis coefficient Fxxs>defect assessment threshold Q, it means that the powder coating currently obtained by intelligent profile scanning has defects. At this time, it is necessary to start the second assessment mechanism to analyze the cause of the current defect; When the comprehensive analysis coefficient Fxxs ≤ the defect assessment threshold Q, it indicates that the powder spraying currently obtained by the intelligent profile scanning is in a normal state.

5. The powder spraying detection system based on intelligent contour scanning according to claim 4 is characterized in that: The second evaluation unit is used to perform a secondary comparative evaluation between the preset environmental evaluation threshold value Y and the obtained environmental impact coefficient Yxxs, and start the second evaluation mechanism. The second evaluation mechanism is triggered by the first evaluation unit and generates a corresponding evaluation result. The specific evaluation scheme is as follows; When the environmental impact coefficient Yxxs> environmental assessment threshold Y, it means that the current powder spraying is abnormal due to environmental factors. At this time, choose powder coatings with good quality and suitable for specific environments, including the weather resistance and anti-pollution ability of the coating. At the same time, use an air purifier to improve the spraying environment, and use appropriate surface treatment technology before spraying; When the environmental impact coefficient Yxxs ≤ the environmental assessment threshold Y, it means that there is an abnormality in the current powder spraying and the environment is in a normal state. This means that there is an abnormality in the spraying equipment, which leads to coating defects in the spraying process. At this time, it is necessary to check and adjust the parameters of the spraying equipment, including spray gun pressure, nozzle cleanliness and powder flow rate, use high-quality spraying materials and accessories, and optimize the occurrence of defects.

Images