Intelligent detection system for leveling property of coating and detection method thereof

Abstract

The invention discloses an intelligent detection system for the leveling property of a coating. A main body of the intelligent detection system is a cuboid with a cavity; the cuboid comprises a front end plane and a rear end plane; the lower part of the front end plane is provided with a sample inlet which is horizontally distributed and is provided with a strip-shaped opening; the lower part of the rear end plane is provided with a sample outlet which is horizontally distributed and is provided with a strip-shaped opening; a conveying device is arranged in the cuboid and between the sample inlet and the sample outlet; a laser light source is arranged on the inner side of the top surface of the cuboid and close to the front end plane; a vision measuring device is arranged on the inner side of the top face of the cuboid and close to the rear end plane. The invention further discloses a detection method adopted by the detection device. By adopting the technical scheme, the rapidness, intelligence and automation of the leveling property detection of the surface of the coating are realized, and the efficiency, accuracy and reliability of the leveling property detection of the coating are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of coating surface detection. More specifically, the present invention relates to an intelligent coating leveling detection system. The present invention also relates to a detection method of the detection system. Background Art

[0002] Powder coating is a "4E" (high efficiency, energy saving, environmental protection, and economy) coating with 100% solid content and zero VOC. It has the characteristics of energy saving, environmental protection and excellent comprehensive performance. It has extremely wide application fields. The leveling property (or flatness) of the powder coating after melting and curing is a key factor in determining the appearance quality of the coating and is also an important indicator for the evaluation of the coating appearance.

[0003] At present, there is no dedicated testing device and method for the leveling of traditional powder coatings. The general method is based on the methods used in the liquid coating or paint industry. One method is to measure it by the time required for the coating film to recover or reach a uniform and smooth surface. The other is to evaluate it by scraping and coating several parallel strips of different thicknesses and observing the number of them that flow completely and partially together.

[0004] The above two traditional leveling test methods mainly use the coating film-forming process to test and evaluate the flowability. Their limitations are: first, the film-forming mechanism of liquid coatings or paints is different from that of 100% solid powder coatings; second, they are not applicable to the testing and evaluation of the appearance of the coating after film formation.

[0005] At present, the common practice in the powder coating industry is that the inspection is mainly based on manual visual inspection. When the leveling is good, the fluorescent light shines on the surface of the test sample, and the shape of the light tends to be a smooth straight line; on the contrary, the shape is more distorted. The grade assessment is also based on manual visual inspection, judging by the visual difference between the test sample and the standard plate.

[0006] This method of relying solely on visual inspection and judgment has great disadvantages. The results are greatly affected by factors such as the inspector's eyesight, observation angle, light in the observation environment, subjective consciousness, technical level, etc., and the reproducibility is also relatively poor. In addition, a standard plate must be used for comparison every time an assessment is made. Some detection devices have complex structures, cumbersome detection methods, and low detection efficiency. The test results can only be judged qualitatively, and quantitative parameter comparison cannot be achieved. The accuracy of the test results is poor, and the versatility of the detection method is also relatively poor.

[0007] The following search results were obtained by searching the existing published technical literature using keywords such as "coating; flatness (or leveling); test method":

[0008] 1. Chinese patent document: "A coating detection method, device and computer storage medium", patent (application) number: 201911236340.3; the technical solution recorded is:

[0009] “Coating detection method, device and computer storage medium, including: S1: detecting the smoothness of the appearance of the fabric coating; S2: detecting the anti-corrosion performance of the coating; S3: detecting the adhesion of the coating; laying the fabric flat on a hard and flat surface to prevent the fabric from deforming during the detection process; fixing an adhesion tester on the fabric, shaking the handle of the adhesion tester in a clockwise direction, so that the needle of the adhesion tester leaves a circular scratch on the fabric, and detecting whether the scratch penetrates the coating; if the scratch does not penetrate the coating, adding weights as appropriate; keeping the needle of the adhesion tester sharp during the operation of the adhesion tester; and obtaining the adhesion performance index of the coating”;

[0010] The technical effects recorded are:

[0011] "It is able to conduct comprehensive tests on the various properties of the coating on the fabric, and the coating detection is accurate and highly automated."

[0012] 2. Chinese patent document: "A method for detecting the coating quality of PCB boards", patent (application) number: 202410250146.5; the technical solution recorded therein is:

[0013] "The PCB coating quality detection method segments the collected PCB image to obtain a three-proof coating coverage area map; performs edge filtering processing to obtain an image containing bubble edge information; extracts bubble features to obtain a bubble feature map set; calculates a bubble result set; and classifies the bubble result set to obtain conclusions and suggestions. The PCB coating quality detection method detects the bubble situation of the PCB board by segmenting the collected PCB image to obtain a three-proof coating coverage area map, and then performs edge filtering processing, extracts bubble features, calculates bubble size and other steps to obtain the bubble situation of the PCB board, and classifies the bubble result set to obtain conclusions and suggestions";

[0014] The technical effects recorded are:

[0015] "It has realized the detection of bubbles in PCB coating, solving the problem of blistering or bubbles on PCB, which affects the flatness and insulation performance of the coating."

[0016] However, the technical solutions recorded in the above-mentioned technical documents and the related technical solutions currently in public application have not been able to solve the problems and defects existing in the prior art, such as "inability to achieve consistency in detection results" and "low detection efficiency". Summary of the invention

[0017] The invention provides an intelligent detection system for coating leveling, the purpose of which is to realize automation and intelligence of detection and improve detection efficiency and accuracy of detection results.

[0018] In order to achieve the above object, the technical solution adopted by the present invention is:

[0019] The coating leveling intelligent detection system of the present invention has a main body which is a hollow cuboid; the cuboid comprises a front end plane and a rear end plane; a sample inlet with a horizontally distributed long strip opening is arranged at the lower part of the front end plane; a sample outlet with a horizontally distributed long strip opening is arranged at the lower part of the rear end plane; a conveying device is arranged inside the cuboid, between the sample inlet and the sample outlet; a laser light source is arranged on the inner side of the top surface of the cuboid, close to the front end plane; and a visual measurement device is arranged on the inner side of the top surface of the cuboid, close to the rear end plane.

[0020] The laser light emitted by the laser light source is directed toward the test sample on the conveying device; after being reflected by the test sample, it is directed toward the visual measuring device.

[0021] The detection system is provided with a control host; the conveying device, the laser light source and the visual measuring device are respectively connected to the control host through a control driving circuit.

[0022] The control host is arranged on the upper part of one side of the cuboid.

[0023] A human-machine interactive LED touch screen is arranged on the outer surface of the top surface of the rectangular parallelepiped, and the LED touch screen is connected to the control host through a signal circuit.

[0024] The size of the LED touch screen is 10.5 inches.

[0025] The laser light source is a long strip-shaped D65 laser light source.

[0026] The visual measurement device is an AI slit filter visual measurement device.

[0027] In order to achieve the same inventive purpose as the above technical solution, the present invention also provides a detection method of the above-mentioned coating leveling intelligent detection system, which comprises the following steps:

[0028] 1) Set relevant program parameters on the touch screen, including the conveying rate of the test sample, the confidence interval, and the coating thickness;

[0029] 2) Place the coated test sample into the sample inlet and press the run button on the touch screen;

[0030] 3) When the sensor senses that the test sample has entered the detection area, the laser light source is turned on to illuminate the surface of the test sample at a set incident angle and light intensity;

[0031] 4) When the laser light source is turned on, the visual measurement device is also turned on to receive the reflected light from the surface of the test sample;

[0032] 5) The test sample is fed into the sample inlet at a constant displacement rate until the test sample is completely output from the sample outlet;

[0033] 6) The detection system transmits the energy E1 of the input light source and the energy E2 collected by the visual measurement device to the control mainboard of the control host in real time;

[0034] 7) The AI ​​chip of the control mainboard of the control host performs data analysis while the data is transmitted back, and displays the real-time monitoring data on the touch screen;

[0035] 8) After the test sample is completely output from the sample outlet, the real-time monitoring data stops transmitting, and the built-in AI artificial intelligence analysis software of the control host uses the pre-set SPC statistical analysis software to perform statistical analysis on the data, and calculate the control limits UCL and LCL, mean value, standard deviation S, E2 within the confidence interval, and E2 / E0 data;

[0036] 9) Evaluate the leveling property of the coating according to the pre-set judgment criteria (if any); output and save the test parameters, judgment results and test sample information in the form of a test data table or test report;

[0037] 10) The test data form or test report is transmitted to the laboratory information management system via a wireless bridge.

[0038] The present invention adopts the above technical scheme to realize rapid, intelligent and automated detection of coating surface leveling, thereby improving the efficiency, accuracy and reliability of coating surface leveling detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The contents shown in the attached drawings and the symbols in the drawings are briefly described as follows:

[0040] Figure 1 It is a structural schematic diagram of the intelligent detection system for coating leveling of the present invention;

[0041] Figure 2 The single value-moving range control chart (I-MR control chart) for E0;

[0042] Figure 3 The individual value-moving range control chart (I-MR control chart) for E2;

[0043] Figure 4 It is the E2 / E0 single value - moving range control chart (I-MR control chart).

[0044] Figure 1 The tags in are:

[0045] 1. Sample inlet, 2. Conveying device, 3. Sample outlet, 4. Laser light source (D65 laser light source), 5. Visual measuring device (AI slit filter visual measuring device), 6. LED touch screen, 7. Control host;

[0046] E0, incident spectral energy, E1, energy of input light source, E2, energy collected by AI slit filter visual measurement device. DETAILED DESCRIPTION

[0047] The specific implementation modes of the present invention are further explained in detail below by describing the embodiments with reference to the accompanying drawings, so as to help those skilled in the art to have a more complete, accurate and in-depth understanding of the inventive concept and technical solution of the present invention.

[0048] like Figure 1 The structure of the present invention is an intelligent detection system for the leveling of powder coatings. The entire detection system consists of a sample inlet 1, a conveying device 2, a sample outlet 3, a D65 laser light source 4, an AI slit filter visual measurement device 5, an LED touch screen 6 and a control host 7.

[0049] The specific analysis is as follows:

[0050] In order to solve the problems existing in the prior art and overcome its defects, and to achieve the invention purpose of improving the detection efficiency and the accuracy of the detection results, the technical solution adopted by the present invention is:

[0051] The coating leveling intelligent detection system of the present invention is as follows: Figure 1 As shown, its main body is a hollow cuboid; the cuboid includes a front plane and a rear plane; at the lower part of the front plane, a horizontally distributed long strip opening inlet 1 is provided; at the lower part of the rear plane, a horizontally distributed long strip opening outlet 3 is provided; inside the cuboid, between the inlet 1 and the outlet 3, a conveying device 2 is provided; on the inner side of the top surface of the cuboid, close to the front plane, a laser light source 4 is provided; on the inner side of the top surface of the cuboid, close to the rear plane, a visual measurement device 5 is provided.

[0052] The intelligent detection system is provided with an opening on one side, namely, an injection port 1, which is used to transport the test sample into the detection system; the other side, which is symmetrically distributed in the horizontal direction with the injection port 1, is provided with an outlet of the same size and shape, namely, an outlet 3, which is used for the test sample to exit the detection system after the detection is completed; the inner side of the injection port 1 horizontally arranged at the bottom of one side of the detection system is connected to the conveying device 2 installed at the bottom of the system; the inner side of the outlet 3 on the other side of the detection system, which is horizontally symmetrical with the injection port 1, is connected to the other end of the conveying device 2.

[0053] A roller is provided at the bottom of the sample inlet 1 and is connected to the right end of the conveying device 2 ; a roller is also provided at the bottom of the sample outlet 3 and is connected to the conveying device 2 .

[0054] A D65 laser light source 4 is provided on the top inner side of the same side as the injection port 1; a slit filter visual measurement device 5 is provided above the opening on the other side and at the same height as the laser light source 4; a visual measurement device 5 is provided at the upper end of the detection system at a position symmetrical to the lamp tube 4 of the D65 light source on one side, and the visual measurement device 5 is an AI slit filter visual measurement device. The laser light source 4 is a long strip-shaped D65 laser light source; the laser light emitted by the laser light source 4 is directed to the test sample on the conveying device 2; after being reflected by the test sample, it is directed to the visual measurement device 5.

[0055] The AI ​​slit filter visual measuring device 5 is arranged at the top position on the left side of the whole detection system, horizontally symmetrical with the laser light source 4. The angle of the measuring device receiving the reflected light can be automatically and synchronously adjusted according to the change of the incident angle.

[0056] A conveyor frame for transmitting the test sample, i.e., a conveyor device 2, is installed at the bottom of the inner cavity of the detection system. The conveyor device 2 is arranged on the inner side of the bottom of the entire detection system, with a micro motor at each end to drive the rollers at both ends to rotate; or a micro motor is provided at one end to move the test sample, and the motor is operated by the control host 7 sending instructions through the data line.

[0057] The present invention uses a conveying device 2 to pass the test sample sprayed with powder coating and completely cured through the detection system at a set speed. After the test sample enters the detection area, the light source (D65 laser light source) begins to continuously irradiate the coating surface.

[0058] The incident angle of the D65 laser light source 4 can be adjusted as needed.

[0059] When measuring the surface leveling of the coating, the present invention introduces an optical instrument. According to the different surface flatness of coatings with different leveling properties, the reflection angles of light at different points (peaks or troughs) are different, and the leveling of the coating is detected by the energy difference of the spectrum entering the detection system. The leveling of the coating after the powder coating is cured is directly measured without studying the film-forming mechanism. The introduced AI visual detection technology, visual algorithm, etc., the test results have better trend consistency compared with manual visual detection under fixed environment and conditions, reduce various interference factors of manual testing, convert subjective judgment into quantifiable digital form, which is more intuitive and has better reproducibility.

[0060] The detection system and detection method of the present invention eliminate the differences and interference caused by traditional detection relying on visual inspection under different environments and different people's perspectives, and transform the traditional reliance on subjective judgment into digital measurement. Through quantifiable data, the detection results are more intuitive, ensuring the reliability and repeatability of the detection.

[0061] The detection system is provided with a control host 7; the conveying device 2, the laser light source 4 and the visual measuring device 5 are connected to the control host 7 through control driving circuits respectively.

[0062] The control host 7 is arranged on the upper part of one side of the cuboid. Alternatively, the control host 7 is arranged on the upper inner side of the entire detection system, directly below the touch screen 6, with data ports on both sides, connected with various related units by data lines.

[0063] On the outer surface of the top surface of the rectangular parallelepiped, a human-machine interactive LED touch screen 6 is arranged, that is, a 10.5-inch LED touch screen 6 is installed on the top of the detection system, and the LED touch screen 6 is connected to the control host 7 through a signal circuit. The LED touch screen 6 is a 10.5-inch capacitive touch screen, which is arranged just above the entire detection system, and the lower end is connected to the circuit board of the control host 7.

[0064] The control host 7 below the LED touch screen is equipped with a control motherboard of an AI artificial intelligence chip. The detection system of the present invention is equipped with an AI artificial intelligence chip. During the detection process, the AI ​​neural network can mark the data of the coating image and convert it into a mathematical equation. The AI ​​intelligent learning will optimize and improve the detection program through statistical analysis of the data and continuous deep learning.

[0065] The AI ​​slit filter visual detection device 5 detects the reflected light, and the detection signal is displayed on the page of the LED touch screen 6 in the form of a real-time curve and data; according to the degree of curvature of the curve deviating from the baseline, a statistical analysis is performed to obtain a data of the leveling property of the test sample; according to the size of the data, the leveling grade and leveling difference of the coating can be evaluated.

[0066] There are many ways to determine the grade: one is to make an intuitive comparison based on the values ​​calculated by the detection system; another is to store the detection data of the PCI standard board in the detection system in advance, and compare the values ​​calculated by the system detection with the values ​​of the PCI standard board for evaluation; the program or method can also be rewritten for testing and evaluation according to the customer's own needs.

[0067] According to the above principle, in order to achieve the same invention purpose as the above technical solution, the present invention also provides a detection method of the above-mentioned coating leveling intelligent detection system;

[0068] It includes the following steps:

[0069] 1) Setting relevant program parameters on the touch screen 6, wherein the program parameters include the conveying rate, confidence interval, and coating thickness of the test sample;

[0070] 2) Place the test sample with coating inlet 1 and press the run button on the touch screen;

[0071] 3) When the sensor senses that the test sample has entered the detection area, the laser light source 4 is turned on to illuminate the surface of the test sample at a set incident angle and light intensity;

[0072] 4) When the laser light source 4 is turned on, the visual measurement device 5 is also turned on to receive the reflected light from the surface of the test sample;

[0073] 5) The test sample is fed into the sample inlet 1 at a constant displacement rate until the test sample is completely output from the sample outlet 3;

[0074] 6) The detection system transmits the energy E1 of the input light source and the energy E2 collected by the visual measurement device 5 to the control mainboard of the control host 7 in real time;

[0075] 7) The AI ​​artificial intelligence chip of the control mainboard of the control host 7 performs data analysis while the data is transmitted back, and displays the real-time monitoring data on the touch screen 6;

[0076] 8) After the test sample is completely output from the sample outlet 3, the real-time monitoring data stops transmitting, and the built-in AI artificial intelligence analysis software of the control host 7 uses the pre-set SPC statistical analysis software to perform statistical analysis on the data, and calculate the control limits UCL and LCL, the mean value, the standard deviation S, the E2 within the confidence interval, and the E2 / E0 data;

[0077] 9) Evaluate the leveling property of the coating according to the pre-set judgment criteria (if any); save and output the test parameters, judgment results and test sample information in the form of a test data table or test report;

[0078] 10) The test data form or test report is transmitted to the laboratory information management system via a wireless bridge.

[0079] Description: The confidence interval is an estimated interval of a population parameter constructed from a sample statistic. It provides an interval estimate, indicating that the true value of the population parameter has a certain probability of falling within the interval.

[0080] The specific contents of the above detection method are:

[0081] First, before starting the test, set the relevant parameters through the touch screen 6, such as the horizontal transport rate V of the test sample, the incident angle ω, the acquisition frequency of the incident spectral energy E0 and the received reflected spectral energy E1, the confidence interval, etc.;

[0082] After the parameters are set, press the Run button and the system starts running; the conveying device 2 starts to move the test sample from the inlet 1 to the outlet 3 at a constant speed at a rate V; the D65 laser light source 4 irradiates at the set incident angle ω, and the AI ​​slit filter visual measurement device 5 starts to receive the reflected light at the same time;

[0083] When the detection system is running at no load, the incident spectral energy E0 and the reflected spectral energy E1 are theoretically equal (ignoring the weak spectral energy lost due to absorption), that is, E0=E1, or E1 / E0=1.

[0084] When the detection system is running at no load, the curve drawn by the AI ​​analysis software is optimized to be a horizontal straight line, which serves as the baseline for detection analysis.

[0085] The coating sample to be tested is fed into the system from the sample inlet 1 at one end. The conveying device 2 of the detection system will carry the coating sample to be tested through the detection system at a uniform speed. At the same time, E0, E2, and E2 / E0 will be displayed in real time on the detection system operation interface.

[0086] After the coating sample to be tested is completely output from the sample outlet 3 on the other side, the test is completed, the light source and the visual measurement device are turned off, and the detection system enters the AI ​​data analysis stage; according to the statistical analysis software preset in the artificial intelligence chip, the collected data E2 and E2 / E0 are plotted with a single value moving range control chart and normal distribution statistical analysis, and the control limits UCL and LCL, mean value, standard deviation S, E2 within the confidence interval, and E2 / E0 and other data are calculated.

[0087] The D of the actual test coating sample is D=E2 / E0, 0<D<1.0. The leveling (or flatness) of the coating is determined by measuring the D value. The larger the D value is or the closer it is to 1.0, the better the leveling (or flatness) of the coating; the smaller the D value is or the closer it is to 0, the worse the leveling (or flatness) of the coating.

[0088] If grade determination is required, the test sample with known leveling grade needs to be operated once according to the test process, and the D value of the test sample with known leveling grade needs to be recorded and saved. When the unknown test sample needs to be tested later, the setting in the program can be compared with the test sample in the database.

[0089] The present invention detects the surface leveling of the coating by irradiating the incident spectral energy on the surface of the coating sample and reflecting it from the surface of the test sample to the AI ​​slit filter visual measurement device, and digitally represents the leveling of the coating, providing intuitive data support for coating surface measurement and research, and also providing the possibility for the operability and standardized determination of coating leveling detection.

[0090] The present invention also provides the following specific implementation examples:

[0091] Embodiment 1:

[0092] A test sample with a size of 150×75mm was selected, and after the powder coating was sprayed and completely cured, it was cooled to room temperature.

[0093] Call up the program on the touch screen 6 and set the horizontal conveying speed of the test sample to V = 30 mm / min;

[0094] The D65 laser light source 4 is set to irradiate at an incident angle of ω = 45°;

[0095] The sampling frequency was 10 / min, and the confidence interval was set at 95%.

[0096] After all parameters are set, place the coated test sample with its wide side facing the inlet 1 and press the Run button at the same time. The conveyor 2 carries the test sample through the detection system at V=30mm / min.

[0097] At the moment the test sample enters the detection system, the D65 laser light source (4) and the AI ​​slit filter visual measurement device are turned on at the same time, the D65 laser light source 4 irradiates the surface of the test sample at an incident angle ω=45°, and the AI ​​slit filter visual measurement device receives the light reflected from the surface of the test sample.

[0098] The AI ​​artificial intelligence software samples E0 and E2 at a sampling frequency of 10 / min and displays them on the touch screen 6 in real time.

[0099] After the test sample completely passes through the sample outlet 3, the light source and the visual measurement device are turned off, the sampling ends, and the system enters the AI ​​data processing stage.

[0100] The touch screen 6-view interface can select to display the single value moving range control chart (I-MR control chart) of E0 according to user needs, such as Figure 2 As shown; you can also choose to display the single value moving range control chart (I-MR control chart) of E2, as shown Figure 3 shown.

[0101] The system will record Figure 2 Control limits UCL (1003.091) and LCL (995.709), mean value (999.4); or Figure 3 The control limits UCL (890.67) and LCL (862.89), the mean value (876.78),

[0102] At the same time, AI data analysis software calculates:

[0103] Figure 2 :The standard deviation of E0 is S = 1.212, and the value of the confidence interval of 95% is E0 = 997.4 ± 2.5 KJ / m 2 .

[0104] Figure 3 :The standard deviation of E2 is S = 6.089, and the value of the confidence interval of 95% is E2 = 876.8 ± 11.9 KJ / m 2 .

[0105] The calculation results are then displayed on the view.

[0106] If further analysis is required, the system AI processing software can draw an E2 / E0 single value moving range control chart, such as Figure 4 As shown:

[0107] Figure 4 The control limits UCL (0.88980) and LCL (0.86481), the average The standard deviation of E2 / E0 is S=0.00566, and the value of the 95% confidence interval is E2 / E0=0.8773±0.0111.

[0108] The leveling property can be expressed intuitively with numbers in various forms such as the average value of E2 or E2 / E0, confidence interval, etc.

[0109] Embodiment 2:

[0110] According to the PCI standard board, the leveling grade of the sample board to be tested is determined. The PCI standard boards of different specifications can be measured according to the steps of Example 1, and the set test program is selected to measure, and the measured data is retained, such as 1#—ES1 、2#—E S2 ,……,N#—E Sn .

[0111] If necessary, AI data analysis software will be combined with 1#—E S1 、2#—E S2 ,……,N#—E Sn , establish a data model, draw a standard curve, and the values ​​of the standard curve correspond to the actual level curve one by one.

[0112] Then test the sample board under the same conditions. After the test, select E2 and PCI standard board E for AI data processing. Sn Compare and determine the level.

[0113] The AI ​​data analysis software will automatically retrieve the pre-tested and stored PCI standard board data from the database for comparison, and give a conclusion on the leveling grade based on the principle of data proximity.

[0114] Embodiment three:

[0115] The customer provides an acceptance standard sample and requires that the leveling of the coating surface be close to that of the standard sample.

[0116] The acceptance standard sample provided by the customer is measured according to the steps of Example 1, and the set test program is selected to measure the measured data E k1 Make reservations.

[0117] Design powder coating formula, spray coating and curing according to customer requirements, test the cured coating samples according to the same test procedures, and compare the test results E2 and E k1 Make a comparison.

[0118] Choose E2 and E from many test formulas k1 We supply customers with powder formulas that are close to theirs, making quality control more intelligent.

[0119] The above-mentioned embodiment of the present invention is based on the difference between the emission spectral energy of the laser light source and the spectral energy reflected back from the surface of the sample to be tested by the AI ​​slit filter visual measuring device, and uses numbers to intuitively represent the leveling (or flatness) of the coating surface, providing quantifiable data support for the research and development and design of powder coatings. The system's built-in AI artificial intelligence chip has an autonomous learning function. According to the gradual accumulation of data, the system will continuously optimize the initial data model, so that the analysis results of the data model tend to be more reliable.

[0120] The present invention is described above by way of example in conjunction with the accompanying drawings. It is obvious that the specific implementation of the present invention is not limited to the above-mentioned method. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention, or the concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.

Claims

1. An intelligent coating leveling detection system, characterized in that: The main body of the detection system is a hollow cuboid; the cuboid comprises a front plane and a rear plane; a sample inlet (1) with a horizontally distributed long strip opening is arranged at the lower part of the front plane; a sample outlet (3) with a horizontally distributed long strip opening is arranged at the lower part of the rear plane; a conveying device (2) is arranged inside the cuboid, between the sample inlet (1) and the sample outlet (3); a laser light source (4) is arranged on the inner side of the top surface of the cuboid, close to the front plane; and a visual measurement device (5) is arranged on the inner side of the top surface of the cuboid, close to the rear plane.

2. The coating leveling intelligent detection system according to claim 1, characterized in that: The laser light emitted by the laser light source (4) is directed toward the test sample on the conveying device (2); after being reflected by the test sample, the laser light is directed toward the visual measuring device (5).

3. The coating leveling intelligent detection system according to claim 2, characterized in that: The detection system is provided with a control host (7); the conveying device (2), the laser light source (4) and the visual measuring device (5) are respectively connected to the control host (7) via a control driving circuit.

4. The coating leveling intelligent detection system according to claim 3, characterized in that: The control host (7) is arranged on the upper part of one side of the rectangular parallelepiped.

5. The coating leveling intelligent detection system according to claim 3, characterized in that: The control host (7) is provided with a control mainboard equipped with an AI artificial intelligence chip.

6. The coating leveling intelligent detection system according to claim 3, characterized in that: A human-machine interactive LED touch screen (6) is arranged on the outer surface of the top surface of the rectangular parallelepiped, and the LED touch screen (6) is connected to a control host (7) via a signal circuit.

7. The coating leveling intelligent detection system according to claim 6, characterized in that: The size of the LED touch screen (6) is 10.5 inches.

8. The intelligent coating leveling detection system according to claim 1, characterized in that: The laser light source (4) is a long strip-shaped D65 laser light source.

9. The coating leveling intelligent detection system according to claim 1, characterized in that: The visual measurement device (5) is an AI slit filter visual measurement device.

10. The detection method of the coating leveling intelligent detection system according to any one of claims 1 to 9, characterized in that: The detection method comprises the following steps: 1) Setting relevant program parameters on the touch screen (6), wherein the program parameters include the conveying rate, confidence interval, and coating thickness of the test sample; 2) Place the test sample with the coating inlet (1) and press the run button on the touch screen; 3) When the sensor senses that the test sample has entered the detection area, the laser light source (4) is turned on to illuminate the surface of the test sample at a set incident angle and light intensity; 4) When the laser light source (4) is turned on, the visual measurement device (5) is also turned on to receive reflected light from the surface of the test sample; 5) The test sample is fed into the sample inlet (1) at a constant displacement rate until the test sample is completely output from the sample outlet (3); 6) The detection system transmits the energy E1 of the input light source and the energy E2 collected by the visual measurement device (5) to the control mainboard of the control host (7) in real time; 7) The AI ​​artificial intelligence chip of the control mainboard of the control host (7) performs data analysis while the data is being transmitted back, and displays the real-time monitoring data on the touch screen (6); 8) After the test sample is completely output from the sample outlet (3), the real-time monitoring data stops transmitting, and the built-in AI artificial intelligence analysis software of the control host (7) uses the pre-set SPC statistical analysis software to perform statistical analysis on the data, and calculate the control limits UCL and LCL, the mean value, the standard deviation S, the E2 within the confidence interval, and the E2 / E0 data; 9) Evaluate the leveling property of the coating according to the pre-set judgment criteria; save and output the test parameters, judgment results and test sample information in the form of a test data table or test report; 10) The test data form or test report is transmitted to the laboratory information management system via a wireless bridge.

Images