A method of calculating a height measurement for inkjet printing

Abstract

The present application is suitable for the field of jet printing, and discloses a jet printing height measurement calculation method, comprising the following steps: emitting a complex color light to a measured substrate and a nozzle surface from the side of the measured substrate far away from the nozzle by using a spectral confocal sensor, and receiving a first light signal reflected from the side of the measured substrate close to the nozzle and a second light signal reflected by the nozzle; determining a first focusing wavelength focused on the side of the measured substrate close to the nozzle and a second focusing wavelength focused on the nozzle and a nozzle incident angle according to the two light signals, substituting into a pre-constructed refractive index calculation function to calculate the refractive index; substituting the double wavelength into a pre-constructed initial thickness calculation function and an error correction model to obtain a theoretical height value and an error compensation amount; if the inclination angle of the measured substrate is not 0 degrees, calculating an actual height value of the measured substrate to the nozzle according to a pre-constructed first real thickness calculation function, and the method is used for realizing online high-precision monitoring of jet printing height.

Description

Technical Field

[0001] This invention relates to the field of inkjet printing, and more particularly to a method for measuring and calculating inkjet printing height. Background Technology

[0002] In the technological development of OLED inkjet printing equipment, real-time and accurate measurement of the printing height is a key parameter to ensure film uniformity and device performance. Printing height refers to the vertical distance from the nozzle surface to the substrate surface. As display technology evolves towards larger sizes and greater flexibility, non-contact inkjet printing is gradually replacing traditional vapor deposition processes. Research shows that a printing height fluctuation of ±5μm can cause uneven droplet spreading, leading to a decrease in device efficiency. However, existing measurement technologies generally suffer from insufficient media penetration, making it impossible to directly obtain the upper surface height through a transparent plate, or requiring multiple measurement conversions and calculations, resulting in accumulated systematic errors. This has become a technological bottleneck restricting the mass production of high-generation OLED lines.

[0003] Current mainstream measurement methods have the following limitations:

[0004] Laser triangulation offers a fast response time, but the optical path is susceptible to refraction interference in transparent media. The measured value reflects the position of the lower surface of the glass, with an error reaching tens of micrometers. The superposition of reflection signals from multiple interfaces increases the difficulty of separating the effective signal. Calculating the printing height through multiple measurements and conversions not only amplifies the systematic error but also makes it difficult to improve the overall measurement accuracy.

[0005] White light interferometry has the advantage of high precision, but its sampling rate is less than 100Hz and it is highly sensitive to external vibrations and temperature fluctuations. It requires vibration isolation and temperature control devices, which increases the size of the equipment and the complexity of the system, making it unsuitable for online integration applications.

[0006] Traditional spectral confocal techniques rely on the perpendicularity of the sensor's optical axis to the object being measured, as well as the precise calibration of the refractive index of the transparent medium. Furthermore, the lack of a multi-layer refractive compensation mechanism at the algorithm level leads to the accumulation of axial errors, making it difficult to meet the requirements of high-precision measurement.

[0007] Contact thickness gauges directly contact the substrate with mechanical probes, which poses risks of damaging the substrate surface and causing contamination. Furthermore, they cannot achieve online detection of moving substrates and have been gradually phased out by the industry.

[0008] Therefore, existing technologies still need improvement and development. Summary of the Invention

[0009] The purpose of this invention is to provide a method for measuring and calculating printing height, which aims to solve the technical problem that existing OLED inkjet printing equipment is unable to accurately measure the printing height in real time during the printing process.

[0010] To achieve the above objectives, the solution provided by the present invention is as follows:

[0011] A method for measuring and calculating printing height includes: emitting polychromatic light from the side of the substrate away from the nozzle to the substrate and nozzle surface using a spectral confocal sensor, and receiving a first light signal reflected from the side of the substrate near the nozzle and a second light signal reflected from the nozzle; determining a first focusing wavelength focused on the side of the substrate near the nozzle and a second focusing wavelength focused on the nozzle, and a nozzle incident angle, respectively, based on the first and second light signals; substituting the first focusing wavelength, the second focusing wavelength, and the nozzle incident angle into a pre-constructed refractive index calculation function to calculate the refractive index; substituting the first focusing wavelength and the second focusing wavelength into a pre-constructed initial thickness calculation function to calculate the theoretical height value from the substrate to the nozzle, and inputting the first focusing wavelength and the second focusing wavelength into a pre-trained error correction model to obtain an error compensation amount; obtaining the tilt angle of the substrate, and if the tilt angle of the substrate is not 0 degrees, substituting the first focusing wavelength, the second focusing wavelength, the nozzle incident angle, the refractive index, the tilt angle, the theoretical height value, and the error compensation amount into a pre-constructed first true thickness calculation function to calculate the actual height value from the substrate to the nozzle.

[0012] Preferably, after obtaining the tilt angle of the substrate under test, the method further includes:

[0013] If the tilt angle of the substrate being tested is 0 degrees, the first focusing wavelength, the second focusing wavelength, the nozzle incident angle, the refractive index, the theoretical height value, and the error compensation amount are substituted into the pre-constructed second true thickness calculation function to calculate the actual height value from the substrate to the nozzle.

[0014] Preferably, the second true thickness calculation function is expressed as:

[0015]

[0016] In the formula, This indicates the actual height value from the measured substrate to the nozzle. This indicates the amount of error compensation predicted by the error correction model based on the first and second focusing wavelengths. This represents the theoretical height from the substrate being tested to the nozzle. Indicates the nozzle incident angle. Indicates the first focusing wavelength. Indicates the second focusing wavelength. .

[0017] Preferably, the polychromatic light emitted by the spectral confocal sensor includes a first monochromatic light and a second monochromatic light. The first monochromatic light is focused on the side of the substrate being tested near the nozzle, and the second monochromatic light is focused on the nozzle. The first monochromatic light reflects a first optical signal, and the second monochromatic light reflects a second optical signal.

[0018] Preferably, the pre-constructed refractive index calculation function is expressed as:

[0019]

[0020] In the formula, Indicates refractive index, Indicates the first focusing wavelength. Indicates the second focusing wavelength. Indicates the nozzle incident angle This indicates the focusing distance corresponding to the first focusing wavelength. The second focusing wavelength corresponds to the focusing distance, and h represents the thickness of the reference plate being measured in the calibration experiment.

[0021] Preferably, the pre-constructed initial thickness calculation function is expressed as:

[0022] =

[0023] In the formula, This represents the theoretical height from the substrate being tested to the nozzle. This indicates the focusing distance corresponding to the first focusing wavelength. This indicates the focusing distance corresponding to the second focusing wavelength.

[0024] Preferably, the pre-trained error correction model employs a three-layer fully connected neural network.

[0025] Preferably, the pre-trained error correction model includes an input layer, a hidden layer, and an output layer. The input layer includes two neurons for receiving the first focusing wavelength and the second focusing wavelength as input features. The hidden layer includes four neurons for fitting a nonlinear relationship between wavelength and error based on the input features. The output layer includes one neuron for mapping the nonlinear relationship between wavelength and error to a numerical value and outputting the error compensation amount.

[0026] Preferably, the first true thickness calculation function is expressed as:

[0027]

[0028]

[0029] In the formula, This indicates the actual height value from the measured substrate to the nozzle. Indicates the first focusing wavelength. Indicates the second focusing wavelength. Indicates the nozzle incident angle This indicates the focusing distance corresponding to the first focusing wavelength. This indicates the focusing distance corresponding to the second focusing wavelength. This represents the angle of refraction when the monochromatic light corresponding to the second focusing wavelength in the polychromatic light is refracted by the surface of the substrate near the nozzle and then incident on the nozzle. Indicates the tilt angle of the substrate being tested. This indicates the amount of error compensation predicted by the error correction model based on the first and second focusing wavelengths. It represents the refractive index.

[0030] The present invention provides a non-contact, cross-medium measurement method that uses a spectral confocal sensor to emit polychromatic light from the side of the substrate away from the nozzle. This allows the sensor to penetrate the substrate and directly acquire reflected light signals from the side near the nozzle and the nozzle surface, avoiding damage to the substrate and nozzle caused by contact measurements and accurately locating the target interface. Furthermore, by calculating the substrate refractive index using the focused wavelength and nozzle incident angle, and combining this with an initial thickness function, the theoretical height value is derived. A pre-trained error correction model is then introduced to obtain compensation. The system can also activate the corresponding real thickness calculation function based on the substrate tilt angle, forming a complete correction chain that includes optical parameter calculation, theoretical value derivation, error compensation, and tilt adaptation. This effectively eliminates measurement deviations caused by refraction, system implicit errors, and substrate tilt. In addition, the entire process relies on optical signals and algorithm models to complete height calculation without the need for complex mechanical adjustments. This method offers high measurement accuracy and strong adaptability, providing reliable height monitoring support for high-precision manufacturing scenarios such as OLED printing. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0032] Figure 1 This is a flowchart of the inkjet printing height measurement and calculation method provided in the embodiments of the present invention;

[0033] Figure 2 This is a schematic diagram of the tilting of the transparent flat plate provided in an embodiment of the present invention;

[0034] Figure 3 This is a schematic diagram of the thickness compensation principle provided in an embodiment of the present invention. Detailed Implementation

[0035] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0036] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 In this embodiment of the invention, a method for measuring and calculating printing height is provided.

[0037] S101. Using a spectral confocal sensor, polychromatic light is emitted from the side of the substrate being tested away from the nozzle to the substrate being tested and the nozzle surface, and a first light signal reflected from the side of the substrate being tested closer to the nozzle and a second light signal reflected from the nozzle are received.

[0038] S102. Determine the first focusing wavelength focused on the side of the substrate near the nozzle and the second focusing wavelength focused on the nozzle, and the nozzle incident angle, respectively, based on the first optical signal and the second optical signal. Substitute the first focusing wavelength, the second focusing wavelength, and the nozzle incident angle into the pre-constructed refractive index calculation function to calculate the refractive index.

[0039] S103. Substitute the first focusing wavelength and the second focusing wavelength into the pre-constructed initial thickness calculation function to calculate the theoretical height value from the substrate to the nozzle, and input the first focusing wavelength and the second focusing wavelength into the pre-trained error correction model to obtain the error compensation amount.

[0040] S104. If the tilt angle of the substrate being tested is not 0 degrees, the first focusing wavelength, the second focusing wavelength, the nozzle incident angle, the refractive index, the tilt angle, the theoretical height value, and the error compensation amount are substituted into the pre-constructed first true thickness calculation function to calculate the actual height value from the substrate being tested to the nozzle.

[0041] The printing height measurement and calculation method in this embodiment uses a spectral confocal sensor to emit polychromatic light from the side of the substrate away from the nozzle. This allows the light to penetrate the substrate and directly acquire reflected light signals from the side near the nozzle and the nozzle surface, achieving non-contact cross-medium measurement. This avoids damage to the substrate and nozzle caused by contact measurement and accurately locates the target interface. Furthermore, the substrate refractive index is calculated by focusing the wavelength and the nozzle incident angle. The theoretical height value is obtained by combining this with the initial thickness function. A pre-trained error correction model is then introduced to obtain the compensation amount. The corresponding real thickness calculation function can also be activated according to the substrate tilt angle, forming a complete correction link of optical parameter calculation, theoretical value derivation, error compensation, and tilt adaptation. This effectively eliminates measurement deviations caused by refraction, system implicit errors, and substrate tilt. In addition, the entire process relies on optical signals and algorithm models to complete the height calculation without complex mechanical adjustments. It has high measurement accuracy and strong adaptability, providing reliable height monitoring support for high-precision manufacturing scenarios such as OLED printing.

[0042] In this embodiment, in step S101, the polychromatic light emitted by the spectral confocal sensor includes a first monochromatic light and a second monochromatic light. The first monochromatic light is focused on the side of the substrate under test near the nozzle, and the second monochromatic light is focused on the nozzle. The first monochromatic light reflects a first optical signal, and the second monochromatic light reflects a second optical signal.

[0043] In this embodiment, the measurement light from the confocal sensor propagates along a predefined path, which sequentially occurs perpendicularly and penetrates the substrate (first thickness segment), crosses the ambient gas gap (second thickness segment, i.e., printing height), and finally reaches the nozzle. The second thickness segment is defined by two planes: the upper surface of the substrate and the lower surface of the nozzle. In other words, in this embodiment, the printing height can be determined by solving for the second thickness segment.

[0044] In this embodiment, in step S102, it can be understood that the working principle of the spectral confocal sensor is as follows: light of different wavelengths is focused at different distances. When light shines perpendicularly onto the surface being measured, only light of a specific wavelength that is precisely focused on that surface will be reflected back to the sensor with high intensity. By analyzing the peak value of the reflection spectrum, the focusing wavelength corresponding to that surface can be determined.

[0045] In this embodiment, instead of directly determining the nozzle incident angle, the exit angle of the second monochromatic light in the air is determined. In a spectral confocal system, a specific wavelength λ corresponds to a specific exit angle. Once the second focusing wavelength is determined, the exit angle of the second monochromatic light, which is the nozzle incident angle, can be determined.

[0046] Define the first focusing wavelength as The second focusing wavelength is The nozzle incident angle is Then the nozzle incident angle Represented as:

[0047]

[0048] In the formula, , , , , , and is a constant coefficient.

[0049] In this embodiment, the focusing distance is established by polynomial fitting. With normalized wavelength The focusing distance-wavelength function is expressed as:

[0050]

[0051] In the formula, is a constant coefficient.

[0052] In this embodiment, the nozzle incident angle The reason for choosing a 6th-order polynomial for the function expression and the focusing distance-wavelength function is that, in precision measurements, to compensate for lens edge distortion and the nonlinearity of the spectrometer itself, the 6th order is considered the upper limit for balancing fitting accuracy and computational stability. Orders exceeding 7 are prone to the Runge phenomenon, leading to severe oscillations at the edges.

[0053] In this embodiment, , , , , , and Specific values ​​and The specific values ​​were obtained through experimental calibration combined with the least squares method.

[0054] In this embodiment, the nozzle incident angle calibration method involves placing the calibration object on a high-precision electric rotary table. At a series of preset angles, a high-precision displacement stage is used to provide a series of standard distances (distances from the sensor to the calibration object). The spectral peak wavelengths are recorded under each state, and finally, the least squares method is used to solve for the spectral peak wavelength. , , , , , and Size.

[0055] In this embodiment, during the focusing distance-wavelength calibration process, a series of standard distances (distances from the sensor to the calibration object) are given along the axial direction using a high-precision displacement stage (such as a nanoscale piezoelectric ceramic stage), and the corresponding spectral peak wavelengths are recorded. Finally, the wavelengths can be solved using the least squares method. Size.

[0056] In this embodiment, the focusing distance corresponding to the first focusing wavelength is defined as... The focusing distance corresponding to the second focusing wavelength is Then the pre-constructed refractive index calculation function Represented as:

[0057]

[0058] In the formula, h represents the thickness of the reference plate being measured in the calibration experiment.

[0059] It should be noted that in the calibration experiment, the thickness of the reference plate under test was measured using other measurement methods, and the specific values ​​of variables such as focusing wavelength and incident angle were obtained. In this way, the true refractive index corresponding to different second focusing wavelengths was calculated, the true refractive index was fitted into a function to form a refractive index pre-calculation function, and then stored in the software for easy subsequent use.

[0060] In this embodiment, in step S103, the pre-constructed initial thickness calculation function is expressed as:

[0061] =

[0062] In the formula, This represents the theoretical height from the substrate being tested to the nozzle.

[0063] In this embodiment, the first focusing wavelength and the second focusing wavelength are substituted into the pre-constructed initial thickness calculation function to calculate the theoretical height value from the substrate to the nozzle.

[0064] Understandably, let the wavelength be... The monochromatic light is focused onto the side of the substrate being tested closest to the nozzle; the wavelength is... The monochromatic light is refracted by the side of the substrate being tested closest to the nozzle and then focused onto the nozzle. If the refractive effect of the substrate is ignored (i.e., its refractive index is assumed to be 1), then the wavelength should be focused at the unrefractive point on the optical axis. This unrefractive point is located between the substrate and the nozzle; under this condition, the calculated thickness is... , which is the axial distance between the two wavelength focal points, represents the initial thickness without refraction compensation.

[0065] In this embodiment, various error sources exist during actual measurement, including but not limited to the focusing error of the confocal optical path system, workpiece installation and manufacturing tolerances (such as measurement equipment installation errors), and random errors caused by environmental factors (such as environmental noise, airflow disturbances, light source instability, and operator intervention). These errors are non-systematic errors. To further improve measurement accuracy and compensate for non-systematic errors, this solution introduces a neural network-based error correction model. This model corrects the non-systematic errors not covered in the initial theoretical height value, ensuring that the measurement error of the final actual height value is ≤ ±0.1 μm. This error correction model uses the focusing wavelength... , As input, it learns and predicts complex error features, and outputs the error compensation amount. .

[0066] In this embodiment, the error correction model employs a three-layer fully connected neural network. The error correction model includes an input layer, a hidden layer, and an output layer. The input layer comprises two neurons, used to receive the first and second focusing wavelengths as input features. The hidden layer comprises four neurons, used to fit the nonlinear relationship between wavelength and error based on the input features. The output layer comprises one neuron, used to map the nonlinear relationship between wavelength and error into a numerical value and output the error compensation amount. The hidden layers use the ReLU activation function, and the output layer uses the linear activation function.

[0067] In this embodiment, experimental data is used as training samples to train and optimize the fully connected neural network, resulting in a pre-trained error correction model.

[0068] Each training sample set includes input features and label values. The input feature is the first focusing wavelength. (Unit: nm) and second focusing wavelength The measured value (unit: nm) is labeled as the difference between the actual printing height and the initial theoretical height (i.e., the true value of the error compensation). The actual printing height is obtained by measuring with a high-precision laser interferometer (measurement accuracy ±0.05μm), and the initial theoretical height is calculated by the initial thickness calculation function. A total of 1000 samples were collected (covering the commonly used measurement ranges of λ1∈[500~550nm] and λ2∈[600~650nm]).

[0069] In this embodiment, in step S104, it is understood that in practical applications, if the substrate being measured is tilted, the incident light will no longer be ideally specularly reflected, especially at larger tilt angles, which will cause spectral signal shift and affect the accuracy of thickness calculation. Even small tilt angles will affect the measurement results. Therefore, if the tilt angle of the substrate being measured is not 0 degrees, a pre-constructed first true thickness calculation function is used to calculate the actual height value from the substrate to the nozzle. The first true thickness calculation function introduces tilt compensation on the basis of error compensation.

[0070] The first true thickness calculation function is expressed as:

[0071]

[0072]

[0073] In the formula, This indicates the actual height value from the measured substrate to the nozzle. Indicates the angle of refraction. Indicates the tilt angle of the substrate being tested. This indicates the amount of error compensation predicted by the error correction model based on the first and second focusing wavelengths.

[0074] In this embodiment, the tilt angle of the substrate under test is measured using an inclinometer.

[0075] Specifically, when constructing the function for calculating the first true thickness, taking a transparent flat plate as an example, such as... Figure 2 As shown, it is a schematic diagram of a transparent flat plate tilted.

[0076] The focusing wavelength remains constant on the upper and lower surfaces of the transparent plate, respectively. and And ignore the peak wavelength shift caused by tilt. The optical axis direction is denoted as... , This indicates the upper surface when it is not tilted, and it is parallel to the optical axis normal. This indicates the angle of deflection of the upper surface around point A relative to the normal direction. The inclined surface, assuming the lower surface Parallel to the upper surface, with an inclination angle of also being .

[0077] Spectral confocal sensors (i.e., confocal probes) emit wavelengths of Monochromatic light at the angle of incidence Focused at point A, wavelength is light at an angle Focusing on point C on the untilted surface, and corresponding to point B on the tilted surface. If refraction is neglected, this ray should focus on point E on the optical axis; however, due to the material's refractive index... The light passed through After refraction, the angle of refraction Incident on the lower surface The light ray ultimately focuses at point D, with the angle between the ray and the optical axis being... . This represents the actual thickness of the transparent plate when the focusing wavelength remains constant on both the upper and lower surfaces. represent and The difference in focusing distance of monochromatic light is the thickness of the plate being measured. This represents the path traveled by the refracted light ray within the transparent plate. and These represent the distance along the straight line AC and the projection length of AB onto the optical axis normal, respectively. and These represent the distance from the perpendicular line AB to point A and the distance from point B, respectively.

[0078] Based on geometric relationships, the following equation holds:

[0079]

[0080]

[0081] The angle of refraction can be obtained. as follows:

[0082]

[0083] Furthermore, based on geometric relationships, the following relation is obtained:

[0084]

[0085]

[0086]

[0087]

[0088]

[0089] Depend on Figure 2 Analysis reveals the true thickness of the transparent plate. Measuring the thickness of a transparent flat plate The true thickness of the transparent plate can be determined. ,in Represented as:

[0090]

[0091] Finally, the expression for the true thickness after tilt compensation can be derived as follows:

[0092]

[0093] Finally, the error compensation amount The first true thickness calculation function is obtained by incorporating the true thickness expression after tilt compensation.

[0094] In this embodiment, if the tilt angle of the substrate being tested is 0 degrees, the first focusing wavelength, the second focusing wavelength, the nozzle incident angle, the refractive index, the theoretical height value, and the error compensation amount are substituted into the pre-constructed second true thickness calculation function to calculate the actual height value from the substrate being tested to the nozzle.

[0095] In this embodiment, since the material actually has a certain refractive index... The theoretical height value needs to be determined. Make corrections to obtain the actual height value from the substrate to the nozzle.

[0096] Specifically, to eliminate measurement errors caused by refraction effects, a second true thickness calculation function was constructed.

[0097] The second true thickness calculation function is expressed as follows:

[0098]

[0099] In the formula, This indicates the actual height value from the measured substrate to the nozzle. This indicates the amount of error compensation predicted by the error correction model based on the first and second focusing wavelengths.

[0100] In this embodiment, Figure 3 The principle of thickness compensation is illustrated using a transparent flat plate as an example. According to the law of refraction, the incident angle of the lower surface is... . Let be the theoretical angle of incidence. Because the light is refracted once at the upper surface of the transparent plate, the light rays that finally reach the lower surface will have a different angle. The actual angle of incidence at the exit.

[0101]

[0102] Considering that the refractive index of air is approximately 1, that is Then the angle of incidence on the lower surface Represented as:

[0103]

[0104] According to geometric relationships, the incident angle and the exit angle satisfy the following formulas respectively:

[0105]

[0106]

[0107] In the formula, This represents the radial displacement of the focused light spot.

[0108] By combining the above relationships, the actual thickness calculation function is derived, which is expressed as:

[0109]

[0110] Finally, the error compensation amount The second true thickness calculation function is obtained by incorporating the actual thickness calculation function.

[0111] The following section demonstrates the complete practical application of the method of this invention using a specific scenario of measuring the printing height on an OLED glass substrate, allowing you to intuitively understand the operation and calculation logic of each step:

[0112] I. The pre-calibration parameters that have been completed are as follows:

[0113]

[0114]

[0115] Test object: Glass substrate for OLED printing (tilt angle) =1°), the printing height from the surface near the nozzle to the nozzle surface needs to be measured.

[0116] II. Specific Measurement Steps

[0117] 1. Optical signal acquisition

[0118] The spectral confocal sensor is activated to emit polychromatic light from the side of the glass substrate away from the nozzle. After the light penetrates the glass substrate, it receives the first light signal from the surface of the substrate near the nozzle and the second light signal from the nozzle surface.

[0119] 2. Determine key optical parameters

[0120] By analyzing the peak light intensity of the optical signal, the first focusing wavelength (substrate side near the nozzle) (λ1=520nm) and the second focusing wavelength (nozzle side) (λ2=610nm) were obtained.

[0121] Substitute the incident angle function to calculate the nozzle incident angle. The result is 3.2°.

[0122] 3. Calculate the refractive index

[0123] λ1=520nm, λ2=610nm and Substituting 3.2° into the pre-built refractive index calculation function, we can calculate the refractive index as 1.51.

[0124] 4. Calculate the theoretical height value

[0125] Substituting λ1=520nm and λ2=610nm into the pre-constructed initial thickness calculation function, we obtain... =304μm.

[0126] 5. Obtain the error compensation amount

[0127] Inputting λ1=520nm and λ2=610nm into the pre-trained error correction model yields the compensation amount. =-0.52μm.

[0128] 6. Calculate the actual printing height under inclined conditions

[0129] Known substrate tilt angle =1°, calculated as follows .

[0130] Then, substituting the values ​​into the first true thickness calculation function, the actual height value is calculated to be 547.98 μm.

[0131] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for measuring and calculating printing height, characterized in that, include: A spectral confocal sensor is used to emit polychromatic light from the side of the substrate away from the nozzle to the substrate and the nozzle surface, and to receive the first light signal reflected from the side of the substrate closer to the nozzle and the second light signal reflected from the nozzle. The first focusing wavelength focused on the side of the substrate near the nozzle and the second focusing wavelength focused on the nozzle and the nozzle incident angle are determined respectively based on the first optical signal and the second optical signal. The first focusing wavelength, the second focusing wavelength and the nozzle incident angle are then substituted into the pre-constructed refractive index calculation function to calculate the refractive index. The first and second focusing wavelengths are substituted into the pre-constructed initial thickness calculation function to calculate the theoretical height from the substrate to the nozzle. The first and second focusing wavelengths are then input into the pre-trained error correction model to obtain the error compensation amount. The tilt angle of the substrate under test is obtained. If the tilt angle of the substrate under test is not 0 degrees, the first focusing wavelength, the second focusing wavelength, the nozzle incident angle, the refractive index, the tilt angle, the theoretical height value, and the error compensation amount are substituted into the pre-constructed first true thickness calculation function to calculate the actual height value from the substrate under test to the nozzle. The first true thickness calculation function is expressed as follows: In the formula, This indicates the actual height value from the measured substrate to the nozzle. Indicates the first focusing wavelength. Indicates the second focusing wavelength. Indicates the nozzle incident angle This indicates the focusing distance corresponding to the first focusing wavelength. This indicates the focusing distance corresponding to the second focusing wavelength. Indicates the angle of refraction. Indicates the tilt angle of the substrate being tested. This indicates the amount of error compensation for the pre-trained error correction model based on the first and second focusing wavelengths in predicting the output. Indicates the refractive index; If the tilt angle of the substrate being tested is 0 degrees, then the first focusing wavelength, the second focusing wavelength, the nozzle incident angle, the refractive index, the theoretical height value, and the error compensation amount are substituted into the pre-constructed second true thickness calculation function to calculate the actual height value from the substrate to the nozzle; the second true thickness calculation function is expressed as: In the formula, This indicates the actual height value from the measured substrate to the nozzle. This indicates the amount of error compensation predicted by the error correction model based on the first and second focusing wavelengths. This represents the theoretical height from the substrate being tested to the nozzle. Indicates the nozzle incident angle. Indicates the first focusing wavelength. Indicates the second focusing wavelength. Indicates the refractive index; The pre-trained error correction model employs a three-layer fully connected neural network to achieve nonlinear compensation for the calculation results of traditional geometric models. The pre-trained error correction model includes an input layer, a hidden layer, and an output layer. The input layer includes two neurons for receiving the first focusing wavelength and the second focusing wavelength as input features. The hidden layer includes four neurons for fitting the nonlinear correlation between wavelength and error based on the input features. The output layer includes one neuron for mapping the nonlinear correlation between wavelength and error to a numerical value and outputting the error compensation amount. The hidden layer uses the ReLU activation function, and the output layer uses the linear activation function.

2. The method for measuring and calculating printing height as described in claim 1, characterized in that, The polychromatic light emitted by the spectral confocal sensor includes a first monochromatic light and a second monochromatic light. The first monochromatic light is focused on the side of the substrate being tested near the nozzle, and the second monochromatic light is focused on the nozzle. The first monochromatic light reflects a first optical signal, and the second monochromatic light reflects a second optical signal.

3. The method for measuring and calculating printing height as described in claim 1, characterized in that, The pre-constructed refractive index calculation function is expressed as: In the formula, Indicates refractive index, Indicates the first focusing wavelength. Indicates the second focusing wavelength. Indicates the nozzle incident angle This indicates the focusing distance corresponding to the first focusing wavelength. The second focusing wavelength corresponds to the focusing distance, and h represents the thickness of the reference plate being measured in the calibration experiment.

4. The method for measuring and calculating printing height as described in claim 1, characterized in that, The pre-constructed initial thickness calculation function is expressed as: = In the formula, This represents the theoretical height from the substrate being tested to the nozzle. This indicates the focusing distance corresponding to the first focusing wavelength. This indicates the focusing distance corresponding to the second focusing wavelength.

Images