Preparation control optimization method of metal ytterbium target for OLED display screen

Abstract

The invention discloses a preparation control optimization method of a metal ytterbium target for an OLED (Organic Light Emitting Diode) display screen. The method comprises the following steps: acquiring impurity spectrum scanning data of a metal ytterbium raw material, analyzing impurity distribution characteristics and determining a melting control temperature; after a cast ingot is formed, the internal pore distribution condition is obtained, a pore density gradient field is constructed, a forging and pressing path planning vector is generated, and precise forging and pressing path design is achieved; in the forging and pressing process, forging and pressing vibration data and displacement data are collected in real time, the influence of vibration on cast ingot position deviation is analyzed, then vibration deviation correction is conducted on a forging and pressing path, and an optimized path plan is formed. And by iteratively optimizing the forging and pressing path, the density of the cast ingot is gradually improved and meets the preset requirement. The method effectively solves the problems of difficult impurity control, non-uniform pore distribution, forging and pressing deviation and the like in the preparation process of the metal ytterbium target material, and the purity and compactness of the target material are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of ytterbium target preparation technology, and particularly to a method for controlling and optimizing the preparation of ytterbium targets for OLED displays. Background Technology

[0002] OLED displays typically consist of a transparent anode, a light-emitting layer, and a metal cathode. The cathode layer has a significant impact on the device's electrical performance, optical performance, and lifespan. Traditional cathode materials often use magnesium or silver, while cathode layers made using ytterbium sputtering targets can further improve the resolution and color performance of OLED screens and reduce energy consumption. Therefore, the quality control of ytterbium sputtering targets is directly related to the consistency and reliability of OLED displays.

[0003] Existing ytterbium target preparation processes mainly employ vacuum casting or extrusion into ingots followed by processing. However, the following problems still exist: First, the melting process is difficult to effectively remove impurity elements, which can easily lead to the formation of compound defects in the cathode film, resulting in abnormal current efficiency, severe localized heating, and consequently shortening the device lifespan. Second, micropores are commonly found inside the ingots, affecting the material density and crystal orientation consistency, leading to uneven film thickness during evaporation and reducing display screen consistency. Third, insufficient cooling and forging control during processing can easily introduce new impurities or cause displacement deviations, further affecting the purity and uniformity of the ytterbium target.

[0004] Therefore, there is an urgent need for a technical method to control impurities, regulate pore distribution, and optimize forging path in the ytterbium target preparation process, so as to improve the purity, density, and overall performance of the metal ytterbium target and thus meet the demand of OLED displays for high-performance cathode materials. Summary of the Invention

[0005] To address at least one of the aforementioned technical problems, this invention proposes a method for controlling and optimizing the fabrication of a metal ytterbium target for OLED displays.

[0006] The first aspect of this invention provides a method for controlling and optimizing the fabrication of a ytterbium target for OLED displays, comprising: Acquire impurity spectral scanning data of ytterbium metal raw material, determine the impurity distribution characteristics of ytterbium metal raw material based on the impurity spectral scanning data, and determine the melting control temperature of ytterbium metal raw material based on the impurity distribution characteristics; The distribution of pores inside the ingot after melting is obtained, a pore density gradient field is constructed, and a forging path planning vector is generated based on the density gradient field. The forging operation is performed on the ingot according to the forging path planning vector. The forging vibration data and the real-time displacement data of the ingot are acquired in real time. The influence of the forging vibration on the ingot position offset is determined based on the forging vibration data and the real-time displacement data, and the influence data is obtained. Based on the influence data, the forging path planning vector is corrected for vibration offset to obtain the vibration offset optimized forging path planning vector. The forging path of the ingot is optimized based on the vibration offset optimization forging path planning vector until the ingot density reaches the preset value.

[0007] In this scheme, the steps of acquiring impurity spectral scanning data of ytterbium metal raw materials, determining the impurity distribution characteristics of ytterbium metal raw materials based on the impurity spectral scanning data, and determining the melting control temperature of ytterbium metal raw materials based on the impurity distribution characteristics are as follows: The surface of ytterbium raw material was scanned in a grid pattern using a laser-induced breakdown spectrometer to obtain spectral scanning data containing characteristic peaks of impurity elements. A mapping model between the intensity of characteristic peaks of impurity elements and impurity concentration was established. A three-dimensional impurity concentration distribution map was generated based on the characteristic peak intensity of each grid point in the spectral scanning data. The overall impurity concentration variance is calculated based on the three-dimensional impurity concentration distribution map. When the variance exceeds a preset threshold, it is determined to be an uneven impurity distribution. For non-uniform impurity distribution, Fourier transform is used to perform spatial frequency analysis on the three-dimensional impurity concentration distribution spectrum to extract the main spatial frequency components and their amplitudes. The temperature rise rate control curve of the melting process is determined based on the wavelength characteristics of the main spatial frequency components. Multiple thermocouple arrays are set in the melting furnace to monitor the temperature gradient of the molten pool in real time. The ytterbium raw material is melted according to the temperature rise rate control curve. When a local temperature fluctuation is detected to exceed the allowable range, dynamic compensation is implemented by adjusting the power distribution of the induction coil.

[0008] In this scheme, the step of obtaining the porosity distribution inside the ingot after melting, constructing a porosity density gradient field, and generating a forging path planning vector based on the density gradient field is as follows: The ingot obtained after melting was scanned by an industrial CT scanning system to obtain three-dimensional porosity distribution data. The morphological image processing method was used to identify the porosity outline and calculate the porosity area ratio of each fault. A porosity volume density distribution matrix was constructed based on the fault spacing. The rate of change of pore density between adjacent faults is calculated based on the pore volume density distribution matrix. When the rate of change exceeds the dynamic threshold, it is marked as a density abrupt change region. The set of pore spatial coordinates of the density abrupt change region is extracted and Deloitte triangulation is performed to generate a pore spatial topology network. Based on the pore space topology network, the nearest neighbor distance variance of each pore node is calculated. When the variance is less than the uniformity threshold, it is determined to be a cluster of pores. Hierarchical clustering analysis is performed on the cluster of pores to obtain the pore aggregation subdomain. The forging main axis direction vector is calculated based on the centroid coordinates of the subdomain. Based on the porosity distribution matrix, isosurfaces are extracted in three-dimensional space to obtain the principal gradient direction of the porosity density gradient field. When the angle between the forging principal axis direction vector and the principal gradient direction is less than the critical angle, the forging principal axis direction vector is used as the initial forging path reference vector. When the angle is not less than the critical angle, principal component analysis is performed on the porosity aggregation subdomain to recalculate the eigenvectors, and the principal gradient direction is superimposed to synthesize the vectors to obtain the optimized forging direction. An initial forging path planning vector is generated based on the optimized forging direction. At the same time, the second derivative of density at each spatial location is calculated based on the pore density gradient field. When the second derivative exceeds the curvature threshold, an auxiliary forging path node is inserted at the corresponding coordinate point, generating a set of forging path planning vectors containing the main path vector and auxiliary nodes.

[0009] In this scheme, the forging operation is performed on the ingot according to the forging path planning vector, and the forging vibration data and real-time displacement data of the ingot during the forging process are acquired in real time. The influence of the forging vibration on the ingot position offset is determined based on the forging vibration data and the real-time displacement data, and the influence data is obtained. Specifically: The forging operation is performed on the ingot according to the forging path planning vector, and the forging vibration data and the three-dimensional spatial displacement time series data of the ingot during the forging process are acquired in real time. The forging vibration waveform data and the three-dimensional spatial displacement time series data are synchronized and aligned with the timestamp. Wavelet packet transform is used to decompose the frequency band energy of the forging vibration data, and the energy ratio of each frequency band is extracted as the vibration feature vector. At the same time, the three-dimensional spatial displacement time series data are subjected to differential operation to obtain the displacement change rate sequence. Based on the Gram angle field algorithm, the vibration feature vector and displacement change rate sequence are converted into a two-dimensional Gram angle field image. The spatiotemporal correlation features in the image are extracted by the convolutional neural network. The mutual information entropy of the spatiotemporal correlation features is calculated to determine the coupling strength between the vibration energy distribution and the displacement change rate. The time series similarity between forging vibration data and three-dimensional spatial displacement time series data is calculated based on the dynamic time warping algorithm, and the vibration-displacement influence matrix is ​​constructed in combination with the coupling strength. The vibration-displacement influence matrix is ​​reduced in dimensionality by singular value decomposition, and the first k principal components are extracted as the influence data of forging vibration on the ingot position displacement. The singular values ​​corresponding to the principal components represent the contribution weight of different vibration modes to the displacement displacement.

[0010] In this scheme, the step of correcting the forging path planning vector for vibration offset based on the influence data to obtain a vibration offset optimized forging path planning vector is specifically as follows: By using the first k principal components in the impact data and their corresponding contribution weights, a mapping relationship model between forging vibration modes and ingot displacement is established. Based on the recurrent neural network, the real-time collected forging vibration data is used for time-series prediction to obtain the predicted value of vibration energy distribution in the future time window at the current forging position. Based on the mapping relationship model, the predicted value of vibration energy distribution is matched to determine the predicted displacement offset vector of the ingot forging within the future time window, and the projection component of the displacement offset prediction vector on the forging path normal plane is calculated as the effective offset. Based on the effective offset, the neighborhood node set of the corresponding coordinate point is found in the forging path planning vector set. The local correction surface of the forging path is constructed by the Kriging interpolation algorithm. The effective offset is superimposed on the forging path planning vector to obtain the vibration displacement compensation vector. Based on the vibration displacement compensation vector, the forging path planning vector within the future time window is corrected by vibration offset, and the forging correction position of the ingot within the future time window is determined, thus obtaining the vibration offset optimized forging path planning vector.

[0011] In this scheme, the optimization of the ingot forging path based on the vibration offset optimization forging path planning vector until the ingot density reaches a preset value specifically involves: The forging path optimization operation of the ingot is performed according to the vibration offset optimization for forging path planning vector. The forged ingot is then scanned in three dimensions using a synchrotron radiation micro-area CT scanning system to obtain three-dimensional volume data containing the distribution of pores and defects. Morphological image processing methods are used to segment the scanned data, calculate the pixel ratio of the material entity in each tomographic image, convert the actual volume based on the tomographic spacing and pixel size, and obtain the ratio of the total volume of the ingot to the volume of the solid material through integral calculation to obtain the ingot density information. When the density reaches the preset value, the forging is stopped, and the surface impurities of the forged ingot are polished to obtain the metal ytterbium target.

[0012] A second aspect of the present invention also provides a system for controlling and optimizing the fabrication of a ytterbium target for an OLED display screen. The system includes a memory and a processor. The memory includes a program for controlling and optimizing the fabrication of a ytterbium target for an OLED display screen. When the processor executes the program for controlling and optimizing the fabrication of a ytterbium target for an OLED display screen, it performs the following steps: Acquire impurity spectral scanning data of ytterbium metal raw material, determine the impurity distribution characteristics of ytterbium metal raw material based on the impurity spectral scanning data, and determine the melting control temperature of ytterbium metal raw material based on the impurity distribution characteristics; The distribution of pores inside the ingot after melting is obtained, a pore density gradient field is constructed, and a forging path planning vector is generated based on the density gradient field. The forging operation is performed on the ingot according to the forging path planning vector. The forging vibration data and the real-time displacement data of the ingot are acquired in real time. The influence of the forging vibration on the ingot position offset is determined based on the forging vibration data and the real-time displacement data, and the influence data is obtained. Based on the influence data, the forging path planning vector is corrected for vibration offset to obtain the vibration offset optimized forging path planning vector. The forging path of the ingot is optimized based on the vibration offset optimization forging path planning vector until the ingot density reaches the preset value.

[0013] This invention discloses a method for controlling and optimizing the preparation of ytterbium targets for OLED displays. By acquiring impurity spectral scanning data of the ytterbium raw material, the impurity distribution characteristics are analyzed, and the melting control temperature is determined. After ingot formation, the internal porosity distribution is acquired, a porosity density gradient field is constructed, and a forging path planning vector is generated to achieve precise forging path design. During forging, forging vibration and displacement data are collected in real time, and the influence of vibration on ingot position offset is analyzed. Vibration offset correction is then applied to the forging path to form an optimized path plan. Through iterative optimization of the forging path, the ingot density is gradually increased to meet preset requirements. This method effectively solves the problems of impurity control, uneven porosity distribution, and forging offset in the preparation of ytterbium targets, significantly improving the purity and density of the target material. Attached Figure Description

[0014] Figure 1 A flowchart of a method for controlling and optimizing the fabrication of a ytterbium target for an OLED display screen according to the present invention is shown. Figure 2 A flowchart illustrating the present invention for determining the melting control temperature of ytterbium raw materials is shown; Figure 3 The flowchart illustrating the vibration offset optimized forging path planning vector obtained by the present invention is shown. Figure 4 A block diagram of a control and optimization system for the fabrication of a metal ytterbium target for an OLED display screen according to the present invention is shown. Detailed Implementation

[0015] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0016] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0017] Figure 1 A flowchart of a method for controlling and optimizing the fabrication of a metal ytterbium target for an OLED display screen according to the present invention is shown.

[0018] like Figure 1 As shown, the first aspect of the present invention provides a method for controlling and optimizing the fabrication of a ytterbium target for OLED displays, comprising: S102, acquire impurity spectral scanning data of ytterbium metal raw material, determine the impurity distribution characteristics of ytterbium metal raw material based on the impurity spectral scanning data, and determine the melting control temperature of ytterbium metal raw material based on the impurity distribution characteristics. S104, Obtain the porosity distribution inside the ingot after melting, construct a porosity density gradient field, and generate a forging path planning vector based on the density gradient field; S106, perform forging operation on the ingot according to the forging path planning vector, acquire forging vibration data and ingot displacement data in real time, determine the influence of forging vibration on ingot position offset according to the forging vibration data and the real-time displacement data, and obtain influence data. S108, Vibration offset correction is performed on the forging path planning vector according to the influence data to obtain the vibration offset optimized forging path planning vector; S110, optimize the forging path of the ingot according to the vibration offset optimization forging path planning vector until the density of the ingot reaches the preset value.

[0019] It should be noted that by acquiring the impurity spectral scanning data of ytterbium raw materials and analyzing their impurity distribution characteristics, a reasonable melting control temperature can be determined, enabling precise control of the melting process. This effectively reduces the residue of high vapor pressure impurities and gaseous impurities, improving the purity of the raw materials. After ingot formation, the internal porosity distribution is detected, and a porosity density gradient field is constructed. This directly reflects the non-uniformity of the spatial distribution of pores. By generating a forging path planning vector, subsequent forging processes can specifically improve the porosity density distribution, increasing the overall density of the material. During the forging operation... Real-time acquisition of forging vibration data and ingot displacement data can accurately capture the impact of vibration on ingot position offset, avoiding uneven densification caused by path offset. Based on the impact data, vibration offset correction is performed on the forging path planning vector, which can compensate for displacement errors caused by vibration in advance and obtain a more stable and accurate forging path. The corrected path is then used to optimize the forging of the ingot, so that porosity is gradually eliminated and the grain arrangement is more consistent until the ingot density reaches the preset value, thereby ensuring that the purity, density and uniformity of the ytterbium metal target are significantly improved.

[0020] Figure 2 A flowchart illustrating the present invention for determining the melting control temperature of ytterbium raw materials is shown.

[0021] According to an embodiment of the present invention, the steps of obtaining impurity spectral scanning data of ytterbium metal raw material, determining the impurity distribution characteristics of ytterbium metal raw material based on the impurity spectral scanning data, and determining the melting control temperature of ytterbium metal raw material based on the impurity distribution characteristics are specifically as follows: S202, the surface of the ytterbium raw material is scanned in a grid pattern using a laser-induced breakdown spectrometer to obtain spectral scanning data containing characteristic peaks of impurity elements, a mapping relationship model between the intensity of characteristic peaks of impurity elements and impurity concentration is established, and a three-dimensional impurity concentration distribution map is generated based on the characteristic peak intensity of each grid point in the spectral scanning data. S204, calculate the overall impurity concentration variance value based on the three-dimensional impurity concentration distribution map, and determine the non-uniform impurity distribution when the variance value exceeds the preset threshold. S206. For non-uniform impurity distribution, Fourier transform is used to perform spatial frequency analysis on the three-dimensional impurity concentration distribution spectrum to extract the main spatial frequency components and their amplitudes. S208, the temperature rise rate control curve of the melting process is determined according to the wavelength characteristics of the main spatial frequency components. Multiple thermocouple arrays are set in the melting furnace to monitor the temperature gradient of the molten pool in real time. The ytterbium raw material is melted according to the temperature rise rate control curve. When a local temperature fluctuation is detected to exceed the allowable range, dynamic compensation is implemented by adjusting the power distribution of the induction coil.

[0022] It is important to note that in the preparation of ytterbium metal targets, the non-uniform distribution of impurities in the raw materials directly affects the purification effect during the melting stage and the performance of the final target material. Traditional smelting processes typically employ fixed temperature rise curves, which cannot adapt to the spatial differences in the impurity distribution within the raw materials. This leads to insufficient volatilization of high vapor pressure impurities or localized overheating causing melt composition segregation. Especially when impurity elements exhibit a non-uniform aggregation state in ytterbium metal, a constant temperature rise rate causes low-melting-point impurities to prematurely vaporize and escape in the high-temperature region, while high-melting-point impurities struggle to diffuse sufficiently due to uneven heat transfer. High-precision gridded scanning of ytterbium metal raw materials using laser-induced breakdown spectroscopy (LAS) constructs a three-dimensional impurity concentration distribution map, enabling precise identification of the spatial distribution characteristics and degree of non-uniformity of impurity elements. Spatial frequency analysis based on Fourier transform can extract the main periodic features of impurity distribution, thereby determining the optimal temperature rise rate control curve. This ensures that the thermal field distribution during the melting process matches the impurity removal requirements, effectively avoiding insufficient impurity volatilization or melt composition segregation caused by unreasonable temperature gradients. By using multiple thermocouple arrays to monitor the molten pool temperature gradient in real time, and combining this with dynamic adjustment of the induction coil power, rapid compensation for local temperature fluctuations can be achieved, ensuring the thermal field stability of the melting process, thereby improving impurity removal efficiency and melt uniformity. High-frequency components correspond to the rapid temperature rise stage, while low-frequency components correspond to the slow temperature rise stage.

[0023] According to an embodiment of the present invention, the step of obtaining the porosity distribution inside the ingot after melting, constructing a porosity density gradient field, and generating a forging path planning vector based on the density gradient field specifically involves: The ingot obtained after melting was scanned by an industrial CT scanning system to obtain three-dimensional porosity distribution data. The morphological image processing method was used to identify the porosity outline and calculate the porosity area ratio of each fault. A porosity volume density distribution matrix was constructed based on the fault spacing. The rate of change of pore density between adjacent faults is calculated based on the pore volume density distribution matrix. When the rate of change exceeds the dynamic threshold, it is marked as a density abrupt change region. The set of pore spatial coordinates of the density abrupt change region is extracted and Deloitte triangulation is performed to generate a pore spatial topology network. Based on the pore space topology network, the nearest neighbor distance variance of each pore node is calculated. When the variance is less than the uniformity threshold, it is determined to be a cluster of pores. Hierarchical clustering analysis is performed on the cluster of pores to obtain the pore aggregation subdomain. The forging main axis direction vector is calculated based on the centroid coordinates of the subdomain. Based on the porosity distribution matrix, isosurfaces are extracted in three-dimensional space to obtain the principal gradient direction of the porosity density gradient field. When the angle between the forging principal axis direction vector and the principal gradient direction is less than the critical angle, the forging principal axis direction vector is used as the initial forging path reference vector. When the angle is not less than the critical angle, principal component analysis is performed on the porosity aggregation subdomain to recalculate the eigenvectors, and the principal gradient direction is superimposed to synthesize the vectors to obtain the optimized forging direction. It should be noted that the centroid coordinates of the pore aggregation subdomains reflect the spatial centroid position of the pore distribution in that region, while the lines connecting the centroids of different subdomains characterize the macroscopic extension direction of the pore aggregation region. By calculating the principal component directions of these centroid points, the main spatial distribution trend of the pore aggregation region can be determined. This principal axis direction vector can guide the forging force direction to be consistent with the longest extension direction of the pore aggregation region, thereby ensuring that the forging force can be applied along the path with the densest pores, maximizing the closure of pores and the densification of the material. The optimized forging direction is achieved by linearly combining the eigenvectors obtained from principal component analysis with the principal gradient direction of the pore density gradient field, using a dynamic weight allocation strategy. The weighting coefficients are adaptively adjusted according to the angle between the two, so that the synthesized optimized forging direction maintains consistency with the extension direction of the pore aggregation region. The dynamic threshold is obtained by statistically learning the porosity change rate of historical qualified ingots using the sliding window method; the uniformity threshold is determined by Monte Carlo simulation of the distance variance distribution of randomly distributed porosity nodes; the critical angle is determined by finite element analysis of the coupling relationship between the forging stress field and the porosity orientation; and the curvature threshold is determined based on the dynamic relationship model between the maximum acceleration of the forging press and the yield strength of the material.

[0024] An initial forging path planning vector is generated based on the optimized forging direction. At the same time, the second derivative of density at each spatial location is calculated based on the pore density gradient field. When the second derivative exceeds the curvature threshold, an auxiliary forging path node is inserted at the corresponding coordinate point, generating a set of forging path planning vectors containing the main path vector and auxiliary nodes.

[0025] It should be noted that high-precision 3D imaging of the porosity distribution inside the ingot is achieved using an industrial CT scanning system, combined with morphological image processing technology to accurately quantify the spatial distribution characteristics of the porosity. Dynamic threshold analysis based on the porosity volume density distribution matrix can automatically identify regions of abrupt density changes, and a spatial topological network of porosity is constructed through Delaunay triangulation, effectively revealing the microstructural characteristics of porosity aggregation. Through hierarchical clustering analysis and principal component analysis, this scheme can intelligently identify porosity aggregation subdomains and calculate the optimal forging axis direction, ensuring that the forging pressure is precisely applied to areas with dense porosity. Simultaneously, through isosurface extraction and second derivative calculation of the porosity density gradient field, this technology can automatically generate an optimized forging path plan including the main path and auxiliary nodes, enabling the forging process to focus on porosity aggregation areas and dynamically adjust the force direction based on local curvature changes. This intelligent forging path planning based on porosity distribution characteristics significantly improves the targeting and efficiency of the forging process, effectively eliminates porosity defects inside the ingot, and improves material density and microstructural uniformity. When the second derivative of the pore density distribution exceeds the curvature threshold, it indicates the presence of a significant density change inflection point or local extremum in that region. These locations often correspond to abrupt boundaries or stress concentration areas in the pore distribution. By inserting auxiliary forging path nodes at the corresponding coordinate points, additional forging forces can be applied to these special locations, effectively eliminating material inhomogeneities caused by abrupt changes in density gradient and preventing stress concentration or microscopic defects during subsequent processing. The weighting coefficient of the main path vector is positively correlated with the local extrema of the pore density gradient field; the set of forging path planning vectors represents the position of each forging operation.

[0026] According to an embodiment of the present invention, the forging operation is performed on the ingot according to the forging path planning vector, and the forging vibration data and real-time displacement data of the ingot during the forging process are acquired in real time. The influence of the forging vibration on the ingot position offset is determined based on the forging vibration data and the real-time displacement data to obtain influence data. Specifically, the following steps are taken: The forging operation is performed on the ingot according to the forging path planning vector, and the forging vibration data and the three-dimensional spatial displacement time series data of the ingot during the forging process are acquired in real time. The forging vibration waveform data and the three-dimensional spatial displacement time series data are synchronized and aligned with the timestamp. Wavelet packet transform is used to decompose the frequency band energy of the forging vibration data, and the energy ratio of each frequency band is extracted as the vibration feature vector. At the same time, the three-dimensional spatial displacement time series data are subjected to differential operation to obtain the displacement change rate sequence. Based on the Gram angle field algorithm, the vibration feature vector and displacement change rate sequence are converted into a two-dimensional Gram angle field image. The spatiotemporal correlation features in the image are extracted by the convolutional neural network. The mutual information entropy of the spatiotemporal correlation features is calculated to determine the coupling strength between the vibration energy distribution and the displacement change rate. The time series similarity between forging vibration data and three-dimensional spatial displacement time series data is calculated based on the dynamic time warping algorithm, and the vibration-displacement influence matrix is ​​constructed in combination with the coupling strength. The vibration-displacement influence matrix is ​​reduced in dimensionality by singular value decomposition, and the first k principal components are extracted as the influence data of forging vibration on the ingot position displacement. The singular values ​​corresponding to the principal components represent the contribution weight of different vibration modes to the displacement displacement.

[0027] It should be noted that during the forging process of ytterbium metal targets, mechanical vibration can cause the ingot to vibrate, resulting in the actual forging path deviating from the preset trajectory, which seriously affects the uniformity of material densification and the final target performance. Therefore, by real-time acquisition of forging vibration waveforms and three-dimensional displacement data of the ingot, wavelet packet transform and differential operation are used to extract the vibration frequency band energy characteristics and displacement change trends, respectively. The Gram angle field algorithm is used to convert the time series data into a two-dimensional feature image. A convolutional neural network is used to mine the deep spatiotemporal correlation between vibration and displacement. Combined with dynamic time warping and mutual information entropy analysis, the contribution of different vibration modes to displacement is quantified. Finally, singular value decomposition is used to obtain the core influence data characterizing the vibration-displacement coupling relationship. The influence data is essentially a quantitative index reflecting the intensity of the effect of different vibration characteristics on displacement. By decomposing the time series vibration signal into physically meaningful frequency band energy characteristics and performing spatiotemporal correlation analysis with the displacement change rate, the contribution law of vibration in a specific frequency band to displacement disturbance can be accurately identified. The vibration-displacement influence matrix represents the influence intensity on ingot displacement under different vibration frequencies and intensities, i.e., the displacement magnitude of the ingot during the forging process under different vibration frequencies and intensities; the vibration energy distribution characterizes the vibration frequency and intensity. k is a preset value. Vibration modes refer to specific frequency vibration components generated by the mechanical movement of the equipment or material deformation during the forging process, reflecting the distribution characteristics of vibration energy in different frequency bands within the forging system.

[0028] Figure 3 The flowchart illustrating the vibration offset optimized forging path planning vector obtained by the present invention is shown.

[0029] According to an embodiment of the present invention, the step of correcting the forging path planning vector for vibration offset based on the influence data to obtain a vibration offset optimized forging path planning vector specifically involves: S302, by using the first k principal components in the influence data and their corresponding contribution weights, a mapping relationship model between forging vibration mode and ingot displacement is established. Based on the recurrent neural network, the real-time collected forging vibration data is used for time-series prediction to obtain the predicted value of vibration energy distribution in the future time window at the current forging position. S304, Based on the mapping relationship model, the predicted value of vibration energy distribution is matched to determine the predicted displacement offset vector of the ingot forging within the future time window, and the projection component of the displacement offset prediction vector on the forging path normal plane is calculated as the effective offset. S306, based on the effective offset, find the neighborhood node set of the corresponding coordinate point in the forging path planning vector set, construct the local correction surface of the forging path through the Kriging interpolation algorithm, and superimpose the effective offset onto the forging path planning vector to obtain the vibration displacement compensation vector. S308, based on the vibration displacement compensation vector, the forging path planning vector within the future time window is corrected by vibration offset, the forging correction position of the ingot within the future time window is determined, and the vibration offset optimized forging path planning vector is obtained.

[0030] It should be noted that by establishing a mapping model between forging vibration modes and ingot displacement, and combining this with a recurrent neural network to predict the vibration energy distribution during the forging process, the potential displacement within a future time window can be predicted in advance, enabling feedforward correction in forging path planning. By superimposing the predicted effective offset onto the original path vector and constructing a local correction surface using the Kriging interpolation algorithm, smooth path compensation can be achieved, avoiding path abrupt changes and processing instability caused by direct correction. This method can significantly reduce the interference of vibration on the accuracy of the forging path, enabling the ingot to maintain higher path consistency and displacement stability during forging, thereby improving the uniformity of the densification process and the consistency of grain orientation, ultimately obtaining a ytterbium metal target with a more uniform internal structure and higher density.

[0031] According to an embodiment of the present invention, the optimization of the ingot forging path based on the vibration offset optimization forging path planning vector until the ingot density reaches a preset value specifically includes: The forging path optimization operation of the ingot is performed according to the vibration offset optimization for forging path planning vector. The forged ingot is then scanned in three dimensions using a synchrotron radiation micro-area CT scanning system to obtain three-dimensional volume data containing the distribution of pores and defects. Morphological image processing methods are used to segment the scanned data, calculate the pixel ratio of the material entity in each tomographic image, convert the actual volume based on the tomographic spacing and pixel size, and obtain the ratio of the total volume of the ingot to the volume of the solid material through integral calculation to obtain the ingot density information. When the density reaches the preset value, the forging is stopped, and the surface impurities of the forged ingot are polished to obtain the metal ytterbium target.

[0032] It should be noted that by introducing a vibration-optimized forging path planning vector to optimize the ingot path, and combining this with a synchrotron radiation micro-area CT scanning system to acquire real-time three-dimensional tomographic data of the forged ingot, the spatial distribution of porosity and defects can be accurately grasped throughout the process, enabling precise evaluation of the densification effect. Through morphological image processing and volume calculation methods, the ratio of the solid volume to the total volume within the ingot can be quantified, thus obtaining objective and traceable density information. Based on density feedback, the decision to continue forging can be dynamically made, avoiding microstructural damage or unnecessary energy consumption caused by over-forging, while ensuring that the process is terminated promptly when the ingot reaches the preset density target.

[0033] According to an embodiment of the present invention, it further includes: The surface of ytterbium raw material is continuously and dynamically scanned using a laser-induced breakdown spectrometer to acquire a real-time updated impurity spectral scanning data stream. A time-series map of impurity concentration distribution with timestamp synchronization is established based on this data stream. The impurity concentration gradient change rate in the currently scanned area is calculated based on this time-series map, and the impurity distribution trend in the unscanned area along the furnace feed direction is predicted using this rate. A high-speed infrared thermal imager is deployed at the furnace feed inlet to capture the surface temperature field distribution of the ytterbium raw material in real time. A melt feedforward control model is constructed based on the spatiotemporal mapping relationship between the temperature field distribution and the impurity concentration gradient change rate. Dynamic temperature control compensation commands are generated through this model to adjust the power distribution of the induction coil to match the predicted melting demand in the impurity area. When the infrared thermal imager detects abnormal local temperature fluctuations, the spectral scanning data stream is resampled, and the impurity concentration distribution time-series map is updated based on the resampling results, and the temperature control compensation commands are recalculated.

[0034] According to an embodiment of the present invention, the step of continuously and dynamically scanning the surface of the ytterbium raw material using a laser-induced breakdown spectrometer to obtain a real-time updated impurity spectral scanning data stream, and establishing a time-series map of impurity concentration distribution with timestamp synchronization based on the impurity spectral scanning data stream, specifically includes: A multi-threaded parallel processing architecture is employed to analyze the scanning data from the laser-induced breakdown spectrometer in real time. The original spectral signal is decomposed into a characteristic peak intensity matrix. The intensity matrix is ​​matched and identified according to a preset impurity element characteristic peak library to generate quantified impurity concentration values ​​for each grid point. A mapping relationship between spatial coordinates and timestamps is established based on the kinematic parameters of the scanning mechanism. The quantified impurity concentration values ​​are then filled into a three-dimensional grid model according to the time sequence, forming a time-series map of impurity concentration distribution. A sliding window Fourier analysis is performed on the time-series map to extract the spatial frequency features of impurity concentration within the window. The periodic or abrupt patterns of impurity distribution are identified based on the spatial frequency features. When an abrupt pattern is detected, the grid density of the scanning path is automatically adjusted, the number of laser sampling points is increased in the abrupt region, and the high-density scanning data is fed back to the melting feedforward control model in real time for temperature rise rate correction.

[0035] It should be noted that during the preparation of ytterbium targets, due to the non-uniform distribution of impurities in the raw materials and the extremely high temperature control requirements of the melting process, traditional static spectral scanning methods struggle to capture the dynamic changes in impurity distribution in real time. This results in a lag in temperature control response during the melting process, which in turn affects the purity and performance of the target material. By using a laser-induced breakdown spectrometer to continuously and dynamically scan the surface of the ytterbium raw materials, real-time impurity spectral data streams are acquired, and a time-series map of impurity concentration distribution with timestamp synchronization is constructed, thereby achieving dynamic monitoring of impurity distribution. Based on this time-series map, the impurity concentration gradient change rate is calculated, and combined with the surface temperature field distribution captured by an infrared thermal imager, a melt feedforward control model is established to dynamically adjust the power distribution of the induction coil, ensuring precise matching between the melting temperature and the impurity distribution. Simultaneously, when a local temperature anomaly or a sudden change in impurity distribution is detected, the system automatically triggers a resampling mechanism to optimize the scanning path and grid density, further improving data real-time performance and control accuracy.

[0036] Figure 4 A block diagram of a control and optimization system for the fabrication of a metal ytterbium target for an OLED display screen according to the present invention is shown.

[0037] A second aspect of the present invention also provides a system for controlling and optimizing the fabrication of a ytterbium target for an OLED display screen. The system includes a memory 401, a processor 402, and a communication interface 403. The memory includes a program for controlling and optimizing the fabrication of a ytterbium target for an OLED display screen. The communication interface is used for data communication between the memory and the processor. When the processor executes the program for controlling and optimizing the fabrication of a ytterbium target for an OLED display screen, it performs the following steps: Acquire impurity spectral scanning data of ytterbium metal raw material, determine the impurity distribution characteristics of ytterbium metal raw material based on the impurity spectral scanning data, and determine the melting control temperature of ytterbium metal raw material based on the impurity distribution characteristics; The distribution of pores inside the ingot after melting is obtained, a pore density gradient field is constructed, and a forging path planning vector is generated based on the density gradient field. The forging operation is performed on the ingot according to the forging path planning vector. The forging vibration data and the real-time displacement data of the ingot are acquired in real time. The influence of the forging vibration on the ingot position offset is determined based on the forging vibration data and the real-time displacement data, and the influence data is obtained. Based on the influence data, the forging path planning vector is corrected for vibration offset to obtain the vibration offset optimized forging path planning vector. The forging path of the ingot is optimized based on the vibration offset optimization forging path planning vector until the ingot density reaches the preset value.

[0038] This invention discloses a method for controlling and optimizing the preparation of ytterbium targets for OLED displays. By acquiring impurity spectral scanning data of the ytterbium raw material, the impurity distribution characteristics are analyzed, and the melting control temperature is determined. After ingot formation, the internal porosity distribution is acquired, a porosity density gradient field is constructed, and a forging path planning vector is generated to achieve precise forging path design. During forging, forging vibration and displacement data are collected in real time, and the influence of vibration on ingot position offset is analyzed. Vibration offset correction is then applied to the forging path to form an optimized path plan. Through iterative optimization of the forging path, the ingot density is gradually increased to meet preset requirements. This method effectively solves the problems of impurity control, uneven porosity distribution, and forging offset in the preparation of ytterbium targets, significantly improving the purity and density of the target material.

[0039] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0040] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0041] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0042] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0043] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0044] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for controlling and optimizing the fabrication of a ytterbium target for OLED displays, characterized in that, Includes the following steps: Acquire impurity spectral scanning data of ytterbium metal raw material, determine the impurity distribution characteristics of ytterbium metal raw material based on the impurity spectral scanning data, and determine the melting control temperature of ytterbium metal raw material based on the impurity distribution characteristics; The distribution of pores inside the ingot after melting is obtained, a pore density gradient field is constructed, and a forging path planning vector is generated based on the density gradient field. The forging operation is performed on the ingot according to the forging path planning vector. The forging vibration data and the real-time displacement data of the ingot are acquired in real time. The influence of the forging vibration on the ingot position offset is determined based on the forging vibration data and the real-time displacement data, and the influence data is obtained. Based on the influence data, the forging path planning vector is corrected for vibration offset to obtain the vibration offset optimized forging path planning vector. The forging path of the ingot is optimized based on the vibration offset optimization forging path planning vector until the ingot density reaches the preset value.

2. The method for controlling and optimizing the fabrication of a metal ytterbium target for an OLED display screen according to claim 1, characterized in that, The process of acquiring impurity spectral scanning data of ytterbium metal raw materials, determining the impurity distribution characteristics of the ytterbium metal raw materials based on the impurity spectral scanning data, and determining the melting control temperature of the ytterbium metal raw materials based on the impurity distribution characteristics, specifically involves: The surface of ytterbium raw material was scanned in a grid pattern using a laser-induced breakdown spectrometer to obtain spectral scanning data containing characteristic peaks of impurity elements. A mapping model between the intensity of characteristic peaks of impurity elements and impurity concentration was established. A three-dimensional impurity concentration distribution map was generated based on the characteristic peak intensity of each grid point in the spectral scanning data. The overall impurity concentration variance is calculated based on the three-dimensional impurity concentration distribution map. When the variance exceeds a preset threshold, it is determined to be an uneven impurity distribution. For non-uniform impurity distribution, Fourier transform is used to perform spatial frequency analysis on the three-dimensional impurity concentration distribution spectrum to extract the main spatial frequency components and their amplitudes. The temperature rise rate control curve of the melting process is determined based on the wavelength characteristics of the main spatial frequency components. Multiple thermocouple arrays are set in the melting furnace to monitor the temperature gradient of the molten pool in real time. The ytterbium raw material is melted according to the temperature rise rate control curve. When a local temperature fluctuation is detected to exceed the allowable range, dynamic compensation is implemented by adjusting the power distribution of the induction coil.

3. The method for controlling and optimizing the fabrication of a metal ytterbium target for an OLED display screen according to claim 1, characterized in that, The process involves obtaining the porosity distribution inside the ingot after melting, constructing a porosity density gradient field, and generating a forging path planning vector based on the density gradient field. Specifically: The ingot obtained after melting was scanned by an industrial CT scanning system to obtain three-dimensional porosity distribution data. The morphological image processing method was used to identify the porosity outline and calculate the porosity area ratio of each fault. A porosity volume density distribution matrix was constructed based on the fault spacing. The rate of change of pore density between adjacent faults is calculated based on the pore volume density distribution matrix. When the rate of change exceeds the dynamic threshold, it is marked as a density abrupt change region. The set of pore spatial coordinates of the density abrupt change region is extracted and Deloitte triangulation is performed to generate a pore spatial topology network. Based on the pore space topology network, the nearest neighbor distance variance of each pore node is calculated. When the variance is less than the uniformity threshold, it is determined to be a cluster of pores. Hierarchical clustering analysis is performed on the cluster of pores to obtain the pore aggregation subdomain. The forging main axis direction vector is calculated based on the centroid coordinates of the subdomain. Based on the porosity distribution matrix, isosurfaces are extracted in three-dimensional space to obtain the principal gradient direction of the porosity density gradient field. When the angle between the forging principal axis direction vector and the principal gradient direction is less than the critical angle, the forging principal axis direction vector is used as the initial forging path reference vector. When the angle is not less than the critical angle, principal component analysis is performed on the porosity aggregation subdomain to recalculate the eigenvectors, and the principal gradient direction is superimposed to synthesize the vectors to obtain the optimized forging direction. An initial forging path planning vector is generated based on the optimized forging direction. At the same time, the second derivative of density at each spatial location is calculated based on the pore density gradient field. When the second derivative exceeds the curvature threshold, an auxiliary forging path node is inserted at the corresponding coordinate point, generating a set of forging path planning vectors containing the main path vector and auxiliary nodes.

4. The method for controlling and optimizing the fabrication of a metal ytterbium target for an OLED display screen according to claim 1, characterized in that, The forging operation is performed on the ingot according to the forging path planning vector, and the forging vibration data and real-time displacement data of the ingot during the forging process are acquired in real time. The influence of the forging vibration on the ingot position offset is determined based on the forging vibration data and the real-time displacement data, and the influence data is obtained, specifically as follows: The forging operation is performed on the ingot according to the forging path planning vector, and the forging vibration data and the three-dimensional spatial displacement time series data of the ingot during the forging process are acquired in real time. The forging vibration waveform data and the three-dimensional spatial displacement time series data are synchronized and aligned with the timestamp. Wavelet packet transform is used to decompose the frequency band energy of the forging vibration data, and the energy ratio of each frequency band is extracted as the vibration feature vector. At the same time, the three-dimensional spatial displacement time series data are subjected to differential operation to obtain the displacement change rate sequence. Based on the Gram angle field algorithm, the vibration feature vector and displacement change rate sequence are converted into a two-dimensional Gram angle field image. The spatiotemporal correlation features in the image are extracted by the convolutional neural network. The mutual information entropy of the spatiotemporal correlation features is calculated to determine the coupling strength between the vibration energy distribution and the displacement change rate. The time series similarity between forging vibration data and three-dimensional spatial displacement time series data is calculated based on the dynamic time warping algorithm, and the vibration-displacement influence matrix is ​​constructed in combination with the coupling strength. The vibration-displacement influence matrix is ​​reduced in dimensionality by singular value decomposition, and the first k principal components are extracted as the influence data of forging vibration on the ingot position displacement. The singular values ​​corresponding to the principal components represent the contribution weight of different vibration modes to the displacement displacement.

5. The method for controlling and optimizing the fabrication of a metal ytterbium target for an OLED display screen according to claim 1, characterized in that, The step of correcting the forging path planning vector based on the influence data to obtain a vibration-optimized forging path planning vector is as follows: By using the first k principal components in the impact data and their corresponding contribution weights, a mapping relationship model between forging vibration modes and ingot displacement is established. Based on the recurrent neural network, the real-time collected forging vibration data is used for time-series prediction to obtain the predicted value of vibration energy distribution in the future time window at the current forging position. Based on the mapping relationship model, the predicted value of vibration energy distribution is matched to determine the predicted displacement offset vector of the ingot forging within the future time window, and the projection component of the displacement offset prediction vector on the forging path normal plane is calculated as the effective offset. Based on the effective offset, the neighborhood node set of the corresponding coordinate point is found in the forging path planning vector set. The local correction surface of the forging path is constructed by the Kriging interpolation algorithm. The effective offset is superimposed on the forging path planning vector to obtain the vibration displacement compensation vector. Based on the vibration displacement compensation vector, the forging path planning vector within the future time window is corrected by vibration offset, and the forging correction position of the ingot within the future time window is determined, thus obtaining the vibration offset optimized forging path planning vector.

6. The method for controlling and optimizing the fabrication of a metal ytterbium target for an OLED display screen according to claim 1, characterized in that, The optimization of the forging path of the ingot based on the vibration offset optimization for the forging path planning vector until the ingot density reaches a preset value is specifically as follows: The forging path optimization operation of the ingot is performed according to the vibration offset optimization for forging path planning vector. The forged ingot is then scanned in three dimensions using a synchrotron radiation micro-area CT scanning system to obtain three-dimensional volume data containing the distribution of pores and defects. Morphological image processing methods are used to segment the scanned data, calculate the pixel ratio of the material entity in each tomographic image, convert the actual volume based on the tomographic spacing and pixel size, and obtain the ratio of the total volume of the ingot to the volume of the solid material through integral calculation to obtain the ingot density information. When the density reaches the preset value, the forging is stopped, and the surface impurities of the forged ingot are polished to obtain the metal ytterbium target.

7. A system for controlling and optimizing the fabrication of a ytterbium target for OLED displays, characterized in that, The ytterbium target fabrication control and optimization system for OLED displays includes a memory and a processor. The memory includes a method program for the preparation control and optimization of ytterbium targets for OLED displays. When the processor executes the method program for the preparation control and optimization of ytterbium targets for OLED displays, it performs the following steps: Acquire impurity spectral scanning data of ytterbium metal raw material, determine the impurity distribution characteristics of ytterbium metal raw material based on the impurity spectral scanning data, and determine the melting control temperature of ytterbium metal raw material based on the impurity distribution characteristics; The distribution of pores inside the ingot after melting is obtained, a pore density gradient field is constructed, and a forging path planning vector is generated based on the density gradient field. The forging operation is performed on the ingot according to the forging path planning vector. The forging vibration data and the real-time displacement data of the ingot are acquired in real time. The influence of the forging vibration on the ingot position offset is determined based on the forging vibration data and the real-time displacement data, and the influence data is obtained. Based on the influence data, the forging path planning vector is corrected for vibration offset to obtain the vibration offset optimized forging path planning vector. The forging path of the ingot is optimized based on the vibration offset optimization forging path planning vector until the ingot density reaches the preset value.

Images