Deterministic Generation Method of Ion Beam Removal Function Based on Online Regulation of Ion Beam Current

Abstract

The present invention discloses a method for deterministically generating an ion beam removal function based on on-line regulation of ion beam current, which includes inputting the peak removal efficiency and full width at half maximum of the required removal function into a first neural network model to obtain the peak beam current density and full width at half maximum of the ion beam current, and then inputting the obtained values into a second neural network model to obtain the target values of the aperture diameter and target distance. The aperture diameter and target distance of the ion beam device are adjusted to the target values, and then the aperture diameter and target distance of the ion beam device are finely adjusted until the beam current density of the ion beam current meets the requirements; compensating for the positioning error of the removal function; and controlling the ion beam device to perform tests on the workpiece to realize the deterministic generation of the required removal function. The present invention aims to solve the problems that the parameters of the removal function in ion beam processing in a vacuum environment are unknown and uncontrollable, and the process of fabricating the removal function is cumbersome, so as to realize the deterministic generation of the ion beam removal function.

Description

Technical Field

[0001] The present invention relates to the field of ion beam ultra-precision machining, and particularly to a method for deterministically generating an ion beam removal function based on on-line regulation of ion beam current. Background Art

[0002] As an advanced ultra-precision machining method, the ion beam shaping technology plays a crucial role in improving the accuracy and machining efficiency of optical elements, and has significant technical advantages in the field of optical element manufacturing. However, under the physical constraints of the vacuum environment, the parameters of the removal function are unknown and uncontrollable during actual machining. The ion beam current intensity, energy distribution, and beam diameter morphology are easily affected by the coupling effects of multiple physical fields such as vacuum degree fluctuation, cathode life attenuation, and electromagnetic field interference, resulting in changes in the removal function parameters during the machining process, thus causing a deviation between the actual removal amount and the simulation calculation. Moreover, to produce the removal function for calculating the dwell time, it is necessary to open and close the vacuum chamber and ion source multiple times. After performing point etching on the optical element and measuring the front and back surface shape errors, the difference is taken to obtain it. And it reduces the machining efficiency and machining accuracy of the ion beam. Summary of the Invention

[0003] The technical problem to be solved by the present invention: Aiming at the above problems of the prior art, a method for deterministically generating an ion beam removal function based on on-line regulation of ion beam current is provided. The present invention aims to solve the problems that the removal function parameters in ion beam machining under a vacuum environment are unknown and uncontrollable, and the process of producing the removal function is cumbersome, and realizes the deterministic generation of the ion beam removal function.

[0004] To solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0005] A method for deterministically generating an ion beam removal function based on on-line regulation of ion beam current, comprising the following steps:

[0006] S1, inputting the peak removal efficiency and full width at half maximum of the required removal function into a pre-trained first neural network model to obtain the corresponding peak beam current density and full width at half maximum of the ion beam current. The first neural network model is pre-trained to establish a mapping relationship between the peak removal efficiency and full width at half maximum of the removal function, and the peak beam current density and full width at half maximum of the ion beam current;

[0007] S2, inputting the obtained peak beam current density and full width at half maximum of the ion beam current into a pre-trained second neural network model to obtain the target values of the corresponding aperture and target distance. The second neural network model is pre-trained to establish a mapping relationship between the peak beam current density and full width at half maximum of the ion beam current, and the target values of the aperture and target distance;

[0008] S3. Adjust the aperture diameter and target distance of the ion beam equipment to the target values of the obtained aperture diameter and target distance, and detect the beam current density of the ion beam. Determine whether the beam current density of the ion beam meets the requirements. If it does not meet the requirements, finely adjust the aperture diameter and target distance of the ion beam equipment until the beam current density of the ion beam meets the requirements;

[0009] S4. Adjust the ion beam equipment to compensate for the positioning error of the removal function;

[0010] S5. Control the ion beam equipment to perform tests on the workpiece to achieve the deterministic generation of the required removal function.

[0011] Optionally, the function expression for compensating the positioning error of the removal function in step S4 is:

[0012] ,

[0013] where, is the center coordinate of the Faraday cup, is the center coordinate of the workpiece, is the offset relationship between the workpiece and the Faraday cup, is the relative coordinate of the measured beam center and the Faraday cup center, is the relative coordinate position of the beam center and the workpiece center.

[0014] Optionally, both the first neural network model and the second neural network model are three-layer BP neural network models. The three-layer BP neural network models both include an input layer, a hidden layer, and an output layer, and both the input layer and the output layer include two neurons, and the hidden layer includes multiple neurons.

[0015] Optionally, the calculation function expression for the number of neurons included in the hidden layer is:

[0016] ,

[0017] where, is the number of neurons included in the hidden layer, is the number of neurons in the input layer, is the number of neurons in the output layer, is a constant, and the value range is between 1 and 10.

[0018] Optionally, before step S1, it further includes:

[0019] S101. Respectively obtain the data samples of the peak beam current density and full width at half maximum of the ion beam, and the peak removal efficiency and full width at half maximum of the corresponding removal function, the aperture diameter, and the target distance;

[0020] S102. Establish a training data set for the first neural network model based on the data samples of the peak beam current density and full width at half maximum (FWHM) of the ion beam, and the peak removal efficiency and FWHM of the corresponding removal function; establish a training data set for the second neural network model based on the data samples of the peak beam current density and FWHM of the ion beam, and the target values of the corresponding aperture and target distance.

[0021] S103. Use the training data set of the first neural network model to train the first neural network model to establish the mapping relationship between the peak removal efficiency and FWHM of the removal function, and the peak beam current density and FWHM of the ion beam; use the training data set of the second neural network model to train the second neural network model to establish the mapping relationship between the peak beam current density and FWHM of the ion beam, and the target values of the aperture and target distance.

[0022] Optionally, in step S101, the acquisition of the data samples of the peak beam current density and FWHM of a single ion beam, and the peak removal efficiency and FWHM of the corresponding removal function, aperture and target distance includes:

[0023] S201. Evacuate the ion beam equipment to a specified pressure and turn on the ion source.

[0024] S202. Adjust the aperture and target distance of the ion beam equipment.

[0025] S203. Perform a raster scan on the Faraday cup to obtain the peak beam current density and FWHM of the ion beam corresponding to the aperture and target distance.

[0026] S204. Keep the aperture and target distance unchanged, perform a spot etching experiment on the workpiece, obtain the removal function by subtracting the front and rear surface shapes of the optical element, and obtain the peak removal efficiency and FWHM of the corresponding removal function.

[0027] Optionally, step S203 includes: moving the ion source of the ion beam equipment to irradiate the ion beam into the small hole of the Faraday cup by raster scanning to obtain the corresponding beam current density. The scanning pitch is equal to the aperture of the small hole of the Faraday cup, and determine the peak beam current density of the ion beam corresponding to the aperture and target distance for the measured beam current density data, and calculate the FWHM by using the method of interpolation fitting or tolerance search. The peak beam current density is the beam current density when the center of the ion source is directly opposite to the center of the Faraday cup, and the FWHM is the diameter of the beam current density position distribution when the beam current density is half of the peak beam current density.

[0028] Optionally, calculating the full width at half maximum (FWHM) using the interpolation fitting method means: performing three-dimensional spatial interpolation fitting on the X-axis and Y-axis corresponding to the beam current density, finding the x and y coordinates corresponding to half of the peak beam current density, and calculating the average value of all the radii as the FWHM; calculating the FWHM using the tolerance search method means: finding the x and y coordinates corresponding to a numerical deviation of half of the peak beam current density and within a preset tolerance threshold, and calculating the average value of all the radii as the FWHM.

[0029] Optionally, obtaining the peak removal efficiency and the full width at half maximum (FWHM) of the corresponding removal function in step S204 includes: calculating the peak removal efficiency and the FWHM of the removal function by using the interpolation fitting or tolerance search method for the measured removal function, where the peak of the removal function is the maximum value of the removal amount, and the FWHM of the removal function is the diameter of the position distribution of the removal amount when the removal amount is half of the peak of the removal function.

[0030] Optionally, calculating the full width at half maximum (FWHM) using the interpolation fitting method means: performing three-dimensional spatial interpolation fitting on the X-axis and Y-axis corresponding to the removal function, finding the x and y coordinates corresponding to half of the peak removal amount, and calculating the average value of all the radii as the FWHM; calculating the FWHM using the tolerance search method means: finding the x and y coordinates corresponding to a numerical deviation of half of the peak removal amount and within a preset tolerance threshold, and calculating the average value of all the radii as the FWHM.

[0031] Compared with the prior art, the present invention can mainly achieve the following beneficial effects: The present invention includes inputting the peak removal efficiency and the full width at half maximum (FWHM) of the required removal function into the first neural network model to obtain the peak beam current density and the FWHM of the ion beam current, and then inputting them into the second neural network model to obtain the target values of the aperture diameter and the target distance, adjusting the aperture diameter and the target distance of the ion beam device to the target values, and then finely adjusting the aperture diameter and the target distance of the ion beam device until the beam current density of the ion beam current meets the requirements; compensating for the positioning error of the removal function; controlling the ion beam device to perform tests on the workpiece to achieve deterministic generation of the required removal function. By mapping the removal function parameters (peak removal efficiency and FWHM) to the ion beam current parameters (peak removal efficiency and FWHM), and mapping the ion beam current parameters to the ion beam process parameters (aperture diameter and target distance), the present invention can achieve rapid positioning of the aperture diameter and the target distance, solve the problems that the removal function parameters in ion beam processing in a vacuum environment are unknown and uncontrollable, and the process of making the removal function is cumbersome, realize deterministic generation of the ion beam removal function, and accelerate the generation efficiency of the removal function. Description of the Drawings

[0032] Figure 1 It is a flowchart of the deterministic generation of the removal function in an embodiment of the present invention.

[0033] Figure 2Schematic diagram of the scanning method of the Faraday cup in the embodiment of the present invention.

[0034] Figure 3 Parameter schematic diagram of the spatial distribution of the ion beam current density in the embodiment of the present invention.

[0035] Figure 4 Schematic diagram of the spatial coordinate system inside the three-axis ion beam machine tool in the embodiment of the present invention, where the component numbers are: 1, workpiece; 2, ion beam current; 3, variable aperture device; 4, ion source; 5, Faraday cup.

[0036] Figure 5 Schematic diagram of the structure of the first neural network model in the embodiment of the present invention. Specific implementation manners

[0037] The above are only the preferred implementation manners of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the principle of the present invention, several improvements and refinements should also be regarded as the protection scope of the present invention. The specific implementation manners of the present invention will be further described in detail below in conjunction with the drawings and embodiments.

[0038] As Figure 1 shown, the method for deterministically generating the ion beam removal function based on the online regulation of the ion beam current in this embodiment includes the following steps:

[0039] S1. Input the peak removal efficiency and full width at half maximum of the required removal function into the pre-trained first neural network model to obtain the corresponding peak beam current density and full width at half maximum of the ion beam current. The first neural network model is pre-trained to establish the mapping relationship between the peak removal efficiency and full width at half maximum of the removal function, and the peak beam current density and full width at half maximum of the ion beam current.

[0040] S2. Input the obtained peak beam current density and full width at half maximum of the ion beam current into the pre-trained second neural network model to obtain the target values of the corresponding aperture and target distance. The second neural network model is pre-trained to establish the mapping relationship between the peak beam current density and full width at half maximum of the ion beam current, and the target values of the aperture and target distance.

[0041] S3. Adjust the aperture and target distance of the ion beam device to the obtained target values of the aperture and target distance and detect the beam current density of the ion beam. Determine whether the beam current density of the ion beam meets the requirements. If it does not meet the requirements, finely adjust the aperture and target distance of the ion beam device until the beam current density of the ion beam meets the requirements.

[0042] S4. Adjust the ion beam device to compensate for the removal function positioning error.

[0043] S5, control the ion beam device to conduct tests on the workpiece to achieve deterministic generation of the required removal function.

[0044] In this embodiment, before step S1, it further includes:

[0045] S101, respectively obtain data samples of the peak beam current density and full width at half maximum (FWHM) of the ion beam current, and the corresponding peak removal efficiency and FWHM of the removal function, the aperture diameter of the aperture, and the target distance;

[0046] S102, according to the data samples of the peak beam current density and FWHM of the ion beam current, and the corresponding peak removal efficiency and FWHM of the removal function, establish a training data set for the first neural network model; according to the data samples of the peak beam current density and FWHM of the ion beam current, and the target values of the corresponding aperture diameter of the aperture and the target distance, and establish a training data set for the second neural network model;

[0047] S103, use the training data set of the first neural network model to train the first neural network model to establish the mapping relationship between the peak removal efficiency and FWHM of the removal function, and the peak beam current density and FWHM of the ion beam current; use the training data set of the second neural network model to train the second neural network model to establish the mapping relationship between the peak beam current density and FWHM of the ion beam current, and the target values of the aperture diameter of the aperture and the target distance.

[0048] In this embodiment, in step S101, the acquisition of the data samples of the peak beam current density and FWHM of a single ion beam current, and the corresponding peak removal efficiency and FWHM of the removal function, the aperture diameter of the aperture, and the target distance includes:

[0049] S201, evacuate the ion beam device to a specified pressure and turn on the ion source;

[0050] S202, adjust the aperture diameter of the aperture and the target distance of the ion beam device;

[0051] S203, conduct a raster scan on the Faraday cup once to obtain the peak beam current density and FWHM of the ion beam current corresponding to the aperture diameter of the aperture and the target distance;

[0052] S204, keep the aperture diameter of the aperture and the target distance unchanged, conduct a point etching experiment on the workpiece once, obtain the removal function by subtracting the front and rear surface shapes of the optical element, and obtain the corresponding peak removal efficiency and FWHM of the removal function.

[0053] On this basis, by continuously adjusting the aperture diameter of the aperture and the target distance of the ion beam device, a large number of data samples of the peak beam current density and FWHM of the ion beam current, and the corresponding peak removal efficiency and FWHM of the removal function, the aperture diameter of the aperture, and the target distance can be obtained for training the first neural network model and the second neural network model.

[0054] In this embodiment, step S203 includes: moving the ion source of the ion beam device to irradiate the ion beam into the small hole of the Faraday cup through raster scanning to obtain the corresponding beam current density. The scanning pitch is equal to the aperture of the small hole of the Faraday cup, and the peak beam current density of the ion beam corresponding to the aperture of the aperture stop and the target distance is determined according to the measured beam current density data, and the full width at half maximum is calculated by using the method of interpolation fitting or tolerance search. The peak beam current density is the beam current density when the center of the ion source is directly opposite the center of the Faraday cup, and the full width at half maximum is the diameter of the beam current density position distribution when the beam current density is half of the peak beam current density. As Figure 2 shown, move the ion source of the ion beam device, and irradiate the ion beam into the small hole of the Faraday cup through raster scanning. The scanning pitch is equal to the aperture of the small hole of the Faraday cup. During this process, the signal acquisition module collects the data measured by the Faraday cup and records the spatial coordinates x and y corresponding to the beam current density value. Due to the scanning characteristics of the Faraday cup, the scanning pitch should be equal to the aperture of the small hole of the Faraday cup to better obtain the spatial distribution of the ion beam current density:

[0055] ,

[0056] In the formula, J is the measured beam current density, I is the output signal, d 0 is the diameter of the small hole of the Faraday cup.

[0057] As Figure 3 shown, when the center of the ion source is directly opposite the center of the Faraday cup, the measured beam current density reaches the peak, and the beam current density at this time is used as the peak beam current density; when the beam current density is half of the peak beam current density, the diameter of the position is the full width at half maximum of the beam. The functional expression of the spatial distribution of the ion beam current density is:

[0058] ,

[0059] In the formula, is the peak beam current density of the ion beam, is the full width at half maximum of the ion beam. In this embodiment, calculating the full width at half maximum by using the method of interpolation fitting means: performing three-dimensional spatial interpolation fitting on the X-axis and Y-axis corresponding to the beam current density, finding the x and y coordinates corresponding to half of the peak beam current density, and taking the average value of all radii as the full width at half maximum; calculating the full width at half maximum by using the method of tolerance search means: finding the x and y coordinates corresponding to the numerical deviation within half of the peak beam current density and the tolerance within the preset tolerance threshold (generally taking a threshold close to 0, such as 0.01), and taking the average value of all radii as the full width at half maximum.

[0060] In this embodiment, obtaining the peak removal efficiency and full width at half maximum (FWHM) of the corresponding removal function in step S204 includes: calculating the peak removal efficiency and FWHM of the removal function by using the method of interpolation fitting or tolerance search for the measured removal function, where the peak of the removal function is the maximum value of the removal amount, and the FWHM of the removal function is the diameter of the position distribution of the removal amount when the removal amount is half of the peak of the removal function. In this embodiment, calculating the FWHM by using the method of interpolation fitting means: performing three-dimensional spatial interpolation fitting on the X-axis and Y-axis corresponding to the removal function, finding the x and y coordinates corresponding to half of the peak removal amount, and calculating the average value of all radii as the FWHM; calculating the FWHM by using the method of tolerance search means: finding the x and y coordinates corresponding to the numerical deviation within half of the peak removal amount and the tolerance within the preset tolerance threshold, and calculating the average value of all radii as the FWHM

[0061] In a three-axis vertical ion beam device as shown in Figure 4 , keeping the relative positions of the workpiece coordinate and the Faraday cup coordinate fixed, the ion source 4 moves relative to the Faraday cup 5 and the workpiece 1. When the Faraday cup 5 measures the beam current density of the ion beam 2, the position of the peak beam current density measured is the center coordinate position of the removal function. When compensating for the positioning error of the removal function, first, the center coordinate position of the workpiece and the center coordinate position of the Faraday cup should be determined, and the offset relationship between the workpiece center and the Faraday cup center is calculated therefrom. Secondly, after starting the vacuum system and the ion source, move the ion source coordinate through the numerical control system to perform a Faraday cup scan, find the position coordinate of the peak beam current density, and thus calculate the relative coordinate position between the beam center and the workpiece center in real time. Therefore, the function expression for compensating the positioning error of the removal function in step S4 of this embodiment is:

[0062] ,

[0063] wherein, is the center coordinate of the Faraday cup, is the center coordinate of the workpiece, is the offset relationship between the workpiece and the Faraday cup, is the relative coordinate between the measured beam center and the Faraday cup center, is the relative coordinate position between the beam center and the workpiece center.

[0064] Considering the function expression of the spatial distribution of the ion beam current density and the removal function of the ion beam can be expressed by the peak removal rate and the full width at half maximum as:

[0065]

[0066] In the formula, is the peak removal efficiency of the removal function, To remove the full width at half maximum (FWHM) of the removal function. In this embodiment, the training dataset of the first neural network model is used to train the first neural network model to establish the mapping relationship between the peak removal efficiency and the FWHM of the removal function, the peak beam density and the FWHM of the ion beam current. The peak removal efficiency of the removal function and the FWHM are used as the input signals of the input layer, and the output layer is the peak beam density and the FWHM of the ion beam current as the output signals. This mapping relationship can be expressed as:

[0067] ,

[0068] where is the peak beam density of the ion beam current, is the FWHM of the ion beam current, is the peak removal efficiency of the removal function, is the FWHM of the removal function, represents the first neural network model. To obtain the data set, dot etching experiments and Faraday cup scanning experiments need to be carried out. Keeping the parameters unchanged, after the Faraday cup scanning is performed first, then a dot is etched on the surface of the optical element. After changing the parameters, the above steps are continued to be repeated. The removal function parameters are obtained by subtracting the front and back surface shapes of the optical element measured.

[0069] The training dataset of the second neural network model is used to train the second neural network model to establish the mapping relationship between the peak beam density and the FWHM of the ion beam current, the target values of the aperture diameter and the target distance. The peak density and the FWHM of the ion beam current are used as the input layer, and the output layer is the aperture diameter and the target distance as the output signals. This mapping relationship can be expressed as:

[0070]

[0071] In the formula, is the target distance, is the aperture diameter, is the peak removal efficiency of the removal function, is the FWHM of the removal function, represents the second neural network model. In a vacuum environment, the on-line variable aperture device of the ion beam can realize the on-line transformation of the aperture diameter. By adjusting the Z-axis of the numerical control system, the target distance can be controlled. By adjusting the aperture diameter and the target distance of the aperture, the adjustment of the beam parameters can be realized.

[0072] It should be noted that the first neural network model and the second neural network model can adopt the required neural network model according to needs. For example, as an optional implementation mode, such as Figure 1As shown, the first neural network model and the second neural network model in this embodiment are both three-layer BP neural network models. The three-layer BP neural network models both include an input layer, a hidden layer, and an output layer. The input layer and the output layer both include two neurons, and the hidden layer includes multiple neurons. For example, as Figure 5 shown is a schematic structural diagram of the BP neural network model adopted by the first neural network model. The input signals of its input layer are the peak removal efficiency of the removal function and the full width at half maximum , and the output signals of the output layer are the peak beam density of the ion beam and the full width at half maximum . In this embodiment, the calculation function expression of the number of neurons included in the hidden layer is:

[0073] ,

[0074] where, is the number of neurons included in the hidden layer, is the number of neurons in the input layer, is the number of neurons in the output layer, is a constant, and its value range is between 1 and 10, and can be taken as needed.

[0075] The above is only the preferred embodiment of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, several improvements and refinements made without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for deterministically generating an ion beam removal function based on online control of ion beam current, characterized in that: The steps include: S1, inputting the peak removal efficiency and half-maximum full width of the required removal function into a pre-trained first neural network model to obtain the corresponding peak beam current density and half-maximum full width of the ion beam current, wherein the first neural network model is pre-trained to establish a mapping relationship between the peak removal efficiency and half-maximum full width of the removal function and the peak beam current density and half-maximum full width of the ion beam current; S2, inputting the obtained peak beam current density and half-maximum full width of the ion beam current into a pre-trained second neural network model to obtain corresponding target values ​​of the aperture and target distance, wherein the second neural network model is pre-trained to establish a mapping relationship between the peak beam current density and half-maximum full width of the ion beam current, the aperture and the target distance; S3, adjusting the aperture diameter and target distance of the ion beam device to the target values ​​of the aperture diameter and target distance, and detecting the beam current density of the ion beam, and judging whether the beam current density of the ion beam meets the requirements. If it does not meet the requirements, fine-tuning the aperture diameter and target distance of the ion beam device until the beam current density of the ion beam meets the requirements; S4, adjusting the ion beam device to compensate for the removal function positioning error; S5, controlling the ion beam device to conduct experiments on the workpiece to achieve deterministic generation of the desired removal function.

2. The method for deterministically generating an ion beam removal function based on online control of ion beam current according to claim 1, characterized in that: The functional expression for compensating and removing the positioning error of the function in step S4 is: ， in, is the coordinate of the center of the Faraday cup, is the workpiece center coordinate, is the offset relationship between the workpiece and the Faraday cup, is the measured relative coordinate of the beam center and the Faraday cup center, is the relative coordinate position between the beam center and the workpiece center.

3. The method for deterministically generating an ion beam removal function based on online control of ion beam current according to claim 1, characterized in that: The first neural network model and the second neural network model are both three-layer BP neural network models, and the three-layer BP neural network models include an input layer, a hidden layer and an output layer, and the input layer and the output layer each include two neurons, and the hidden layer includes multiple neurons.

4. The method for deterministically generating an ion beam removal function based on online control of ion beam current according to claim 3, characterized in that: The calculation function expression of the number of neurons included in the hidden layer is: ， in, is the number of neurons in the hidden layer, is the number of neurons in the input layer, is the number of neurons in the output layer, is a constant, The value range is from 1 to 10.

5. The method for deterministically generating an ion beam removal function based on online ion beam current control according to claim 1, characterized in that: Before step S1, the method further includes: S101, respectively obtaining data samples of the peak beam current density and the half-maximum full width of the ion beam current, the peak removal efficiency and the half-maximum full width of the corresponding removal function, the aperture of the aperture, and the target distance; S102, establishing a training data set for a first neural network model based on data samples of the peak beam current density and the half-maximum full width of the ion beam current, and the peak removal efficiency and half-maximum full width of the corresponding removal function; establishing a training data set for a second neural network model based on data samples of the peak beam current density and the half-maximum full width of the ion beam current, and the corresponding aperture and target distance; S103, using the training data set of the first neural network model to train the first neural network model to establish a mapping relationship between the peak removal efficiency and half-maximum full width of the removal function, and the peak beam current density and half-maximum full width of the ion beam current; using the training data set of the second neural network model to train the second neural network model to establish a mapping relationship between the peak beam current density and half-maximum full width of the ion beam current, the aperture and the target value of the target distance.

6. The method for deterministically generating an ion beam removal function based on online control of ion beam current according to claim 5, characterized in that: In step S101, the acquisition of data samples of the peak beam current density and the half-maximum full width of a single ion beam, the peak removal efficiency and the half-maximum full width of the corresponding removal function, the aperture of the aperture and the target distance includes: S201, evacuating the ion beam device to a specified pressure and turning on the ion source; S202, adjusting the aperture of the ion beam device and the target distance; S203, performing a raster scan on the Faraday cup to obtain the peak beam current density and the full width at half maximum of the ion beam corresponding to the aperture and the target distance; S204, keeping the aperture and target distance unchanged, performing a point etching experiment on the workpiece, obtaining a removal function by subtracting the front and back shapes of the optical element, and obtaining the peak removal efficiency and half-maximum full width of the corresponding removal function.

7. The method for deterministically generating an ion beam removal function based on online ion beam current control according to claim 6, characterized in that: Step S203 includes: the ion source of the mobile ion beam device injects the ion beam into the small hole of the Faraday cup by raster scanning to obtain the corresponding beam current density, the scanning interval is equal to the aperture of the small hole of the Faraday cup, and the peak beam current density of the ion beam corresponding to the aperture of the aperture and the target distance is determined according to the measured beam current density data, and the half-maximum full width is calculated by interpolation fitting or tolerance finding method, wherein the peak beam current density is the beam current density when the center of the ion source is facing the center of the Faraday cup, and the half-maximum full width is the diameter of the beam current density position distribution when the beam current density is half of the peak beam current density.

8. The method for deterministically generating an ion beam removal function based on online ion beam current control according to claim 7, characterized in that: The interpolation fitting method is used to calculate the half-maximum full width, which means: three-dimensional space interpolation fitting is performed on the X-axis and Y-axis corresponding to the beam density, and the x, y coordinates corresponding to half of the peak beam density are found, and the average value of all radii is calculated as the half-maximum full width; the tolerance search method is used to calculate the half-maximum full width, which means: by finding the x, y coordinates corresponding to the numerical deviation of half of the peak beam density and the tolerance within the preset tolerance threshold, the average value of all radii is calculated as the half-maximum full width.

9. The method for deterministically generating an ion beam removal function based on online ion beam current control according to claim 6, characterized in that: Obtaining the peak removal efficiency and half-maximum full width of the corresponding removal function in step S204 includes: calculating the peak removal efficiency and half-maximum full width of the removal function by interpolation fitting or tolerance finding method for the measured removal function, wherein the peak value of the removal function is the maximum value of the removal amount, and the half-maximum full width of the removal function is the diameter of the removal amount position distribution when the removal amount is half of the peak value of the removal function.

10. The method for deterministically generating an ion beam removal function based on online ion beam current control according to claim 9, characterized in that: The method of interpolation fitting is used to calculate the half-maximum full width, which means: for the X-axis and Y-axis corresponding to the removal function, three-dimensional space interpolation fitting is performed to find the x, y coordinates corresponding to half of the peak removal amount, and the average value of all radii is calculated as the half-maximum full width; the method of tolerance search is used to calculate the half-maximum full width, which means: by finding the x, y coordinates corresponding to the numerical deviation of half of the peak removal amount and the tolerance within the preset tolerance threshold, the average value of all radii is calculated as the half-maximum full width.

Images