Electron beam evaporation rate self-adaptive control method and system and storage medium

Abstract

The invention discloses an electron beam evaporation rate adaptive control method and system and a storage medium, and relates to the technical field of electron beam evaporation equipment control. The method comprises the following steps: constructing a material parameter library, and calling a PID initial parameter interval in response to a target material; collecting the actual evaporation rate and the vacuum degree of the vacuum chamber in real time; and carrying out iterative optimization on the PID parameters by utilizing a swarm intelligence optimization algorithm. A hierarchical strategy is executed in the optimization process; when the vacuum degree variation amplitude or the rate deviation exceeds a threshold value, optimization is carried out in a relatively large first search range; when the rate deviation is within a second deviation threshold range, optimization is performed within a smaller second search range. And finally outputting the target PID parameter to adjust the power of the electron gun, and updating the parameter to a material parameter library. The method can sense a multi-dimensional environment state, adapts to a complex working condition through a self-adaptive adjustment strategy, and effectively improves the stability and precision of rate control.

Description

Technical Field

[0001] This invention relates to the field of electron beam evaporation equipment control technology, and in particular to an adaptive control method, system and storage medium for electron beam evaporation rate. Background Technology

[0002] Electron beam evaporation technology uses a high-energy electron beam to bombard a target, causing the material to vaporize and deposit as a film on a substrate. The stability of the evaporation rate directly determines the microstructure and physical properties of the film. Current electron beam evaporation rate control mostly employs the classic PID algorithm, which relies on manually presetting proportional, integral, and derivative parameters based on the target material characteristics. However, in actual multilayer film fabrication processes, different materials have vastly different melting points and latent heats of vaporization. Frequent manual parameter switching is not only inefficient but also difficult to accurately match material properties.

[0003] More importantly, environmental factors such as vacuum fluctuations within the vacuum chamber during evaporation, changes in liquid level due to target material consumption, and aging of the electron gun filament dynamically alter the system transfer function characteristics of the controlled object. Fixed PID parameters cannot adapt to such time-varying systems, easily leading to rate overshoot, oscillations, or response hysteresis, resulting in decreased film quality. Although some studies have attempted to introduce optimization algorithms to tune PID parameters, most are based on single error feedback, lacking feedforward perception of environmental disturbances such as vacuum levels, and lacking a rapid response adjustment mechanism when facing sudden environmental changes, making it difficult to balance steady-state accuracy and dynamic response speed.

[0004] Therefore, how to construct a rate control method capable of sensing multi-dimensional environmental states and adaptively adjusting control strategies to adapt to complex operating conditions has become an urgent technical challenge. Summary of the Invention

[0005] The main objective of this invention is to provide an adaptive control method, system, and storage medium for electron beam evaporation rate, aiming to construct a rate control method capable of sensing multi-dimensional environmental states and adaptively adjusting control strategies to adapt to complex operating conditions.

[0006] To achieve the above objectives, this invention proposes an adaptive control method for electron beam evaporation rate, comprising the following steps: Step S1: Construct a material parameter library containing multiple evaporation materials. In response to the target material selected in the current evaporation process, call the corresponding PID initial parameter range from the material parameter library. Step S2: During the evaporation process, the actual evaporation rate of the electron beam evaporation equipment and the vacuum level of the vacuum chamber are collected in real time. Step S3: Iteratively optimize the PID parameters using a swarm intelligence optimization algorithm to correct the deviation between the actual evaporation rate and the target evaporation rate; during the iterative optimization process, a hierarchical optimization strategy is executed based on the monitored data: When the change in vacuum level is detected to exceed a preset threshold, or the absolute value of the deviation between the actual evaporation rate and the target evaporation rate exceeds a first deviation threshold, the search space is determined to be a first search range, and optimization is performed within the first search range. When the absolute value of the deviation between the actual evaporation rate and the target evaporation rate is detected to be within the range of the second deviation threshold, the search space is determined to be the second search range, and optimization is performed within the second search range; Wherein, the parameter range covered by the first search range is larger than the parameter range covered by the second search range; Step S4: Output the target PID parameters obtained by iterative optimization to adjust the power of the electron gun, and update the target PID parameters to the material parameter library to cover the original PID initial parameter range of the target material.

[0007] Preferably, in step S2, the real-time acquisition of the actual evaporation rate of the electron beam evaporation device and the vacuum level of the vacuum chamber further includes: Real-time acquisition of the remaining target material of the evaporation material and the actual power of the electron gun; In step S3, the iterative optimization of the PID parameters using a swarm intelligence optimization algorithm is based on the actual evaporation rate, the vacuum level, the remaining amount of the target material, and the actual power.

[0008] Preferably, the remaining amount of the target material is obtained through a target material remaining amount monitoring unit; The target material remaining quantity monitoring unit adopts at least one of the following methods: detecting the real-time weight of the target material through a weight sensor, and calculating it based on the density of the evaporating material and the cumulative evaporation time.

[0009] Preferably, in step S1, the PID initial parameter range in the material parameter library is constructed based on historical swarm intelligence optimization results and material physical properties; The physical properties of the material include at least its melting point and vapor pressure.

[0010] Preferred options also include: When the electron beam evaporation equipment detects a target replacement signal, it identifies the new evaporation material after the replacement. Step S1 is automatically triggered, which retrieves the PID initial parameter range corresponding to the new evaporation material from the material parameter library and starts the adaptive control process for the new evaporation material.

[0011] Preferably, in step S3, the swarm intelligence optimization algorithm is a particle swarm optimization algorithm; The particle swarm optimization algorithm uses a fitness function to evaluate whether the PID parameters meet preset conditions. The calculation parameters of the fitness function include at least: the absolute value of the deviation between the actual evaporation rate and the target evaporation rate, the system settling time, and the overshoot.

[0012] Preferably, in the hierarchical optimization strategy of step S3: The preset threshold is 10%; The second deviation threshold ranges from 0 to 2%.

[0013] Preferably, it also includes emergency response procedures: During the iterative optimization process in step S3, if it is determined that any of the following conditions are met: The swarm intelligence optimization algorithm fails to converge to the target parameter within a preset time, or the absolute value of the deviation between the actual evaporation rate and the target evaporation rate continues to exceed the safety threshold. Then immediately terminate the iterative optimization, call the preset backup PID parameter group to control the power of the electron gun, and issue a warning signal; Wherein, the safety threshold is greater than the first deviation threshold.

[0014] Preferably, in step S2: The actual evaporation rate is acquired using a rate detection unit, which includes a quartz crystal oscillator; The vacuum level of the vacuum chamber is collected using a vacuum level detection unit, which includes a vacuum gauge.

[0015] Preferably, step S5 is also included: After the evaporation process is completed, a process report is generated and stored; The process report records the following: optimization iteration process data of the swarm intelligence optimization algorithm, historical data of PID parameter adjustment, rate change curves, and environmental status data.

[0016] This application also discloses an adaptive control system for electron beam evaporation rate, including: The parameter library module is used to construct a material parameter library containing various evaporation materials, and in response to the target material selected in the current evaporation process, to call the corresponding PID initial parameter range from the material parameter library; The monitoring module is used to collect the actual evaporation rate of the electron beam evaporation equipment and the vacuum level of the vacuum chamber in real time during the evaporation process. The control module is used to iteratively optimize the PID parameters using a swarm intelligence optimization algorithm to correct the deviation between the actual evaporation rate and the target evaporation rate. The control module is configured to execute a hierarchical optimization strategy: When the change in vacuum level is detected to exceed a preset threshold, or the absolute value of the deviation between the actual evaporation rate and the target evaporation rate exceeds a first deviation threshold, optimization is performed within a first search range; when the absolute value of the deviation between the actual evaporation rate and the target evaporation rate is detected to be within a second deviation threshold range, optimization is performed within a second search range; the first search range is greater than the second search range; the control module is also used to output the target PID parameters obtained by iterative optimization to adjust the power of the electron gun, and update the target PID parameters to the material parameter library.

[0017] This application also discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any of the preceding claims.

[0018] The above technical solution has the following advantages: This invention achieves automatic parameter matching and rapid cold start during multi-material process switching by constructing a material parameter library containing the physical properties of various evaporation materials and initial PID parameter ranges. During the evaporation process, the system collects multi-dimensional environmental data, including vacuum level, in real time and uses a swarm intelligence optimization algorithm to iteratively optimize the PID parameters.

[0019] Specifically, this invention introduces a hierarchical optimization strategy, dynamically switching the search space based on the amplitude of vacuum fluctuations and the magnitude of rate deviations. When the environment changes abruptly or deviates significantly, a large-scale coarse-tuning search is activated for rapid response; when approaching the target value, a small-scale fine-tuning search is activated to ensure steady-state accuracy. This mechanism resolves the contradiction between traditional PID control's inability to adapt to environmental changes and the difficulty of balancing speed and accuracy with a single optimization strategy. Furthermore, the optimized target parameters are fed back to update the parameter library, enabling the system to possess self-evolving capabilities based on historical data, continuously improving control performance as the number of processes increases. Attached Figure Description

[0020] The present invention will now be described in detail with reference to specific embodiments and accompanying drawings, wherein: Figure 1 A flowchart of an adaptive control method for electron beam evaporation rate provided in an embodiment of the present invention. Figure 2 The structural block diagram of the electron beam evaporation rate adaptive control system provided in the embodiment of the present invention is shown. Detailed Implementation

[0021] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] This application provides an adaptive control system and method for electron beam evaporation rate. This system is mainly used for rate control of the thin film deposition process in electron beam evaporation equipment. Figure 2 As shown, the adaptive control system for electron beam evaporation rate includes a parameter library module 101, a monitoring module 102, and a control module 103. These modules work together to control the electron gun 200, which is located in the vacuum chamber 400, to heat and evaporate the evaporation material in the crucible 300.

[0023] The parameter library module 101 is used to construct and maintain a material parameter library containing various evaporation materials. This material parameter library is not a simple static data set, but a dynamic database constructed based on historical swarm intelligence optimization results and material physical properties. The material physical properties recorded in it include at least the melting point and vapor pressure of the material. For example, the library contains the initial PID parameter ranges for aluminum and silicon dioxide, and marks the melting point of aluminum as 660 degrees Celsius and the melting point of silicon dioxide as 1713 degrees Celsius. Specifically, in this embodiment, the initial PID parameter ranges for aluminum (Al) are: Kp: 1.5-3.0, Ki: 0.8-2.0, Kd: 0.3-1.0; and the initial PID parameter ranges for silicon dioxide (SiO2) are: Kp: 2.0-4.0, Ki: 0.5-1.5, Kd: 0.2-0.8. This pre-set method based on physical characteristics allows the system to provide a scientifically sound cold start range with a high probability of convergence when encountering a new material or changing material batches for the first time, avoiding the oscillation risk caused by completely random initialization. When responding to the target material selected in the current evaporation process, the parameter library module 101 can quickly retrieve the PID initial parameter range corresponding to the target material from the library, providing a search basis for subsequent optimization algorithms.

[0024] The monitoring module 102 is responsible for comprehensively sensing the system status during the evaporation process. Unlike traditional devices that only monitor a single rate feedback, the monitoring module 102 in this embodiment collects in real-time the actual evaporation rate of the electron beam evaporation equipment, the vacuum level of the vacuum chamber 400, the remaining amount of target material, and the actual power of the electron gun 200. Specifically, the actual evaporation rate is collected by the rate detection unit 501, which uses a quartz crystal oscillator to accurately capture minute changes in the deposition rate. The vacuum level is collected by the vacuum level detection unit 502, which includes a vacuum gauge for real-time feedback of the pressure environment within the vacuum chamber 400. The remaining amount of target material is obtained by the target material remaining amount monitoring unit 503. Depending on the equipment structure, the target material remaining amount monitoring unit 503 can use a weight sensor located at the bottom of the crucible 300 for real-time weighing, or, if a weight sensor cannot be installed, calculate the remaining amount based on the density of the evaporating material and the cumulative evaporation time. Vacuum level and target material remaining amount are introduced as monitoring dimensions because these two physical quantities directly change the transfer function characteristics of the evaporation process. Sudden changes in vacuum level will change the mean free path of the electron beam, while changes in target height will change the distance between the electron beam bombardment point and the substrate and the heat distribution. PID error feedback alone is insufficient to cope with such fundamental drifts in system characteristics.

[0025] The control module 103 is the core of the system, utilizing a swarm intelligence optimization algorithm to iteratively optimize the PID parameters. In this embodiment, the swarm intelligence optimization algorithm selected is Particle Swarm Optimization (PSO). To ensure the convergence efficiency and control effect of the algorithm, this embodiment configures the PSO algorithm with 30 particles, a maximum number of iterations of 50, and a search precision of 0.01. Based on multi-dimensional data such as the actual evaporation rate, vacuum level, remaining target material, and actual power, the control module 103 calculates and outputs the PID parameters to correct the deviation between the actual evaporation rate and the target evaporation rate. During the iterative optimization process, the control module 103 uses a fitness function to evaluate whether the PID parameters meet preset conditions. To achieve a comprehensive quantitative evaluation of the control effect, the fitness function... The weighted summation is shown in the following formula:

[0026] in, This represents the absolute value of the deviation between the actual evaporation rate and the target evaporation rate. Indicates the system settling time. Indicates overshoot; , , These are the corresponding weighting coefficients, for example, in this embodiment, we take... By minimizing this fitness function, the algorithm is mathematically guided to evolve towards faster response, smaller overshoot, and lower steady-state error.

[0027] To address the contradiction between rapid response under large disturbances and high-precision control under steady-state conditions, a single optimization strategy is insufficient to simultaneously achieve both. The control module 103 is configured to execute a hierarchical optimization strategy. This strategy dynamically adjusts the algorithm's search space based on the state of the monitored data. Specifically, the system defaults to or uses a first search range (coarse-tuning mode) in non-steady-state conditions. When the detected change in vacuum level exceeds a preset threshold, or the absolute value of the deviation between the actual evaporation rate and the target evaporation rate exceeds a first deviation threshold, the control module 103 locks the search space into the first search range and performs global optimization within this range. In this embodiment, the preset threshold for the change in vacuum level is set to 10%. This means that when a leak occurs in the vacuum system or fluctuations in pumping capacity cause a drastic change in vacuum level exceeding 10%, or when the rate deviation is extremely large, the algorithm determines that the system is in a non-steady-state or environmentally abrupt state. Activating the first search range, which covers a larger parameter range, allows the PID parameters to quickly jump to the new operating point, preventing system instability due to parameter adjustment lag.

[0028] The control module 103 switches and locks the search space to the second search range (fine-tuning mode) only when the absolute value of the deviation between the actual evaporation rate and the target evaporation rate converges to within the second deviation threshold range, performing local optimization. In this embodiment, the second deviation threshold range is set to 0 to 2%. That is, when the rate is close to the target value, the algorithm uses the second search range with a smaller parameter range for fine-tuning, thereby controlling the rate fluctuation within a very small range and avoiding steady-state oscillations caused by a large-range search. If the deviation exceeds the range again during the fine-tuning process, for example, due to environmental interference, the system will automatically revert to the first search range. The parameter range covered by the first search range is larger than that covered by the second search range. This layered strategy from coarse to fine and from large to small effectively simulates the logic of finding a general range and then fine-tuning in manual debugging, but with speed and accuracy far exceeding that of manual work.

[0029] The control module 103 also has an emergency handling function. During the iterative optimization process, if it is determined that the swarm intelligence optimization algorithm fails to converge to the target parameter within a preset time, or if the absolute value of the deviation between the actual evaporation rate and the target evaporation rate continuously exceeds a safety threshold, the control module 103 will immediately terminate the iterative optimization. To prevent coating failure or equipment damage due to prolonged optimization, the control module 103 will directly call the preset backup PID parameter group to control the power of the electron gun 200 and issue a warning signal. The safety threshold is set to be greater than the first deviation threshold, for example, 10%, to ensure that the circuit breaker mechanism is only triggered in the event of algorithm failure or severe loss of control.

[0030] After a complete control process, the control module 103 outputs the target PID parameters obtained through iterative optimization to the power controller 600 to adjust the power of the electron gun 200, and performs a crucial self-learning step: updating the target PID parameters to the material parameter library in the parameter library module 101 to cover the original initial PID parameter range of the target material. This mechanism enables the system to evolve; as the number of processes increases, the initial range in the parameter library will increasingly match the current equipment state, such as aging filaments or deposited chamber walls, making the optimization for the next process start-up more rapid. Simultaneously, after the evaporation process is completed, the system also generates and stores a process report, recording the optimization iteration process data of the swarm intelligence optimization algorithm, historical data of PID parameter adjustments, rate change curves, and environmental state data for easy traceability.

[0031] The following is combined Figure 1 Taking the preparation of aluminum and silica multilayer thin films as an example, the working process of this embodiment is described in detail. At the beginning of the process, in response to the aluminum material selection command, the parameter library module 101 retrieves the corresponding PID initial parameter range for aluminum from the library. Subsequently, the electron gun 200 is activated, and the monitoring module 102 begins real-time data acquisition. Assume that the target velocity of aluminum is set to 1.0 nm / s, the initial actual velocity is 0.7 nm / s, the deviation is 0.3 nm / s (exceeding the threshold of 0.1 nm / s), and the vacuum degree is stable. Since the absolute value of the deviation exceeds the first deviation threshold, the control module 103 runs a particle swarm optimization algorithm within the first search range for coarse optimization, quickly finding a set of parameters that increase the velocity. For example, after 15 iterations, the parameter combination Kp=2.2, Ki=1.3, Kd=0.6 is found, causing the actual velocity value to rise to 0.95 nm / s.

[0032] As the velocity approaches 1.0 nm / s, the absolute value of the deviation falls within 0.02 nm / s, or 2%, satisfying the second deviation threshold condition. The control module 103 automatically switches to the second search range for fine-tuning. At this point, the search step size decreases, and the parameters are fine-tuned, narrowing the search range to Kp: 2.0-2.4, Ki: 1.2-1.4, and Kd: 0.5-0.7. After eight iterations, the target parameters Kp=2.3, Ki=1.35, and Kd=0.62 are obtained, stabilizing the velocity between 0.99 and 1.01 nm / s, achieving high-precision steady-state control.

[0033] During the evaporation process, if the vacuum level in the vacuum chamber 400 suddenly fluctuates, for example from 5.0× Pa decreased to 6.5× The change in pressure (Pa) reached 30%, causing a sharp drop in the actual velocity, for example, to 0.92 nm / s. Monitoring module 102 detected this sudden environmental change. Although the velocity deviation might not yet be fully realized, based on the preset threshold of a 10% change in vacuum, control module 103 immediately determined that the current PID parameters were no longer suitable for the current atmospheric pressure environment and forcibly switched back to the first search range for coarse adjustment. The algorithm quickly searched for new PID parameters suitable for the low vacuum environment within a larger parameter space; for example, after 12 iterations, the parameters were updated to Kp=2.4, Ki=1.4, and Kd=0.65, rapidly pulling the velocity back to the target value of 0.98-1.02 nm / s, avoiding prolonged oscillations caused by relying solely on error feedback adjustment.

[0034] When the aluminum layer evaporation is complete and it's time to switch to the silica layer, the electron beam evaporation equipment detects the target replacement signal and identifies the new evaporation material as silica. The system automatically triggers the initialization step, and the parameter library module 101 automatically calls up the PID initial parameter range corresponding to silica: Kp: 2.0-4.0, Ki: 0.5-1.5, Kd: 0.2-0.8. Since silica has a much higher melting point than aluminum, its PID characteristics are completely different. This automatic calling mechanism avoids the tediousness and error risk of manual resetting. Subsequently, the adaptive control process for silica is initiated, repeating the above-described layered optimization process. With a target SiO2 velocity of 0.5 nm / s, PSO is first coarsely tuned, for example, after 18 iterations with parameters Kp=3.1, Ki=0.9, and Kd=0.4, resulting in a velocity of 0.48 nm / s. Then, fine-tuning is performed, for example, after 10 iterations with parameters Kp=3.2, Ki=0.95, and Kd=0.42, stabilizing the velocity between 0.49 and 0.51 nm / s. After the process is complete, the optimal parameters for aluminum and silicon dioxide obtained in this process are written back to the parameter library, optimizing the starting point for the next process.

[0035] Furthermore, this embodiment also verified the emergency response function by simulating a sudden leak in the vacuum system. When the simulated vacuum level suddenly increased to 1.0× When the actual velocity deviation reaches 0.15 nm / s and fails to converge after 10 PSO optimizations, the system automatically triggers an emergency strategy, calling the backup PID parameters for SiO2: Kp=3.5, Ki=1.0, Kd=0.45. Experimental results show that the system corrects the velocity to 0.50±0.02 nm / s within 5 seconds and successfully sends a "vacuum anomaly" warning signal to the host computer.

[0036] In summary, the embodiments of this application solve the problem that traditional PID control cannot adapt to sudden environmental changes and multiple material switching in electron beam evaporation by introducing multi-dimensional environmental perception such as vacuum degree and target material amount, and combining it with a hierarchical optimization strategy based on working conditions, thus achieving high-precision, automated and evolutionary rate control.

[0037] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method of electron beam evaporation rate adaptive control, characterized in that, The method comprises the following steps: Step S1, constructing a material parameter library containing a plurality of evaporation materials, calling a corresponding PID initial parameter interval from the material parameter library in response to a target material selected by a current evaporation process; Step S2, collecting the actual evaporation rate of the electron beam evaporation equipment and the vacuum degree of the vacuum chamber in real time during the evaporation process; Step S3, using a swarm intelligence optimization algorithm to iteratively optimize the PID parameters to correct the deviation of the actual evaporation rate from the target evaporation rate; In the iterative optimization process, a hierarchical optimization strategy is executed based on the monitored data: When the change amplitude of the vacuum degree is monitored to exceed a preset threshold, or the absolute value of the deviation of the actual evaporation rate from the target evaporation rate exceeds a first deviation threshold, it is determined that the search space is a first search range, and optimization is performed within the first search range; When the absolute value of the deviation of the actual evaporation rate from the target evaporation rate is monitored to be within a second deviation threshold range, it is determined that the search space is a second search range, and optimization is performed within the second search range; Wherein, the parameter interval covered by the first search range is greater than the parameter interval covered by the second search range; Step S4, outputting the target PID parameters obtained by iterative optimization to adjust the power of the electron gun, and updating the target PID parameters to the material parameter library to cover the original PID initial parameter interval of the target material.

2. The method of claim 1, wherein the method further comprises: In step S2, the real-time collection of the actual evaporation rate of the electron beam evaporation equipment and the vacuum degree of the vacuum chamber further comprises: Real-time collection of the target material remaining amount of the current evaporation material and the actual power of the electron gun; In step S3, the iterative optimization of the PID parameters using the swarm intelligence optimization algorithm is based on the actual evaporation rate, the vacuum degree, the target material remaining amount, and the actual power.

3. The method of claim 2, wherein the method further comprises: The target material remaining amount is obtained by a target material remaining amount monitoring unit; The target material remaining amount monitoring unit uses at least one of the following methods: detecting the real-time weight of the target material by a weight sensor, or calculating and obtaining based on the density of the evaporation material and the cumulative evaporation time.

4. The method of claim 1, wherein the method further comprises: In step S1, the PID initial parameter interval in the material parameter library is constructed based on historical swarm intelligence optimization results and material physical properties; The material physical properties at least include the melting point and vapor pressure of the material.

5. The method of claim 1, wherein the method further comprises: Further comprising: When the electron beam evaporation equipment detects a target material replacement signal, a new evaporation material after replacement is identified; Automatically triggering step S1, calling the PID initial parameter interval corresponding to the new evaporation material from the material parameter library, and starting the adaptive control process for the new evaporation material.

6. The method of claim 1, wherein the method further comprises: In step S3, the swarm intelligence optimization algorithm is a particle swarm optimization algorithm; The particle swarm optimization algorithm uses a fitness function to evaluate whether the PID parameters meet the preset conditions; The calculation parameters of the fitness function at least include the absolute value of the deviation of the actual evaporation rate from the target evaporation rate, the system adjustment time, and the overshoot.

7. The method of claim 1, wherein the method further comprises: In the hierarchical optimization strategy of step S3: The preset threshold is 10%; The second deviation threshold range is 0 to 2%.

8. The method of claim 1, wherein the method further comprises: Further comprising an abnormal emergency handling step: In the iterative optimization process of step S3, if any of the following conditions is met: The swarm intelligence optimization algorithm does not converge to the target parameter within a preset time, or the absolute value of the deviation of the actual evaporation rate from the target evaporation rate continuously exceeds a safety threshold value; The iterative optimization is immediately terminated, a preset backup PID parameter group is called to control the power of the electron gun, and a warning signal is issued; The safety threshold value is greater than the first deviation threshold value.

9. The method of claim 1, wherein the method further comprises: In step S2: The actual evaporation rate is collected by a rate detection unit, which includes a quartz crystal oscillator; The vacuum degree of the vacuum chamber is collected by a vacuum degree detection unit, which includes a vacuum gauge.

10. The method of claim 1, wherein the method further comprises: Step S5 is further included: After the evaporation process is completed, a process report is generated and stored; The process report records the following contents: optimization iteration process data of the swarm intelligence optimization algorithm, PID parameter adjustment history data, rate change curve, and environmental state data.

11. An electron beam evaporation rate self-adapting control system, characterized in that, The method comprises: A parameter library module is configured to build a material parameter library containing multiple evaporation materials, and to call corresponding PID initial parameter intervals from the material parameter library in response to a target material selected in a current evaporation process; A monitoring module is configured to collect the actual evaporation rate of the electron beam evaporation device and the vacuum degree of the vacuum chamber in real time during the evaporation process; A control module is configured to use a swarm intelligence optimization algorithm to perform iterative optimization on the PID parameters to correct the deviation of the actual evaporation rate from the target evaporation rate; The control module is configured to implement a hierarchical optimization strategy: When the change amplitude of the vacuum degree exceeds a preset threshold value, or the absolute value of the deviation of the actual evaporation rate from the target evaporation rate exceeds a first deviation threshold value, optimization is performed in a first search range; when the absolute value of the deviation of the actual evaporation rate from the target evaporation rate is within a second deviation threshold value range, optimization is performed in a second search range; the first search range is greater than the second search range; the control module is further configured to output the target PID parameters obtained through iterative optimization to adjust the power of the electron gun, and to update the target PID parameters to the material parameter library.

12. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by a processor to implement the method of any one of claims 1-10.

Images