MPCVD parameter virtual optimization control system and method based on digital twinning

Abstract

The invention discloses an MPCVD parameter virtual optimization control system and method based on digital twinning, and the system comprises a digital twinning module which is used for obtaining the operation data of an MPCVD physical entity system in real time, and constructing a dynamic virtual model which is synchronously mapped with the state of the MPCVD physical entity system; the virtual optimization module is connected with the digital twinning module, and is used for performing simulation prediction and optimization analysis on the preprocessed MPCVD process parameters by utilizing a virtual prediction algorithm based on the dynamic virtual model, and generating an optimal parameter control scheme by comparing a prediction result with a preset process target; the man-machine interaction and control module comprises an upper computer controller and a lower computer controller; the upper computer is used for system state monitoring, parameter setting and operation instruction issuing; and the lower computer controller is used for receiving an instruction of the upper computer and executing optimization control on the MPCVD physical entity system according to the optimal parameter control scheme. According to the method, virtual optimization and dynamic display of the MPCVD parameters are realized, and the process quality stability is improved.

Description

Technical Field

[0001] This invention relates to the field of microwave plasma chemical vapor deposition (MPCVD) technology, specifically to a virtual optimization control system and method for MPCVD parameters based on digital twins. Background Technology

[0002] MPCVD technology, with its unique process of using microwave-excited reactive gases to form plasma and performing chemical vapor deposition within a reaction chamber, has been widely used in the fabrication of high-quality diamond thin films and semiconductor materials. However, the complexity of the MPCVD process presents numerous challenges. The process parameters are numerous and interrelated, encompassing key elements such as microwave power, gas flow rate ratio, reaction chamber temperature, and vacuum level.

[0003] Traditional parameter optimization methods rely primarily on operator experience and repeated practical experiments, which have significant drawbacks, mainly as follows: 1. Efficiency: Actual experiments require frequent parameter adjustments and waiting for reaction results, which consumes a lot of time and manpower, resulting in a significant extension of the production cycle; 2. Cost aspect: Each experiment involves the consumption of raw materials and energy, and multiple experiments lead to a sharp increase in costs; 3. Accuracy level: Relying solely on experience-based judgment lacks precise quantitative analysis, making it difficult to obtain the optimal parameter combination and seriously affecting the stability of process quality.

[0004] In the prior art, Chinese invention patent application CN 120560360 A discloses a method for controlling the gas pressure in microwave plasma chemical vapor deposition (MPCVD) of diamond, specifically relating to the field of MPCVD technology. The method includes S1, real-time spectral monitoring; S2, machine learning prediction; S3, multi-stage gas pressure regulation; S4, multi-parameter collaborative control; and S5, closed-loop gas pressure control. It uses real-time spectral monitoring combined with a machine learning model to predict the optimal gas pressure value, and improves the diamond deposition quality by employing multi-stage gas pressure regulation and multi-parameter collaborative control strategies, thereby enhancing surface smoothness and crystallinity. However, the machine learning model established by the above method only predicts the gas pressure value and cannot comprehensively regulate the MPCVD process quality. Furthermore, it cannot intuitively display the prediction results. When the prediction results deviate from the actual process target, directly adjusting the gas pressure can easily lead to malfunctions or quality problems, seriously affecting the stability of the process quality.

[0005] Therefore, in the field of MPCVD process parameter optimization, there is a lack of mature, intuitive, and dynamic systems that can reflect the prediction and optimization effects of MPCVD equipment parameters. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art, to maturely, intuitively and dynamically reflect the prediction and optimization effect of MPCVD device parameters, improve the stability of process quality, and provide a virtual optimization control system and method for MPCVD parameters based on digital twins.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: A virtual optimization control system for MPCVD parameters based on digital twins, comprising: The digital twin module is used to acquire the operational data of the MPCVD physical entity system in real time and construct a dynamic virtual model that is synchronously mapped with the state of the physical entity system. The virtual optimization module, connected to the digital twin module, is used to perform simulation prediction and optimization analysis on the preprocessed MPCVD process parameters based on the dynamic virtual model and using a virtual prediction algorithm. By comparing the prediction results with the preset process target, the optimal parameter control scheme is generated. The human-computer interaction and control module includes a host computer and a slave controller; the host computer is used for system status monitoring, parameter setting, and issuing operation commands; the slave controller is used to receive commands from the host computer and perform optimized control on the MPCVD physical entity system according to the optimal parameter control scheme.

[0008] A digital twin dynamic virtual model of the MPCVD physical entity system is generated through a data twin module, displaying the system's operational data in real time. A virtual optimization module uses virtual prediction algorithms to simulate, predict, and optimize MPCVD process parameters. By comparing the prediction results with preset process targets, an optimal parameter control scheme is generated. A lower-level controller then executes optimized control on the MPCVD physical entity system according to this optimal parameter control scheme. This achieves virtual optimization and dynamic display of MPCVD parameters, providing a mature, intuitive, and dynamic reflection of the MPCVD device parameter prediction and optimization effects, thereby improving process quality stability.

[0009] Preferably, the MPCVD physical entity system includes: The reaction chamber is used to provide a sealed space for MPCVD reactions; A microwave generating and transmitting unit is used to generate microwaves and couple energy into the reaction cavity. A vacuum and gas supply unit is used to provide and maintain the required vacuum environment and process gas for the reaction chamber; A cooling unit is used to dissipate heat from the reaction chamber and the microwave generating and transmitting unit; The measurement and control unit is connected to the microwave generation and transmission unit, the vacuum and gas supply unit, and the cooling unit, respectively, and is used to execute control commands. The data acquisition unit is used to acquire the process and status data of the MPCVD physical entity system in real time; The power distribution unit is used to provide power to the various electrical components of the system.

[0010] Preferably, the microwave generating and transmitting unit includes a solid-state microwave power supply, a microwave generator, a three-pin tuner, and a mode converter connected in sequence, for generating adjustable microwaves and transmitting them to the reaction cavity.

[0011] Preferably, the vacuum and gas supply unit includes: Vacuum pump unit, used to evacuate the reaction chamber; A multi-channel gas delivery pipeline is provided for supplying process gases to the reaction chamber; each pipeline is equipped with a gas path solenoid valve, a pressure valve, and a mass flow controller; the mass flow controller is used to control the flow rate of at least one gas selected from hydrogen, methane, oxygen, argon, and nitrogen; the gas path solenoid valve is used to control the on / off state of the pipeline, and the pressure valve is used to regulate the pipeline pressure.

[0012] Preferably, the data acquisition unit includes at least one of the following sensors: A microwave power sensor is used to monitor the microwave absorbed power within the reaction chamber. Temperature sensors are used to monitor the temperature of key parts of the reaction chamber; Pressure sensors are used to monitor the pressure inside the reaction chamber; Water temperature and water pressure sensors are used to monitor the operating status of the cooling unit; A vacuum gauge is used to monitor the vacuum level of a cavity. Gas flow sensors are used to monitor the real-time flow rate of various process gases.

[0013] Preferably, the data acquisition unit further includes at least one of the following sensors: An emission spectrometer is used for online detection of the concentration of active species in plasma; Langmuir probes are used to measure the electron temperature and density of plasmas. Gas composition analyzer, used to detect the concentration of reaction products and impurity gases; Intracavitary microflow sensor is used to assess the uniformity of gas distribution within the reaction chamber; A microwave phase sensor is used to monitor the propagation phase characteristics of microwaves within a cavity. Thin film growth monitoring sensor for online measurement of thin film thickness and surface morphology; Gas leak detection sensors are used to monitor leaks of hydrogen or methane. Infrared thermal imagers are used for non-contact monitoring of temperature field distribution.

[0014] Preferably, the virtual prediction algorithm is a neural network-based model; the neural network model takes a set of MPCVD process parameters as input and the corresponding process result indicators as output, and is used to establish a nonlinear mapping relationship between process parameters and process results; the virtual prediction algorithm is trained with historical process data and coupled with an optimization algorithm to perform iterative search in a preset parameter space, the neural network evaluates the performance of the parameter combination, and finally converges to generate the optimal parameter scheme that meets the preset target.

[0015] Preferably, the process parameters include at least microwave power, reactive gas flow rate ratio, cavity temperature and working pressure, and the process result indicators include at least film quality, deposition rate and process energy consumption.

[0016] Preferably, it also includes a data storage and analysis module for storing historical operating data of the MPCVD physical entity system, model data of the digital twin module, and optimization process and result data of the virtual optimization module, and providing support for data analysis and model updates.

[0017] A virtual optimization control method for MPCVD parameters based on digital twins, applied to the system described above, the method comprising: S1: Real-time acquisition of the operational data of the MPCVD physical entity system, and construction of a dynamic virtual model that is synchronously mapped to its state; S2: Based on the dynamic virtual model, the MPCVD process parameters are simulated and predicted using a neural network-based virtual prediction algorithm; by comparing and optimizing the prediction results with the preset process targets, the optimal parameter control scheme is generated. S3: Based on the optimal parameter control scheme, optimize the control of the MPCVD physical entity system through the lower-level controller.

[0018] This invention constructs a precise dynamic virtual model of the MPCVD physical entity system and combines it with a virtual prediction algorithm to achieve efficient optimization of MPCVD process parameters in a virtual environment. This breaks through the limitations of traditional process parameter optimization, significantly improves the quality and efficiency of MPCVD process in applications such as thin film preparation and material processing, and effectively reduces production costs, thus realizing intelligent and efficient MPCVD process. Attached Figure Description

[0019] The present invention will now be described in further detail with reference to the accompanying drawings: Figure 1 This is a block diagram of the system of the present invention; Figure 2 This is a schematic diagram of the MPCVD physical entity system of the present invention; Explanation of reference numerals in the attached figures: 1: Reaction chamber; 2: Mass flow controller; 3: Microwave generator; 4: Power distribution unit; 5: Solid-state microwave power supply; 6: Industrial control integrated computer. Detailed Implementation

[0020] like Figure 1 As shown, the present invention provides a virtual optimization control system for MPCVD parameters based on digital twins, comprising: The digital twin module is used to acquire the operational data of the MPCVD physical entity system in real time and build a dynamic virtual model that is synchronously mapped with the state of the MPCVD physical entity system. The virtual optimization module, connected to the digital twin module, is used to perform simulation prediction and optimization analysis on the pre-processed MPCVD process parameters based on a dynamic virtual model and using virtual prediction algorithms. By comparing the prediction results with the preset process targets, the optimal parameter control scheme is generated. The human-computer interaction and control module includes a host computer and a slave controller. The host computer is used for system status monitoring, parameter setting, and issuing operation commands. The slave controller is used to receive commands from the host computer and perform optimized control on the MPCVD physical entity system according to the optimal parameter control scheme.

[0021] It also includes a data storage and analysis module, which stores historical operating data of the MPCVD physical entity system, model data of the digital twin module, and optimization process and result data of the virtual optimization module, and provides support for data analysis and model updates.

[0022] like Figure 2 As shown, the MPCVD physical entity system includes: Reaction chamber 1 is used to provide a sealed space for MPCVD reactions; A microwave generating and transmitting unit is used to generate microwaves and couple the energy into the reaction cavity 1; The vacuum and gas supply unit is used to provide and maintain the required vacuum environment and process gas for the reaction chamber 1; A cooling unit is used to dissipate heat from the reaction chamber 1 and the microwave generation and transmission unit; The measurement and control unit is connected to the microwave generation and transmission unit, the vacuum and gas supply unit, and the cooling unit, respectively, and is used to execute control commands. The data acquisition unit is used to acquire process and status data of the MPCVD physical entity system in real time. Power distribution unit 4 is used to provide power to all electrical components of the system.

[0023] The microwave generation and transmission unit includes a solid-state microwave power supply 5, a microwave generator 3, a three-pin tuner and a mode converter connected in sequence, which are used to generate adjustable microwaves and transmit them to the reaction cavity 1.

[0024] The water-cooling unit includes a water chiller and a water distributor. The water chiller generates chilled water, and the water distributor distributes the chilled water to multiple branches that require water cooling. The chilled water is then sent to the reaction chamber 1 and the microwave generation and transmission unit to provide reliable heat dissipation, maintain stable system temperature, and ensure normal operation of the equipment.

[0025] The vacuum and gas supply unit includes: Vacuum pump unit, used to evacuate reaction chamber 1; A multi-channel gas delivery pipeline is used to supply process gases to reaction chamber 1. Each pipeline is equipped with a gas solenoid valve, a pressure valve, and a mass flow controller 2. The gas solenoid valve controls the on / off state of the pipeline, and the pressure valve regulates the pipeline pressure. The mass flow controller 2 controls the flow rates of hydrogen, methane, oxygen, argon, and nitrogen. The mass flow controller 2 can be of model FCST100SFC.

[0026] The vacuum and gas supply unit also includes an electromagnetic vacuum valve installed at the inlet of the vacuum pump unit, which is used to control the opening and closing of the vacuum pump unit and multiple gas delivery pipelines by electromagnetic force.

[0027] The data acquisition unit includes the following sensors: A microwave power sensor is used to monitor the microwave absorbed power within reaction chamber 1; Temperature sensor used to monitor the temperature of key parts of reaction chamber 1; A pressure sensor is used to monitor the pressure inside reaction chamber 1; Water temperature and water pressure sensors are used to monitor the operating status of the cooling unit; A vacuum gauge is used to monitor the vacuum level of a cavity. Gas flow sensors are used to monitor the real-time flow rate of various process gases.

[0028] To further enable precise mapping and monitoring, the data acquisition unit also includes at least one of the following sensors: An emission spectrometer is used for online detection of the concentration of active species in plasma; Langmuir probes are used to measure the electron temperature and density of plasmas. Gas composition analyzer, used to detect the concentration of reaction products and impurity gases; Intracavitary microflow sensor is used to assess the uniformity of gas distribution within the reaction chamber; A microwave phase sensor is used to monitor the propagation phase characteristics of microwaves within a cavity. Thin film growth monitoring sensor for online measurement of thin film thickness and surface morphology; Gas leak detection sensors are used to monitor leaks of hydrogen or methane. Infrared thermal imagers are used for non-contact monitoring of temperature field distribution.

[0029] Thin film growth monitoring sensors can be laser reflection sensors or ellipsometrists.

[0030] Plasma state was monitored using an emission spectrometer and a Langmuir probe; the gas environment was monitored using a gas composition analyzer and an intracavity microflow sensor; and film quality was monitored using a thin film growth monitoring sensor.

[0031] Microwave energy is monitored using microwave power and phase sensors. Since microwaves are the energy source for MPCVD, their power density and transmission efficiency within reaction chamber 1 directly affect plasma generation efficiency. The microwave power sensor monitors the "actual absorbed power" of microwaves within the reaction chamber in real time (rather than the "output power" of the microwave generation and transmission units), avoiding energy loss due to microwave reflection (such as increased reflectivity caused by aging of the chamber's inner wall material), thus ensuring stable plasma excitation energy. If a decrease in absorbed power is detected, the impedance matching of the three pins is adjusted using a three-pin tuner to improve microwave utilization. The microwave phase sensor monitors the transmission phase of microwaves within the chamber to ensure the "resonant coupling" effect between microwaves and plasma (phase shift leads to decreased plasma stability).

[0032] The data acquisition unit collects process and status data of the MPCVD physical entity system in real time and transmits it to the digital twin module. The digital twin module obtains the operating data of the MPCVD physical entity system in real time and dynamically updates the dynamic virtual model based on the real-time operating data to ensure a high degree of consistency between the dynamic virtual model and the MPCVD physical entity system, thereby achieving accurate virtual simulation.

[0033] The process objective of the virtual optimization module is to maximize the deposition rate and control energy consumption while ensuring the quality of the diamond film. Preset target thresholds for film quality, deposition rate, and process energy consumption are provided.

[0034] The virtual prediction algorithm is based on a neural network model. The neural network model takes a set of MPCVD process parameters as input and corresponding process result indicators as output, establishing a nonlinear mapping relationship between process parameters and process results. The virtual prediction algorithm is trained using historical process data and coupled with an optimization algorithm to iteratively search within a preset parameter space. The neural network evaluates the performance of the parameter combinations and ultimately converges to generate the optimal parameter scheme that satisfies the preset objective. The optimization algorithm employs a genetic algorithm or particle swarm optimization algorithm to improve global optimization efficiency.

[0035] Specifically, MPCVD process parameters include microwave power, reactant gas flow rate ratio, chamber temperature, and operating pressure. The reactant gas flow rate ratio is the proportion of hydrogen, methane, and oxygen. Process results include film quality, deposition rate, and process energy consumption.

[0036] Real-time microwave power, reactant gas flow rate ratio, cavity temperature, and operating pressure are obtained from the operational data of the MPCVD physical entity system acquired by the data acquisition unit, and preprocessed. Preprocessing includes data cleaning and data normalization. Data cleaning identifies and corrects null values, duplicate values, erroneous values, and inconsistent data formats; data normalization eliminates the influence of differences between data features.

[0037] The pre-processed microwave power, reactant gas flow rate ratio, cavity temperature, and working pressure are input into the neural network model of the virtual prediction algorithm. The trained neural network model performs prediction and evaluation to obtain the predicted process results. These predicted results are fed back to the optimization algorithm, which adjusts its search strategy based on their proximity to the target. After multiple iterations, the optimal parameter scheme is generated. This invention uses a neural network model for simulation calculations to predict process results. Through continuous iterative calculations by the optimization algorithm, it seeks the optimal parameter combination and generates a precise optimized parameter scheme, providing reliable guidance for actual production.

[0038] The optimal parameter scheme is fully implemented through a dynamic virtual model. Simulation results show that the process is stable, with no risks such as plasma instability or abnormal temperature rise. The host computer displays the detailed parameters, prediction results, and virtual operation report of the optimal parameter scheme, which are then reviewed and confirmed by the operator before execution. The host computer's interface provides an intuitive and dynamic view of the MPCVD device parameter prediction and optimization effects.

[0039] The host computer includes an industrial control all-in-one computer, a monitor, a mouse, and a keyboard, providing users with an intuitive and user-friendly interface for easy monitoring of system operation status and input of operation commands. The host computer not only displays the real-time operating status of the MPCVD physical entity system and the simulation results of the virtual model, but also has remote access capabilities. Users can monitor and operate the system in real-time on remote devices via the network, and receive timely system status and optimization result push notifications, achieving convenient remote management.

[0040] The host computer sends the parameter set of the optimal parameter scheme to the lower-level controller, which then parses the instructions and sends them to: Microwave generating and transmitting unit: Adjusting microwave power; Vacuum and gas supply unit: Adjust the pressure valve to control the working pressure of the chamber, adjust the mass flow controller 2, and adjust the flow ratio of the reaction gas; Water-cooled unit to adjust cavity temperature.

[0041] The lower-level controller, model ZMC408M, features 32 input / output interfaces. It connects to each unit in the MPCVD physical system via RS-485 communication, receiving data feedback from the digital twin module and executing the optimal parameter control scheme generated by the virtual optimization module. The upper-level computer communicates with the lower-level controller via Modbus / TCP protocol for data exchange and command transmission. The upper-level computer can display the real-time operating status of the MPCVD physical system and the simulation results of the dynamic virtual model.

[0042] This invention constructs a precise dynamic virtual model of the MPCVD physical entity system and combines it with a virtual prediction algorithm to achieve efficient optimization of MPCVD process parameters in a virtual environment. This breaks through the limitations of traditional process parameter optimization, significantly improves the quality and efficiency of MPCVD process in applications such as thin film preparation and material processing, and effectively reduces production costs, thus realizing intelligent and efficient MPCVD process.

[0043] This invention also provides a virtual optimization control method for MPCVD parameters based on digital twins, applied to the above-mentioned system, the method comprising: S1: Real-time acquisition of the operational data of the MPCVD physical entity system, and construction of a dynamic virtual model that is synchronously mapped to its state; S2: Based on a dynamic virtual model, a virtual prediction algorithm based on neural networks is used to simulate and predict MPCVD process parameters; by comparing and optimizing the prediction results with preset process targets, the optimal parameter control scheme is generated. S3: Based on the optimal parameter control scheme, optimize the control of the MPCVD physical entity system through the lower-level controller.

[0044] The above are merely preferred embodiments 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 scope of the technology 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 defined in the claims.

Claims

1. A virtual optimization control system for MPCVD parameters based on digital twins, characterized in that, include: The digital twin module is used to acquire the operational data of the MPCVD physical entity system in real time and construct a dynamic virtual model that is synchronously mapped with the state of the MPCVD physical entity system. The virtual optimization module, connected to the digital twin module, is used to perform simulation prediction and optimization analysis on the preprocessed MPCVD process parameters based on the dynamic virtual model and using a virtual prediction algorithm. By comparing the prediction results with the preset process target, the optimal parameter control scheme is generated. The human-computer interaction and control module includes a host computer and a slave controller; the host computer is used for system status monitoring, parameter setting, and issuing operation commands; the slave controller is used to receive commands from the host computer and perform optimized control on the MPCVD physical entity system according to the optimal parameter control scheme.

2. The system according to claim 1, characterized in that, The MPCVD physical entity system includes: The reaction chamber is used to provide a sealed space for MPCVD reactions; A microwave generating and transmitting unit is used to generate microwaves and couple energy into the reaction cavity. A vacuum and gas supply unit is used to provide and maintain the required vacuum environment and process gas for the reaction chamber; A cooling unit is used to dissipate heat from the reaction chamber and the microwave generating and transmitting unit; The measurement and control unit is connected to the microwave generation and transmission unit, the vacuum and gas supply unit, and the cooling unit, respectively, and is used to execute control commands. The data acquisition unit is used to acquire the process and status data of the MPCVD physical entity system in real time; The power distribution unit is used to provide power to the various electrical components of the system.

3. The system according to claim 2, characterized in that, The microwave generation and transmission unit includes a solid-state microwave power supply, a microwave generator, a three-pin tuner, and a mode converter connected in sequence, used to generate adjustable microwaves and transmit them to the reaction cavity.

4. The system according to claim 2, characterized in that, The vacuum and gas supply unit includes: Vacuum pump unit, used to evacuate the reaction chamber; A multi-channel gas delivery pipeline is provided for supplying process gases to the reaction chamber; each pipeline is equipped with a gas path solenoid valve, a pressure valve, and a mass flow controller; the mass flow controller is used to control the flow rate of at least one gas selected from hydrogen, methane, oxygen, argon, and nitrogen; the gas path solenoid valve is used to control the on / off state of the pipeline, and the pressure valve is used to regulate the pipeline pressure.

5. The system according to claim 2, characterized in that, The data acquisition unit includes at least one of the following sensors: A microwave power sensor is used to monitor the microwave absorbed power within the reaction chamber. Temperature sensors are used to monitor the temperature of key parts of the reaction chamber; Pressure sensors are used to monitor the pressure inside the reaction chamber; Water temperature and water pressure sensors are used to monitor the operating status of the cooling unit; A vacuum gauge is used to monitor the vacuum level of a cavity. Gas flow sensors are used to monitor the real-time flow rate of various process gases.

6. The system according to claim 5, characterized in that, The data acquisition unit also includes at least one of the following sensors: An emission spectrometer is used for online detection of the concentration of active species in plasma; Langmuir probes are used to measure the electron temperature and density of plasmas. Gas composition analyzer, used to detect the concentration of reaction products and impurity gases; Intracavitary microflow sensor is used to assess the uniformity of gas distribution within the reaction chamber; A microwave phase sensor is used to monitor the propagation phase characteristics of microwaves within a cavity. Thin film growth monitoring sensor for online measurement of thin film thickness and surface morphology; Gas leak detection sensors are used to monitor leaks of hydrogen or methane. Infrared thermal imagers are used for non-contact monitoring of temperature field distribution.

7. The system according to claim 1, characterized in that, The virtual prediction algorithm is a neural network-based model. The neural network model takes a set of MPCVD process parameters as input and the corresponding process result indicators as output, and is used to establish a nonlinear mapping relationship between process parameters and process results. The virtual prediction algorithm is trained using historical process data and coupled with an optimization algorithm to iteratively search within a preset parameter space. The performance of the parameter combination is evaluated by a neural network, and finally converges to generate the optimal parameter scheme that satisfies the preset objective.

8. The system according to claim 7, characterized in that, The process parameters include at least microwave power, reaction gas flow rate ratio, cavity temperature and working pressure, and the process result indicators include at least film quality, deposition rate and process energy consumption.

9. The system according to claim 1, characterized in that, It also includes a data storage and analysis module, which stores the historical operating data of the MPCVD physical entity system, the model data of the digital twin module, and the optimization process and result data of the virtual optimization module, and provides support for data analysis and model updates.

10. A virtual optimization control method for MPCVD parameters based on digital twins, characterized in that, The method, applied to the system as described in any one of claims 1 to 9, comprises: S1: Real-time acquisition of the operational data of the MPCVD physical entity system, and construction of a dynamic virtual model that is synchronously mapped to its state; S2: Based on the dynamic virtual model, the MPCVD process parameters are simulated and predicted using a neural network-based virtual prediction algorithm; by comparing and optimizing the prediction results with the preset process targets, the optimal parameter control scheme is generated. S3: Based on the optimal parameter control scheme, optimize the control of the MPCVD physical entity system through the lower-level controller.

Images