Efficient management electronic device production optimization system

Abstract

The invention discloses a high-efficiency management electronic device production optimization system, which applies a high-efficiency management electronic device production optimization method, and comprises the following steps of: taking structural design parameters of an electric vacuum device as input and empirical simulation response data as output; performing training prediction of a regression prediction model of the electric vacuum device based on the input data and the output data; then combining the prediction model with an optimization algorithm, analyzing and iterating through the optimization algorithm to meet an optimization calculation result of the electric vacuum device, performing analysis and calculation based on microwave transmission and method characteristics in the electric vacuum device, and optimizing physical characteristics simulated by an electromagnetic field of the vacuum electronic device through parallel calculation to obtain a simulation result of the electromagnetic field of the electric vacuum device. And then real-time online monitoring is carried out based on the internal vacuum degree of the electronic vacuum device, that is, real-time leakage detection is carried out on the electronic vacuum device, the reliability of the electronic vacuum device is controlled, the numerical simulation performance of electronic device production is improved, and the efficiency of the manufacturing process is also efficiently controlled.

Description

Technical Field

[0001] The invention relates to the technical field of vacuum device parameter optimization, and in particular to an electronic device production optimization system with efficient management. Background Art

[0002] With the increasing demand for high-performance vacuum devices, various new vacuum devices have been rapidly developed. The internal structure of vacuum devices is precise and complex, and the materials are numerous, which makes the design and generation process of their structures very cumbersome and lengthy. Since the interaction area of ​​the device itself involves a very strong nonlinear process, it is very difficult to conduct theoretical research and analysis on the entire process. Therefore, after completing the preliminary design of the new vacuum device, it is necessary to further process and experiment it. Since the processing and experiment take a long time, and the experimental equipment and human resources consumed are relatively expensive, computer numerical simulation is used to simulate the new vacuum device before processing and experiment. However, the structure generated by the vacuum device design is usually difficult to meet the specific performance indicators at one time, and it is necessary to constantly modify the size parameters of each structure of the vacuum device. If the structural parameters selected for modification are incorrect, it may even cause the overall structure of the vacuum device to be adjusted, which is extremely cumbersome, time-consuming and labor-intensive, and the efficiency of the overall simulation is severely limited. Therefore, it is very necessary to design an efficient management system for optimizing the production of electronic devices to improve the performance of numerical simulation of electronic device production and efficiently control the efficiency of the manufacturing process. Summary of the invention

[0003] The object of the present invention is to provide an electronic device production optimization system with efficient management to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: an efficient management method for optimizing the production of electronic devices, comprising the following steps:

[0005] Step 1: Taking the structural design parameters of the electric vacuum device as input and the empirical simulation response data as output, the regression prediction model of the electric vacuum device is trained and predicted based on the input data and the output data;

[0006] Step 2: Combine the prediction model with the optimization algorithm, analyze and iterate the optimization calculation results of the electric vacuum device through the optimization algorithm, compare the calculation results obtained by numerical simulation with the results obtained by experiment, find out the similarities and differences and list them for output;

[0007] Step 3: Analyze and calculate the microwave transmission and method characteristics inside the electric vacuum device;

[0008] Step 4: Further utilize the time-varying electromagnetic field simulation analysis inside the vacuum electronic device to optimize the physical characteristics of the electromagnetic field simulation of the vacuum electronic device through parallel computing;

[0009] Step 5: Perform real-time online monitoring based on the internal vacuum degree of the electronic vacuum device, that is, perform real-time leak detection on the electronic vacuum device to control the reliability of the electronic vacuum device;

[0010] The specific experimental method for training and predicting the regression prediction model of the electric vacuum device based on input data and output data includes:

[0011] Step 11: Perform structural analysis and modeling on the internal structure and performance indicators of the electric vacuum device, design and analyze the high-frequency characteristics using three-dimensional electromagnetic simulation software, determine the initial parameter starting values, including establishing a single-cycle model, setting the eigensolver, boundary conditions and frequency range parameters, parameter scanning and optimization, simulation result analysis, and perform simulation based on the above settings through a computer;

[0012] Step 12: Based on a large amount of existing input data and output data, the structural design parameters of the electric vacuum device are used as the input data of the electric vacuum device regression prediction model, and the corresponding empirical simulation response data are used as the output data of the electric vacuum device regression prediction model. These input data and output data are divided into training set data and test set data of the electric vacuum device, and training is performed on the training set data of the electric vacuum device;

[0013] Step 13: Train the deep neural network through data and build a deep neural network regression prediction model, and use the test set data of the electric vacuum device to evaluate these regression models and select the optimal electric vacuum device regression prediction model;

[0014] The method of analyzing and iterating the optimization calculation results of the electric vacuum device through the optimization algorithm is:

[0015] Step 21: Use the prediction result of the electric vacuum device as the fitness function of the particle swarm optimization algorithm, initialize the particle speed and position in the population, and construct an optimization objective function that meets the target design requirements of the electric vacuum device, that is, construct the fitness function in the particle swarm optimization algorithm, calculate the fitness value of each particle in the population based on the fitness function, and then update the individual optimal solution and the global optimal solution in turn;

[0016] Step 22: Determine whether the optimization calculation result of the optimization algorithm meets the target. When it is determined that the target design requirements of the electric vacuum device are met, the optimization value of the particle swarm optimization algorithm is directly output. When it is determined that the target design requirements of the electric vacuum device are met, the particle speed and the particle position are updated in sequence.

[0017] Step 23: After the speed and position of the particle are updated, the new fitness value corresponding to the new speed and position of the particle is calculated again based on the fitness function, and then the updated individual optimal solution of the particle and the global optimal solution are judged again whether they meet the target requirements until the target is met;

[0018] The analysis method for analyzing and calculating microwave transmission and method characteristics is:

[0019] Step 31: Based on the target design requirements in the given production process of the electric vacuum device, a simulation model for the high-frequency characteristics of microwaves inside the electric vacuum device is established;

[0020] Step 32: Analyze and fine-tune the high-frequency characteristics of the microwave signal according to the updated structural parameters, and determine the simulation parameters of the interaction based on the adjusted structural parameters;

[0021] Step 33: Conducting simulation analysis of transmission characteristics, i.e., interacting the microwave signal corresponding to the structural parameters inputted by the external update of the electric vacuum device with the electrons injected by the electron injection channel;

[0022] The step of optimizing the physical characteristics of the electromagnetic field simulation of the vacuum electron device by parallel computing comprises:

[0023] By using a multi-core CPU to simultaneously process the simulation calculations of structural parameters, numerical discrete parameters, microwave parameters, and particle parameters in particle simulation, the calculation boundary is processed by reducing the calculation area and setting an absorption boundary at the boundary. In the process of numerical simulation design of vacuum electronic devices, the physical structure needs to be modified, that is, the boundary conditions are modified. Based on this physical characteristic, the changes of electric field, magnetic field, current density over time and the distribution of particle velocity are compared through serial and parallel simulation. Before parallelization, the time occupied by each program module of the vacuum device serial program is analyzed to optimize the positioning for multi-core parallelization.

[0024] The steps of controlling the reliability of the electric vacuum device are:

[0025] According to the structural characteristics and load requirements of the electric vacuum device, the control variable method is adopted to control the value range of the quantitative parameters under the condition that the structural parameters remain unchanged. The leakage detection platform is connected with the leakage detection hole attached to the leakage detector. Based on the airtight parts of the electric vacuum device to be tested, the airtight parts can be loaded into the leakage detection platform. According to the model size parameters of the vacuum device and the micro-battery field data, the appropriate current excitation parameters are selected, the sawtooth wave signal bias voltage and amplitude value of the signal generator are adjusted, and the appropriate high-frequency modulation output amplitude and frequency of the signal generator are selected and adjusted. The two signals are connected in series and then connected to the input end of the voltage-current conversion to perform the airtightness detection of the vacuum degree inside the electronic vacuum device.

[0026] According to the above technical solution, the step of evaluating these regression models using the test set data of the electric vacuum device and selecting the optimal electric vacuum device regression prediction model also includes:

[0027] By selecting the best trained deep neural network regression prediction model of the electric vacuum device, the nonlinear mapping relationship between the structural design parameters of the electric vacuum device and its corresponding simulation response data is predicted.

[0028] According to the above technical solution, the step of comparing the calculation results obtained by numerical simulation with the results obtained by experiment includes:

[0029] It is simultaneously determined whether the calculated result of the fitness function can meet the target requirements of the electric vacuum device. If the target requirements are met, the optimization process is directly terminated, and the optimal value that meets the target requirements and the corresponding structural design parameters are output.

[0030] According to the above technical solution, the step of judging further includes:

[0031] If it is judged that the target requirements are not met, the structural design parameters of the electric vacuum device are updated and re-input into the regression prediction model, and the particle swarm optimization algorithm is used again for optimization calculation until the target requirements are finally met and the optimization of the electric vacuum device is completed.

[0032] According to the above technical solution, the step of establishing a simulation model for the high-frequency characteristics of microwaves inside the electric vacuum device also includes:

[0033] The microwave electromagnetic field changes the motion state of charged particles by applying acceleration to them. The motion of the charged particles produces a current source effect that reacts to the microwave electromagnetic field, and energy is exchanged through repeated actions.

[0034] According to the above technical solution, the step of exchanging energy through repeated actions also includes:

[0035] When the forward speed of the electron reaches a reasonable matching state with the forward speed of the microwave signal, the electron beam will transfer its own energy to the microwave signal, and finally amplify the microwave signal. Based on the microwave electromagnetic field simulation method of the electric vacuum device, by performing a complete simulation calculation of the physical process of the vacuum electronic device, the corresponding results are obtained according to the actual needs, which greatly improves the production efficiency of the vacuum electronic device and also makes it have higher accuracy and efficiency.

[0036] According to the above technical solution, the step of parallel computing optimization also includes:

[0037] The electromagnetic field of vacuum electronic devices is parallelized by using a grid method. Each thread of the GPU calculates the electromagnetic field value on a grid. Each field component no longer appears as a property of the grid structure, but as a separately declared array variable. The input signal solution and the port surface information interaction are all executed by a single core on the GPU.

[0038] In the solution of the electromagnetic field, the electric field components at different grid points remain independent of each other, that is, the solution of the electric field component at a certain grid point at the current moment will not affect the solution of the electric field components on other grids, making it highly parallel. In the parallel solution of the electromagnetic field module, through this optimization method, the multi-core CPU only needs to obtain the required parameters from the video memory for each call to the solution function, reducing the number of registers occupied by each thread, thereby increasing the number of active threads overall.

[0039] According to the above technical solution, the electronic device production optimization system with efficient management includes:

[0040] A parameter data regression prediction module is used to train and predict the regression prediction model of the electric vacuum device based on input data and output data;

[0041] Numerical simulation calculation module, used to analyze and iterate the optimization calculation results of electric vacuum devices through optimization algorithm;

[0042] Transmission characteristics analysis module, used to analyze and calculate the microwave transmission and method characteristics inside the electric vacuum device;

[0043] The physical properties optimization module is used to optimize the physical properties of electromagnetic field simulations of vacuum electronic devices through parallel computing.

[0044] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention, by structural analysis and modeling of the internal structure and performance indicators of the electric vacuum device, uses three-dimensional electromagnetic simulation software to design and analyze the high-frequency characteristics, inputs the structural design parameters of the electric vacuum device into the regression prediction model, quickly obtains the predicted response data corresponding to the group of structural design parameters, improves the overall design efficiency and optimizes and reduces the overall time consumption, then compares the calculation results obtained by numerical simulation with the results obtained by experiment, judges whether the calculation results can meet the target requirements of the electric vacuum device, performs a cycle optimization, and further performs complete simulation calculations on the physical process of the vacuum electronic device to improve the production efficiency of the vacuum electronic device, and also makes it have higher accuracy and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0046] Figure 1 A flowchart of an electronic device production optimization method for efficient management provided in Embodiment 1 of the present invention;

[0047] Figure 2 A schematic diagram of the module composition of an efficient management electronic device production optimization system provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Embodiment 1: Figure 1 This is a flow chart of an efficient management electronic device production optimization method provided in Embodiment 1 of the present invention. This embodiment can be applied to the scenario of numerical simulation of electric vacuum devices. This method can be executed by an efficient management electronic device production optimization system provided in this embodiment. Figure 1 As shown, the method specifically comprises the following steps:

[0050] Step 1: Taking the structural design parameters of the electric vacuum device as input and the empirical simulation response data as output, the regression prediction model of the electric vacuum device is trained and predicted based on the input data and the output data;

[0051] In an embodiment of the present invention, a specific experimental method for training and predicting a regression prediction model of an electric vacuum device based on input data and output data includes:

[0052] Step 11: Perform structural analysis and modeling on the internal structure and performance indicators of the electric vacuum device, design and analyze the high-frequency characteristics using three-dimensional electromagnetic simulation software, determine the initial parameter starting values, including establishing a single-cycle model, setting the eigensolver, boundary conditions and frequency range parameters, parameter scanning and optimization, simulation result analysis, and perform simulation based on the above settings through a computer;

[0053] Step 12: Based on a large amount of existing input data and output data, the structural design parameters of the electric vacuum device are used as the input data of the electric vacuum device regression prediction model, and the corresponding empirical simulation response data are used as the output data of the electric vacuum device regression prediction model. These input data and output data are divided into training set data and test set data of the electric vacuum device, and training is performed on the training set data of the electric vacuum device;

[0054] Step 13: Train the deep neural network through data and build a deep neural network regression prediction model, and use the test set data of the electric vacuum device to evaluate these regression models and select the optimal electric vacuum device regression prediction model.

[0055] Exemplarily, the nonlinear mapping relationship between the structural design parameters of the electric vacuum device and its corresponding simulation response data is predicted by selecting the best trained deep neural network regression prediction model of the electric vacuum device.

[0056] Step 2: Combine the prediction model with the optimization algorithm, analyze and iterate the optimization calculation results of the electric vacuum device through the optimization algorithm, compare the calculation results obtained by numerical simulation with the results obtained by experiment, find out the similarities and differences and list them for output;

[0057] In an embodiment of the present invention, the method of analyzing and iterating the optimization calculation results of the electric vacuum device through the optimization algorithm is:

[0058] Step 21: Use the prediction result of the electric vacuum device as the fitness function of the particle swarm optimization algorithm, initialize the particle speed and position in the population, and construct an optimization objective function that meets the target design requirements of the electric vacuum device, that is, construct the fitness function in the particle swarm optimization algorithm, calculate the fitness value of each particle in the population based on the fitness function, and then update the individual optimal solution and the global optimal solution in turn;

[0059] Step 22: Determine whether the optimization calculation result of the optimization algorithm meets the target. When it is determined that the target design requirements of the electric vacuum device are met, the optimization value of the particle swarm optimization algorithm is directly output. When it is determined that the target design requirements of the electric vacuum device are met, the particle speed and the particle position are updated in sequence.

[0060] Step 23: After the speed and position of the particle are updated, the new fitness value corresponding to the new speed and position of the particle is calculated again based on the fitness function, and then the updated individual optimal solution of the particle and the global optimal solution are judged again whether they meet the target requirements until the target is met;

[0061] Step 24: synchronously determine whether the fitness function calculation result can meet the target requirements of the electric vacuum device. If the target requirements are met, directly end the optimization process and output the optimal value that meets the target requirements and the corresponding structural design parameters;

[0062] Step 25: When it is determined that the target requirements are not met, the structural design parameters of the electric vacuum device are updated and re-input into the regression prediction model, and the particle swarm optimization algorithm is used again for optimization calculation until the target requirements are finally met, thereby completing the optimization of the electric vacuum device.

[0063] Step 3: Analyze and calculate the microwave transmission and method characteristics inside the electric vacuum device;

[0064] In the embodiment of the present invention, the analysis method for analyzing and calculating the microwave transmission and method characteristics is:

[0065] Step 31: Based on the target design requirements in the given production process of the electric vacuum device, a simulation model for the high-frequency characteristics of microwaves inside the electric vacuum device is established;

[0066] Step 32: Analyze and fine-tune the high-frequency characteristics of the microwave signal according to the updated structural parameters, and determine the simulation parameters of the interaction based on the adjusted structural parameters;

[0067] Step 33: Conducting simulation analysis of transmission characteristics, i.e., interacting the microwave signal corresponding to the structural parameters inputted by the external update of the electric vacuum device with the electrons injected by the electron injection channel;

[0068] Exemplarily, the microwave electromagnetic field changes the motion state of the charged particles by applying acceleration to the charged particles, and the motion of the charged particles generates a current source effect that reacts to the microwave electromagnetic field, and energy exchange is performed through repeated action;

[0069] When the forward speed of the electron reaches a reasonable matching state with the forward speed of the microwave signal, the electron beam will transfer its own energy to the microwave signal, and finally amplify the microwave signal. Based on the microwave electromagnetic field simulation method of the electric vacuum device, by performing a complete simulation calculation of the physical process of the vacuum electronic device, the corresponding results are obtained according to the actual needs, which greatly improves the production efficiency of the vacuum electronic device and also makes it have higher accuracy and efficiency.

[0070] Step 4: Further utilize the time-varying electromagnetic field simulation analysis inside the vacuum electronic device to optimize the physical characteristics of the electromagnetic field simulation of the vacuum electronic device through parallel computing;

[0071] In an embodiment of the present invention, the step of optimizing the physical characteristics of the electromagnetic field simulation of the vacuum electron device by parallel computing includes:

[0072] Step 41: using a multi-core CPU to simultaneously process the simulation calculations of structural parameters, numerical discrete parameters, microwave parameters, and particle parameters in particle simulation, wherein the calculation boundary is processed by reducing the calculation area and setting an absorption boundary at the boundary. In the process of numerical simulation design of vacuum electronic devices, if the physical structure needs to be modified, that is, the boundary conditions are modified, and the changes of the electric field, magnetic field, and current density over time and the distribution of particle velocity are compared through serial and parallel simulations for this physical characteristic. Before parallelization, the time occupied by each program module of the vacuum device serial program is analyzed to optimize the positioning for multi-core parallelization.

[0073] Step 42: Use a grid approach to parallelize the electromagnetic field of vacuum electronic devices, so that each thread of the GPU calculates the electromagnetic field value on a grid. Each field component no longer appears as an attribute of the grid structure, but as a separately declared array variable. The input signal solution and the port surface information interaction are all placed on the GPU side and executed by a single core.

[0074] Step 43: In solving the electromagnetic field, the electric field components at different grid points remain independent of each other, that is, the solution of the electric field component at a certain grid point at the current moment will not affect the solution of the electric field component on other grids, making it highly parallel. In the parallel solution of the electromagnetic field module, through this optimization method, each call of the multi-core CPU to the solution function only needs to obtain the required parameters from the video memory, reducing the number of registers occupied by each thread, thereby improving the number of active threads overall.

[0075] Step 5: Perform real-time online monitoring based on the internal vacuum degree of the electronic vacuum device, that is, perform real-time leak detection on the electronic vacuum device to control the reliability of the electronic vacuum device;

[0076] In the embodiment of the present invention, according to the structural characteristics and load requirements of the electric vacuum device, a control variable method is adopted to control the value range of the quantitative parameter when the structural parameters remain unchanged, and the leakage detection platform is connected to the leakage detection hole attached to the leakage detector. Based on the airtight parts of the electric vacuum device parts to be detected, the leakage detection platform can be loaded with airtight parts. According to the model size parameters of the vacuum device and the micro-battery field data, appropriate current excitation parameters are selected, the sawtooth wave signal bias voltage and amplitude value of the signal generator are adjusted, and the signal generator is selected to output an appropriate high-frequency modulation output amplitude and frequency. The two signals are connected in series and then connected to the input end of the voltage-current conversion to perform airtightness detection of the vacuum degree inside the electronic vacuum device.

[0077] Embodiment 2: Embodiment 2 of the present invention provides an electronic device production optimization system with efficient management. Figure 2 A schematic diagram of the module composition of an efficient management electronic device production optimization system provided in the second embodiment of the present invention is shown in FIG. Figure 2 As shown, the system includes:

[0078] A parameter data regression prediction module is used to train and predict the regression prediction model of the electric vacuum device based on input data and output data;

[0079] Numerical simulation calculation module, used to analyze and iterate the optimization calculation results of electric vacuum devices through optimization algorithm;

[0080] Transmission characteristics analysis module, used to analyze and calculate the microwave transmission and method characteristics inside the electric vacuum device;

[0081] The physical properties optimization module is used to optimize the physical properties of electromagnetic field simulations of vacuum electronic devices through parallel computing.

[0082] In some embodiments of the present invention, the parameter data regression prediction module includes:

[0083] Structural analysis and modeling module, used to perform structural analysis and modeling on the internal structure and performance indicators of electric vacuum devices;

[0084] High-frequency characteristics analysis module, used to design and analyze high-frequency characteristics using 3D electromagnetic simulation software;

[0085] A simulation module is used to perform simulation based on parameter structural analysis settings through a computer;

[0086] A training prediction module is used for training based on input data and output data;

[0087] Regression prediction model building module, used to build deep neural network regression prediction model;

[0088] The nonlinear mapping relationship module is used to predict the nonlinear mapping relationship between the structural design parameters of the electric vacuum device and its corresponding simulation response data.

[0089] In some embodiments of the present invention, the numerical simulation calculation module includes:

[0090] Algorithm combination module, used to combine the prediction model with the optimization algorithm;

[0091] The fitness function module is used to use the prediction results of the electric vacuum device as the fitness function of the particle swarm optimization algorithm;

[0092] The calculation result judgment module is used to judge whether the optimization calculation result of the optimization algorithm meets the target.

[0093] In some embodiments of the present invention, the transmission characteristic analysis module includes:

[0094] A structural parameter updating module, used for analyzing the high-frequency characteristics of microwave signals and fine-tuning the parameters according to the updated structural parameters;

[0095] Transmission characteristics simulation analysis module, used for simulation analysis of transmission characteristics;

[0096] In some embodiments of the present invention, the physical property optimization module includes:

[0097] Parallel computing optimization module, used to perform parallel optimization calculation of electromagnetic fields of vacuum electronic devices using a grid approach;

[0098] The reliability control module is used to control the reliability of the electric vacuum device.

[0099] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0100] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An efficient management method for optimizing the production of electronic devices, characterized in that: The method comprises the following steps: Step 1: Taking the structural design parameters of the electric vacuum device as input and the empirical simulation response data as output, the regression prediction model of the electric vacuum device is trained and predicted based on the input data and the output data; Step 2: Combine the prediction model with the optimization algorithm, analyze and iterate the optimization calculation results of the electric vacuum device through the optimization algorithm, compare the calculation results obtained by numerical simulation with the results obtained by experiment, find out the similarities and differences and list them for output; Step 3: Analyze and calculate the microwave transmission and method characteristics inside the electric vacuum device; Step 4: Further utilize the time-varying electromagnetic field simulation analysis inside the vacuum electronic device to optimize the physical characteristics of the electromagnetic field simulation of the vacuum electronic device through parallel computing; Step 5: Perform real-time online monitoring based on the internal vacuum degree of the electronic vacuum device, that is, perform real-time leak detection on the electronic vacuum device to control the reliability of the electronic vacuum device; The specific experimental method for training and predicting the regression prediction model of the electric vacuum device based on input data and output data includes: Step 11: Perform structural analysis and modeling on the internal structure and performance indicators of the electric vacuum device, design and analyze the high-frequency characteristics using three-dimensional electromagnetic simulation software, determine the initial parameter starting values, including establishing a single-cycle model, setting the eigensolver, boundary conditions and frequency range parameters, parameter scanning and optimization, simulation result analysis, and perform simulation based on the above settings through a computer; Step 12: Based on a large amount of existing input data and output data, the structural design parameters of the electric vacuum device are used as the input data of the electric vacuum device regression prediction model, and the corresponding empirical simulation response data are used as the output data of the electric vacuum device regression prediction model. These input data and output data are divided into training set data and test set data of the electric vacuum device, and training is performed on the training set data of the electric vacuum device; Step 13: Train the deep neural network through data and build a deep neural network regression prediction model, and use the test set data of the electric vacuum device to evaluate these regression models and select the optimal electric vacuum device regression prediction model; The method of analyzing and iterating the optimization calculation results of the electric vacuum device through the optimization algorithm is: Step 21: Use the prediction result of the electric vacuum device as the fitness function of the particle swarm optimization algorithm, initialize the particle speed and position in the population, and construct an optimization objective function that meets the target design requirements of the electric vacuum device, that is, construct the fitness function in the particle swarm optimization algorithm, calculate the fitness value of each particle in the population based on the fitness function, and then update the individual optimal solution and the global optimal solution in turn; Step 22: Determine whether the optimization calculation result of the optimization algorithm meets the target. When it is determined that the target design requirements of the electric vacuum device are met, the optimization value of the particle swarm optimization algorithm is directly output. When it is determined that the target design requirements of the electric vacuum device are met, the particle speed and the particle position are updated in sequence. Step 23: After the speed and position of the particle are updated, the new fitness value corresponding to the new speed and position of the particle is calculated again based on the fitness function, and then the updated individual optimal solution of the particle and the global optimal solution are judged again whether they meet the target requirements until the target is met; The analysis method for analyzing and calculating microwave transmission and method characteristics is: Step 31: Based on the target design requirements in the given production process of the electric vacuum device, a simulation model for the high-frequency characteristics of microwaves inside the electric vacuum device is established; Step 32: Analyze and fine-tune the high-frequency characteristics of the microwave signal according to the updated structural parameters, and determine the simulation parameters of the interaction based on the adjusted structural parameters; Step 33: Conducting simulation analysis of transmission characteristics, i.e., interacting the microwave signal corresponding to the structural parameters inputted by the external update of the electric vacuum device with the electrons injected by the electron injection channel; The step of optimizing the physical characteristics of the electromagnetic field simulation of the vacuum electron device by parallel computing comprises: By using a multi-core CPU to simultaneously process the simulation calculations of structural parameters, numerical discrete parameters, microwave parameters, and particle parameters in particle simulation, the calculation boundary is processed by reducing the calculation area and setting an absorption boundary at the boundary. In the process of numerical simulation design of vacuum electronic devices, the physical structure needs to be modified, that is, the boundary conditions are modified. Based on this physical characteristic, the changes of electric field, magnetic field, current density over time and the distribution of particle velocity are compared through serial and parallel simulation. Before parallelization, the time occupied by each program module of the vacuum device serial program is analyzed to optimize the positioning for multi-core parallelization. The steps of controlling the reliability of the electric vacuum device are: According to the structural characteristics and load requirements of the electric vacuum device, the control variable method is adopted to control the value range of the quantitative parameters under the condition that the structural parameters remain unchanged. The leakage detection platform is connected with the leakage detection hole attached to the leakage detector. Based on the airtight parts of the electric vacuum device to be tested, the airtight parts can be loaded into the leakage detection platform. According to the model size parameters of the vacuum device and the micro-battery field data, the appropriate current excitation parameters are selected, the sawtooth wave signal bias voltage and amplitude value of the signal generator are adjusted, and the appropriate high-frequency modulation output amplitude and frequency of the signal generator are selected and adjusted. The two signals are connected in series and then connected to the input end of the voltage-current conversion to perform the airtightness detection of the vacuum degree inside the electronic vacuum device.

2. The method for optimizing the production of electronic devices with high efficiency management according to claim 1, characterized in that: The step of evaluating these regression models using the test set data of the electric vacuum device to select the optimal electric vacuum device regression prediction model also includes: By selecting the optimal trained deep neural network regression prediction model of the electric vacuum device, the nonlinear mapping relationship between the structural design parameters of the electric vacuum device and its corresponding simulation response data is predicted.

3. The method for optimizing the production of electronic devices with high efficiency management according to claim 1, characterized in that: The step of comparing the calculation results obtained by numerical simulation with the results obtained by experiment includes: It is simultaneously determined whether the calculated result of the fitness function can meet the target requirements of the electric vacuum device. If the target requirements are met, the optimization process is directly terminated, and the optimal value that meets the target requirements and the corresponding structural design parameters are output.

4. The method for optimizing the production of electronic devices with high efficiency management according to claim 3, characterized in that: The step of judging further includes: If it is judged that the target requirements are not met, the structural design parameters of the electric vacuum device are updated and re-input into the regression prediction model, and the particle swarm optimization algorithm is used again for optimization calculation until the target requirements are finally met and the optimization of the electric vacuum device is completed.

5. The method for optimizing the production of electronic devices with high efficiency management according to claim 4, characterized in that: The step of establishing a simulation model for the high-frequency characteristics of microwaves inside the electric vacuum device also includes: The microwave electromagnetic field changes the motion state of charged particles by applying acceleration to them. The motion of the charged particles produces a current source effect that reacts to the microwave electromagnetic field, and energy is exchanged through repeated actions.

6. The method for optimizing the production of electronic devices with high efficiency management according to claim 5, characterized in that: The step of exchanging energy by repeated action also includes: When the forward speed of the electron reaches a reasonable matching state with the forward speed of the microwave signal, the electron beam will transfer its own energy to the microwave signal, and finally amplify the microwave signal. Based on the microwave electromagnetic field simulation method of the electric vacuum device, by performing a complete simulation calculation of the physical process of the vacuum electronic device, the corresponding results are obtained according to the actual needs, which greatly improves the production efficiency of the vacuum electronic device and also makes it have higher accuracy and efficiency.

7. The method for optimizing the production of electronic devices with high efficiency management according to claim 1, characterized in that: The step of parallel computing optimization also includes: The electromagnetic field of vacuum electronic devices is parallelized by using a grid method. Each thread of the GPU calculates the electromagnetic field value on a grid. Each field component no longer appears as a property of the grid structure, but as a separately declared array variable. The input signal solution and the port surface information interaction are all executed by a single core on the GPU. In the solution of the electromagnetic field, the electric field components at different grid points remain independent of each other, that is, the solution of the electric field component at a certain grid point at the current moment will not affect the solution of the electric field components on other grids, making it highly parallel. In the parallel solution of the electromagnetic field module, through this optimization method, the multi-core CPU only needs to obtain the required parameters from the video memory for each call to the solution function, reducing the number of registers occupied by each thread, thereby increasing the number of active threads overall.

8. An efficient management electronic device production optimization system for executing the efficient management electronic device production optimization method of claim 1, characterized in that: The system comprises: A parameter data regression prediction module is used to train and predict the regression prediction model of the electric vacuum device based on input data and output data; Numerical simulation calculation module, used to analyze and iterate the optimization calculation results of electric vacuum devices through optimization algorithm; Transmission characteristics analysis module, used to analyze and calculate the microwave transmission and method characteristics inside the electric vacuum device; The physical properties optimization module is used to optimize the physical properties of electromagnetic field simulations of vacuum electronic devices through parallel computing.

9. The electronic device production optimization system for efficient management according to claim 8, characterized in that: The parameter data regression prediction module includes: Structural analysis and modeling module, used to perform structural analysis and modeling on the internal structure and performance indicators of electric vacuum devices; High-frequency characteristics analysis module, used to design and analyze high-frequency characteristics using 3D electromagnetic simulation software; A simulation module is used to perform simulation based on parameter structural analysis settings through a computer; A training prediction module is used for training based on input data and output data; Regression prediction model building module, used to build deep neural network regression prediction model; A nonlinear mapping relationship module is used to predict the nonlinear mapping relationship between the structural design parameters of the electric vacuum device and its corresponding simulation response data; The numerical simulation calculation module comprises: Algorithm combination module, used to combine the prediction model with the optimization algorithm; The fitness function module is used to use the prediction results of the electric vacuum device as the fitness function of the particle swarm optimization algorithm; A calculation result judgment module is used to judge whether the optimization calculation result of the optimization algorithm meets the target; The transmission characteristic analysis module comprises: A structural parameter updating module, used for analyzing the high-frequency characteristics of microwave signals and fine-tuning the parameters according to the updated structural parameters; Transmission characteristics simulation analysis module, used for simulation analysis of transmission characteristics; The physical property optimization module includes: Parallel computing optimization module, used to perform parallel optimization calculation of electromagnetic fields of vacuum electronic devices using a grid approach; The reliability control module is used to control the reliability of the electric vacuum device.

Images