Parallel multi-objective optimization method for traveling wave tube component

Abstract

The invention relates to the field of traveling wave tube design and optimization, in particular to a parallel multi-objective optimization method for traveling wave tube components. According to the method, decision variable optimization and target optimization selection and setting are carried out on traveling wave tube components through optimization setting, a universal parallel computing processor is introduced to adaptively carry out parallel computing on various traveling wave tube component projects and obtain corresponding simulation calculation results, and finally non-dominated sorting, optimal solution updating and population evolution are executed. On the basis of the method, the traveling wave tube part can be designed and optimized more flexibly, the method has higher optimization design freedom degree, and the design optimization and development technology of the traveling wave tube is further iterated and updated; the problem that according to an existing traveling wave tube component multi-objective optimization method, independent serial optimization execution needs to be carried out on each calculation project, and the efficiency is low is effectively solved.

Description

Technical Field

[0001] The invention relates to the field of traveling wave tube design and optimization, and in particular to a parallel multi-objective optimization method for traveling wave tube components. Background Art

[0002] Traveling wave tubes are widely used in military applications, communication systems and other fields due to their wide operating frequency band, high output power and high efficiency. Traveling wave tubes are composed of electron guns, high-frequency circuits, collectors, magnetic systems, and input window structures. In the design process of traveling wave tubes, each component of the traveling wave tube will encounter multi-objective optimization problems.

[0003] For example, the electron gun that generates electron beams needs to optimize the cathode radius and the distance between the cathode and cathode to obtain a larger total cathode emission current and a smaller beam waist radius to generate more electron beam energy; the high-frequency circuit that provides dispersion characteristics needs to optimize the structural parameters to obtain a flatter dispersion characteristic within the working frequency band, and requires better coupling impedance to improve the electron efficiency of the injection-wave interaction; after the injection-wave interaction is completed, the collector needs to optimize the structure and voltage at each level to meet higher recovery efficiency and collector efficiency. The optimization decision variables and optimization objectives of different components in the traveling wave tube are different. At present, there is still a lack of a unified optimization framework that can optimize all components of the traveling wave tube with multiple objectives.

[0004] In the previous multi-objective optimization method for traveling wave tube components, each calculation project needs to be executed separately, and the next calculation project can only be executed after the calculation of one project is completed. This calculation method is a serial optimization calculation method. This serial optimization method has a slow calculation speed and takes more time. Summary of the invention

[0005] In view of the above-mentioned problems or shortcomings, in order to solve the problem of slow serial optimization calculation speed, form a unified architecture for multi-objective optimization of all traveling wave tube components, and improve the flexibility of optimization design of traveling wave tube components, the present invention proposes a parallel multi-objective optimization method for traveling wave tube components.

[0006] A parallel multi-objective optimization method for traveling wave tube components, the specific steps are as follows:

[0007] Step 1. Set the optimization target of the traveling wave tube components;

[0008] According to the design requirements of the traveling wave tube components, the optimization decision variables of the traveling wave tube components are set in the optimization setting interface, including whether the traveling wave tube optimization decision variables are optimized, the lower limit and upper limit of the optimization decision variables; the optimization target of the traveling wave tube components is set, including the optimization target value and the optimization direction. The optimization direction includes two directions: greater than and less than.

[0009] Specifically, the optimization decision variables of the electron gun include the electron gun cathode position, anode position and electron gun voltage; the optimization objectives of the electron gun include cathode emission total current, backfly cathode current, interception current, injection waist radius and injection waist position.

[0010] Specifically, the optimization decision variables of the high-frequency structure include the inner radius and pitch of the helix; the optimization targets of the high-frequency structure include the normalized phase velocity, coupling impedance and attenuation constant.

[0011] Specifically, the optimization decision variables of the collector include the collector inlet energy distribution, the collector structure parameters, and the collector electrode voltages; the optimization objectives of the collector include the collector efficiency and the recirculation rate.

[0012] Specifically, the optimization decision variables of the magnetic system include the structural parameters of the magnetic steel, the structural parameters of the pole shoe, and the material; the optimization targets of the magnetic system include the magnetic field period, the peak magnetic field on the axis, and the magnetic system performance parameters.

[0013] Specifically, the optimization decision variables of the input-output window include the structural design parameters of the input-output window; and the optimization targets of the input-output window include S parameters and standing wave ratio.

[0014] Step 2. Set the optimization algorithm;

[0015] Select a multi-objective optimization algorithm, including NSGA-II or MOPSO optimization algorithm, and set corresponding optimization algorithm parameters according to the selected optimization algorithm.

[0016] Specifically, the NSGA-II multi-objective optimization algorithm parameters include the maximum number of iterations N Iter , the size of the population N Pop , crossover rate d cross , mutation rate d mutate and variable precision N ps .

[0017] The parameters of the MOPSO multi-objective optimization algorithm include the maximum number of iterations N Iter , the size of the population N Pop , inertia weight factor d w , individual learning factor d c1 , mutation rate d c2 and variable precision N ps .

[0018] Specifically, the variable precision N mentioned ps The number of significant digits after the decimal point for constrained optimization decision variables is N ps bit, that is, when N ps =2, indicating that the number of significant digits of the optimization decision variable is 2 decimal places.

[0019] Step 3. Generate optimization algorithm to initialize the population;

[0020] According to the optimization decision variables and optimization objectives of the traveling wave tube components set in Step 1 and the optimization algorithm set in Step 2, N are randomly generated within the parameter range of the optimization decision variables of the traveling wave tube components. pop Optimization algorithm individuals, generate the initial population Pop(I Iter ), record I Iter =0.

[0021] Specifically, use I Iter Indicates the current iteration number of the optimization algorithm, that is, I Iter =0 means the current population is the initialized population, I Iter =1 means the current population Pop(I Iter ) is the first generation population.

[0022] Step 4. Based on the current population Pop(I Iter ) to send a "start parallel computing" instruction to the parallel computing processor.

[0023] Step 5. The parallel computing processor starts executing;

[0024] The parallel computing processor receives the instruction "Start parallel computing" from Step 4, identifies the type of the traveling wave tube component computing project, and the parallel computing processor calculates the number of cells according to the population size N in Step 2. Pop Generate the number of processes N corresponding to the setting Thread , that is, N Thread =N pop And update the optimization decision variables of the traveling wave tube components accordingly, and then start the corresponding solver for parallel calculation. When the solver calculation is completed, obtain the current population Pop(I Iter ), the parallel computing processor sends a “parallel computing completed” instruction to Step 6.

[0025] Specifically, the types of computational engineering of traveling wave tube components that can be processed by the parallel processor include electron gun, high frequency circuit, collector, magnetic system and input / output window, and the corresponding solvers are EOS, HFCS, EOS, MFS and WS respectively. The computational engineering of electron gun and collector are simulated and calculated by EOS solver.

[0026] Step 6. Perform non-dominated sorting and update the optimal solution;

[0027] Receive the "Parallel Computation Completed" instruction sent by the parallel processor in Step 5, and in the current population Pop(I Iter ) performs fast non-dominated sorting and congestion calculation, updates the optimal solution, and determines the Pareto frontier.

[0028] Step 7. Execute population evolution and generate new population;

[0029] For the current population Pop(I Iter ) executes elite strategy, crossover and mutation to form a new generation of population Pop(I Iter +1), Record I Iter =I Iter +1.

[0030] Step8. Execute termination condition judgment;

[0031] Determine the current number of iterations I Iter Is it greater than the maximum number of iterations N? Iter , if I Iter Greater than N Iter Then the optimization calculation of the traveling wave tube components is terminated and Step 9 is executed; otherwise, Step 4 is returned to execute and a "start parallel calculation" instruction is sent to the parallel computing processor until I Iter Greater than N Iter .

[0032] Step 9. Result analysis and display;

[0033] According to the currently obtained optimization results, the optimization decision variables and optimization objectives of each generation are stored in the data file, and the saved data are displayed.

[0034] In summary, the present invention proposes a parallel multi-objective optimization method for traveling wave tube components, which optimizes decision variables and selects and sets optimization targets for traveling wave tube components through optimization settings, and then introduces a general parallel computing processor to adaptively parallelly calculate multiple traveling wave tube component projects and obtain corresponding simulation calculation results, and finally performs non-dominated sorting and updates the optimal solution and population evolution. Based on the present invention, traveling wave tube components can be designed and optimized more flexibly, with a higher degree of freedom in optimization design, and the design optimization and development technology of traveling wave tubes can be further iterated and updated; the problem of low efficiency in the existing multi-objective optimization method for traveling wave tube components, which requires separate serial optimization execution of each calculation project, is effectively solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic block diagram of the process of the present invention;

[0036] Figure 2 is a schematic block diagram of a parallel computing processor in the present invention;

[0037] Figure 3 is a schematic diagram of a parallel computing engineering of an electron gun embodiment of the present invention;

[0038] Figure 4 It is the result of the optimized design of the electron gun embodiment of the present invention;

[0039] Figure 5 It is a block diagram comparing the parallel execution mechanism of the present invention with the serial execution mechanism of the prior art;

[0040] Figure 6 is a diagram showing the results of the optimal design of the electron gun by implementing the parallel execution mechanism of the embodiment;

[0041] Figure 7 This is a diagram showing the optimization design result of the electron gun implemented by the serial mechanism of the prior art. DETAILED DESCRIPTION

[0042] To facilitate understanding of the technical content of the present invention, the content of the present invention is described in detail below by taking the parallel multi-objective optimization of an electron gun as an example with reference to the accompanying drawings.

[0043] A parallel multi-objective optimization method for traveling wave tube components, such as Figure 1 As shown, the following steps are included:

[0044] Step 1. Set the optimization target of the traveling wave tube components;

[0045] The optimization decision variable selected is the cathode curvature radius Rc, the current value is 9.3mm, and the optimization range is set to [8.37,10.23]; the lateral distance Z of the focusing pole g The current value is 1.71mm, and the optimization range is set to [1.539,1.881]; the longitudinal distance R g1 The current value is 6.1, and the optimization range is set to [5.49, 6.71]; the axial distance Z of the anode a The current value of is 9.51mm, and the optimization range is set to [8.559,10.461].

[0046] The optimization objectives include setting the cathode emission total current I total , set the optimization direction to be greater than, set the optimization target value to 200mA; set the injection waist radius R b As the optimization target, set the optimization direction to be less than, and set the optimization target value to 0.4mm, that is, the optimization objective function is min{I total >200mA,R b <0.4mm}.

[0047] Step 2. Set the optimization algorithm;

[0048] Select the NSGA-II multi-objective optimization algorithm, and set the optimization algorithm parameters to the maximum number of iterations N. Iter =50, population size N Pop =16, cross rate dcross =0.6, mutation rate d mutate = 0.2 and variable precision N ps =2.

[0049] Step 3. Generate optimization algorithm to initialize the population;

[0050] According to the optimization decision variables and optimization objectives of the traveling wave tube components set in Step 1 and the optimization algorithm set in Step 2, click Start Optimization on the optimization setting interface to randomly generate N optimization decision variables within the parameter range. Pop = 16 individuals, generating the initial population Pop(I Iter ), record I Iter =I Iter +1.

[0051] Step 4. Based on the current population Pop(I Iter ) to send a "start parallel computing" instruction to the parallel computing processor.

[0052] Step 5. Perform non-dominated sorting and update the optimal solution;

[0053] like Figure 2 The figure shows the execution flow of the parallel computing processor. The parallel computing processor receives the instruction "Start parallel computing" from Step 4 and identifies the type of the traveling wave tube component computing project as the electron gun computing project. The parallel computing processor generates the corresponding number of processes N according to the set population size. Thread , that is, N Thread =N pop And update the structural parameters of the traveling wave tube components accordingly, and then start the corresponding solver EOS for parallel calculation. When the solver calculation is completed, obtain the current population Pop (I Iter ), the parallel computing processor sends a “parallel computing completed” instruction to Step 6.

[0054] like Figure 3 The figure shows the N generated by the parallel computing processor. Thread =16 electron gun calculation projects.

[0055] Step 6. Perform non-dominated sorting and update the optimal solution;

[0056] Receive the "Parallel Computation Completed" instruction sent by the parallel processor in Step 5, and in the current population Pop(I Iter ) performs fast non-dominated sorting and congestion calculation, updates the optimal solution, and determines the Pareto frontier.

[0057] Step 7. Execute population evolution and generate new population;

[0058] For the current population Pop(I Iter ) executes elite strategy, crossover and mutation to form a new generation of population Pop(I Iter +1), Record I Iter =I Iter +1.

[0059] Step8. Termination condition judgment;

[0060] Determine the current number of optimization iterations I Iter Is it greater than the maximum number of iterations N? Iter , if I Iter Greater than N Iter Then the optimization calculation of the traveling wave tube components is terminated and Step 9 is executed; otherwise, Step 4 is returned to send a "start parallel calculation" instruction to the parallel computing processor until I Iter Greater than N Iter .

[0061] Step 9. Result analysis and display;

[0062] According to the optimization results, the optimization decision variables and optimization objectives of each generation are stored in the data file, and the saved data is displayed. Figure 4 The results and display of the electron gun optimization design scheme of this embodiment are shown. The design parameters of the traveling wave tube components and the corresponding performance indicators can be seen in the optimization results, and the total cathode emission current I total Greater than the set optimization target value 200mA, injection waist radius R b Smaller than the set optimization target value of 0.4mm.

[0063] Figure 5 This is a block diagram comparing the startup calculation order of the serial optimization calculation and the parallel optimization calculation of the prior art and the present invention. Taking the serial multi-objective optimization of the electron gun as an example, the calculation project of each electron gun needs to be executed separately, and the calculation of one electron gun project can only be executed after the calculation of the next electron gun project is completed. This calculation method is the serial optimization calculation method. The parallel optimization calculation can start multiple electron gun projects for calculation at the same time, which can improve the calculation speed of the electron gun multi-objective optimization.

[0064] exist Figure 6 The optimization start time and optimization end time of the electron gun optimization design scheme of this embodiment can be viewed in the description box in the optimization interface. The parallel optimization takes 0 hours, 2 minutes, and 58 seconds, or 178 seconds. If a serial optimization calculation iteration is adopted, 16 electron gun calculation projects need to be calculated serially. And when the number of optimization iterations increases, more electron gun calculation projects need to be optimized and calculated serially, which will consume more time. Figure 7The calculation time of the serial optimization of the electron gun example design optimization scheme is 0 hours, 12 minutes and 48 seconds, or 768 seconds. From the optimization time of the two figures, it can be seen that parallel optimization calculation can significantly improve the multi-objective optimization calculation speed of traveling wave tube components.

[0065] It can be seen from the above embodiments that the present invention can design and optimize traveling wave tube components more flexibly, has a higher degree of freedom in optimization design, and further iterates and updates the design optimization and development technology of the traveling wave tube; and effectively solves the problem of low efficiency of the existing multi-objective optimization method of traveling wave tube components, which requires separate serial optimization execution of each calculation project.

Claims

1. A parallel multi-objective optimization method for traveling wave tube components, characterized in that: The specific steps are as follows: Step 1. Set the optimization target of the traveling wave tube components; According to the design requirements of the traveling wave tube components, the optimization decision variables of the traveling wave tube components are set in the optimization setting interface, including whether the traveling wave tube optimization decision variables are optimized, the lower limit value and the upper limit value of the optimization decision variables; the optimization target of the traveling wave tube components is set, including the optimization target value and the optimization direction; Step 2. Set the optimization algorithm; Select a multi-objective optimization algorithm, including NSGA-II or MOPSO optimization algorithm, and set corresponding optimization algorithm parameters according to the selected optimization algorithm; Step 3. Generate optimization algorithm to initialize the population; According to the optimization decision variables and optimization objectives of the traveling wave tube components set in Step 1 and the optimization algorithm set in Step 2, N are randomly generated within the parameter range of the optimization decision variables of the traveling wave tube components. pop Optimization algorithm individuals, generate the initial population Pop(I Iter ), record I Iter =0; Specifically, use I Iter Indicates the current iteration number of the optimization algorithm, that is, I Iter =0 means the current population is the initialized population, I Iter =1 means the current population Pop(I Iter ) is the population of the first generation; Step 4. Based on the current population Pop(I Iter ), sending an instruction to start parallel computing to the parallel computing processor; Step 5. The parallel computing processor starts executing; The parallel computing processor receives the instruction to start parallel computing from Step 4, identifies the type of the traveling wave tube component computing project, and the parallel computing processor calculates the number of cells according to the population size N in Step 2. Pop Generate the number of processes N corresponding to the setting Thread , that is, N Thread =N pop and updating the optimization decision variables of the traveling wave tube components accordingly, and then starting the corresponding solver for parallel calculation; When the solver calculation is completed, get the current population Pop(I Iter ), the parallel computing processor sends a parallel computing completion instruction to Step 6; Step 6. Perform non-dominated sorting and update the optimal solution; Receive the parallel computing completion instruction sent by the parallel processor in Step 5, and in the current population Pop(I Iter ) Perform fast non-dominated sorting and congestion calculation, update to obtain the optimal solution, and determine the Pareto frontier; Step 7. Execute population evolution and generate new population; For the current population Pop(I Iter ) executes elite strategy, crossover and mutation to form a new generation of population Pop(I Iter +1), Record I Iter =I Iter +1; Step8. Execute termination condition judgment; Determine the current number of iterations I Iter Is it greater than the maximum number of iterations N? Iter , if I Iter Greater than N Iter Then the optimization calculation of the traveling wave tube components is terminated and Step 9 is executed; otherwise, Step 4 is returned to execute to send a start parallel computing instruction to the parallel computing processor until I Iter Greater than N Iter ; Step 9. Result analysis and display; According to the currently obtained optimization results, the optimization decision variables and optimization objectives of each generation are stored in the data file, and the saved data are displayed.

2. The parallel multi-objective optimization method for traveling wave tube components according to claim 1, characterized in that: The optimization direction described in Step 1 includes two directions: greater than and less than.

3. The parallel multi-objective optimization method for traveling wave tube components according to claim 1, characterized in that: The optimization decision variables of the electron gun in Step 1 include the electron gun cathode position, anode position and electron gun voltage; the optimization objectives of the electron gun include cathode emission total current, backflash cathode current, interception current, injection waist radius and injection waist position.

4. The parallel multi-objective optimization method for traveling wave tube components according to claim 1, characterized in that: In the Step 1, the optimization decision variables of the high-frequency structure include the inner radius and pitch of the helix; the optimization targets of the high-frequency structure include but are not limited to normalized phase velocity, coupling impedance, and attenuation constant.

5. The parallel multi-objective optimization method for traveling wave tube components according to claim 1, characterized in that: The optimization decision variables of the collector in Step 1 include the collector inlet energy distribution, the collector structure parameters, and the collector electrode voltages; the optimization objectives of the collector include the collector efficiency and the recirculation rate.

6. The parallel multi-objective optimization method for traveling wave tube components according to claim 1, characterized in that: The optimization decision variables of the magnetic system in Step 1 include the structural parameters of the magnetic steel, the structural parameters of the pole shoe, and the material; the optimization targets of the magnetic system include the magnetic field period, the peak magnetic field on the axis, and the magnetic system performance parameters.

7. The parallel multi-objective optimization method for traveling wave tube components according to claim 1, characterized in that: The optimization decision variables of the input-output window in Step 1 include the structural design parameters of the input-output window; and the optimization targets of the input-output window include S parameters and standing wave ratio.

8. The parallel multi-objective optimization method for traveling wave tube components according to claim 1, characterized in that: In Step 1: NSGA-II multi-objective optimization algorithm parameters include the maximum number of iterations N Iter , the size of the population N Pop , crossover rate d cross , mutation rate d mutate and variable precision N ps ; MOPSO multi-objective optimization algorithm parameters include the maximum number of iterations N Iter , the size of the population N Pop , inertia weight factor d w , individual learning factor d c1 , mutation rate d c2 and variable precision N ps ; The variable precision N ps The number of significant digits after the decimal point for constrained optimization decision variables is N ps bit, that is, when N ps =2, indicating that the number of significant digits of the optimization decision variable is 2 decimal places.

9. The parallel multi-objective optimization method for traveling wave tube components according to claim 1, characterized in that: In the Step 5, the types of computational engineering of traveling wave tube components that can be processed by the parallel processor include electron gun, high frequency circuit, collector, magnetic system and input / output window, and the corresponding solvers are EOS, HFCS, EOS, MFS and WS respectively; the computational engineering of electron gun and collector are both simulated and calculated by EOS solver.

Images