Rapid prediction method for performance of whole traveling wave tube

Abstract

The invention relates to the field of design and optimization of traveling wave tubes, in particular to a method capable of predicting the performance of the whole traveling wave tube. The method comprises the steps of firstly establishing a traveling wave tube part prediction model according to design parameters; then, the beam-wave interaction of the traveling wave tube is taken as a core, related performance parameters of the prediction model of each part of the traveling wave tube are transmitted to a beam-wave interaction solver, and calculation of performance indexes such as output power, electronic efficiency and gain of the traveling wave tube is completed through calculation of the beam-wave interaction solver; and calculating through the collector prediction model to obtain a traveling wave tube whole tube efficiency performance index, and establishing a traveling wave tube whole tube prediction model in a data flow mode. According to the invention, the data chain for transmitting the full parameters of the traveling wave tube can be established, and the performance prediction of each component and the performance prediction of the whole traveling wave tube are completed by using the proxy model based on the full-parameter data chain. According to the method, the design work of the traveling wave tube has higher flexibility, and the design and development technology of the traveling wave tube is further iterated and updated.

Description

Technical Field

[0001] The invention relates to the field of traveling wave tube design and optimization, and in particular to a method for quickly predicting the performance of a traveling wave tube. Background Art

[0002] Traveling wave tubes are widely used in communication systems and other fields due to their wide operating frequency band, high output power and high efficiency. At present, the design of traveling wave tubes is mainly completed with the help of electromagnetic simulation software such as HFSS, CST, and MTSS. However, the simulation calculations of these software are all for each component of the traveling wave tube. Modeling and simulation calculations are performed separately, which indirectly reflects the performance of the entire tube. There is a lack of a traveling wave tube data link that transmits all the data of the traveling wave tube components, which makes it impossible to directly realize the rapid prediction of the performance of the entire tube.

[0003] When predicting the performance of the entire traveling wave tube, designers need to go back to the process of manually adjusting the structural parameters of each traveling wave tube component, and then design and optimize the entire tube after the component design is completed. This manual method of adjusting parameters will significantly increase the design cycle of the entire traveling wave tube and consume more computing resources. In addition, this manual method of adjusting parameters cannot connect the data of different traveling wave tube components and it is difficult to take into account all the key parameters of the traveling wave tube components at the same time. Summary of the invention

[0004] In view of the above problems or deficiencies, in order to realize the rapid performance prediction of the whole traveling wave tube, the present invention provides a method for rapid prediction of the performance of the whole traveling wave tube. The method establishes a data link for transmitting all parameters of the traveling wave tube, and uses a proxy model based on the full parameter data link to complete the performance prediction of each component of the traveling wave tube and the performance prediction of the whole tube.

[0005] A method for quickly predicting the performance of a traveling wave tube as a whole, the specific steps are as follows:

[0006] Step1. Design indicators determine design parameters;

[0007] According to the design index requirements of the traveling wave tube, the design parameters of the electron gun, high-frequency structure, magnetic system, energy transmission window, and collector traveling wave tube components are first determined through the traveling wave tube forward design tool;

[0008] Step 2. Establish a prediction model for traveling wave tube components based on design parameters;

[0009] According to the design parameters of the traveling wave tube components determined in Step 1, prediction models of various components of the traveling wave tube are established using prediction tools; according to the organizational structure of the traveling wave tube, the prediction models of various components of the traveling wave tube include five prediction models, namely, electron gun prediction model, high-frequency structure prediction model, energy transmission window prediction model, magnetic system prediction model and collector prediction model.

[0010] Specifically, x represents design parameters, y represents performance parameters, i represents input, and o represents output, then the data link format of the electron gun prediction model is [x@guni, y@guno]. The data link format of the high-frequency structure prediction model is [x@hfcsi, y@hfcso]; the data link format of the energy transmission window prediction model is [x@wsi, y@wso]; the data link format of the magnetic system prediction model is [x@mfsi, y@mfso]; the data link format of the collector prediction model is [x@mdci, y@mdco].

[0011] Specifically, the prediction tools include neural networks, Kriging proxy models and autoencoders.

[0012] Step 3. Establish the TWTATPM prediction model based on the full parameter data link;

[0013] According to the complete process of electron emission, forming, focusing, transmission, energy exchange and collection in the traveling wave tube, with the injection-wave interaction of the traveling wave tube as the core, the prediction models of various components of the traveling wave tube are connected through the data flow in the whole traveling wave tube, and the traveling wave tube whole tube prediction model TWTATPM based on the traveling wave tube full parameter data link is established.

[0014] Step 4. Predict the performance of the entire traveling wave tube;

[0015] Input data into the TWT whole tube prediction model TWTATPM constructed in Step 3, and use the TWT whole tube prediction model TWTATPM to predict the performance of the TWT whole tube.

[0016] Furthermore, the Step 3 also includes that the data format of the injection-wave interaction prediction model is [x@bwii, y@bwio], and the data format of the established whole-tube prediction model TWTATPM is [x@TWTATPMi, y@TWTATPMo].

[0017] Furthermore, the predictable traveling wave tube performance in Step 4 includes the output power of the whole tube, the efficiency of the whole tube, the gain, the frequency and the bandwidth.

[0018] In summary, the present invention first establishes prediction models for each component of the traveling wave tube according to the design parameters; then, with the injection-wave interaction of the traveling wave tube as the core, the relevant performance parameters of the prediction models for each component of the traveling wave tube are passed to the injection-wave interaction solver, and the calculation of the performance indicators such as the output power, electronic efficiency, and gain of the traveling wave tube is completed by the injection-wave interaction solver, and the efficiency performance indicators of the entire traveling wave tube are calculated by the collector prediction model, and the prediction model of the entire traveling wave tube is established in the form of data flow; finally, by establishing a traveling wave tube full-parameter data link, the performance prediction of each component of the traveling wave tube and the performance prediction of the entire tube are completed using an agent model based on the traveling wave tube full-parameter data link. The present invention allows the design of the traveling wave tube to have higher flexibility, and further iterates and updates the design and development technology of the traveling wave tube. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic flow chart of the method proposed by the present invention;

[0020] Figure 2 It is the overall architecture diagram of the method proposed by the present invention;

[0021] Figure 3 It is the data flow diagram of the whole traveling wave tube of the method proposed by the present invention. DETAILED DESCRIPTION

[0022] In order to facilitate those skilled in the art to understand the technical content of the present invention, the present invention is described in detail below by taking a helical traveling wave tube as an example with reference to the accompanying drawings.

[0023] A method for quickly predicting the performance of a traveling wave tube, the process is as follows: Figure 1 As shown, it mainly includes four parts: Step 1 design indicators to determine structural parameters, Step 2 establishes a traveling wave tube component prediction model according to the design parameters, Step 3 establishes a traveling wave tube whole tube prediction model TWTATPM based on a full parameter data link, and Step 4 predicts the performance of the traveling wave tube whole tube.

[0024] Step1. Design indicators determine design parameters;

[0025] The design indicators of the helical traveling wave tube in this embodiment include: operating frequency F0, bandwidth W0, output power P out , saturation gain G, total efficiency η all The design parameters of each traveling wave tube component, including the electron gun, high frequency structure, magnetic system, energy transmission window, and collector, are obtained using the traveling wave tube forward design tool.

[0026] Step 2. Establish a prediction model for traveling wave tube components based on design parameters;

[0027] According to the design parameters of the traveling wave tube components obtained in Step 1, the prediction model tool is used to establish the prediction model of each component of the traveling wave tube respectively; according to the organizational structure of the traveling wave tube, the prediction model of the traveling wave tube components in Step 2 includes the electron gun prediction model, the high-frequency structure prediction model, the magnetic system prediction model, the energy transmission window prediction model and the collector prediction model.

[0028] The input of the electron gun prediction model is determined by the structural position of the cathode and anode, including the cathode curvature radius, the anode position and the voltage at each level. The input of the helical traveling wave tube electron gun prediction model in this embodiment is [Z c , R c , r c , Zg, U gun ]@guni electron gun model design parameters, where Z c , R c 、r c etc. affect the cathode position, Z g Affects the anode position, U gun is the voltage value; the output of the electron gun prediction model is the total cathode emission current, the cathode back-bombardment current, the interception current, the injection waist radius, and the injection waist position, that is, [I allc , I rc , I gc , R b , Z m ]@guno.

[0029] The input of the high-frequency structure prediction model includes the solution frequency, the inner radius of the helix, and the pitch. The input of the high-frequency structure prediction model of the helix traveling wave tube in this embodiment is the four design parameters of the solution frequency, the inner radius of the helix, the pitch, and the shell radius, namely [Fre, R, pitch, Ra]@hfcsi. Regarding the solution frequency Fre@hfcsi, it should at least set the corresponding value near the design indicator center frequency F0 and the design indicator bandwidth W0, and should at least include the three values ​​of the center frequency, the design indicator frequency lower limit F0-0.5*W0, and the design indicator frequency upper limit F0+0.5*W0. The output of the high-frequency structure prediction model of the helix traveling wave tube in this embodiment is the normalized phase velocity, coupling impedance, and attenuation constant, namely [V pc , K c ,α]@hfcso.

[0030] The input of the magnetic system prediction model includes the structural parameters of the magnetic steel, the structural parameters of the pole shoe, and the material. The input of the magnetic system prediction model of the helical traveling wave tube in this embodiment is [s, b, m]@mfsi. The output of the magnetic system prediction model is the magnetic system performance parameters such as the magnetic field period and the peak value of the magnetic field on the axis. The output of the magnetic system prediction model of the helical traveling wave tube in this embodiment is [B1, B2, ..., B i , ..., B n]@mfso, where B i represents the magnetic field amplitude on the magnetic field axis, and n represents the number of cycles of the selected magnetic system.

[0031] The input of the energy transmission window prediction model includes the solution frequency and the energy transmission window structure design parameters. The input of the energy transmission window prediction model of the helical traveling wave tube in this embodiment is [a, b, R t ]@wsi]; The output of the energy transmission window prediction model is the S parameter. The output of the energy transmission window prediction model of the helical traveling wave tube in this embodiment is [S 11 , S 21 ]@wso.

[0032] The input of the collector prediction model includes the collector inlet energy, the collector structure parameters and the collector electrode voltages. The output of the collector prediction model is the collector efficiency, the total efficiency and the reflux rate. The input of the helical traveling wave tube collector prediction model in this embodiment is [R, U1, U2, U3, U4]@mdci, and the output of the collector prediction model is [η mdc , η all , η rc ]@mdco.

[0033] Step 3. Establish the TWTATPM prediction model based on the full parameter data link;

[0034] like Figure 2 The figure shows the architecture of the whole-tube prediction model of the traveling wave tube that needs to be established in Step 3. Taking the injection-wave interaction of the traveling wave tube as the core, the relevant performance parameters [y@guno, y@wso, y@hfcso, y@mfso] of the electron gun prediction model, input window prediction model, high-frequency structure prediction model, and magnetic system prediction model in Step 2 are passed to the injection-wave interaction solver. The performance indicators such as the output power, electronic efficiency, and gain of the traveling wave tube are calculated by the injection-wave interaction solver.

[0035] Then the electron energy distribution calculated by the injection-wave interaction solution is used as input and passed to the collector prediction model in Step 2. The collector efficiency η is obtained by performing the prediction function using the collector prediction model in Step 2. mdc , and total efficiency η all , reflux rate η rc , the traveling wave tube whole tube prediction model TWTATPM is established through data stream.

[0036] Step 4. Predict the performance of the entire traveling wave tube;

[0037] like Figure 3The figure shows the data flow of the whole-tube prediction model of the traveling wave tube. The input data [x@i] of the electron gun prediction model, input window prediction model, high-frequency structure prediction model, magnetic system prediction model, input window prediction model, collector prediction model of the traveling wave tube components established in Step 2 are input into the whole-tube prediction model TWTATPM established in Step 3. The whole-tube prediction model TWTATPM established in Step 3 can predict the whole-tube performance of the traveling wave tube and predict the output power P of the whole tube. out 、The whole tube efficiency η all , gain G, frequency F0, bandwidth W0.

[0038] It can be seen from the above embodiments that the present invention first establishes prediction models for each component of the traveling wave tube according to the design parameters; then, with the injection-wave interaction of the traveling wave tube as the core, the relevant performance parameters of the prediction models for each component of the traveling wave tube (electron gun prediction model, input window prediction model, high-frequency structure prediction model, magnetic system prediction model) are passed to the injection-wave interaction solver, and the calculation of the performance indicators such as the output power, electronic efficiency, and gain of the traveling wave tube is completed by the injection-wave interaction solver. The efficiency performance index of the entire traveling wave tube is calculated by the collector prediction model, and the prediction model of the entire traveling wave tube is established in the form of a data stream. The present invention can establish a data link for transmitting all parameters of the traveling wave tube, and use a proxy model based on the full-parameter data link to complete the performance prediction of each component of the traveling wave tube and the performance prediction of the entire tube. The present invention allows the design of the traveling wave tube to have higher flexibility, and further iterates and updates the design and development technology of the traveling wave tube.

Claims

1. A method for quickly predicting the performance of a traveling wave tube, characterized in that: The following steps are involved: Step1. Design indicators determine design parameters; According to the design index requirements of the traveling wave tube, the design parameters of the electron gun, high-frequency structure, magnetic system, energy transmission window, and collector traveling wave tube components are first determined through the traveling wave tube forward design tool; Step 2. Establish a prediction model for traveling wave tube components based on design parameters; According to the design parameters of the traveling wave tube components determined in Step 1, prediction models of the various components of the traveling wave tube are established using prediction tools; according to the organizational structure of the traveling wave tube, the prediction models of the various components of the traveling wave tube include five prediction models, namely, the electron gun prediction model, the high-frequency structure prediction model, the energy transmission window prediction model, the magnetic system prediction model and the collector prediction model; Step 3. Establish the TWTATPM prediction model based on the full parameter data link; According to the complete process of electron emission, shaping, focusing, transmission, energy exchange and collection in the traveling wave tube, with the injection-wave interaction of the traveling wave tube as the core, the prediction models of various components of the traveling wave tube are connected through the data flow in the whole traveling wave tube, and the whole traveling wave tube prediction model TWTATPM based on the full parameter data link is established; Step 4. Predict the performance of the entire traveling wave tube; Input data into the TWT whole tube prediction model TWTATPM constructed in Step 3, and use the TWT whole tube prediction model TWTATPM to predict the performance of the TWT whole tube.

2. The method for rapidly predicting the performance of a traveling wave tube as claimed in claim 1, characterized in that: In the Step 2, x represents the design parameter, y represents the performance parameter; i represents the input, and o represents the output, then the data link format of the electron gun prediction model is [x@guni, y@guno]; the data link format of the high-frequency structure prediction model is [x@hfcsi, y@hfcso]; the data link format of the energy transmission window prediction model is [x@wsi, y@wso]; the data link format of the magnetic system prediction model is [x@mfsi, y@mfso]; the data link format of the collector prediction model is [x@mdci, y@mdco].

3. The method for rapidly predicting the performance of a traveling wave tube as claimed in claim 1, characterized in that: The prediction tools in Step 2 include neural networks, Kriging surrogate models and autoencoders.

4. The method for rapidly predicting the performance of a traveling wave tube as claimed in claim 1, characterized in that: The Step 3 also includes that the data format of the injection-wave interaction prediction model is [x@bwii, y@bwio], and the data format of the established whole-tube prediction model TWTATPM is [x@TWTATPMi, y@TWTATPMo].

5. The method for rapidly predicting the performance of a traveling wave tube as claimed in claim 1, characterized in that: The predictable performance of the traveling wave tube as a whole in Step 4 includes the output power of the whole tube, the efficiency of the whole tube, the gain, the frequency and the bandwidth.

Images