CN113221432A - Artificial intelligence-based dynamic prediction method for service life of grid electrode of ion thruster - Google Patents
Artificial intelligence-based dynamic prediction method for service life of grid electrode of ion thruster Download PDFInfo
- Publication number
- CN113221432A CN113221432A CN202110551028.4A CN202110551028A CN113221432A CN 113221432 A CN113221432 A CN 113221432A CN 202110551028 A CN202110551028 A CN 202110551028A CN 113221432 A CN113221432 A CN 113221432A
- Authority
- CN
- China
- Prior art keywords
- grid
- ion
- life
- artificial intelligence
- corrosion rate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 17
- 238000004088 simulation Methods 0.000 claims abstract description 40
- 238000005260 corrosion Methods 0.000 claims abstract description 32
- 230000007797 corrosion Effects 0.000 claims abstract description 32
- 238000013507 mapping Methods 0.000 claims abstract description 26
- 238000013528 artificial neural network Methods 0.000 claims abstract description 24
- 238000004364 calculation method Methods 0.000 claims abstract description 15
- 150000002500 ions Chemical class 0.000 claims description 51
- 230000001133 acceleration Effects 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 14
- 238000011144 upstream manufacturing Methods 0.000 claims description 9
- 230000007935 neutral effect Effects 0.000 claims description 7
- 238000005530 etching Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 5
- 238000003062 neural network model Methods 0.000 claims description 4
- 238000012805 post-processing Methods 0.000 claims description 3
- 102000002029 Claudin Human genes 0.000 claims description 2
- 108050009302 Claudin Proteins 0.000 claims description 2
- 238000005299 abrasion Methods 0.000 claims description 2
- 150000001875 compounds Chemical class 0.000 claims description 2
- 238000010992 reflux Methods 0.000 claims description 2
- 238000004544 sputter deposition Methods 0.000 claims description 2
- 238000002834 transmittance Methods 0.000 claims description 2
- 238000013461 design Methods 0.000 description 4
- 238000012937 correction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000003628 erosive effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000010884 ion-beam technique Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/25—Design optimisation, verification or simulation using particle-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention discloses an artificial intelligence based dynamic prediction method for the grid life of an ion thruster. Firstly, a grid three-dimensional numerical simulation model is utilized to perform simulation analysis on a plurality of typical working conditions of an ion thruster grid system, then databases under different parameter conditions are established according to simulation results, and finally the databases are utilized to train a neural network to form mapping of grid multi-parameters and grid wall surface corrosion rate. The service life change of the grid can be estimated in real time according to the working condition parameter change of the grid by utilizing the mapping. Compared with the existing method for estimating the service life of the grid based on single working condition, the method is more suitable for actual conditions, is more reliable and reasonable, and solves the problem of overlarge simulation calculation amount of multiple working conditions and multiple parameters. Finally, the service life prediction requirement of spacecrafts such as satellites and the like on the variable thrust of the ion thruster in a wide range can be met.
Description
Technical Field
The invention belongs to the technical field of ion thrusters, and relates to an artificial intelligence-based dynamic prediction method for the grid life of an ion thruster.
Background
In order to meet the requirement of a satellite on wide-range variable thrust, the grid system of the ion thruster needs to work under the condition of multi-mode real-time switching during actual operation. At this time, the gate system parameters, such as upstream plasma density, aperture, voltage, etc., also change in real time. Therefore, the method for predicting the service life of the grid electrode by using the traditional fixed working condition simulation has a large error compared with the actual working condition. On the other hand, the real-time dynamic change of parameters caused by the gate multi-mode switching poses a great challenge to the numerical simulation of a gate system.
The operation states of the grid can be roughly classified into an over-focus state, a normal focus state and an under-focus state according to whether the acceleration grid has a direct current interception state or not. The simulation grid required for each operating state is very different, typically increasing in multiples with the upstream plasma density. Therefore, if the actual multi-mode operation process of the grid is directly simulated by a single example, the method cannot be realized by the conventional computer capability. For the problems, the current common idea is iterative correction, namely, the grid operation is calculated to a steady state based on a group of initial parameters, then grid parameters are corrected according to the steady state result of the initial parameters, and then simulation calculation is continued under the corrected parameters until the total iteration time is equal to the actual grid operation time. However, this concept still satisfies the problem of excessive calculation. Although a single example in this method meets the computer computing power requirements, the amount of computation to complete the entire iteration is too large.
Therefore, on the basis of iterative calculation of partial representative working conditions, direct calculation of the grid representative working conditions can be considered, and then the functional relation between different grid parameters and the service life of the grid is fitted based on the representative working condition results. However, the gate parameters and the gate lifetime have a highly non-linear relationship, and the gate parameters are numerous, and it is difficult to find a suitable fitting coefficient by using a conventional linear fitting or polynomial fitting. Meanwhile, an artificial intelligence method which is developed at present is a means which is particularly suitable for searching multi-parameter and nonlinear functions.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and solve the problem of dynamic prediction of the grid service life of the ion thruster. The invention provides a grid life dynamic estimation model established by combining a grid simulation model and an artificial intelligence method. The specific idea is as follows: firstly, an existing grid simulation model is utilized to carry out simulation on representative working conditions, and a simulation result of the representative working conditions is obtained. Then, a database is formed by using the simulation result of all the representative working conditions, and the neural network is trained by using the database to obtain the functional relation between the grid electrode parameters and the grid electrode corrosion rate (namely the grid electrode service life). And finally, predicting the service life of the grid under any working condition combination by using the trained neural network.
The technical scheme of the invention is as follows: an artificial intelligence based dynamic prediction method for the service life of an ion thruster grid,
first, a gate simulation program based on a macro-particle-Monte Carlo collision (PIC-MCC) algorithm is established. And secondly, establishing a post-processing program for solving the grid corrosion rate, and solving the corrosion rate distribution and the average corrosion rate of the end surface and the hole wall of the grid. Thirdly, forming a database by each typical working condition parameter, each failure limit parameter and the corresponding grid wall surface corrosion rate. Then, a neural network model based on a BP neural network algorithm is established, the neural network is trained by utilizing the established database, and the trained neural network can establish the sub-working condition multi-parameter mapping. Then, aiming at the single-mode working grid, input parameters at the final moment can be obtained by repeatedly calling mapping and continuously correcting the input parameters of the PIC-MCC model, and then the grid life end performance parameters are obtained through simulation calculation and the grid life is given. And finally, in the multi-mode wear etching estimation, iteration is carried out by adopting different mappings through different modes to realize mode switching, and meanwhile, the final service life of the grid is obtained by adding a grid failure criterion in the whole iteration process.
The method is characterized by comprising the following steps:
step 1, establishing a grid simulation model based on a macro particle-Monte Carlo collision algorithm;
step 3, forming a database by using each typical working condition parameter, each failure limit parameter and the corresponding grid wall surface corrosion rate;
step 4, establishing a neural network model based on a BP neural network algorithm, training a neural network by using the established database in the step 3, and establishing working condition-based multi-parameter mapping of the trained neural network;
step 5, aiming at the single-mode working grid, continuously correcting input parameters based on the macro particle-Monte Carlo collision model by repeatedly calling mapping to obtain input parameters of the final moment, and then performing simulation calculation to obtain performance parameters at the end of the grid service life and simultaneously give the grid service life;
and 6, in the multi-mode wear etching estimation, iteration is carried out by adopting different mappings through different modes to realize mode switching, and meanwhile, the final service life of the grid is obtained by adding a grid failure criterion in the whole iteration process.
Compared with the prior art, the invention has the beneficial effects that:
in the algorithm, on the basis of iterative computation of partial representative working conditions, a nonlinear function between grid multi-parameters and grid wall surface corrosion rate is obtained by adopting a time-rising artificial intelligence method. The method solves the problem of huge simulation calculation amount of multi-mode, multi-working condition and multi-parameter of the grid, greatly improves the accuracy of multi-mode grid service life estimation, and can estimate the grid service life at any time, under any working condition or under any working condition combination in real time. After evaluation and verification, the simulation method considers that the algorithm can effectively and dynamically predict the service life of the grid electrode of the ion thruster in the radial direction, and after comparison with an experimental result, the calculation precision is good, and the simulation method has important instructive significance for verifying the performance of the grid electrode of the thruster at the initial stage of design, optimizing parameters of the thruster at the later stage and adjusting the direction and the design range of the experimental parameters. The design method is suitable for early design and later experimental guidance of the ion thruster grid system.
Drawings
Figure 1 is a flow chart of the PIC-MCC algorithm.
Fig. 2 is a BP neural network algorithm flow.
Fig. 3 is a schematic diagram of finding a mapping f using a neural network.
Fig. 4 is a schematic diagram of an iteration using a mapping f.
FIG. 5 is a schematic diagram of a mode-switching simulation scheme.
FIG. 6 is a schematic illustration of adding a failure criterion during an iteration process.
Detailed Description
The invention relates to an artificial intelligence-based dynamic prediction method for the grid life of an ion thruster. The main objective of the invention is to form a set of simulation tools capable of dynamically predicting the grid service life of the multi-mode ion thruster aiming at the grid abrasion service life simulation analysis of the multi-mode ion thruster. In the actual working process of the grid, the upstream plasma density, the neutral atom density, the grid voltage, the aperture, the hole spacing, the grid thickness and the grid spacing are all in dynamic changes. And in each operating mode, the grid electrode can be in completely different working conditions. From the view point of numerical simulation, the essence of the method is a feedback correction process, namely, the grid input parameters are checked through feedback correction so as to accurately reflect the grid running state in real time. Therefore, the idea of the invention is to iteratively correct the gate wear estimation model based on the dynamic feedback of the neural network algorithm. In order to obtain a grid corrosion rate which can reflect any parameter, a neural network needs to be trained. Therefore, firstly, a few samples are obtained by utilizing the simulation of a grid program, and a database is built; then, importing the database into a neural network system, and training a neural network; and finally, the trained neural network is the mapping. For the single mode situation, a few samples can be generated by using numerical simulation, and then the mapping is found by training a neural network; and then, continuously correcting the input parameters by repeatedly calling mapping to obtain the input parameters at the final moment, and then carrying out simulation calculation. For multi-mode switching, the mapping of different modes is obtained first, and then iteration is performed by adopting different mappings for different modes. For the life prediction dynamic model, the problem can be solved by adding a grid failure criterion in the whole iteration process.
The implementation steps are as follows:
(1) first, as shown in fig. 1, a gate simulation model based on a macro-particle-monte carlo collision (PIC-MCC) algorithm is established. This set of models consists of two parts: a model simulating the movement of the ion beam current in the optical system, a model simulating the generation of charge-exchanged (CEX) ions. The first part of the program is used for simulating the extraction process of the beam ions, the collision among the ions is little and can not be considered, and the simulation is finished when the total number of the ions is stable. In the simulation process, the ratio of divalent ions to monovalent ions is calculated according to the following formula:
in the formula (I), the compound is shown in the specification,
divalent ion and monovalent ion densities, respectively; i isd,IbRespectively discharge current and beam current; v. of0,viAtomic and ionic velocities, respectively; t isa,ηcRespectively the grid transmittance and the claudin coefficient of the atoms; sigma++,σ+Respectively a divalent ion ionization collision section and a monovalent ion ionization collision section; gamma is the ionization rate. In addition, the neutral atoms are not charged, and thus the distribution of the neutral atoms does not affect the electric field in the simulation region. This embodiment assumes that neutral atoms are uniformly distributed in the calculation region. The second partial model was used to simulate the CEX ion distribution. Since the amount of CEX ions is a small amount relative to the beam ions, it is believed that the CEX ion density has no effect on the overall potential distribution. The generation and movement of charge-exchanged ions in the simulation region are simulated by utilizing beam ion distribution and neutral atom distribution. In the hundreds of hours of initial operation of the gate, it is believed that the amount of accelerated gate wall and hole wall erosion is a small amount,the corrosion process has no influence on the overall potential distribution. The corrosion depth calculation model mainly utilizes the relevant physical quantities of beam ions and CEX ions to calculate the corrosion size of the hole wall and the downstream surface of the accelerating grid.
(2) And establishing a post-processing model for solving the grid corrosion rate, and solving the corrosion rate distribution and the average corrosion rate of the end surface and the hole wall of the grid. The definition of the corrosion rate obtains the expression:
in the formula REAs the etching rate, J is the ion current density of the gate wall surface, Y is the ion sputtering yield, q is the ion charge amount, MgIs the atomic mass of the gate material, pgIs the gate material density. From the expression of the corrosion rate, it is found that: obtaining the ion current density on the surface of the grid by using the statistics of a PIC-MCC grid simulation model, and obtaining the R of the wall surface of the gridEThe rate of corrosion.
(3) The typical working condition and the corresponding working parameters of the comb grid multi-mode mainly comprise: geometric parameters such as acceleration gate thickness, acceleration gate aperture, deceleration gate thickness, deceleration gate aperture, screen-acceleration spacing, acceleration-deceleration spacing; electrical parameters such as screen gate voltage, acceleration gate voltage, deceleration gate voltage; plasma parameters such as upstream plasma density, neutral atom density, anode voltage (which can be characterized as a divalent ion ratio). Meanwhile, the estimated grid failure limit parameter, for example, the accelerated grid electron reflux failure limit radius, can be calculated according to the following formula:
in the formula, VNThe net acceleration voltage is the sum of the screen grid voltage and the plasma relative screen grid potential in the ionization chamber; vTIs the total acceleration voltage, which is the difference between the net acceleration voltage and the acceleration gate voltage; lg，rs，taAnd raThe gate pitch, the screen gate hole radius, the accelerated gate thickness, and the accelerated gate hole radius, respectively.
And then, simulating each typical working condition parameter and each failure limit parameter by using a PIC-MCC grid simulation model to obtain the grid wall surface corrosion rate under each typical working condition parameter and each failure limit parameter. Finally, a database is formed by the typical working condition parameters, the failure limit parameters and the corresponding grid wall surface corrosion rate.
(4) As shown in fig. 2, a neural network model based on the BP neural network algorithm is established. The neural network is then trained using the established database, the training process being as shown in fig. 3. The trained neural network can establish the split-working-condition multi-parameter mapping. For example, the map of the accelerating grid aperture, upstream plasma density, atomic density, screen grid voltage and erosion rate can be expressed as:
f(ra,n0,nn,Us)→Re (4)
in the formula, ra,n0,nn,Us,ReRespectively, the accelerating grid aperture, upstream plasma density, atomic density, screen grid voltage, and etch rate.
(5) For the single-mode working grid, the input parameters of the final moment can be obtained by repeatedly calling the mapping f and continuously correcting the input parameters of the PIC-MCC model, and then the grid life end performance parameters are obtained through simulation calculation and the grid life is given. Fig. 4 shows a process for iteratively correcting the predicted end-of-gate-life performance parameters every 100 hours.
(6) For the multi-mode wear etch estimation, the essence of the mode switching is the switching of the mapping f, so a simulation scheme as shown in fig. 5 is adopted for the mode switching, that is, the mapping of different modes is obtained first, and then the iteration is performed by adopting different mappings for different modes.
For the life prediction dynamic model, the final simulation tool required by the part of contents can be replaced by using a PIC-MCC program for simulation calculation, so that the life of the grid can be rapidly predicted by giving any group of mode combinations. This part of the content can be solved by adding a gate failure criterion throughout the iteration process, as shown in fig. 6.
Claims (8)
1. A dynamic prediction method for the service life of an ion thruster grid based on artificial intelligence is characterized by comprising the following steps: the method comprises the following steps:
step 1, establishing a grid simulation model based on a macro particle-Monte Carlo collision algorithm;
step 2, establishing a post-processing model for solving the grid corrosion rate, and solving the corrosion rate distribution and the average corrosion rate of the end surface and the hole wall of the grid;
step 3, forming a database by using each typical working condition parameter, each failure limit parameter and the corresponding grid wall surface corrosion rate;
step 4, establishing a neural network model based on a BP neural network algorithm, training a neural network by using the established database in the step 3, and establishing working condition-based multi-parameter mapping of the trained neural network;
step 5, aiming at the single-mode working grid, continuously correcting input parameters based on the macro particle-Monte Carlo collision model by repeatedly calling mapping to obtain input parameters of the final moment, and then performing simulation calculation to obtain performance parameters at the end of the grid service life and simultaneously give the grid service life;
and 6, in the multi-mode wear etching estimation, iteration is carried out by adopting different mappings through different modes to realize mode switching, and meanwhile, the final service life of the grid is obtained by adding a grid failure criterion in the whole iteration process.
2. The method for dynamically predicting the grid life of the ion thruster based on the artificial intelligence of claim 1, wherein in the step 1, a grid simulation model based on a macro-particle-monte carlo collision algorithm is established, wherein the ratio of the divalent ions to the monovalent ions is calculated according to the formula (1):
in the formula (I), the compound is shown in the specification,
divalent ion and monovalent ion densities, respectively; i isd,IbRespectively discharge current and beam current; v. of0,viAtomic and ionic velocities, respectively; t isa,ηcRespectively the grid transmittance and the claudin coefficient of the atoms; sigma++,σ+Respectively a divalent ion ionization collision section and a monovalent ion ionization collision section; gamma is the ionization rate.
3. The method for dynamically predicting the grid life of the ion thruster based on the artificial intelligence of claim 1, wherein in the step 2, a post-processing program for solving the grid corrosion rate is established, and the corrosion rate distribution and the average corrosion rate of the grid end surface and the hole wall are solved, wherein the grid corrosion rate is calculated by the formula (2):
in the formula REAs the etching rate, J is the ion current density of the gate wall surface, Y is the ion sputtering yield, q is the ion charge amount, MgIs the atomic mass of the gate material, pgIs the gate material density.
4. The method for dynamically predicting the grid life of the artificial intelligence-based ion thruster according to claim 1, wherein in step 3, the typical working conditions and corresponding working parameters comprise: acceleration grid thickness, acceleration grid aperture, deceleration grid thickness, deceleration grid aperture, screen grid-acceleration spacing, acceleration-deceleration spacing, screen grid voltage, acceleration grid voltage, deceleration grid voltage, upstream plasma density, neutral atom density, anode voltage.
5. The method for dynamically predicting the grid life of the ion thruster based on the artificial intelligence of claim 1, wherein in the step 3, grid failure limit parameters are estimated, and the accelerated grid electron reflux failure limit radius is calculated according to the following formula:
in the formula, VNThe net acceleration voltage is the sum of the screen grid voltage and the plasma relative screen grid potential in the ionization chamber; vTIs the total acceleration voltage, which is the difference between the net acceleration voltage and the acceleration gate voltage; lg，rs，taAnd raThe gate pitch, the screen gate hole radius, the accelerated gate thickness, and the accelerated gate hole radius, respectively.
6. The method for dynamically predicting the grid life of the ion thruster based on the artificial intelligence of claim 1, wherein in the step 4, the trained neural network establishes a split-working-condition multi-parameter mapping, which comprises a mapping table of an accelerated grid aperture, an upstream plasma density, an atomic density, a screen grid voltage and a corrosion rate as follows:
f(ra,n0,nn,Us)→Re (4)
in the formula, ra,n0,nn,Us,ReRespectively, the accelerating grid aperture, upstream plasma density, atomic density, screen grid voltage, and etch rate.
7. The method for dynamically predicting the grid life of the ion thruster based on the artificial intelligence of claim 1, wherein the step 5 is to iteratively correct the procedure of predicting the performance parameters at the end of the grid life every 100 hours.
8. The method for dynamically predicting the grid life of the ion thruster based on the artificial intelligence of claim 1, wherein in the step 6, in the estimation of the multi-mode abrasion etching, iteration is performed by adopting different mappings through different modes to realize mode switching, and meanwhile, the final grid life is obtained by adding a grid failure criterion in the whole iteration process, wherein the failure criterion is calculated by a formula (3).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110551028.4A CN113221432A (en) | 2021-05-20 | 2021-05-20 | Artificial intelligence-based dynamic prediction method for service life of grid electrode of ion thruster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110551028.4A CN113221432A (en) | 2021-05-20 | 2021-05-20 | Artificial intelligence-based dynamic prediction method for service life of grid electrode of ion thruster |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113221432A true CN113221432A (en) | 2021-08-06 |
Family
ID=77093278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110551028.4A Pending CN113221432A (en) | 2021-05-20 | 2021-05-20 | Artificial intelligence-based dynamic prediction method for service life of grid electrode of ion thruster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113221432A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114329919A (en) * | 2021-12-09 | 2022-04-12 | 兰州空间技术物理研究所 | Simulation analysis method for sputter etching of screen grid of ion thruster |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190164046A1 (en) * | 2017-11-29 | 2019-05-30 | Anaflash Inc. | Neural network circuits having non-volatile synapse arrays |
| CN111044822A (en) * | 2019-12-25 | 2020-04-21 | 兰州空间技术物理研究所 | Simulation method for discharging process of discharge chamber of Kaufman ion thruster |
-
2021
- 2021-05-20 CN CN202110551028.4A patent/CN113221432A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190164046A1 (en) * | 2017-11-29 | 2019-05-30 | Anaflash Inc. | Neural network circuits having non-volatile synapse arrays |
| CN111044822A (en) * | 2019-12-25 | 2020-04-21 | 兰州空间技术物理研究所 | Simulation method for discharging process of discharge chamber of Kaufman ion thruster |
Non-Patent Citations (4)
| Title |
|---|
| CHANG LU等: "Barrel Erosion of Ion Thruster Accelerator Grid Under Different Operating Conditions", IEEE TRANSACTIONS ON PLASMA SCIENCE, vol. 46, no. 12, 31 December 2018 (2018-12-31), pages 4062 - 4077, XP011696670, DOI: 10.1109/TPS.2018.2865600 * |
| 权路路等: "霍尔推力器预电离对低频振荡及壁面腐蚀影响的研究", 中国空间科学技术, no. 01, 25 February 2016 (2016-02-25), pages 51 - 57 * |
| 林兴发: "两阶段离子推进器栅极系统性能及寿命的研究", 中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑, 15 February 2020 (2020-02-15), pages 031 - 1614 * |
| 邹心遥等: "基于BP神经网络的电子元器件寿命预测", 微电子学与计算机, vol. 26, no. 1, 31 January 2009 (2009-01-31), pages 52 - 54 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114329919A (en) * | 2021-12-09 | 2022-04-12 | 兰州空间技术物理研究所 | Simulation analysis method for sputter etching of screen grid of ion thruster |
| CN114329919B (en) * | 2021-12-09 | 2023-05-23 | 兰州空间技术物理研究所 | Simulation analysis method for sputter etching of screen grid of ion thruster |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108520325B (en) | Integrated life prediction method based on accelerated degradation data in variable environment | |
| US7847247B2 (en) | Method of plasma particle simulation, storage medium, plasma particle simulator and plasma processing apparatus | |
| CN109214708B (en) | Electric power system risk assessment method based on cross entropy theory optimization support vector machine | |
| CN111199099A (en) | Method for evaluating operation life of ion thruster based on grid corrosion | |
| CN109165400B (en) | Design method of focused ion beam etching processing technological parameters | |
| CN112884237A (en) | Power distribution network prediction auxiliary state estimation method and system | |
| Wirz et al. | Time-dependent erosion of ion optics | |
| CN111948563B (en) | Electric forklift lithium battery residual life prediction method based on multi-neural network coupling | |
| CN113221432A (en) | Artificial intelligence-based dynamic prediction method for service life of grid electrode of ion thruster | |
| CN109066651B (en) | Method for calculating limit transmission power of wind power-load scene | |
| CN111950205A (en) | Lithium battery SOH prediction method based on FWA optimization extreme learning machine | |
| CN114004133B (en) | Unequal weight macro particle correction method of electron impact ionization Monte Carlo model | |
| CN111105005B (en) | Wind power prediction method | |
| CN110851911A (en) | Terminal state calculation model training method, control sequence searching method and device | |
| CN111160772A (en) | Large power grid risk rapid assessment method | |
| CN114186709A (en) | Energy prediction method for optimizing key parameters of gray model based on emperor butterfly algorithm | |
| CN114094599A (en) | Multi-station fusion participating peak-shaving frequency-modulation voltage-regulation potential evaluation method and device | |
| CN112114254B (en) | Power battery open-circuit voltage model fusion method | |
| Zhou et al. | Online estimation of state of charge of Li-ion battery using an iterated extended Kalman particle filter | |
| CN111044822B (en) | Simulation method for discharging process of discharge chamber of Kaufman ion thruster | |
| CN110210072B (en) | Method for solving high-dimensional optimization problem based on approximate model and differential evolution algorithm | |
| CN110675276B (en) | Method and system for inversion droop control of direct current power transmission system | |
| CN112347610A (en) | Hall thruster life evaluation method based on hollow cathode | |
| Lim et al. | Evolutionary optimization with dynamic fidelity computational models | |
| Yasin et al. | Optimal sizing of distributed generation by using quantum-inspired evolutionary programming |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |