Hall thruster oscillation control method and system based on neural network

Abstract

The application provides a Hall thruster oscillation control method and system based on a neural network, and relates to the field of Hall thrusters. The method comprises the following steps: in response to the operation of a Hall thruster, taking the initial power of the Hall thruster operation as a performance constraint condition, and collecting the discharge oscillation current waveform of the Hall thruster operation in real time; based on the discharge oscillation current waveform, performing calculation to obtain oscillation characteristic parameters corresponding to the discharge oscillation current waveform; according to the oscillation characteristic parameters, judging the current oscillation state of the Hall thruster through an oscillation characteristic mapping neural network model and generating a target operating condition for the Hall thruster; according to the current oscillation state, the target operating condition and the performance constraint condition, generating an oscillation regulation instruction, and according to the oscillation regulation instruction, respectively controlling and adjusting the anode voltage of the anode power supply of the Hall thruster, the magnetic field current of the excitation coil power supply and the propellant flow of the flow regulation unit to perform coordinated control of oscillation suppression and performance maintenance.

Description

Technical Field

[0001] This application relates to the field of Hall thruster technology, and in particular to a Hall thruster oscillation control method and system based on neural networks. Background Technology

[0002] Currently, Hall thrusters are widely used in satellite constellation deployments and microsatellite missions due to their advantages such as high specific impulse, large thrust-to-power ratio, and reliable structure. However, complex low-frequency current oscillations are still prevalent during the discharge process of Hall thrusters. These oscillations can reduce discharge stability, affect propulsion performance, and pose potential risks to the power system and overall satellite operation. Therefore, it is of great significance to construct a comprehensive Hall thruster discharge oscillation model, establish an oscillation characteristic database covering multiple operating conditions, and conduct research on oscillation control methods.

[0003] Because low-frequency oscillations in Hall thrusters significantly impact the thruster's operational stability and the overall satellite power supply, numerous studies have been conducted on control methods for these oscillations. Based on the control methods, these can be broadly categorized into passive feedback control of electrical operating parameters and active intervention control. The controlled object is typically a single-component low-frequency plasma oscillation. Existing oscillation control methods largely rely on empirical trial and error, lacking adaptive adjustment capabilities to address the complex oscillations within Hall thrusters, resulting in low operational stability. Summary of the Invention

[0004] The purpose of this invention is to provide a Hall thruster oscillation control method and system based on neural networks, so as to solve the technical problem that the oscillation control method in the prior art leads to low working stability of the Hall thruster.

[0005] In a first aspect, this application provides a Hall thruster oscillation control method based on a neural network, applied to a low-frequency discharge oscillation control system for a Hall thruster, the method comprising:

[0006] In response to the operation of the Hall thruster, the initial power of the Hall thruster is used as a performance constraint, and the discharge oscillation current waveform of the Hall thruster is collected in real time.

[0007] Based on the discharge oscillation current waveform, calculations are performed to obtain the oscillation characteristic parameters corresponding to the discharge oscillation current waveform; wherein, the oscillation characteristic parameters include the oscillation frequency and the oscillation amplitude;

[0008] The current oscillation state of the Hall thruster is determined by an oscillation feature mapping neural network model based on the oscillation feature parameters, and a target operating condition for the Hall thruster is generated; wherein, the oscillation feature mapping neural network model includes a mapping relationship between the thruster operating condition and the oscillation frequency and the oscillation amplitude;

[0009] Based on the current oscillation state, the target operating condition, and the performance constraints, an oscillation control command is generated. Based on the oscillation control command, the anode voltage of the anode power supply of the Hall thruster, the magnetic field current of the excitation coil power supply, and the propellant flow of the flow regulation unit are controlled and adjusted respectively to perform coordinated control of oscillation suppression and performance maintenance.

[0010] In one possible implementation, the Hall thruster includes an anode and a cathode; the Hall thruster low-frequency discharge oscillation control system includes a time-domain current signal acquisition unit and a low-frequency oscillation feature extraction unit; the time-domain current signal acquisition unit includes a current probe and a digital oscilloscope; the current probe is disposed in the power output circuit of the anode; the digital oscilloscope is connected to the current probe through a shielded signal line; the real-time acquisition of the discharge oscillation current waveform of the Hall thruster includes:

[0011] The transient change of the discharge current of the Hall thruster is obtained by the current probe, and the time-domain waveform of the discharge current of the Hall thruster is obtained by the digital oscilloscope. The transient change of the discharge current and the time-domain waveform of the discharge current are transmitted to the low-frequency oscillation feature extraction unit in real time.

[0012] In one possible implementation, the low-frequency oscillation feature extraction unit includes a data caching unit and a fast Fourier transform (FFT) algorithm unit;

[0013] The calculation based on the discharge oscillation current waveform to obtain the oscillation characteristic parameters corresponding to the discharge oscillation current waveform includes:

[0014] Based on the original current signal corresponding to the discharge oscillation current waveform, the Fast Fourier Transform (FFT) algorithm unit performs spectral analysis on the original current signal using a low-frequency oscillation feature extraction algorithm to obtain spectral analysis results. Based on the spectral analysis results, the dominant frequency of the discharge current oscillation, the amplitude of the discharge current oscillation, and several time-frequency characteristic parameters are extracted. The low-frequency oscillation feature extraction algorithm includes the Fast Fourier Transform (FFT) algorithm.

[0015] In one possible implementation, the oscillation feature mapping neural network model includes a neural network, a model storage unit, and a parameter update unit; the step of determining the current oscillation state of the Hall thruster and generating a target operating condition for the Hall thruster based on the oscillation feature parameters through the oscillation feature mapping neural network model includes:

[0016] By running a pre-trained oscillation feature evaluation model based on historical oscillation characteristics, a mapping relationship is established between the thruster operating conditions and the oscillation frequency and the oscillation amplitude. Based on the oscillation feature parameters, the oscillation characteristics of the Hall thruster under the current operating conditions are judged in real time through the mapping relationship, and a basis is provided for optimizing the control strategy of the subsequent thruster operating conditions, thereby obtaining the oscillation state prediction result. The thruster operating conditions include at least one of anode voltage, excitation coil current, and propellant flow rate.

[0017] In one possible implementation, the Hall thruster low-frequency discharge oscillation control system further includes a control command generation and communication unit, which includes a decision processing unit, a control command generation unit, and a communication interface.

[0018] The step of generating oscillation control commands based on the current oscillation state, the target operating condition, and the performance constraints, and controlling and adjusting the anode voltage of the Hall thruster's anode power supply, the magnetic field current of the excitation coil power supply, and the propellant flow rate of the flow regulation unit according to the oscillation control commands to perform coordinated control of oscillation suppression and performance maintenance, includes:

[0019] The optimal control parameters are calculated by the decision processing unit based on the oscillation state prediction results and the performance constraints.

[0020] Based on the optimal control parameters, the control command generation unit generates control commands for the power supply voltage for the anode, the magnetic field current of the excitation coil power supply, and the propellant flow rate of the flow control unit.

[0021] The communication interface uses a standard serial bus to interact with the power supply unit and flow regulation unit corresponding to the Hall thruster, so as to perform real-time oscillation control of the Hall thruster.

[0022] In one possible implementation, the performance constraint is expressed as follows:

[0023] ;

[0024] ;

[0025] in, V , q , I These represent the voltage at the anode, the current flow at the anode, and the coil current, respectively. A This represents the amplitude of the low-frequency oscillation, which is an independent variable. V , q , I multivariate functions; α Indicates the error limit;P Represents power, which is an independent variable. V , q , I multivariate functions; To control the initial power of the Hall thruster, a parameter is used to measure power changes; st represents the power constraint condition.

[0026] In one possible implementation, the anode is fabricated from a conductive metal material and connected to a power source; the cathode is a hollow cathode structure, used to supply electrons to the Hall thruster via a power source; the Hall thruster is mechanically fixed by a bracket, and the anode, the cathode, and the excitation coil corresponding to the excitation coil power source are respectively connected to the power supply unit corresponding to the Hall thruster via wires;

[0027] The power supply unit and flow regulation unit corresponding to the Hall thruster include the power supply of the anode, the power supply of the excitation coil, the power supply of the cathode, and the anode propellant flow meter; the power supply of the anode is used to provide the working voltage, the power supply of the excitation coil is used to adjust the magnetic field strength by adjusting the current of the excitation coil, and the power supply of the cathode is used to heat the cathode and maintain electron emission;

[0028] The power supply unit corresponds to a propellant supply system, which includes a mass flow meter and a flow regulating valve, used to regulate the propellant flow under control commands to switch between different thruster operating conditions.

[0029] Secondly, this application provides a Hall thruster oscillation control system based on a neural network, applied to a low-frequency discharge oscillation control system for Hall thrusters, comprising:

[0030] The acquisition module is used to respond to the operation of the Hall thruster, take the initial power of the Hall thruster as a performance constraint, and acquire the discharge oscillation current waveform of the Hall thruster in real time.

[0031] The calculation module is used to perform calculations based on the discharge oscillation current waveform to obtain the oscillation characteristic parameters corresponding to the discharge oscillation current waveform; wherein, the oscillation characteristic parameters include the oscillation frequency and the oscillation amplitude;

[0032] The generation module is used to determine the current oscillation state of the Hall thruster based on the oscillation characteristic parameters through an oscillation feature mapping neural network model and generate a target operating condition for the Hall thruster; wherein, the oscillation feature mapping neural network model includes a mapping relationship between the thruster operating condition and the oscillation frequency and the oscillation amplitude.

[0033] The control module is used to generate oscillation control commands based on the current oscillation state, the target operating condition, and the performance constraints, and to control and adjust the anode voltage of the anode power supply, the magnetic field current of the excitation coil power supply, and the propellant flow rate of the flow regulation system according to the oscillation control commands, so as to perform coordinated control of oscillation suppression and performance maintenance.

[0034] Thirdly, this application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the method described in the first aspect above.

[0035] Fourthly, this application also provides a computer-readable storage medium storing computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method described in the first aspect above.

[0036] This application brings the following beneficial effects:

[0037] This application provides a Hall thruster oscillation control method and system based on neural networks. Responding to the operation of the Hall thruster, the method uses the initial power of the Hall thruster as a performance constraint and collects the discharge oscillation current waveform of the Hall thruster in real time. Based on the discharge oscillation current waveform, it calculates the corresponding oscillation characteristic parameters, including oscillation frequency and oscillation amplitude. According to the oscillation characteristic parameters, it uses an oscillation feature mapping neural network model to determine the current oscillation state of the Hall thruster and generates a target operating condition for the Hall thruster. The oscillation feature mapping neural network model includes a mapping relationship between the thruster operating condition and the oscillation frequency and amplitude. Based on the current oscillation state, the target operating condition, and the performance constraint, it generates oscillation control commands and controls the adjustment of the anode voltage of the anode power supply, the magnetic field current of the excitation coil power supply, and the propellant flow rate of the flow regulating unit of the Hall thruster to perform coordinated control of oscillation suppression and performance maintenance. In this solution... The control of Hall thruster oscillations is no longer achieved through a single circuit structure, but rather by directly evaluating oscillation characteristics through a neural network and then adjusting the thruster's operating conditions to control the thruster's discharge oscillations. The selection process of the control target is intelligently achieved through the neural network, moving away from the simple target selection using electrical components found in existing control technologies. Instead, it automatically adjusts the Hall thruster's operating conditions for different oscillation control targets, achieving real-time identification and control of oscillation characteristics. Furthermore, in terms of control variables, it achieves multi-parameter optimization of flow rate, voltage, and excitation coil current, involving variables across multiple time scales. Simultaneously, it introduces power performance parameters as constraints, reducing the problem of thruster malfunction or power drift caused by control oscillations, thus improving control reliability. By adopting a data-driven approach, it bypasses complex physical modeling and directly learns the mapping relationship between oscillations and operating conditions from data, achieving high robustness and adaptive real-time control. This overcomes the limitations of traditional control methods, improves the working stability of the Hall thruster, and solves the technical problem of low working stability of Hall thrusters caused by existing oscillation control methods.

[0038] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0040] Figure 1 This is a schematic diagram of a segmented anode external circuit feedback regulation system in the prior art;

[0041] Figure 2 A schematic diagram of a self-excited Hall thruster structure in the prior art;

[0042] Figure 3 This is a schematic diagram of segmented anodes and the applied active modulation signal conditioning in the prior art;

[0043] Figure 4 This is a schematic diagram of a pulse boost chopper circuit in the prior art;

[0044] Figure 5 A flowchart illustrating the Hall thruster oscillation control method based on neural networks provided in this application embodiment;

[0045] Figure 6 A front view of the Hall thruster used in the Hall thruster low-frequency discharge oscillation control system provided in the embodiments of this application;

[0046] Figure 7 Another schematic flowchart of the Hall thruster oscillation control method based on neural networks provided in the embodiments of this application;

[0047] Figure 8 A schematic diagram of a Hall thruster oscillation control system based on a neural network is provided for an embodiment of this application;

[0048] Figure 9 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0050] The terms "comprising" and "having," and any variations thereof, used in the embodiments of this application, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0051] Currently, passive feedback control is mainly achieved by connecting a resistor or coil in series in the anode working circuit. Negative feedback control can be achieved by designing segmented anodes and connecting resistors in series in the circuit, such as... Figure 1 As shown, this causes the segmented circumferential voltage of the anode to decrease proportionally with the current, ultimately significantly reducing the amplitude of the segmented circumferential current oscillation, thus verifying the effectiveness of resistance regulation in improving discharge stability and efficiency.

[0052] In existing technologies, there are also methods to achieve "self-excitation" by connecting a coil in series in the anode circuit, and to optimize the magnetic circuit by adjusting the number of turns of the electromagnetic coil, such as... Figure 2 As shown, the ionization region of the Hall thruster in self-excitation mode is more concentrated and the axial peak position is stable, which can effectively reduce the amplitude of breathing oscillation.

[0053] By actively intervening and applying a periodically modulated signal to the discharge power supply, targeted control of discharge oscillations is achieved. This is achieved by designing circumferentially segmented anodes and applying a periodic square wave voltage with a 90° phase difference to each segment to construct a rotating potential difference that controls the circumferential low-frequency oscillations. Figure 3 As shown in the figure. The study found that the ion saturation current, brightness, and electron temperature in the near-anodine region are strongly correlated with the phase of the driving voltage. The rotating potential difference can induce or suppress circumferential spoke oscillations, verifying a new approach to the active control of plasma oscillations.

[0054] In existing technologies, the operating characteristics of a Hall thruster change when two different waveform modulation signals—square wave and pulsed boost chopper—are input to the anode, as shown in Figure 4. Studies have shown that a periodic square wave signal can synchronize the anode current and voltage waveforms, thereby enabling external control of discharge oscillations and achieving stable thruster operation at an adjustable frequency, but it does not improve thrust performance. A pulsed boost chopper signal, on the other hand, can achieve stable synchronization between the anode current and the driving frequency by periodically modulating the anode voltage, effectively controlling plasma discharge oscillations. Under the same average power, pulsed boost chopper drive can improve ion acceleration efficiency, with the highest thrust efficiency improvement reaching 30% in experiments.

[0055] Although various oscillation control methods have been proposed, including magnetic field control, loop parameter optimization, and active control, most of these methods rely on empirical trial and error and lack adaptive adjustment capabilities to address the complex oscillations in Hall thrusters, resulting in low operational stability. Furthermore, the parameter tuning and robustness of traditional fixed-gain PID controllers remain significant challenges when dealing with the complex nonlinear and chaotic dynamics of Hall thrusters. In addition, current Hall thruster control parameters are relatively simple, lacking multi-parameter, cross-timescale, and multi-objective optimization control methods for discharge oscillations. Therefore, there is an urgent need to explore intelligent control strategies for low-frequency discharge oscillations in Hall thrusters.

[0056] Based on this, the present application provides a Hall thruster oscillation control method and system based on neural networks. This method can solve the technical problems such as the low working stability of Hall thrusters caused by the oscillation control methods in the prior art.

[0057] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0058] Figure 5 This is a flowchart illustrating a Hall thruster oscillation control method based on a neural network, provided as an embodiment of this application. The method is applied to a low-frequency discharge oscillation control system for a Hall thruster. Figure 5 As shown, the method includes:

[0059] In step S110, in response to the operation of the Hall thruster, the initial power of the Hall thruster is used as a performance constraint, and the discharge oscillation current waveform of the Hall thruster is acquired in real time.

[0060] As one possible implementation method, such as Figure 6 As shown, the Hall thruster includes an anode 1 and a cathode 2; the low-frequency discharge oscillation control system of the Hall thruster includes a time-domain current signal acquisition unit and a low-frequency oscillation feature extraction unit; the time-domain current signal acquisition unit includes a current probe and a digital oscilloscope; the current probe is set in the power output circuit of the anode; the digital oscilloscope is connected to the current probe through a shielded signal line, and the sampling frequency can reach several MHz; the discharge oscillation current waveform of the Hall thruster is acquired in real time, which may specifically include the following steps:

[0061] The transient changes in the discharge current of the Hall thruster are obtained by a current probe, and the time-domain waveform of the discharge current of the Hall thruster is obtained by a digital oscilloscope. The transient changes in the discharge current and the time-domain waveform of the discharge current are then transmitted in real time to the low-frequency oscillation feature extraction unit.

[0062] For example, the Hall thruster low-frequency discharge oscillation control system includes: the aforementioned time-domain current signal acquisition unit, oscillation feature extraction unit, neural network mapping unit, oscillation control command generation unit, execution control unit, and strategy optimization unit under performance constraints.

[0063] In one alternative implementation, such as Figure 6 As shown, the anode 1 is formed by processing conductive metal material and is connected to the power supply of the anode; the cathode 2 is a hollow cathode structure, which is used to provide electrons to the Hall thruster through the power supply of the cathode; the Hall thruster is mechanically fixed by the bracket 3, and the excitation coils corresponding to the power supply of the anode, cathode and excitation coil are respectively connected to the power supply unit corresponding to the Hall thruster through wires.

[0064] The power supply unit and flow regulation unit corresponding to the Hall thruster include the power supply of the anode, the power supply of the excitation coil, the power supply of the cathode, and the anode propellant flow meter; the power supply of the anode is used to provide the working voltage, the power supply of the excitation coil is used to adjust the magnetic field strength by adjusting the current of the excitation coil, and the power supply of the cathode is used to heat the cathode and maintain electron emission.

[0065] The power supply unit is equipped with a propellant supply system, which includes a mass flow meter and a flow regulating valve. This system is used to regulate the propellant flow under control commands in order to switch between different thruster operating conditions.

[0066] In practical applications, the operation of a Hall thruster corresponds to a power supply unit and a flow regulation unit. The power supply unit includes an anode power supply, an excitation coil power supply, and a cathode power supply, while the flow regulation unit includes an anode propellant flow meter. Each power supply in the power supply unit is housed in a shielded metal casing.

[0067] In this step, such as Figure 7 As shown, when the thruster is running, the initial power is recorded by the computer as a performance constraint, and at the same time, the time-domain current signal acquisition unit acquires the discharge oscillation current waveform in real time and transmits it to the low-frequency oscillation feature extraction unit.

[0068] Step S120: Calculate the oscillation characteristic parameters corresponding to the discharge oscillation current waveform based on the discharge oscillation current waveform.

[0069] The oscillation characteristic parameters include the oscillation frequency and the oscillation amplitude. In this step, as... Figure 7 As shown, the feature extraction unit calculates the oscillation frequency, amplitude and other feature parameters based on the discharge oscillation current waveform and then transmits them to the oscillation feature mapping neural network system.

[0070] In one possible implementation, the low-frequency oscillation feature extraction unit includes a data buffer unit and a Fast Fourier Transform (FFT) algorithm unit; based on the discharge oscillation current waveform, the oscillation feature parameters corresponding to the discharge oscillation current waveform are calculated, which may specifically include the following steps:

[0071] The low-frequency oscillation feature extraction unit receives the original current signal corresponding to the discharge oscillation current waveform transmitted by the aforementioned digital oscilloscope. Based on the original current signal corresponding to the discharge oscillation current waveform, the low-frequency oscillation feature extraction unit performs spectral analysis on the original current signal using the Fast Fourier Transform (FFT) algorithm to obtain the spectral analysis results. Based on the spectral analysis results, the dominant frequency of the discharge current oscillation, the amplitude of the discharge current oscillation, and several time-frequency characteristic parameters are extracted. The low-frequency oscillation feature extraction algorithm includes the Fast Fourier Transform (FFT) algorithm.

[0072] In practical applications, the execution program corresponding to the low-frequency oscillation feature extraction unit is installed in the host industrial control computer. By introducing a time-domain signal feature extraction method, the original current oscillation data is reduced in dimensionality, which reduces the amount of data and the processing power of the neural network. This significantly improves the recognition accuracy and generalization ability of the low-frequency oscillation behavior of the Hall thruster, providing a new data-driven approach for oscillation feature evaluation.

[0073] Step S130: Based on the oscillation characteristic parameters, determine the current oscillation state of the Hall thruster through the oscillation characteristic mapping neural network model and generate the target operating conditions for the Hall thruster.

[0074] The oscillation feature mapping neural network model corresponds to the mapping relationship between thruster operating conditions and oscillation frequency and amplitude. In this step, as... Figure 7 As shown, the neural network model determines the current oscillation state and the target operating condition.

[0075] As an optional implementation, the oscillation feature mapping neural network model includes a neural network, a model storage unit, and a parameter update unit. Based on the oscillation feature parameters, the oscillation feature mapping neural network model determines the current oscillation state of the Hall thruster and generates target operating conditions for the Hall thruster. Specifically, this may include the following steps:

[0076] By running a pre-trained oscillation characteristic evaluation model based on historical oscillation characteristics, a mapping relationship between thruster operating conditions and oscillation frequency and amplitude is established. Based on the oscillation characteristic parameters, the oscillation characteristics of the Hall thruster under the current operating conditions are judged in real time through the mapping relationship, and the basis for optimizing the control strategy of subsequent thruster operating conditions is provided, thus obtaining the oscillation state prediction results. Among them, the thruster operating conditions include at least one of anode voltage, excitation coil current, and propellant flow rate.

[0077] The execution program corresponding to the oscillation feature mapping neural network model is installed in the host industrial control computer. In this embodiment, the construction of the oscillation feature and nonlinear mapping model based on deep learning is as follows: Addressing the characteristics of strong noise in the Hall thruster oscillation features and strong nonlinear coupling with the operating conditions, this solution constructs a nonlinear mapping model based on a deep neural network to achieve the correlation learning between oscillation features and operating conditions.

[0078] Step S140: Generate oscillation control commands based on the current oscillation state, target operating conditions, and performance constraints. Then, control and adjust the anode voltage of the Hall thruster's anode power supply, the magnetic field current of the excitation coil power supply, and the propellant flow rate of the flow regulation unit according to the oscillation control commands to perform coordinated control of oscillation suppression and performance maintenance.

[0079] In this step, such as Figure 7 As shown, the control command generation and communication unit generates appropriate adjustment commands based on the target operating conditions and performance constraints generated by the neural network system, and sends them to the anode power supply, excitation coil power supply and flow regulation unit respectively; the power supply and gas supply unit adjusts the anode voltage, magnetic field current and propellant flow according to the commands, thereby achieving coordinated control of oscillation suppression and performance maintenance.

[0080] In one optional implementation, the Hall thruster low-frequency discharge oscillation control system further includes a control command generation and communication unit, which includes a decision processing unit, a control command generation unit, and a communication interface.

[0081] The above-mentioned oscillation control command is generated based on the current oscillation state, target operating conditions, and performance constraints. Based on the oscillation control command, the anode voltage of the Hall thruster's anode power supply, the magnetic field current of the excitation coil power supply, and the propellant flow rate of the flow regulation unit are controlled and adjusted respectively to perform coordinated control of oscillation suppression and performance maintenance, including:

[0082] Based on the oscillation state prediction results output by the aforementioned oscillation feature mapping neural network model and the performance constraints, the optimal control parameters are calculated by the decision processing unit. Based on the optimal control parameters, the control command generation unit generates control commands for the power supply voltage of the anode, the magnetic field current of the excitation coil power supply, and the propellant flow of the flow regulation unit. Through the communication interface, the power supply unit and the flow regulation unit corresponding to the Hall thruster are used to exchange data to perform real-time oscillation control of the Hall thruster via a standard serial bus.

[0083] For the intelligent oscillation closed-loop control and adaptive optimization strategy of Hall thrusters: breaking through the limitations of traditional empirical adjustment and fixed gain control, this scheme establishes an intelligent control framework based on deep learning inversion mechanism, proposes an intelligent low-frequency oscillation closed-loop control method combining deep learning and PID, constructs a real-time closed-loop control system, and realizes adaptive adjustment of discharge voltage, excitation coil current and anode flow.

[0084] In this embodiment, rapid identification and adaptive suppression of oscillations in the Hall thruster under a wide operating condition are achieved, solving the problems of traditional control methods' strong reliance on physical models and poor robustness. Based on this, an intelligent optimization algorithm under performance constraints is introduced, forming an adaptive control strategy that can achieve oscillation suppression and automatic parameter optimization under power maintenance conditions. This effectively improves the discharge stability of the Hall thruster and ensures its operational performance.

[0085] As an example, performance constraints can be expressed as follows:

[0086] ;

[0087] ;

[0088] in, V , q , I These represent the voltage at the anode, the current flow at the anode, and the coil current, respectively. A This represents the amplitude of the low-frequency oscillation, and is an independent variable. V , q , I multivariate functions; α Indicates the error limit; P Represents power, which is an independent variable. V , q , I multivariate functions; To control the initial power of the Hall thruster, a parameter is used to measure power changes; 'st' represents the power constraint condition. Expressing performance constraints in this way allows for more precise data representation, achieving accurate constraint control.

[0089] In this embodiment, the control of Hall thruster oscillation is no longer achieved through a single circuit structure. Instead, it directly evaluates oscillation characteristics through a neural network and adjusts the thruster's operating conditions to control the thruster's discharge oscillation. The selection process of the control target is intelligently achieved through the neural network, moving away from the simple target selection using electrical devices found in existing control technologies. Instead, it automatically adjusts the Hall thruster's operating conditions for different oscillation control targets, achieving real-time identification and control of oscillation characteristics. Furthermore, it optimizes multiple parameters such as flow rate, voltage, and excitation coil current, involving variables across multiple time scales, and introduces power performance parameters as constraints. This reduces the problem of thruster malfunction or power drift caused by oscillation control, improving control reliability. By adopting a data-driven approach, it bypasses complex physical modeling and directly learns the mapping relationship between oscillation and operating conditions from data, achieving high robustness and adaptive real-time control. This overcomes the limitations of traditional control methods, improves the working stability of the Hall thruster, and solves the technical problem of low working stability of Hall thrusters caused by existing oscillation control methods.

[0090] By employing oscillation feature identification, neural network training, real-time oscillation evaluation based on neural networks, control optimization, and closed-loop execution, intelligent control of low-frequency discharge oscillations is achieved, suppressing oscillation intensity and improving discharge stability while ensuring thruster performance stability. Furthermore, by establishing a Hall thruster low-frequency discharge oscillation control system, the system automatically adjusts the Hall thruster's operating conditions for different oscillation control objectives, achieving real-time identification and control of oscillation features. By combining an oscillation feature database and a mapping model, a closed-loop control algorithm is constructed to achieve dynamic optimization and stable control of key parameters for low-frequency discharge oscillations, improving thruster operational stability. Moreover, this scheme, while maintaining the stability of the thruster's main performance parameters, uses a multi-objective optimization algorithm to collaboratively adjust multiple operating parameters, exploring the optimal parameter combination for low-frequency discharge oscillation features, achieving optimized control of oscillation features, and establishing an adaptive optimization strategy for discharge oscillations while maintaining thruster performance. The above-described structure and process in this embodiment enable real-time monitoring, feature extraction, target analysis, and closed-loop control of low-frequency oscillations in the Hall thruster, improving the stability and reliability of thruster operation.

[0091] Figure 8 A schematic diagram of a Hall thruster oscillation control system based on a neural network is provided. This device can be applied to a low-frequency discharge oscillation control system for Hall thrusters. Figure 8 As shown, the Hall thruster oscillation control system 800 based on a neural network includes:

[0092] The acquisition module 801 is used to respond to the operation of the Hall thruster, take the initial power of the Hall thruster as a performance constraint, and acquire the discharge oscillation current waveform of the Hall thruster in real time.

[0093] The calculation module 802 is used to perform calculations based on the discharge oscillation current waveform to obtain the oscillation characteristic parameters corresponding to the discharge oscillation current waveform; wherein, the oscillation characteristic parameters include the oscillation frequency and the oscillation amplitude;

[0094] The generation module 803 is used to determine the current oscillation state of the Hall thruster based on the oscillation characteristic parameters through an oscillation feature mapping neural network model and generate a target operating condition for the Hall thruster; wherein, the oscillation feature mapping neural network model includes a mapping relationship between the thruster operating condition and the oscillation frequency and the oscillation amplitude.

[0095] The control module 804 is used to generate oscillation control commands based on the current oscillation state, the target operating condition, and the performance constraints, and to control and adjust the anode voltage of the anode power supply, the magnetic field current of the excitation coil power supply, and the propellant flow rate of the flow regulation system according to the oscillation control commands, so as to perform coordinated control of oscillation suppression and performance maintenance.

[0096] The Hall thruster oscillation control system based on neural networks provided in this application has the same technical features as the Hall thruster oscillation control method based on neural networks provided in the above embodiments, so it can also solve the same technical problems and achieve the same technical effects.

[0097] An electronic device provided in this application embodiment, such as Figure 9 As shown, the electronic device 900 includes a processor 902 and a memory 901. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the method provided in the above embodiments.

[0098] See Figure 9 The electronic device also includes a bus 903 and a communication interface 904. The processor 902, the communication interface 904 and the memory 901 are connected via the bus 903. The processor 902 is used to execute executable modules, such as computer programs, stored in the memory 901.

[0099] The memory 901 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 904 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0100] Bus 903 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 9 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0101] The memory 901 is used to store programs. After receiving an execution instruction, the processor 902 executes the program. The method executed by the apparatus defined by the process disclosed in any of the preceding embodiments of this application can be applied to the processor 902 or implemented by the processor 902.

[0102] The processor 902 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 902 or by instructions in software form. The processor 902 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 901, and processor 902 reads the information from memory 901 and, in conjunction with its hardware, completes the steps of the above method.

[0103] Corresponding to the above-described Hall thruster oscillation control method based on neural networks, this application embodiment also provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are invoked and executed by a processor, the computer-executable instructions cause the processor to perform the steps of the above-described Hall thruster oscillation control method based on neural networks.

[0104] The Hall thruster oscillation control system based on neural networks provided in this application can be specific hardware on a device or software or firmware installed on the device. The device provided in this application has the same implementation principle and technical effects as the aforementioned method embodiments. For the sake of brevity, any parts not mentioned in the device embodiments can be referred to the corresponding content in the aforementioned method embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can all be referred to the corresponding processes in the above method embodiments, and will not be repeated here.

[0105] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0106] For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0107] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0108] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0109] If the aforementioned function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the Hall thruster oscillation control method based on neural networks described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0110] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0111] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A Hall thruster oscillation control method based on neural networks, characterized in that, The method, applied to a low-frequency discharge oscillation control system for Hall thrusters, includes: In response to the operation of the Hall thruster, the initial power of the Hall thruster is used as a performance constraint, and the discharge oscillation current waveform of the Hall thruster is collected in real time. Based on the discharge oscillation current waveform, calculations are performed to obtain the oscillation characteristic parameters corresponding to the discharge oscillation current waveform; wherein, the oscillation characteristic parameters include the oscillation frequency and the oscillation amplitude; The current oscillation state of the Hall thruster is determined by an oscillation feature mapping neural network model based on the oscillation feature parameters, and a target operating condition for the Hall thruster is generated; wherein, the oscillation feature mapping neural network model includes a mapping relationship between the thruster operating condition and the oscillation frequency and the oscillation amplitude; Based on the current oscillation state, the target operating condition, and the performance constraints, an oscillation control command is generated. Based on the oscillation control command, the anode voltage of the anode power supply of the Hall thruster, the magnetic field current of the excitation coil power supply, and the propellant flow of the flow regulating unit are controlled and adjusted respectively to perform coordinated control of oscillation suppression and performance maintenance. The Hall thruster includes an anode and a cathode; the Hall thruster low-frequency discharge oscillation control system includes a time-domain current signal acquisition unit and a low-frequency oscillation feature extraction unit; the time-domain current signal acquisition unit includes a current probe and a digital oscilloscope; the current probe is disposed in the power output circuit of the anode; the digital oscilloscope is connected to the current probe through a shielded signal line; the real-time acquisition of the discharge oscillation current waveform of the Hall thruster includes: acquiring the transient change of the discharge current of the Hall thruster through the current probe, acquiring the time-domain waveform of the discharge current of the Hall thruster through the digital oscilloscope, and transmitting the transient change of the discharge current and the time-domain waveform of the discharge current to the low-frequency oscillation feature extraction unit in real time; The oscillation feature mapping neural network model includes a neural network, a model storage unit, and a parameter update unit. The step of determining the current oscillation state of the Hall thruster and generating a target operating condition for the Hall thruster based on the oscillation feature parameters using the oscillation feature mapping neural network model includes: establishing a mapping relationship between the thruster operating condition and the oscillation frequency and amplitude by running a pre-trained oscillation feature evaluation model based on historical oscillation features; determining the oscillation characteristics of the Hall thruster under the current operating condition in real time based on the oscillation feature parameters and the mapping relationship; and providing a basis for optimizing the control strategy of the subsequent thruster operating condition to obtain an oscillation state prediction result. The thruster operating condition includes at least one of anode voltage, excitation coil current, and propellant flow rate. The performance constraints are expressed as follows: ； ； in, V , q , I These represent the voltage at the anode, the current flow at the anode, and the coil current, respectively. A This represents the amplitude of the low-frequency oscillation, and is an independent variable. V , q , I multivariate functions; α Indicates the error limit; P Represents power, which is an independent variable. V , q , I multivariate functions; To control the initial power of the Hall thruster, a parameter is used to measure power changes; st represents the power constraint condition.

2. The method according to claim 1, characterized in that, The low-frequency oscillation feature extraction unit includes a data buffer unit and a Fast Fourier Transform (FFT) algorithm unit; the calculation based on the discharge oscillation current waveform to obtain the oscillation feature parameters corresponding to the discharge oscillation current waveform includes: Based on the original current signal corresponding to the discharge oscillation current waveform, the Fast Fourier Transform (FFT) algorithm unit performs spectral analysis on the original current signal using a low-frequency oscillation feature extraction algorithm to obtain spectral analysis results. Based on the spectral analysis results, the dominant frequency of the discharge current oscillation, the amplitude of the discharge current oscillation, and several time-frequency characteristic parameters are extracted. The low-frequency oscillation feature extraction algorithm includes the Fast Fourier Transform (FFT) algorithm.

3. The method according to claim 1, characterized in that, The Hall thruster low-frequency discharge oscillation control system also includes a control command generation and communication unit, which includes a decision processing unit, a control command generation unit, and a communication interface. The step of generating oscillation control commands based on the current oscillation state, the target operating condition, and the performance constraints, and controlling and adjusting the anode voltage of the Hall thruster's anode power supply, the magnetic field current of the excitation coil power supply, and the propellant flow rate of the flow regulation unit according to the oscillation control commands to perform coordinated control of oscillation suppression and performance maintenance, includes: The optimal control parameters are calculated by the decision processing unit based on the oscillation state prediction results and the performance constraints. Based on the optimal control parameters, the control command generation unit generates control commands for the power supply voltage for the anode, the magnetic field current of the excitation coil power supply, and the propellant flow rate of the flow control unit. The communication interface uses a standard serial bus to interact with the power supply unit and flow regulation unit corresponding to the Hall thruster, so as to perform real-time oscillation control of the Hall thruster.

4. The method according to claim 1, characterized in that, The anode is formed by processing a conductive metal material and is connected to the power source of the anode; the cathode is a hollow cathode structure, and the cathode is used to provide electrons to the Hall thruster through the power source of the cathode; the Hall thruster is mechanically fixed by a bracket, and the anode, the cathode and the excitation coil corresponding to the excitation coil power source are respectively connected to the power supply unit corresponding to the Hall thruster through wires; The power supply unit and flow regulation unit corresponding to the Hall thruster include the power supply of the anode, the power supply of the excitation coil, the power supply of the cathode, and the anode propellant flow meter; the power supply of the anode is used to provide the working voltage, the power supply of the excitation coil is used to adjust the magnetic field strength by adjusting the current of the excitation coil, and the power supply of the cathode is used to heat the cathode and maintain electron emission; The power supply unit corresponds to a propellant supply system, which includes a mass flow meter and a flow regulating valve, used to regulate the propellant flow under control commands to switch between different thruster operating conditions.

5. A Hall thruster oscillation control system based on a neural network, characterized in that, Applications include: Low-frequency discharge oscillation control systems for Hall thrusters, including: The acquisition module is used to respond to the operation of the Hall thruster, take the initial power of the Hall thruster as a performance constraint, and acquire the discharge oscillation current waveform of the Hall thruster in real time. The calculation module is used to perform calculations based on the discharge oscillation current waveform to obtain the oscillation characteristic parameters corresponding to the discharge oscillation current waveform; wherein, the oscillation characteristic parameters include the oscillation frequency and the oscillation amplitude; The generation module is used to determine the current oscillation state of the Hall thruster based on the oscillation characteristic parameters through an oscillation feature mapping neural network model and generate a target operating condition for the Hall thruster; wherein, the oscillation feature mapping neural network model includes a mapping relationship between the thruster operating condition and the oscillation frequency and the oscillation amplitude. The control module is used to generate oscillation control commands based on the current oscillation state, the target operating condition, and the performance constraints, and to control and adjust the anode voltage of the anode power supply, the magnetic field current of the excitation coil power supply, and the propellant flow rate of the flow regulation system according to the oscillation control commands, so as to perform coordinated control of oscillation suppression and performance maintenance. The Hall thruster includes an anode and a cathode; the Hall thruster low-frequency discharge oscillation control system includes a time-domain current signal acquisition unit and a low-frequency oscillation feature extraction unit; the time-domain current signal acquisition unit includes a current probe and a digital oscilloscope; the current probe is disposed in the power output circuit of the anode; the digital oscilloscope is connected to the current probe through a shielded signal line; the acquisition module is specifically used to: acquire the transient change of the discharge current of the Hall thruster through the current probe, and acquire the time-domain waveform of the discharge current of the Hall thruster through the digital oscilloscope, and transmit the transient change of the discharge current and the time-domain waveform of the discharge current to the low-frequency oscillation feature extraction unit in real time; The oscillation feature mapping neural network model includes a neural network, a model storage unit, and a parameter update unit; the generation module is specifically used to: establish a mapping relationship between the thruster operating condition and the oscillation frequency and the oscillation amplitude by running a pre-trained oscillation feature evaluation model based on historical oscillation features; determine the oscillation characteristics of the Hall thruster under the current operating condition in real time according to the oscillation feature parameters through the mapping relationship and provide a basis for optimizing the control strategy of the subsequent thruster operating condition, thereby obtaining the oscillation state prediction result; wherein, the thruster operating condition includes at least one of anode voltage, excitation coil current, and propellant flow rate; The performance constraints are expressed as follows: ； ； in, V , q , I These represent the voltage at the anode, the current flow at the anode, and the coil current, respectively. A This represents the amplitude of the low-frequency oscillation, and is an independent variable. V , q , I multivariate functions; α Indicates the error limit; P Represents power, which is an independent variable. V , q , I multivariate functions; To control the initial power of the Hall thruster, a parameter is used to measure power changes; st represents the power constraint condition.

6. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 4.

Images