Neural network control method of pulse plasma thruster

Abstract

The invention discloses a neural network control method of a pulse plasma thruster. The method comprises the steps of: establishing a multiple-input-single-output single-layer neural network model foranalyzing the relationship between the thrust of the plasma thruster and pulse current, and establishing a multiple-input-single-output multi-layer neural network model, wherein an input layer is pulse current with different amplitudes and frequencies, a middle layer is a pulse voltage, the output layer is thrust, the pulse voltage is proportional to the pulse current, and the square of the pulsevoltage is proportional to the thrust value, so that the relationship between the pulse current and the plasma thruster is determined to obtain an accurate value of the thrust. By adopting the technical scheme disclosed by the invention, the relationship between the pulse current of the pulse plasma and the thrust is analyzed, the rate of convergence of network learning and the stability of network learning are improved by optimizing a neural network algorithm, and real-time control of the thrust of the thruster is realized, thereby improving the ability to deal with thrust uncertainty.

Description

technical field [0001] The invention belongs to the technical field of electric propulsion plasma control, in particular to a neural network control method of a pulse plasma thruster. Background technique [0002] Electric propulsion is a kind of advanced propulsion method that uses electric energy to directly heat the propellant or ionizes and accelerates the propellant to obtain propulsion power. It has high specific impulse, thrust and efficiency. It has broad application prospects in space missions such as space navigation and interstellar navigation. [0003] The pulsed ion thruster is a kind of electromagnetic electric thruster, which has been widely used in the main propulsion system of satellites and deep space probes. [0004] The pulse control of the pulsed plasma thruster is of great significance to optimize the design of the engine and improve the performance of the engine. The advantage of the neural network control method is that it is suitable for nonlinear ...

AI Technical Summary

Problems solved by technology

The advantage of the neural network control method is that it is suitable for nonlinear approximation. It has important academic value and engineering significance for the study of the pulse control of the pulsed plasma thruster. At present, there is no neural network control method specifically for the pulse control of the pulsed plasma thruster. Research

Images