Solid rocket engine thrust control method based on radial basis function neural network

Abstract

The invention provides a solid rocket engine thrust control method based on a radial basis function neural network. The method is characterized in that the method comprises the following steps: step 1, determining a control object, enabling an execution mechanism to employ a gas adjustment system and a pneumatic servo system, and selecting a solid rocket engine as a power device; step 2, establishing a mathematical model of a controlled object of the solid rocket engine; step 3, determining a reference model and an actual model according to expected control performance requirements; step 4, designing an RBF neural network controller for solid rocket engine thrust self-adaptive control, and selecting learning indexes of a neural network; and step 5, optimizing RBF neural network controllerparameters by using a genetic algorithm. According to the method, the neural network parameter setting is optimized through the genetic algorithm, so that the solid rocket engine control system basedon the RBF neural network has good tracking capability on the characteristics of the given reference model; and the method has relatively good anti-interference performance and high robustness.

Description

technical field [0001] The invention relates to the field of automatic control, in particular to a thrust control method of a solid rocket motor based on a radial basis neural network. Background technique [0002] Solid rocket motor, also known as solid propellant rocket motor, uses solid propellant, which can meet the requirements of the new generation of tactical missiles for its power plant to the greatest extent. It has the advantages of small size, light weight, fast speed and good maneuverability. An ideal propulsion system is a strategic weapon that countries give priority to development. [0003] A large number of theories and experiments have been done for the solid rocket motor with controllable thrust at home and abroad. In order to ensure that the solid rocket can fly according to the predetermined orbit, it is necessary to make corresponding guidance and control. The attitude control of the rocket is realized through the thrust vector control system, and the ...

AI Technical Summary

Problems solved by technology

In recent years, due to the development of neural networks, many studies have applied radial basis neural networks to the control of dynamic objects, but from the perspective of control methods, it is still the first to use radial basis neural networks for solid rocket control. For rocket engine control, its parameters will change drastically with the environment, and the unmodeled dynamics will also bring certain difficulties to the control, and the radial basis neural network has a good adaptive function, which can better overcome this problem

Images