An Intelligent Damage-mitigating Control Method For Liquid-propellant Rocket Engines
- Paper number
IAC-07-C4.3.09
- Author
Dr. Jianjun Wu, National University of Defense Technology, China
- Coauthor
- Year
2007
- Abstract
The Damage-Mitigating Control (DMC) is an important approach to improve the safety and reliability of Liquid-propellant Rocket Engine(LRE) by achieving a desired tradeoff level between the performance of the engine and the damage of critical components in the condition of no effect on the accomplishment of the launch mission. However, two critical problems need to be solved in the DMC of LRE. The first is the global optimization of the best tradeoff between the damages in several critical components and the performance of engines through a feedback loop of control. The second is the design and realization of an autonomic and real-time Damage-Mitigating Controller(DMCer) which is able to monitor the damages of critical components continuously and take damage-mitigating controls in time if necessary. So in this paper an intelligent DMC method for LRE is developed based on a certain LOX/Kerosene rocket engine. Firstly, the fundamental concepts and principles on DMC are clarified. Then, the key technologies including the modeling of LRE dynamics, the modeling of damage and estimation of critical components, and techniques on design and realization of the DMCer are analysed in details. Secondly, a hierarchical intelligent control system is designed and developed for the autonomic and real-time DMC of LRE. It includes three structural layers such as the bottom damage-monitoring controller designed to monitor the damages of critical components in a real-time manner, the middle decision-making controller designed to isolate the damaged component and decide whether the arisen damages will lead to the abrupt failure, and the top damage-mitigating controller designed as an actuator to make a final tradeoff under the order of the decision-making controller. Thirdly, an analysis and synthesis of the DMC law is presented and clarified based on the multi-objective optimization. Then an intelligent method and procedure of applying the Fuzzy Neural Network(FNN) in the design of a DMCer is proposed. The neural network is used to learn from the information and then found inference rules in the decision-making controller. The fuzzy inference mechanism is utilized to make the decisions of whether, what and how damage-mitigating controls should be taken in time. Results with the simulation computation show that the proposed controller achieves the expected goal of damage-mitigating for the LRE. It provides important practical value not only for improving the safety and reliability of the LRE, and also for the thermo-dynamic systems of aero-plane, automobile, chemical plant because their service life is also critical.
- Abstract document
- Manuscript document
IAC-07-C4.3.09.pdf (🔒 authorized access only).
To get the manuscript, please contact IAF Secretariat.