Particle swarm algorithm-based damage-mitigating control law analysis and synthesis for liquid-propellant rocket engine

Authors:
Publisher website
Google Scholar
Abstract
The damage-mitigating control is a novel technique to ameliorate the reliability and safety of liquid-propellant rocket engines by achieving an optimized trade-off level between overall dynamic performance of the liquid-propellant rocket engine and structural durability of some selected critical damageable components under the condition of no impact on the achievement of the launch and flight mission. Thus, it is needed to be solved for the damage-mitigating control that the global optimization of the best trade-off between the damage of the critical damageable components and the performance of rocket engine. The major challenge should focus on: (i) to construct model of a certain rocket engine system dynamics, critical components structural dynamics, and damage dynamics; (ii) to optimize open loop feed-forward control law based on liquid-propellant rocket engine system dynamic model, structural and damage dynamics model, by using particle swarm optimization algorithm; (iii) to synthesize an intelligent damage-mitigating control system using the optimized open loop control law. In this paper, synthesis procedure of damage mitigation is introduced; structure and damage dynamic model of damageable components are formulated. The results of the simulation computation show that the synthesized control laws are implemented and achieve the effect of damage mitigating for the liquid-propellant rocket engine. It can provide important theoretical and practical value not only for improving the safety and reliability of the liquid-propellant rocket engine, but also for the complex thermo-flow-mechanical systems such as airplane engines, automobile engines, and fossil-fueled power plant because their service life is very critical too.
Funding Information
  • National Natural Science Foundation of China (Nos.50276068 and 51206181)

This publication has 16 references indexed in Scilit: