基于聚类分析的液体火箭发动机稳态过程故障程度评估方法

Discussion on the Fault Level Evaluation Method for the Steady Process of Liquid Propellant Rocket Engine Based on Cluster Analysis

目前液体火箭发动机稳态过程的故障检测与评估的研究相对较少,挖掘发动机稳态过程故障信息数据对运载火箭保障发射任务成功率、降低运行成本、实现可重复使用等具有重要意义,提出基于聚类分析的故障程度评估方法,构建了故障程度评估算法,通过采用液体火箭发动机的仿真数据,验证评估单一故障程度与二度故障程度。验证结果表明,该算法评估效果满足要求。

There is relatively less study focus on the fault detection and evaluation of liquid rocket engine steady-state process. The fault information data mining of engine steady-state process is very important for raising future launch vehicle mission success rate, reducing launch cost and realizing rocket reusable. This paper puts forward a fault level evaluation method based on clustering analysis. A fault level evaluation method by using the simulation data of liquid rocket engine was presented, and the single fault evaluation and second-degree failure evaluation were validated. The results show that the evaluation effect of the algorithm is satisfactory.