基于蒙特卡罗发动机竞争失效的下发仿真模型

针对常规的解析法预测发动机下发时间建模过程复杂且不易求解的问题,提出了一种基于蒙特卡罗仿真预测发动机首次下发时间的方法。通过分析发动机机队的历史数据,研究排气温度裕度(EGTM)的衰退规律以及性能衰退超标的寿命分布,统计各主要部件发生首次部件损伤的时间分布,并计算偶然性损伤的发生概率,确定偶然性损伤的时间分布,对性能衰退、部件损伤、偶然性损伤三种失效模式进行竞争分析,建立蒙特卡罗仿真模型,预测发动机的首次下发时间规律。结合该机队提供的实际下发数据,利用Kolmogorov-Smirnov检验确定首次下发时间的分布类型,经分析,仿真结果的可靠度误差甚小,在-1%~2%之间,从而验证了该方法的合理性和可行性。 

Monte Carlo-based competitive failure delivery simulation model of engine

To solve the problem of the complicated modeling process and the slving difficulty when using conventional analytical methods to predict the engine delivery time,a method based on Monte Carlo simulation to predict the first delivery time of the engine was proposed. By analyzing the historical data of the engine fleet, the exhaust gas temperature margin (EGTM) decline rule and the lifetime distribution of performance degradation excess were studied. The time distribution of the first hardware damage occurred of each major hardware was counted, the probability of occurrence of accidental damage was calculated, and the time distribution of accidental damage was determined. Performance degradation, hardware damage and accidental damage were subjected to competitive analysis, and a Monte Carlo simulation model was established to predict the law of time when the engine is first delivered. Combined with the actual data provided by the engine fleet, the Kolmogorov-Smirnov test was used to determine the distribution type of first delivery time. After analysis, the reliability error of the simulation results was very small, between -1% and 2%, and the rationality and feasibility of the method was verified.