智能分布式武器装备系统寿命估计方法研究

Research on life estimation method of intelligent distributed Weapon and equipment system

武器装备包含传感器、控制器、执行机构等众多关键部件,研究关键部件的寿命预测及与系统寿命之间的关系尤为重要。采用面积最大化值法提取关键部件综合性能指标的急速退化期。对关键部件的急速退化期建立GM(1,1)-ELM模型。GM(1,1)模型能够捕捉综合性能指标的变化趋势,ELM模型对GM(1,1)的预测残差进行预测,弥补了GM(1,1)模型预测精度不高与ELM模型训练数据不完备的问题。实验结果显示,GM(1,1)-ELM模型与单一模型比较,可提高对综合性能指标的预测精度,并且对多个关键部件寿命预测具有一定的鲁棒性。根据全部关键部件的预测寿命,建立寿命分布模型,按照实际需求估计智能分布式武器系统的寿命及使用状态。

Weapons and equipment include many key components such as sensors,controllers,and actuators.It is particularly important to study the life prediction of key components and the relationship between them and the life of the weapon’s system.The rapid degradation period of the comprehensive performance index of key components was extracted by the area maximization method.The GM(1,1)-ELM model was established for the rapid degradation period of key components.The GM(1,1)model can capture the change trend of the comprehensive performance indicators.The ELM model can predict the prediction residuals of GM(1,1),which makes up for the problem of incomplete training data.The experimental results show that compared with a single model,the GM(1,1)-ELM model can improve the prediction accuracy of comprehensive performance indicators,and has a certain robustness for the life prediction of multiple key components.According to the predicted life of all key components,a life distribution model was established,and the life and use status of the intelligent distributed weapon system were estimated according to the actual needs.