改进的灰色预测模型在导弹中的应用

Improved Grey Forecasting Model of Fault Prediction in Missile Applications

灰色预测模型的预测精度往往依赖于原始数据序列的光滑度,然而在实验中得到的原始数据序列通常并不满足光滑性要求。为了提高灰色预测模型的预测精度,保障武器系统的可靠性,提出了一种提高原始数据序列光滑度的方法。方法基于传统灰色GM(1,1)模型,利用更具广泛性的对数——幂函数变换法来改善原始数据序列的光滑度。并应用于导弹故障预测中,取得了良好的效果。仿真结果表明,方法有效地改善了原始数据序列的光滑度,明显地提高了故障预测的精度,可以有效地预测导弹故障。

Usually the prediction accuracy of Grey Prediction Model relies on the smoothness of the raw data sequence.Nevertheless,the raw data sequence obtained from experiments can not always satisfy the smoothness requirement.In order to improve the prediction accuracy of Grey Prediction Model,this paper proposes a method of enhancing the smoothness of raw data sequence.Based upon the traditional grey GM（1,1） model,the method makes use of the power function-logarithmic transformation method to improve the smoothness of raw data sequence.This paper applied the method to the missile failure prediction,and achieved favorable effect.Simulation results show that this method has effectively improved the smoothness of raw data sequence and markedly enhanced the accuracy of failure prediction,therefore,it can predict the missile failure effectively.