灰色直接迭代建模在武器故障预报中的应用

武器系统故障数据常呈现出非等时距、随机性及非准光滑特性,不满足GM(1,1)的建模条件。为此,引入故障信息序列的导数作为建模的白化值,并使用迭代优化算法进行参数优化,从而构建灰色直接迭代GM(1,1)模型。同时,直接迭代建模可有效利用数据第一点的信息,建模时不需累加生成变换和还原处理,算法简单可行,可以有效解决武器系统贫信息状态下的故障预报问题。仿真结果验证了该模型对于实际武器故障序列具有较高的拟合和预测精度。为了解和掌握武器质量状态的发展规律提供了可行的方法。

Application of Non Isometric Iterative Model in the Fault Prediction of the Arms

Based on the system fault data of weapons for the original time series,aiming at the characteristics of non equal interval,through the establishment of GM(1,1)a whitening method of the original gray information,the gradual iteration of GM (1,1)model is established. Gradual Iteration of direct modeling method makes full use of the first information,modeling without accumulated generating transformation and reduction treatment. This paper using MATLAB simulation,the model has higher for unequal time interval sequence fitting and forecasting accuracy is verified,which provides a new method for understanding and grasping the development law of weapon quality.