论文部分内容阅读
故障预测对保障武器装备安全可靠工作具有重要意义。但是,用于武器装备故障诊断和预测的数据往往是小样本、多特征参数数据,当前主要的故障预测方法在实际故障预测中虽取得了一定的效果,但均存在不足之处。本文基于灰色预测建模理论,分析了GM(1,1)预测建模中的不足,考虑多个特征参数间的相互关系以及预测序列的实际特点,修正了初始值和背景值,建立了小样本情况下的自适应多特征参数预测模型,并以某型飞机发动机的多特征参数的仿真数据为例进行了预测分析,结果表明该模型具有很好的预测精度,证明了该模型的有效性。
Fault prediction is of great significance to ensure the safe and reliable operation of weaponry and equipment. However, the data used for fault diagnosis and prediction of weaponry equipment are often small samples and multi-characteristic parameter data. At present, the main fault prediction methods have achieved some effects in actual fault prediction, but they all have some shortcomings. Based on the gray forecasting modeling theory, this paper analyzes the shortcomings of the GM (1,1) forecasting model, considers the interrelationship between multiple characteristic parameters and the actual characteristics of the forecasting sequence, modifies the initial value and the background value, and establishes a small The results show that the model has good prediction accuracy and the validity of the model is proved. Key words: adaptive multi-parameter prediction model; .