基于支持向量机的航空发动机滑油监控分析
Aeroengine Lubrication Monitoring Analysis Via Support Vector Machines
-
摘要: 提出了一种基于支持向量机的航空发动机滑油金属含量预测方法。详细分析了支持向量机用于时间序列预测的理论基础,并给出了运用支持向量回归进行多步预测的一般公式,提出了用最终预报误差(FPE)准则优化选取嵌入维数。与传统的AR预测模型相比,支持向量机由于采用了新型的结构风险最小化准则表现出优秀的推广能力。经过数值仿真得出自回归(AR)模型仅适合于短期预测;支持向量机预测推广能力强、具有较强的鲁棒性和容错性,对较长区间预测仍具有较好的效果。最后,将其应用于某型发动机滑油的铁金属含量预测,取得了较好的效果。Abstract: A novel aeroengine lubrication monitoring method based on support vector machines is presented in this paper.Basic theory analysis of support vector regression in time series forecasting is introduced in detail and a multi-step forecasting formula is presented,Final Prediction Error (FPE) principle is suggested to select the embedding dimension.Compared with general autoregressive forecasting method it adopts new type of structural risk minimization principle and thus it owns excellent generalization ability.During numerical simulations,we infer that Auto-Regressive (AR) forecasting method is suitable to short intervals while Support Vector Machines (SVM) still possesses good robustness and fault-tolerant virtue in metaphase intervals forecasting.Finally,some typs of aeroengine's lubrication metal content have been monitored for feasibility validation and test results are satisfactory.
点击查看大图
计量
- 文章访问数: 1929
- HTML浏览量: 4
- PDF量: 442
- 被引次数: 0