排气温度是最能反映航空发动机运行状态的性能参数之一.对连续飞行班次的起飞排气温度裕度(EGTM,Exhaust Gas Temperature Margin)参数进行预测分析,有助于判知航空发动机将来的工作性能,为预防和排除故障提供充分的时间和决策依据.在...排气温度是最能反映航空发动机运行状态的性能参数之一.对连续飞行班次的起飞排气温度裕度(EGTM,Exhaust Gas Temperature Margin)参数进行预测分析,有助于判知航空发动机将来的工作性能,为预防和排除故障提供充分的时间和决策依据.在依据具有非线性、非平稳特征的起飞EGTM历史监测值序列构建预测模型时,基于奇异值分解滤波算法提出了一种联合径向基函数预测网络(RBFPN,Radial Basis Function Prediction Networks)和函数系数自回归模型(FAR,Functional-coefficient Auto Regressive model)的预测方案,充分发挥RBFPN和FAR在预测EGTM参数值变动趋势成分和随机成分的各自优势,使其互为补充,协同处理.实验结果表明该联合预测方案能够有效抑制RBFPN或FAR单独采用时所呈现出的不足,提高预测性能.展开更多
阐述了装备可用度预测的重要性,并以此为需求牵引,构建了具有非线性、非平稳的装备可用度时间序列。基于奇异值分解滤波算法将其分解为尽量平滑的趋势成分和平稳的随机成分,分别应用粒子群训练的径向基神经网络和函数系数自回归模型进...阐述了装备可用度预测的重要性,并以此为需求牵引,构建了具有非线性、非平稳的装备可用度时间序列。基于奇异值分解滤波算法将其分解为尽量平滑的趋势成分和平稳的随机成分,分别应用粒子群训练的径向基神经网络和函数系数自回归模型进行预测,充分体现了2类模型各自的优势,使其相互补充,并进行实例分析,验证了模型和算法的有效性。实验与应用结果表明,该组合方法的预测性能和效果比单一使用RBF(radial basis function neural network)和FAR(functional-coefficient auto regressive model)进行预测更好,为装备可用度研究提供了一条新的思路。展开更多
This paper proposes an extended model based on ACR nmdel： Functional coefficient autoregressive conditional root model （FCACR）. Under some assumptions, the authors show that the process is geometrically ergodic, st...This paper proposes an extended model based on ACR nmdel： Functional coefficient autoregressive conditional root model （FCACR）. Under some assumptions, the authors show that the process is geometrically ergodic, stationary and all moments of the process exist. The authors use the polynomial spline function to approximate the functional coefficient, and show that the estimate is consistent with the rate of convergence Op（hv＋1 ＋ n-1/3）. By simulation study, the authors discover the proposed method can approximate well the real model. Furthermore, the authors apply the model to real exchange rate data analysis.展开更多
文摘排气温度是最能反映航空发动机运行状态的性能参数之一.对连续飞行班次的起飞排气温度裕度(EGTM,Exhaust Gas Temperature Margin)参数进行预测分析,有助于判知航空发动机将来的工作性能,为预防和排除故障提供充分的时间和决策依据.在依据具有非线性、非平稳特征的起飞EGTM历史监测值序列构建预测模型时,基于奇异值分解滤波算法提出了一种联合径向基函数预测网络(RBFPN,Radial Basis Function Prediction Networks)和函数系数自回归模型(FAR,Functional-coefficient Auto Regressive model)的预测方案,充分发挥RBFPN和FAR在预测EGTM参数值变动趋势成分和随机成分的各自优势,使其互为补充,协同处理.实验结果表明该联合预测方案能够有效抑制RBFPN或FAR单独采用时所呈现出的不足,提高预测性能.
文摘阐述了装备可用度预测的重要性,并以此为需求牵引,构建了具有非线性、非平稳的装备可用度时间序列。基于奇异值分解滤波算法将其分解为尽量平滑的趋势成分和平稳的随机成分,分别应用粒子群训练的径向基神经网络和函数系数自回归模型进行预测,充分体现了2类模型各自的优势,使其相互补充,并进行实例分析,验证了模型和算法的有效性。实验与应用结果表明,该组合方法的预测性能和效果比单一使用RBF(radial basis function neural network)和FAR(functional-coefficient auto regressive model)进行预测更好,为装备可用度研究提供了一条新的思路。
基金supported by the National Nature Science Foundation of China under Grant Nos.10961026, 11171293,71003100,70221001,70331001,and 10628104the Ph.D.Special Scientific Research Foundation of Chinese University under Grant No.20115301110004+2 种基金Key Fund of Yunnan Province under Grant No.2010CC003the Fundamental Research Funds for the Central Universitiesthe Research Funds of Renmin University of China under Grant No.11XNK027
文摘This paper proposes an extended model based on ACR nmdel： Functional coefficient autoregressive conditional root model （FCACR）. Under some assumptions, the authors show that the process is geometrically ergodic, stationary and all moments of the process exist. The authors use the polynomial spline function to approximate the functional coefficient, and show that the estimate is consistent with the rate of convergence Op（hv＋1 ＋ n-1/3）. By simulation study, the authors discover the proposed method can approximate well the real model. Furthermore, the authors apply the model to real exchange rate data analysis.
