回归分析在房地产市场方面的应用

本文通过建立的回归模型，对商品住宅的销售量进行了预测。     

高超声速飞行器前体/进气道参数重要性测度

针对机体推进高度一体化的吸气式高超声速飞行器前体/进气道相关参数对其性能指标的独立和交互作用问题,基于回归模型和二阶混合中心矩理论,以总压恢复系数为性能指标,提出了参数重要性测度新方法。设计了二维混压楔形前体/进气道变几何构型方案并给出重要性测度模型,定量和定性地对参数重要性进行了对比和分析。研究结果表明:前体/进气道相关参数间具有强的非线性作用,交互作用非常显著,独立作用可以忽略不计,且重要性测度模型易于扩展。     

基于高斯过程的航天器自适应滑模姿态控制

针对存在模型不确定性和外界干扰的刚性航天器,提出了一种基于高斯过程回归(GPR)的新型自适应滑模姿态控制算法。该算法具有自学习能力,在不同的姿态控制任务下都能够实现高精度、强鲁棒和高效率的姿态跟踪。首先,在航天器的四元数标称系统动态模型基础上,应用在线稀疏高斯过程回归(SOGP)方法学习系统的未知动态;其次,结合高斯过程的预测均值设计滑模控制算法,利用高斯过程的预测方差自适应调节控制增益,并应用李雅普诺夫方法严格证明闭环系统的稳定性,保证了航天器姿态跟踪误差的渐进收敛性;最后,通过数值仿真验证了所设计控制器的有效性。结果表明,该自学习控制算法与自适应滑模控制(ASMC)与神经网络自适应控制等算法相比,具有更快的收敛速度、更高的跟踪精度以及更低的控制成本。     

The relationship between TEC and Earth’s magnetic field during quiet and disturbed days over Istanbul,Turkey

In this study, the impact of Earth’s magnetic field on total electron content (TEC) was studied by using statistical multiple linear regression model and co-integration method. TEC values were measured over the Turkey-Istanbul (ista) station using date of global positioning system (GPS), and the magnetic field components of the Earth were measured from Boğaziçi University, Kandilli Observatory and Earthquake Research Institute, Geomagnetic observatory Istanbul (ISK) station. This examination has been carried out during the dates of March 14–19, 2015 covering the dates of March 17–18, 2015 known in the literature as St. Patrick's Day geomagnetic storm. The three days before the storm (March 14–16) were named as quiet days, whereas the other days (March 17–19) were named as disturbed days after which the two periods were examined separately. It was observed as a result of the examination that the x-component (south-north direction) of the magnetic field had a negative impact on TEC on quiet days, whereas the impact was positive on disturbed days. However, the y-component (east–west direction) of the magnetic field had an inverse relationship of the x-component on the quiet and disturbed days. In addition, it was deduced that the impact coefficient of the x and y-component of the magnetic field was higher on disturbed days in comparison with those on quiet days. The correlation coefficient between the TEC and the components of the Earth’s magnetic field was 0.11 on quiet days and 0.95 on disturbed days. Therefore, it can be stated that the relationship of the TEC values with the geomagnetic field are higher on disturbed days.     

定时截尾数据回归分析方法

提出定时截尾数据回归分析方法,建立定时截尾数据回归方程,给出回归系数和标准差的最佳无偏整体估计、百分位值的点估计及其置信限估计.详细讨论了工程中常见的极值分布、Weibull分布、正态分布以及一般位置-尺度分布场合下定时截尾数据的回归分析问题.该方法可以充分利用截尾时刻的试验信息,并将各个状态下的定时截尾数据作为一个整体进行统计分析,具有信息量大、精度高的特点,从而可以在试验时间、经费等有限的情况下对产品进行高精度的寿命预测和可靠性评估.     

粒子群优化在直升机旋翼动平衡调整中的应用

传统的直升机旋翼调整方法没有考虑调整参数与振动信号之间的非线性关系,针对这一缺点,提出将广义回归神经网络(GRNN,General Regression Neural Network)和粒子群算法相结合的旋翼调整方法,采用GRNN网络建立旋翼动平衡调整模型,以桨叶的调整参数作为神经网络的输入,以旋翼转轴和机身的三向的加速度测量值作为网络输出,建立调整参数与直升机振动信号间的模型.以直升机振动作为目标函数,采用粒子群优化算法对桨叶的调整参数进行寻优,获得当直升机振动最小时的桨叶的调整量. 飞行实验结果表明,此方法可通过飞行测试获得的新数据对神经网络进行更新,使系统在使用过程中不断完善,并可在较少的飞行调整下完成旋翼的动平衡调整.      

基于小波神经网络的直升机主减速器故障诊断系统

应用离散小波变换(DWT)和神经网络相结合构建直升机主减速器速器故障诊断系统: DWT对振动信号进行特征提取,神经网络对故障进行辨识和分类。阐述了DWT、帕塞瓦尔定理和广义回归神经网络(GRNN)基本理论,提出了直升机主减速器的故障诊断系统流程图,最后用某型直升机飞行时主减速器上的振动数据对该系统进行验证。实验使用了BPNN(back-propagation neural network)和GRNN两种神经网络,结果表明:提出的故障诊断系统能对主减速器故障进行较好的辨识和分类,这将为直升机主减速器故障诊断系统的进一步开发提供新的技术参考。      

A Novel PF-LSSVR-based Framework for Failure Prognosis of Nonlinear Systems with Time-varying Parameters

Particle filtering (PF) is being applied successfully in nonlinear and/or non-Gaussian system failure prognosis. However, for failure prediction of many complex systems whose dynamic state evolution models involve time-varying parameters, the traditional PF-based prognosis framework will probably generate serious deviations in results since it implements prediction through iterative calculation using the state models. To address the problem, this paper develops a novel integrated PF-LSSVR framework based on PF and least squares support vector regression (LSSVR) for nonlinear system failure prognosis. This approach employs LSSVR for long-term observation series prediction and applies PF-based dual estimation to collaboratively estimate the values of system states and parameters of the corresponding future time instances. Meantime, the propagation of prediction uncertainty is emphatically taken into account. Therefore, PF-LSSVR avoids over-dependency on system state models in prediction phase. With a two-sided failure definition, the probability distribution of system remaining useful life (RUL) is accessed and the corresponding methods of calculating performance evaluation metrics are put forward. The PF-LSSVR framework is applied to a three-vessel water tank system failure prognosis and it has much higher prediction accuracy and confidence level than traditional PF-based framework.     

超燃冲压发动机尾喷管性能对型面参数的回归研究

为了研究型面参数对基于三次曲线构型的超燃冲压发动机尾喷管性能的影响规律,采用均匀设计和回归分析的方法开展了型面参数化研究,得到了尾喷管性能参数与构型参数间的回归模型.研究表明下壁面高度对尾喷管的升力性能影响不显著,其余型面参数对尾喷管性能有影响且为交互作用.同时,上下壁面出口高度增加能提高尾喷管推力性能,上壁面长度增加对升力性能有益,而上壁面出口高度和角度的增加则反之;上壁面出口高度对俯仰力矩性能的影响与升力性能类似,但俯仰力矩性能与上壁面入口角度呈正相关,而出口角度在大部分情况下对俯仰力矩性能呈负相关;下壁面在一定范围内截短对上壁面压强分布影响不显著.      

基于PLSR-CER模型的大飞机成本风险控制

大型飞机的研制往往存在多种难以预见的不确定性因素,而且目前国内大型客机研制经验缺乏,相关的性能参数和研制成本的历史数据稀少,并存在多重相关性等问题,因此,文中引入了偏最小二乘回归方法,以客机的机体特征及性能特征作为多因变量,建立了一种多因变量对单一自变量的成本预测回归模型,并将部分客机性能参数和成本数据导入成本预测回归模型进行验证。实例分析表明,基于偏最小二乘回归法(Partial least-squares reqression,PLSR)构建的费用估算关系(Cost estimating relationship,CER)模型相对传统预测模型有很大的改善,预测的精度更高,更能体现客机研制成本与前期的性能要求之间的关系。