基于在线LS-SVM算法的变参数混沌时间序列预测

研究利用最小二乘支持向量机(LS-SVM)预测变参数混沌时间序列。变参数混沌系统适合于描述现实中的复杂混沌现象,但由于参数的慢变导致系统动力学特性不断发生变化,基于Tan-kens嵌入定理的建模预测方法难以适用,其时间序列预测可以看作是小样本学习问题。最小二乘支持向量机是在二次损失函数下采用等式约束求解问题的一种支持向量机,保留支持向量机优点同时计算量大大减少。提出用一种具有遗忘机制的最小二乘支持向量机在线递推算法,并引入历史数据的高次项预测变参数混沌时间序列。对典型变参数混沌时间序列的预测结果表明,该方法具有较高预测精度,能快速跟踪预测变参数混沌时间序列。     

高分辨率多径时延测量算法

提出一种基于最小二乘准则的高分辨率多径信号时延测量方法。对参考信号作三阶样条插值获得每条多径信号最小二乘意义的最优估计,用每条多径信号的测量值重构接收信号,通过反复迭代使重构信号是接收信号最小二乘意义的最优估计,实现多径信号的时延测量。仿真结果表明：新算法在低信噪比、窄带信号条件下仍能获得较高的时延测量精度,性能优于MUSIC类时差测量算法。     

一种提高热电阻测温准确性方法研究

分析了工业铂、铜热电阻电阻-温度特性,以工业铂电阻温度传感器为例,给出了一种提高热电阻测温准确度的方法。通过对工业铂电阻进行多点测试,利用最小二乘法对测得数据进行拟合,得出新的电阻-温度转换公式和分度表。通过试验数据的比较表明,使用该方法可以提高传感器的测温准确性。     

PN码相位精确测量的累加最小二乘法

利用最小二乘（LS）法可以实现PN码相位的精确测量,但该方法对粗同步要求高,难以满足实际工程需求。给出了一种新的累加LS法的实现结构,在该结构下根据鉴相曲线的波形变化特征,利用累加后的离散鉴相曲线的最大值和最小值点作为特征基准点,取最值之间的数据进行LS线性拟合,实现粗同步于半个码元宽度以内的伪码相位精确测量。理论分析表明,该方法的鉴相特性呈线性特征,可以有效使用累加LS线性拟合方法。仿真结果表明,该方法相比于其他方法使用范围更广,测量精度更高。      

深空探测光学导航敏感器在轨几何定标方法

光学导航敏感器是光学自主导航的一个核心器件,它所获得的导航目标源的光线指向的精度将直接影响自主导航的精度。设计了一个分步式的光学导航敏感器在轨几何定标方法,该方法先求解外定标参数,然后在外定标所确定的广义相机坐标系下求解内定标参数,从而完成对内外定标参数的标定。为了在星上计算资源与能力有限的环境下,利用更多的参考星图实现对定标参数的高精度估计,利用逐行法化最小二乘方法估计定标参数。实验表明,通过高精度的在轨几何定标,可以有效提高光学导航敏感器的指向量测精度,使其满足光学自主导航的需求。     

An adaptive turbo-shaft engine modeling method based on PS and MRR-LSSVR algorithms

In order to establish an adaptive turbo-shaft engine model with high accuracy, a new modeling method based on parameter selection (PS) algorithm and multi-input multi-output recursive reduced least square support vector regression (MRR-LSSVR) machine is proposed. Firstly, the PS algorithm is designed to choose the most reasonable inputs of the adaptive module. During this process, a wrapper criterion based on least square support vector regression (LSSVR) machine is adopted, which can not only reduce computational complexity but also enhance generalization performance. Secondly, with the input variables determined by the PS algorithm, a mapping model of engine parameter estimation is trained off-line using MRR-LSSVR, which has a satisfying accuracy within 5&. Finally, based on a numerical simulation platform of an integrated helicopter/ turbo-shaft engine system, an adaptive turbo-shaft engine model is developed and tested in a certain flight envelope. Under the condition of single or multiple engine components being degraded, many simulation experiments are carried out, and the simulation results show the effectiveness and validity of the proposed adaptive modeling method.     

A point-wise least squares spectral analysis (LSSA) of the Caspian Sea level fluctuations,using TOPEX/Poseidon and Jason-1 observations

The Caspian Sea has displayed considerable fluctuations in its water level during the past century. Knowledge of such fluctuation is vital for understanding the local hydrological cycles, climate of the region, and construction activities within the sea and along its shorelines. This study established a point-wise satellite altimetry approach to monitor the fluctuations of the Caspian Sea using a complete dataset of TOPEX/Poseidon for the period 1993 to the middle of 2002, and its follow-on Jason-1 for the period 2002 to August 2009. Therefore, 280 virtual time-series were constructed to monitor the fluctuations. The least squares spectral analysis (LSSA) method is, then employed to find the most significant frequencies of the time-series, while the statistical method of principle component analysis (PCA) is applied to extract the dominant variability of level variations. The study also used the observations of TOPEX/Poseidon and Jason-1 over the Volga River along with 5 years of Volga’s water discharge to study its influence on the Caspian Sea level changes. The LSSA results indicate that the lunar semidiurnal (M2) and the Sun semidiurnal (S2) frequencies are the main tidal frequencies of the Caspian Sea with the mean amplitude of 4.2 and 2.8 cm, respectively. A statistically significant long-term frequency (12.5-years period) is also found from altimetry and tide gauge observations. A phase lag, related to the inter-annual frequencies of the Volga River was detected from the point-wise time-series showing level propagation from the northwest to the southeast of the sea. The cross-correlation between the power spectrum of Volga and that of the northern-most, middle, and southern-most points within the Caspian Sea were respectively 0.63, 0.51 and 0.4 of zero-lag correlation, corroborating the influence of the Volga River. The result of PCA also shows that different parts of the Caspian Sea exhibit different amplitudes of level variations, indicating that the point-wise approach, when employing all available satellite measurements could be a suitable method for a preliminary monitoring of this inland water resource as it gives accurate local fluctuations.     

民用飞机系统部件维修间隔确定方法研究

维修方案中规定了各种部件的维修类型和维修间隔,由于各部件的维修类型相对固定,所以维修间隔的确定是制定维修方案的关键。针对这种特点,同时为了能更准确的反应部件实际情况,本文采用飞机部件的使用可靠性数据来确定其维修间隔。为此,针对符合威布尔分布的部件,研究了利用中位秩及最小二乘法对两参数威布尔分布的参数进行估计,并根据其寿...     

最小二乘支持向量机的在线学习算法研究

针对最小二乘支持向量机在线逼近过程中求解矩阵维数逐渐扩大的问题,提出了一种最小二乘支持向量机的在线学习算法。借助滚动窗口的思想,建立一个随时间滑动的建模数据区间,在线逼近中通过接受新数据删除旧数据来保持数据区间长度不变,同时数据不断更新,从而实现模型的在线学习。仿真结果表明了这种学习算法的有效性。     

用户星天线空间扫描策略

 天线空间扫描过程是建立中继卫星系统星间链路的关键阶段之一。首先分析比较了水平行扫、螺旋扫描和分行螺旋扫描策略,并根据分析结果,结合轨迹预处理算法,设计了一种变速分行螺旋扫描策略。然后采用蒙特卡罗仿真分析了用户星天线开环回扫指向精度,近似确定了天线空间扫描的范围。最后通过仿真比较等速分行螺旋扫描和变速分行螺旋扫描策略,并仿真了用户星天线捕获中继卫星的过程,通过仿真结果证明所设计的变速分行螺旋扫描具有一定的工程应用价值。