Real-time filter based on the quasi-accurate detection of gross errors

Because of the sensitivity of the Kalman framework to gross errors, proper techniques for detection of gross errors are necessary. By integrating the selection of quasi-accurate observations and the Kalman framework, a new filter called the quasi-accurate filter (QUAF) is developed. The expansibility and implementation scheme of the new algorithm are then discussed in detail, and the reliability matrix for the Kalman filter is proposed to analyze the reliability of the filters with different detection technologies. Finally, the experimental results from a real world case study are used to validate the conclusions. The QUAF carries out the preliminary selection of the quasi-accurate observations (QAOs) using the innovation of the Kalman filter, and use the check QAOs to determine reasonable observations. This causes the QUAF to handle more easily and possess wider expansibility. QUAF can be reformulated to the special cases of several common detection methods, such as the innovation method, robust estimation and quasi-accurate detection (QUAD). Since only reasonable observations are used, the QUAF has better detection accuracy and stronger avoidance of gross errors than the innovation method and robust estimation. Meanwhile, compared with QUAD methods, QUAF introduces the state-predicted model, requiring fewer quasi-accurate observations and making it more suitable for systems with complicated observation structures or sparse observations.     

The Predictive Performance Evaluation of Biased Regression Predictors With Correlated Errors

When the interdependence of disturbances is present in a regression model, the pattern of sample residuals contains information which is useful in the prediction of post‐sample drawings and when multicollinearity among regressors is also present, it is useful to use biased regression estimators. This information is exploited in the biased predictors derived here. Also, the predictive performance of various biased predictors with correlated errors is discussed and all pair‐wise comparisons are made among these predictors. The theoretical results are illustrated by a numerical example. Copyright © 2015 John Wiley & Sons, Ltd.     

渐近伪压缩型映象迭代序列的强收敛定理

在没有任何有界条件下建立了渐近伪压缩型映象带混合型误差修改的Ishikawa和Mann迭代序列收敛到不动点的充要条件,所得结果本质推广和改进了有关文献中的相关结果.     

赣方言区普通话水平测试中的语音偏误调查

本文在对赣方言区普通话水平测试成绩调查的基础上,分析了赣方言区考生常见的语音偏误,以及各等级的偏误特点,指出了调查分析结果对赣方言区普通话水平测试和普通话教学的积极作用。     

2004年以来河南省粮食生产变化趋势及增产途径探析

2004—2013年这10年中,河南粮食连年增产,为保障国家粮食安全作出了重要贡献。但我们也必须清醒地看到,河南粮食生产中还存在许多问题,主要是：种粮成本上升的幅度超过粮食生产效益上升的幅度,影响农民种粮积极性;粮食主产区的生产能力接近极限,增产压力较大;粮食主产区抵御自然灾害的能力不强,基本上还是靠天收粮;土地流转总量不大,粮食生产规模效益偏低,等等。对此,应加强农田基础设施建设,提高粮食生产者抵御自然灾害的能力;依靠科技进步来提高粮食单产水平;提高粮食收购价格,控制农资涨价幅度,切实保护粮食生产者的利益;建立现代规模化、集约化粮食生产体系。总之,只有充分利用粮食生产这一大优势,才能真正实现河南经济社会的可持续发展。     

不动点问题和零点问题的公共元的具误差的迭代算法

在自反的严格凸的具有K-K性质的光滑Banach空间中,设计了一种新的具误差的投影算法,用以逼近拟φ-非扩张映射的不动点集和极大单调算子的零点集的公共元素,并利用所设计的算法证明了公共元的强收敛定理.作为应用,给出寻找凸泛函的极小值问题.     

缺失数据下带约束条件的部分线性变系数EV模型的估计

用一般级数估计方法研究了带约束及响应变量缺失条件下的部分线性变系数EV模型的参数与非参数估计,并讨论了参数估计的一致性和渐进正态性及非参数估计的收敛速度,且通过数值模拟验证了所提方法的估计效果.     

神经网络集成模型在广东省海洋生产总值预测中的应用

海洋生产总值预测是一项极其复杂但又非常重要的工作,本文探讨了神经网络集成模型在广东省海洋生产总值预测中的应用.通过采集广东省2000—2010年海洋统计年鉴数据,借助Bagging方法生成仿真数据训练网络个体,将12个个体网络集成对广东省2010年海洋生产总值进行预测分析,该模型解决了海洋经济数据非线性、时变性、样本量少和偏差大等问题.与线性回归方法的比较结果表明,神经网络集成模型预测海洋生产总值的结果更可靠.     

锯齿强化前后的技术-经济比较

<正>强化后的带锯齿在制材时能减少单位锯材面积上的电耗和锯材加工误差,并能延长带锯条的工作时间,从而提高了制材效率。特别适于锯解硬原木和大径原木的制材厂。     

X射线成像望远镜像质评价的计算机仿真

高分辨率X射线掠入射望远镜的研究和应用越来越广泛。镜片刚体运动误差、表面误差、表面粗糙度、表面散射和衍射是影响X射线成像望远镜成像质量的主要因素。详细分析了刚体运动误差和低频表面误差对成像质量的影响;使用空间解析几何学的二次曲面方程建立WolterI、II型X射线掠入射望远镜镜片的数学模型,降低光线追踪的计算复杂度;借助自主开发的光学仿真软件分析了国内研制的X射线成像望远镜的刚体运动误差和低频表面误差对成像质量的影响,计算出像点艾里斑半径均方根值随误差变化的增长率,对工程上提出合理的制造公差的误差容限有重要的指导意义。