Selecting the forgetting factor in subset autoregressive modelling

Abstract. Conventional methods to determine the forgetting factors in autoregressive (AR) models are mostly based on arbitrary or personal choices. In this paper, we present two procedures which can be used to select the forgetting factor in subset AR modelling. The first procedure uses the bootstrap to determine the value of a fixed forgetting factor. The second procedure starts from this base and applies the time-recursive maximum likelihood estimation to a variable forgetting factor. In one illustration using real exchange rates, we demonstrate the effect of the forgetting factor in subset AR modelling on ex ante forecasting of non-stationary time series. In a second illustration, these two procedures are applied to time-update forecasts for a stock market index. Subset AR models not including a forgetting factor act as a set of benchmarks for assessing ex ante forecasting performance, and consistently improved forecasting performance is demonstrated for these proposed procedures.     

Particle filtering for sensor fault diagnosis and identification in nonlinear plants

We propose a novel method for sensor monitoring and fault-tolerant estimation in systems described by general stochastic nonlinear and/or non-Gaussian state-space models. Faults are defined as abruptly occurring calibration errors, causing the sensor readings to be biased or scaled. Actuators and the process itself are assumed to be fault free. The main novelty of the work is an adaptive particle filter, whose configuration changes in order to diagnose sensor faults and to compensate for their effects. The presence, type and magnitude of sensor faults are determined through hypothesis testing and maximum likelihood estimation, based on the difference between the measurements and the particle filter estimates. The validity of the proposed approach was demonstrated through simulations on a drum-boiler model, although its effectiveness is not conditioned on any particular feature of the considered example.     

无约束非线性lp问题的区间极大熵方法

针对信号处理、系统识别等领域中涉及到的无约束非线性lp问题,为减小由于二进制编码的舍入误差对该问题计算结果的影响,对求解该问题的极大熵方法进行了区间扩张.证明了区间扩张后的极大熵函数至少具有二阶收敛性,并设计了具有多项式时间复杂度的区间算法进行求解,举例进行了数值计算.数值计算结果显示,该区间算法可靠,计算结果与区间扩张前相比,结果更加精确.     

云数据库中基于极大熵差分进化的负载评估算法

由于分布式关系型数据库基于两阶段提交协议的设计方式,使得系统如出现单节点瓶颈问题,数据库事务将全部回滚,从而造成巨大的系统开销,影响数据库在大数据环境下的应用。针对这一现状,提出一种基于极大熵差分进化的负载评估算法,利用评价函数法,将多目标优化问题转化为不可微的单目标优化问题,再利用极大熵函数,将不可微优化问题转化为一个带有参数的无约束优化问题,最后用差分进化算法对其进行求解,找出节点资源最优集,从而为过载节点的数据迁移提供了理论依据,也进一步实现了对云数据库的设计。实验结果表明,该算法能够提高系统的整体性能,有效避免单节点瓶颈问题。     

机械激活矿渣水泥

用于式高能振动磨机械激活矿渣（炉渣和／或硅灰）水泥。研究表明：经机械激活后．矿渣水泥的颗粒粒度降低，颗粒的比表面积增大。在掺合了矿渣的水泥中硅微粒聚团经机械激活后可有效地分散并作用于水泥中．使矿渣水泥的水化反应括性增强，促进水泥水化反应，降低水泥制品的孔隙度．从而可以提高水泥制品的力学性能。通过Fourier变换红外光谱和核磁共振等方法分析了机械激活作用的机理。     

一种检测单晶炉热场温度的新方法

为去除单晶炉热场温度测量信号中的低频干扰、提高温度信号的估计精度,本文提出了一种基于跨维模拟退火(trans dimensional simulated annealing,TDSA)的单晶炉热场温度检测方法.该方法首先将AIC(Akaike information criterion)信号个数判断准则纳入采样机制中以便对干扰个数采样估计,进而利用所设计的基于信号频谱和轮盘赌思想的Metropolis Hastings机制对相应个数的干扰频率进行采样,最后设计了混合均匀和高斯采样机制对单晶炉热场温度进行采样估计.仿真和单晶炉工程实验结果表明,该方法在干扰个数未知的情况下,能够有效地抑制低频干扰、准确检测单晶炉热场的温度.     

基于GPGPU的实体表面实时提取

为实时提取三维实体表面,提出一种基于GPGPU并行计算的实体表面实时提取方法。在分析深度剥离算法原理和GPU图形绘制管线的基础上,给出在GPU上利用深度剥离算法实现实时提取三维实体表面的算法;通过OpenGL的高级着色语言GLSL控制GPU的图形绘制管线实现了该算法,给出其伪代码。以龙、叶轮和刀具扫描体的模型为应用实例验证了该算法效果良好,特别是对于刀具扫描体表面的提取,可满足实时性要求。     

单向Cf/SiC复合材料的弯曲疲劳性能

对单向Cf/SiC复合材料进行了三点弯曲疲劳性能测试,得到了复合材料的应力－寿命曲线（S－N曲线）,并对其进行线性拟合,得到疲劳最大应力与复合材料疲劳寿命的关系;考察了疲劳过程中刚度下降和疲劳裂纹产生情况。结果表明在疲劳过程中复合材料的弯曲模量有3个变化阶段：首先在疲劳加载初期,弯曲模量的下降速度及幅度都较大;其次在弯曲模量下降到原始弯曲模量的85％（133GPa)后,其变化方式没有明显的规律可循,有时甚至可能上升;最后复合材料发生疲劳断裂时,模量将发生突变。显微结构分析表明：基体横向裂纹群的产生是疲劳断裂的独有特征。它的产生是由于基体SiC的断裂应变小于碳纤维的断裂应变,基体首先开裂并导致应力重新分布的结果。     

FAST LINEAR ESTIMATION METHODS FOR VECTOR AUTOREGRESSIVE MOVING-AVERAGE MODELS

Abstract. Three linear methods for estimating parameter values of vector auto-regressive moving-average (VARMA) models which are in general at least an order of magnitude faster than maximum likelihood estimation are developed in this paper. Simulation results for different model structures with varying numbers of component series and observations suggest that the accuracy of these procedures is in most cases comparable with maximum likelihood estimation. Procedures for estimating parameter standard error are also discussed and used for identification of nonzero elements in the VARMA polynomial structures. These methods can also be used to establish the order of the VARMA structure. We note, however, that the primary purpose of these estimates is to generate initial estimates for the nonzero parameters in order to reduce subsequent computational time of more efficient estimation procedures such as exact maximum likelihood.     

最小费用最大流问题的一种新算法

现有的最小费用最大流算法都有自身的缺陷,增广链的选取不当会给计算带来不便,同时费用也达不到理想的效果。鉴于对最小费用最大流算法的增广链选取和最小费用的探索,文章通过对费用差的定义给出了一种求最小费用最大流的新算法。新算法的原则是优先选择费用差最小的有向路径进行增广,当费用差相同时就选择修正后的路径。通过对最小费用最大流算法的改进,新算法易理解且便于计算。通过实例说明了新算法的有效性和执行效率。