典型有偏估计方法均方误差极小值一致性分析

针对高斯-马尔科夫定理只能保证最小二乘估计在线性无偏估计类中具有最小方差,和目前有偏估计对最小二乘估计的改善程度研究甚少的现状,提出研究典型有偏估计方法的均方误差极小值的一致性。结果表明：尽管不同有偏估计方法具有不同的结构形式,但是这些有偏估计方法对最小二乘估计方差的改善程度是相同的,即均方误差的极小值总是相等的。实例分析结果也表明：典型有偏估计方法的均方误差极小值是一致的。     

统计模型及其参数的LS估计

本文首先整理了统计模型,它包括线性模型、广义线性模型和非线性模型。接着给出了它们的最小二乘估计和高斯-马尔可夫估计。最后讨论了它们的优良性。     

改进LS估计的若干新成果

本文整理了关于改进线性模型(Y,Xθ,σ~2I)参数θ的最小二乘估计的若干新成果,它包括岭型估计、Stein压缩估计、主成分估计、Cassella估计等等。最后给出了杈估计与杈概括。     

关于有偏估计提高测量可靠性的探讨

可靠性是评价化工过程测量数据优劣的重要指标。理论与实践中普遍采用的观测数据平均值或加权平均作为待测参数测量值方法的本质是最小二乘无偏估计,而无偏测量数据可靠性与方差一一对应,根据高斯-马尔科夫定理可知无偏测量方差有下界,所以无偏测量数据不一定具有高可靠性。提出采用有偏估计改善测量数据的可靠性,首先分析未知参数测量与参数估计过程之间的等价性;其次给出同时采用方差和偏差定量表示有偏测量数据可靠性的方法;最后研究偏参数对有偏测量数据可靠性的影响,并采用数值法求解了偏参数的最优值。仿真结果表明偏参数合理取值范围内有偏测量的可靠性总是优于无偏测量的可靠性。     

(LG)型参数估计的若干讨论

在已知(LG)型参数的最小二乘估计与GM估计的基础上,本文提出了改进最小二乘估计的两大估计——Stein型压缩估计与主成分估计。这是线性模型参数估计的重要问题,而改进(LG)型参数的最小二乘估计的专题研究,至今还是一个有待进一步深入的新论题,本文只是初步结果。     

半参数广义线性模型惩罚似然估计的几个性质

本文在给出了半参数广义线性模型惩罚似然估计及惩罚准似然估计的基础上,得到了下面几个性质:(Ⅰ)极小惩罚似然估计及惩罚准似然估计等价于惩罚加权最小二乘估计;(Ⅱ)估计的收敛性及加速收敛法;(Ⅲ)估计的一种 Bayes 解释。     

PLS回归软测量方法在催化重整稳定油组分估计中的应用

提出扰动分类法和线性部分最小二乘（PLS）回归相结合的建立软测量模型的方法，并将它用于催化重整稳定油组分的估计中。仿真结果表明扰动分类法和PLS回归相结合建立的软测量模型简单、实用。     

基于有偏估计的数据融合提高双传感器测量可靠性

为了提高测量数据可靠性,过程控制领域广泛采用双冗余传感器测量生产状态信息,而目前处理双传感器测量数据常用的方法为线性无偏估计融合理论。通过估计理论得到的测量数据同时涉及方差和偏差两个统计特性。由于无偏测量数据具有无偏性的优良性质,所以测量方差能够全面地描述无偏测量数据的可靠性,然而不能由此认为无偏测量数据一定具有高可靠性。为了进一步提高双传感器测量数据的可靠性,提出有偏估计数据融合方法。首先,证明了有偏测量能够改善单传感器测量数据的可靠性;其次,采用凸组合方法推导了双传感器有偏估计融合表达式;最后,证明了有偏估计融合的均方误差小于任意单传感器的均方误差。仿真分析与实例应用均表明有偏估计数据融合可以有效地提高双传感器测量数据的可靠性。     

基于有偏估计的数据融合提高双传感器测量可靠性

为了提高测量数据可靠性,过程控制领域广泛采用双冗余传感器测量生产状态信息,而目前处理双传感器测量数据常用的方法为线性无偏估计融合理论。通过估计理论得到的测量数据同时涉及方差和偏差两个统计特性。由于无偏测量数据具有无偏性的优良性质,所以测量方差能够全面地描述无偏测量数据的可靠性,然而不能由此认为无偏测量数据一定具有高可靠性。为了进一步提高双传感器测量数据的可靠性,提出有偏估计数据融合方法。首先,证明了有偏测量能够改善单传感器测量数据的可靠性;其次,采用凸组合方法推导了双传感器有偏估计融合表达式;最后,证明了有偏估计融合的均方误差小于任意单传感器的均方误差。仿真分析与实例应用均表明有偏估计数据融合可以有效地提高双传感器测量数据的可靠性。     

脆性材料强度统计分析中Weibull模数估计的试样数量的优化

为了全面的评价脆性材料强度特性指标,研究了强度统计方法中Weibull模数估计的试样数的优化问题。文中利用无偏极大似然估计理论结果,用置信度和相对误差双参数,计算出Weibull模数估计中的最优试样数,以便在Weibull模数估计试验中根据研究对象的要求,正确选用试样数量。文中结果在脆性材料强度特性评价与可靠性评价及其它适合Weibull统计的对象均可应用。     

IDENTIFICATION AND NON LINEAR POLE PLACEMENT SELF TUNING PID CONTROL OF A JACKETED WELL MIXED REACTOR

This paper describes the application of Non Linear Self Tuning PID (NLSTPID) system with the intention of controlling the temperature of a cooling jacketed polymer reactor containing toluene and styrene mixture. The use of polynomial Nonlinear AutoRegressive Moving Average with eXternal input (NARMAX) model related with tank temperature and heat input for nonlinear control was emphasised. The first part of the paper presents an identification algorithm for the construction of polynomial NARMAX and AutoRegressive Moving Average with external input (ARMAX) models. A Pseudo Random Binary Sequence (P.R.B.S) signal was utilised as a forcing function in order to determine the parameters of the models. Levenberg Marquardt algorithm was used to estimate the relevant parameters of NARMAX model. Similar work was carried out for ARMAX model using Bierman, Kalman and Least Square Estimation algorithms. The time response of the tank temperature obtained from computer simulation, identified models and experimental data to a unit step change in manipulated variable were compared. Next, linear and non linear models were used with STPID algorithm to demonstrate the performance of the available control in response to disturbances. All theoretical works were compared with experimental data.     

宽相对分子质量分布聚丙烯技术进展

介绍了相对分子质量分布对聚丙烯(PP)性能的影响以及加宽相对分子质量分布在PP新产品开发中的作用。通过改变聚合工艺或采用新型催化剂加宽PP的相对分子质量来克服普通PP刚性差、热变形温度低、熔体强度差的缺点，提高PP的性能，拓宽了PP应用领域。     

ESTIMATION OF UPPER BOUNDS FOR THE RATES OF ENZYMATIC REACTIONS

A general methodology that allows the estimation of maximum rates of enzymatic reactions is described. For a typical mechanism of an enzymatic reaction, the rate is a function of kinetic parameters which are unknown but required to obey certain constraints. Specifically, the ratio of forward to backward rate constants must be consistent with the equilibrium constant, and the rate of each bimolecular reaction-step must be less than the rate of collision of the two reactant species. If additional information is available on the reaction rate, more constraints can be introduced. By maximizing the rate expression with respect to the kinetic parameters, subject to all applicable constraints, a first-principles upper bound is obtained for the reaction rate. If the reaction rate is actually known, the methodology can alternatively estimate an extremum for the concentration of the enzyme, a substrate, or a product. Simple thermodynamic arguments could also provide bounds for concentrations or the direction (but not the magnitude) of the rate, by examining only the overall transformation of reactants to products and completely ignoring the mechanism. The collision-limit treatment proposed here exploits basic internal characteristics of enzymatic reaction mechanisms to predict better bounds for the concentrations and the thermodynamically allowable maximum magnitude of the rate.     

用计算机分峰法研究复合超细纤维的结晶结构——第Ⅰ报 峰形曲线的表征函数及分峰程序

本文采用C语言在小型机MV3300上编制WAXD图谱分峰程序,算法采用阻尼最小二乘法,并建立峰形表征函数库函数。     

ON THE UNBIASEDNESS PROPERTY OF AIC FOR EXACT OR APPROXIMATING LINEAR STOCHASTIC TIME SERIES MODELS

Abstract. A rigorous analysis is given of the asymptotic bias of the log maximum likelihood as an estimate of the expected log likelihood of the maximum likelihood model, when a linear model, such as an invertible, gaussian ARMA ( p, q ) model, with or without parameter constraints, is fit to stationary, possibly non-gaussian observations. It is assumed that these data arise from a model whose spectral density function either (i) coincides with that of a member of the class of models being fit, or, that failing, (ii) can be well-approximated by invertible ARMA ( p, q ) model spectral density functions in the class, whose ARMA coefficients are parameterized separately from the innovations variance. Our analysis shows that, for the purpose of comparing maximum likelihood models from different model classes, Akaike's AIC is asymptotically unbiased, in case (i), under gaussian or separate parametrization assumptions, but is not necessarily unbiased otherwise. In case (ii), its asymptotic bias is shown to be of the order of a number less than unity raised to the power max { p, q } and so is negligible if max { p, q } is not too small. These results extend and complete the somewhat heuristic analysis given by Ogata (1980) for exact or approximating autoregressive models.     

BP神经网络补偿快速成型系统闭环误差的研究

提出快速成型系统闭环误差反馈的必要性,将BP神经网络引入快速成型技术,建立了快速成型误差补偿模型,探求出一种快速成型系统闭环误差的反馈方法,并通过实例进行了验证。结果表明,基于BP神经网络控制的最大绝对误差和相对误差均比传统误差补偿方法大为降低。     

复合材料层合板强度界限

本文提出了复合材料层合板强度上下限的概念，给出了层合板强度上下限的计算方法，并以复合材料单向板的强度准则的为基础，对一些特殊铺层的层合板在几种受力状态下的强度限进行了讨论。     

ON THE EFFICIENCY OF THE SAMPLE MEAN IN LONG-MEMORY NOISE

Abstract. When estimating the unknown mean of a stationary time series, the best linear unbiased estimator is often a significantly better estimator than the ordinary least squares estimates _n. The relative efficiency of these two estimators is investigated for time series whose spectrum behaves like a power at the origin (e.g., fractional Gaussian noise and fractional ARIMA).     

相对湿度对麻杆水泥复合板蠕变的影响

介绍了麻杆增强水泥复合板,在承受弯曲载荷时,相对湿度对蠕变性能的影响,并建立了一年生植物增强水泥复合板的蠕变方程。     

相对湿度对低阶煤低温氧化过程的影响

采用pulse calorimeter仪器,分别研究了在湿的氮气气氛下和湿的氧气气氛下的相对湿度对低阶煤低温氧化过程的影响.结果表明,在湿的氮气气氛下,相对湿度对凝结热(包括吸附热)有显著的影响,低阶煤低温氧化过程释放的热随着湿度的升高而增加;同样在湿的氧气气氛下,随着湿度的增加体系放出的热明显增加;另一方面,随着温度的增加,体系的凝结热和吸附热减少;而在湿氧气下体系释放的热随温度的增加起始先减少后增加.