预测

Model 2:Model 1 can predicts various indicators of the regional population, but the parameters trends to uncertainty during the long-term, and the cumulative effect of error is significant.

模型一能得到关于各区域人口的各项指标，但在对人口长期趋势进行预测时，模型一中各参数存在较大的不确定性、且误差的累积效应明显，在进行长期预测时达到人口高峰后一直下降，不再有波动，这与实际历史上的三次高峰不符合，故在模型二中挖掘数据内在的相关信息，利用2001-2005年中国国家统计年鉴数据及模型一中短期预测的人口总数建立灰GM(1，1)模型对我国人口总数进行长期预测。

For the limitations of the traditional algorithms that can not be reasonably used to predict the traffic, this paper adopted the BP neural network and curvilinear regression coupling algorithms to add the inadequate.

摘 要：针对传统交通量预测方法中的局限性，采用BP神经网络与曲线回归耦合算法，用BP神经网络对历史数据训练，利用曲线回归对各因素待测年份之值进行预测，将各因素的预测值带入已训练好的BP神经网络中，即可得到未来交通量的预测值。

Results Job demands were positively related to emotional exhaustion, job resources were negatively related to depersonalization and positively to reduced personal accomplishment.

结果工作要求对情绪衰竭有正向预测作用，工作资源对情绪衰竭和去人性化有负向预测作用、对个人成就感有正向预测作用；工作资源对工作投入的3个维度都有正向预测作用。

Based on the research work of basic theories including phase space reconstruction,embedding theorem,correlation dimension,local dynamics,Lyapunov exponents,surrogate data etc,based on the research work of general methods such as principal component analysis,correlation dimension GP algorithm,false neighbors method,nonlinear time series prediction,local prediction, adaptive prediction,neural network model,support vector machines regression model, prediction power,nonlinear detection,coarse-graining methodology,conditional entropy and so on,the framework of nonlinear time series analysis are constructed.

在对包括相空间重构、嵌入定理、关联维数、局部动力学、Lyapunov指数、替代数据(来源：A97BC论文网www.abclunwen.com)、等基本理论与其物理意义的研究和讨论基础上；在对包括主分量分析、关联维数GP算法、伪邻近点法、非线性时间序列预测、局域预测、自适应预测、神经网络模型、支持向量回归模型、预测效果、非线性检测、粗粒化方法、条件熵等非线性时间序列一般分析方法的原理和算法研究基础上；构建了新的非线性时间序列分析的理论体系，归纳总结了非线性时间序列分析的基本问题和主要研究方面。2。

The results demonstrate that: ARMA(1,1) model,auto-regressive model with GDP as exogenous variable,auto-regressive model with policy as exogenous dummy variable,and logarithm linearity and removal average model are all valid models for forecasting tax revenue.

结果表明：ARMA(1，1)模型中，以GDP为外生变量的自回归模型、以政策因素为虚拟外生变量的自回归模型以及对数线性移动平均模型都是预测税收收入的有效模型，但以GDP为外生变量的自回归模型在预测2005年税收收入时，预测值与实际值的预测偏差仅有1。

In this paper the structure principle of GM(1, 1) was analyzed and the defect was pointed out that the traditional model only adapt to the data with exponential law, and only can fit for the monotonic data. The prediction is defected when there are two consecutive data unchanged, and 'a' is equal to 0.5 in prediction formula.

从传统GM(1，1)预测模型构造原理出发分析其存在的理论缺陷，此模型适用于具有较强指数规律的序列，只能描述单调变化过程，而且预测数据序列中出现连续两个相等值时将使得a等于0，会使预测数列发散致使预测失效。

Besides, the database of models andalgorithms has been suggested here and been shared by data mining engine and functionsimulating. The idea of disjunction between models and algorithms is proposed in acreative way. With it, much efficiency, flexility and resource saving will be made.

在预测模式和模型算法的选择上，本文也提出了一种新的策略，运用不同预测模式下，各种模型算法组合的历史测试记录和预测记录，为预测模式和模型算法的选择提供参考。

It was the first time the sequence stratigraphy study had been done all across Naimanqi Depression. Jiufotang Formation of Naimanqi Depression was divided into two third-class sequences (SQ1 and SQ2) and five system tracts (SQ1 has LST, TST and HST; SQ2 has TST and HST), and in High System Tract of SQ2, two fourth-class sequences were defined (Ssq1 and Ssq2).2. According to the analysis of core and logging data, we considered that it is the delta fan facies and established the electrofacies model.3. The distribution of reservoir sand bodies had been predicted in the areas without well data by combining the methods of sequence stratigraphy and reservoir inversion. In addition, we summarized the method and process of combining geology and geophysics to predict reservoirs and we made a good prediction by applying it.4. Combining with sequence and depositional types, sandstone thicknesses and reservoir properties, based on Petrel Software platform, we used fuzzy mathematics to comprehensively evaluate reservoirs and define favorable areas. We classified the reservoirs in Naimanqi Depression into five types (TypeⅠ, TypeⅡ, TypeⅢ, TypeⅣand TypeⅤ) and pointed out four favorable areas: Block Nai 1, Block of western Naican 1-Zhangjia Depression, Block Nai 6 and Block of Xinfa-Zhangjia Depression.

其主要创新点如下：1、首次在整个奈曼旗凹陷范围内进行了层序地层学的研究，将奈曼旗凹陷九佛堂组地层分为2个三级层序(SQ1和SQ2)和5个体系域(SQ1由LST、TST和HST组成，SQ2由TST和HST组成)，并在SQ2的高位体系域中又识别出两个四级层序(Ssq1和Ssq2)。2、根据岩芯分析和测井资料识别，定义了研究区扇三角洲沉积相类型，并建立了测井相模板。3、通过层序地层学的研究，结合储层反演，对奈曼旗凹陷层序格架内没有井的区域进行了储层砂体的空间展布预测，并总结了地质-地球物理综合储层预测的方法与思路流程，在应用中取得了良好的预测效果。4、结合层序、沉积类型、砂岩厚度和储层物性等方面，以Petrel软件为基础平台，利用模糊数学的方法进行储层综合评价及有利区划分，将奈曼旗凹陷储层划分为五个级别(Ⅰ类、Ⅱ类、Ⅲ类、Ⅳ类和Ⅴ类)，并预测了四个有利区块，分别为：奈1区块、奈参1西-张家洼子区块、奈6区块和新发-张家洼子区块

The general orthogonal polynomial is adopted for model approximation of DPS and the predictive control of DPS is translated into the lumped one. Model algorithmic control is applied to deal with the lumped parameter system and a control law is given. The predictive controller of DPS is obtained by inverse transformation. The method and steps are discussed in detail.

该方法采用广义正交多项式对分布参数系统的模型进行逼近，将离散时间分布参数系统的预测控制问题转化为集中参数系统的预测控制问题，运用模型算法控制方法对集中参数系统进行预测控制，求出控制律，经反演变换得出分布参数系统的预测控制。

In order to meet the discharge norm of the contamination concentration limit, zone generalized predictive control is adopted which transforms the conventional quadratic objective of the MPC into a linear programming with soft constraints; besides, considering the wastewater treatment plants energy-saving targets, a multi-objective predictive control is formulated and finally hierarchical multi-objective optimization algorithm is applied for on-line optimization solution.

根据污水处理过程的实际要求，为了满足氮浓度的排放标准区间要求，研究了污水处理过程的区间广义预测控制策略，把预测控制的常规二次目标转化为带软约束问题的线性优化问题；同时考虑到污水处理的能耗，在出水水质目标之上建立了能耗最小目标，得到了污水处理的多目标预测控制，最后采用分层多目标优化算法进行在线优化求解，得到污水处理过程的多目标预测控制。

In the United States, chronic alcoholism and hepatitis C are the most common ones.

在美国，慢性酒精中毒，肝炎是最常见的。

If you have any questions, you can contact me anytime.

如果有任何问题，你可以随时联系我。