α-因子缓冲算子的构造及其性质的研究

为了解决灰色预测模型中的随机挠动序列问题,本文在灰色系统理论以及灰色预测建模技术的基础上,通过梳理历史文献的研究结果,提出了一个具有缓冲作用强度可调整的新型缓冲算子,即α-因子缓冲算子,该算子具有强度可调整、形式统一、预测结果优良等特性.具体来说就是通过调整强度系数可以改变其缓冲作用的强度大小、形式上兼顾了弱化缓冲算子和强化缓冲算子的统一性、在一定范围内的应用中预测结果优于其他的缓冲算子.实证的结果的确显示出了这些优良性质.     

一种新型对数弱化缓冲算子的构造及其应用

缓冲算子是灰色预测研究中的重要内容.通过对现有缓冲算子的研究,结合对数函数的性质,提出基于对数函数的新型对数弱化缓冲算子和变权对数弱化缓冲算子;从理论上证明其为弱化缓冲算子,同时研究了关于序列调节度和改变序列光滑性方向的性质.通过实例验证,所提出的对数缓冲算子能够提高预测模型的精度.     

基于残差GM(1,1)模型的强(弱)化缓冲算子还原模型

残差GM(1,1)模型预测效果相对于GM(1,1)模型较好,但是其要求残差尾段符号一致的自身缺陷常常存在,在实际工作中难以运用,需要解决其自身缺陷,故本篇提出新的模型,即基于残差尾段的强(弱)化缓冲算子还原模型.在残差GM(1,1)模型的基础上,以残差尾段序列作为原始数据,判断其是否满足灰色建模条件,如满足,则直接建模;如不满足,需要对其进行序列算子强(弱)化处理,进行GM(1,1)建模,之后进行强(弱)化缓冲算子还原.以实例为证,最终结果表明强(弱)化缓冲算子还原模型的预测精度稍有提高,且解决了自身缺陷和允许序列不符合灰色建模(诱发缺陷)的情况出现.     

基于变权缓冲算子的灰色G(1,1)模型在地铁能耗预测中的应用

准确预测城市轨道交通系统能源消耗量对城市轨道交通建设规划、环境保护及节能减排具有重要意义.城市轨道交通系统能源消耗量逐年上升且呈指数型变化趋势,因而适合用灰色理论G(1,1)模型描述其变化规律.为了提高模型的预测精度,采用了变权缓冲算子对历史建模数据进行预处理.实例分析表明,经遗传算法权值优化的强化缓冲算子明显增强了建模数据的指数变化特征,提高了灰色G(1,1)模型的预测精度,使预测结果更贴近实际.     

分数阶灰色累加生成算子与累减生成算子及互逆性

针对现有灰色预测模型主要以一阶累加生成序列作为建模序列,再累减还原为原始序列预测值,本文通过Gamma函数将累加生成算子和累减生成算子拓展到正实数领域,给出分数阶累加生成算子和分数阶累减生成算子的解析表达式,一阶和整数阶均是其特例,证明了两算子之间的互逆性.为建立分数阶灰色预测模型和拓宽灰色预测模型的应用范围提供理论基础.     

基于GFM-VAR的我国通用航空产业演进阶段识别及政策强度有效性分析

为揭示通用航空产业系统演化的动态和有序规律,挖掘产业形成和可持续发展的阶段识别证据,针对产业演化数据的信息不完全和连续聚类特性,基于灰色系统理论和最优分割理论构建通用航空产业演进阶段识别的灰色生成序列最优分割模型(Grey sequence generation-Fisher Model,GFM),并运用VAR模型分析通航产业政策强度的有效性。首先运用灰色系统理论的(位置加权)几何平均强化缓冲算子(W)GASBO((Weight)Geometry average strengthening buffer operator)扩张演化数据波形特征并生成灰色序列矩阵;运用最优分割理论的组内离差平方和最小判断原则判定通用航空产业演化的有序递进阶段;然后,通过我国2004~2013年通用航空产业演进阶段的实证研究,发现:我国通用航空产业演化以2010年为分界点,先后经历了初创和成长两个阶段;基于VAR模型的脉冲响应函数确定我国通航产业政策强度受“倒逼出台”、“时滞限制”的影响呈弱有效性;基于政策强度(WGASBO算子)的灰色生成序列最优分割模型更有效、更具有可行性。     

GM(1,1)模型在高校选修课开课规模中的应用

针对多所高校普遍存在的选修课供求比例失衡、课程资源配置不够合理等问题,笔者以浙江某高校的历史选课数据为例,利用GM(1,1)灰色预测模型对高校选修课开课数量及其规模进行了预测,并对比了四种模型的预测误差,结果表明二阶算子作用序列下的预测数据可信度较高,并以此为基础讨论了不同类型选修课布局及课程计划管理.     

基于分数阶累加的灰色GM(1,1|sin)模型及其应用

本文研究了灰色模型对振荡序列的预测问题.在已有GM(1,1|sin)模型的基础上,利用分数阶算子对原始序列进行累加生成的方法,获得了分数阶累加GM(1,1|sin)模型的表达式;以平均相对误差最小化为目标,利用粒子群算法求解非线性优化问题,获得了模型的最优参数.最后以城市交通流的模拟预测为例,结果表明本文提出的模型比GM(1,1|sin)模型具有更高的模拟精度,推广了GM(1,1|sin)预测模型的结果.     

南四湖人湖河流水质面板数据分析的灰色速率模型

针对面板数据灰色关联决策评价模型的关联度计算问题,在传统的面板数据灰色关联决策评价模型的基础上,构建评价对象的指标时间数据的累加序列,通过指标时间累加生成速率序列构造面板数据灰色生成速率关联决策模型,采用生成速率序列的接近性表征原始数据序列的动态变化趋势.通过灰色累加生成速率关联决策方法扩展到面板数据分析,解决了小样本面板数据的灰色评价分析问题,将这种方法应用于南四湖入湖河流水质面板数据质量评价中,经实例计算验证了面板数据灰色累加生成速率关联决策评价模型的稳定性、合理性和实用性为水环境面板数据质量分析评价提供了可行的计算思路和方法.     

分数阶灰色累减生成算子及其性质研究

给出了分数阶灰色累减生成算子的详细推导过程,并证明了分数阶灰色累减生成算子的不动点定理、信息优先原理、交换律与指数律,为分数阶灰色预测模型提供了理论基础.算例验证了分数阶灰色累减生成算子的特征,在灰色预测模型GM(1,1)中的应用证明了分数阶灰色累减生成算子的有效性.     

A new method to mitigate data fluctuations for time series prediction

Although the classic exponential-smoothing models and grey prediction models have been widely used in time series forecasting, this paper shows that they are susceptible to fluctuations in samples. A new fractional bidirectional weakening buffer operator for time series prediction is proposed in this paper. This new operator can effectively reduce the negative impact of unavoidable sample fluctuations. It overcomes limitations of existing weakening buffer operators, and permits better control of fluctuations from the entire sample period. Due to its good performance in improving stability of the series smoothness, the new operator can better capture the real developing trend in raw data and improve forecast accuracy. The paper then proposes a novel methodology that combines the new bidirectional weakening buffer operator and the classic grey prediction model. Through a number of case studies, this method is compared with several classic models, such as the exponential smoothing model and the autoregressive integrated moving average model, etc. Values of three error measures show that the new method outperforms other methods, especially when there are data fluctuations near the forecasting horizon. The relative advantages of the new method on small sample predictions are further investigated. Results demonstrate that model based on the proposed fractional bidirectional weakening buffer operator has higher forecasting accuracy.     

Multi-variable grey model based on dynamic background algorithm for forecasting the interval sequence

The multi-variable grey model based on dynamic background algorithm improves the forecasting performance of the multi-variable grey model on the precise number sequence. In order to make this model suitable for the interval sequence, the matrix form of the multi-variable grey model based on dynamic background algorithm is proposed in the paper. In the modeling process, the interval is treated as a two-dimensional column vector, the parameters of the multi-variable grey model are replaced by matrices, and the dynamic background algorithm for interval sequences is proposed. The analysis results of the matrix algorithm for the dynamic background value and the prediction formula show that the new model is essentially a way to predict one of the two bounds of an interval by combining them, reflecting the integrity and interaction between the lower and upper bounds. The interval predictions of industrial electricity consumption of Zhejiang Province, China national electricity consumption and consumer price index show that the new model can well predict the minimum and maximum values of the interval sequence and has better prediction performance compared with the method of predicting each boundary sequence separately.     

“奖优罚劣”的区间数多指标决策的关联分析法

针对区间数多指标系统的决策特点,对指标数据初始化处理时,利用“奖优罚劣”原则,提出了一种易于计算且实用的[-1,1]线性变换算子,然后定义正、负理想方案,结合灰色关联分析方法,建立一种新的区间数多指标的灰色关联决策模型.该模型为区间数多指标决策提供了一种科学、实用的方法,并利用现有的实例来证实此方法的科学性与可行性.     

基于非等时距加权灰色模型与神经网络的组合预测算法

非等时距预测算法在不等时间间隔序列的趋势分析与预测方面具有重要作用．在传统灰色预测理论的基础上,提出一种基于非等时距加权灰色模型和神经网络的组合预测算法．通过构建非等时距加权灰色预测模型,将原始数据序列的平均值作为累加序列初值,将连续累积函数的积分面积作为背景值,对累加序列进行加权处理,以真实反映时间序列发展对预测结果的影响．在此基础上,引入BP神经网络对灰色预测的残差序列进行修正,进一步提高了预测精度．经算例验证,该算法预测精度达到1级,且高于类似算法．     

An extragradient-type algorithm for non-smooth variational inequalities

In this paper we study extragradient-type methods to solve variational inequality problems involving maximal monotone point-to-set operators. We will show, using an example, that, in this case, to achieve convergence to the solution set of sequences generated by extragradient methods it is necessary, in the search of the auxiliary point and its suitable image by the involved operator, to enlarge the set of possible search directions. Such an enlargement, a generalization of the ε-subgradient, is known and we use it to overcome this failure of the extragradient-methods. We propose a new algo-rithm and prove that it generates a sequence convergent to a solution point of the problem also when the operator is point-to-set     

Mann-Type Steepest-Descent and Modified Hybrid Steepest-Descent Methods for Variational Inequalities in Banach Spaces

In this paper, we propose three different kinds of iteration schemes to compute the approximate solutions of variational inequalities in the setting of Banach spaces. First, we suggest Mann-type steepest-descent iterative algorithm, which is based on two well-known methods: Mann iterative method and steepest-descent method. Second, we introduce modified hybrid steepest-descent iterative algorithm. Third, we propose modified hybrid steepest-descent iterative algorithm by using the resolvent operator. For the first two cases, we prove the convergence of sequences generated by the proposed algorithms to a solution of a variational inequality in the setting of Banach spaces. For the third case, we prove the convergence of the iterative sequence generated by the proposed algorithm to a zero of an operator, which is also a solution of a variational inequality.     

A proximal method with logarithmic barrier for nonlinear complementarity problems

We study the proximal method with the regularized logarithmic barrier, originally stated by Attouch and Teboulle for positively constrained optimization problems, in the more general context of nonlinear complementarity problems with monotone operators. We consider two sequences generated by the method. We prove that one of them, called the ergodic sequence, is globally convergent to the solution set of the problem, assuming just monotonicity of the operator and existence of solutions; for convergence of the other one, called the proximal sequence, we demand some stronger property, like paramonotonicity of the operator or the so called “cut property” of the problem.     

改进灰导数和时间响应函数的新灰色预测模型

以白化方程为基础,利用梯形公式白化灰导数,同时根据最小二乘法求出时间响应函数中的常数c,得到一种改进的灰色预测模型.应用实例说明改进模型具有较高的预测和拟合精度.