An Adaptive Multiscale Ensemble Learning Paradigm for Nonstationary and Nonlinear Energy Price Time Series Forecasting

For forecasting nonstationary and nonlinear energy prices time series, a novel adaptive multiscale ensemble learning paradigm incorporating ensemble empirical mode decomposition (EEMD), particle swarm optimization (PSO) and least square support vector machines (LSSVM) with kernel function prototype is developed. Firstly, the extrema symmetry expansion EEMD, which can effectively restrain the mode mixing and end effects, is used to decompose the energy price into simple modes. Secondly, by using the fine‐to‐coarse reconstruction algorithm, the high‐frequency, low‐frequency and trend components are identified. Furthermore, autoregressive integrated moving average is applicable to predicting the high‐frequency components. LSSVM is suitable for forecasting the low‐frequency and trend components. At the same time, a universal kernel function prototype is introduced for making up the drawbacks of single kernel function, which can adaptively select the optimal kernel function type and model parameters according to the specific data using the PSO algorithm. Finally, the prediction results of all the components are aggregated into the forecasting values of energy price time series. The empirical results show that, compared with the popular prediction methods, the proposed method can significantly improve the prediction accuracy of energy prices, with high accuracy both in the level and directional predictions. Copyright © 2016 John Wiley & Sons, Ltd.     

Exact inference for two exponential populations with competing risks data

In a reliability comparative test, the joint censoring model is usually adopted to evaluate the performances of units with the same facility. However, most researchers ignore the pos- sibility that there is more than one factor for the failure when a test unit fails. To solve this problem, we consider a joint Type-II hybrid censoring model for the analysis of exponential competing failure data. Based on the maximum likelihood theory, we compute the maximum likelihood estimators （MLEs） of parameters and then obtain the condition ensuring MLEs existence for every unknown parameter. Then we derive the conditional exact distributions and corresponding moment properties for parameters by the moment generating function （MGF）. A Monte-Carlo simulation is conducted to compare the performances of different ways. And finally, we conduct a numerical example to illustrate the proposed method.     

碘化N-正丁基氟哌啶醇对大鼠缺氧复氧心肌细胞钠钙交换体电流的抑制作用

碘化N-正丁基氟哌啶醇(N-n-butyl haloperidol iodide,F2)为本研究室改造合成的新化合物。前期研究发现F2作为L-型钙通道拮抗剂,能剂量依赖地拮抗缺血再灌注所导致的大鼠心脏损伤。研究F2对缺氧复氧(hypoxia/reoxygenation,H/R)大鼠心肌细胞钠钙交换体电流的作用并探讨其保护机制。采用Langendorff灌流系统灌流SD大鼠心脏,标准酶解法消化分离得到单个心室肌细胞。正常台式液灌流5min,立即灌流充90%N2-10%CO2的缺氧液,建立体外心肌细胞H/R模型,采用全细胞膜片钳技术记录对照、模型以及不同浓度F2(0.1、1、10μmol/L)对心肌细胞钠钙交换体电流,观察H/R状态F2对心肌细胞钠钙交换体电流的影响。结果显示:缺氧抑制钠钙交换体电流主要是抑制外向电流;H/R引起钠钙交换体电流增大,尤其是外向电流的增大。F2呈浓度依赖地抑制钠钙交换体电流,钠钙交换体电流I-V曲线上移。以上表明:F2能抑制钠钙交换体电流,尤其是外向电流,防止H/R时心肌细胞的钙超载,保护心肌细胞。     

概念精准与逻辑严密视角下对"密度"概念教学思路的再认识

"密度"作为初中物理引入最早、抽象程度最高的概念,历来都是教学的重点和难点,针对"密度"在日常教学中存在的症结,联系学生朴素认识,导入密度物理意义;直接比较物体质量,成功引发认知冲突;找出谬误产生根源,合理修正先前假设;结合具体实验数据,尝试诠释比值内涵;巧妙安排问题顺序,完成密度学习进阶4个环节,并对教学效果进行了相关的实证分析,教学效果很好.     

虚假财务报表的FCM算法识别

财会人员借助财务惯例和财务准则等财务技术手段篡改财务报表,与常规财务数据对照会发现,所篡改的财务数据在财务数据指标方面往往缺少相互关联性,甚至是自相矛盾。同时在报表模式上与常规财会数据也呈现内在差别,但是这些差别人工检验较难。本文采用模糊C均值聚类方法将这些差异放大,从而识别出虚假财务报表。该方法具有识别准确率高和快速的特点,具有理论方法创新意义。     

考虑连续特性的水平井钻柱轴向振动分析

为精确的描述水平井钻柱轴向受迫振动规律,本文以连续性波动理论为基础,考虑了钻柱连续特性、库伦阻尼及钻井液粘性阻尼作用和轴向振动工具对钻柱的位移激励,并采用等效粘性阻尼法对库仑阻尼进行线性化处理,建立了轴向振动工具作用下的水平井钻柱运动的动态解析模型;通过Burnett等人发表的轴向振动工具测试实验验证了本文所建模型,并利用数值模拟分析了轴向振动工具的频率及钻柱与井壁间摩擦系数对水平井钻柱轴向受迫振动的影响。结果表明：在一定范围内增加轴向振动工具的振荡频率可提高水平井钻柱的轴向受迫振动响应;钻柱与井壁间摩擦系数对水平井钻柱轴向受迫振动的影响随着钻柱长度的增加而降低。本文的研究方法和模型可为改善轴向振动工具在水平井作业中的作用提供理论指导。     

磷酸-硫酸活化法制备木屑活性炭工艺

以林业废弃物杨木屑为原料,采用正交试验法探讨以磷酸为主活化剂,浓硫酸为辅助活化剂,在不同工艺条件下制备活性炭,测定其亚甲基蓝脱色力和碘的吸附值,考虑活化因素对活性炭得率和吸附性能的影响,确定最佳工艺参数.试验结果表明:磷酸-硫酸活化法制备木屑活性炭的最佳工艺条件为浸渍比1∶2.5,浸渍浓度60%,活化时间90 min,活化温度550℃.     

基于相关矢量法L型阵列相干信号DOA估计

针对L型阵列提出了一种基于互相关矩阵的相关矢量(cross correlation matrix correlation vector method,CCM-CVM)重构解相干的波达方向(direction of arrival,DOA)估计算法。为了进一步提高估计精度,在此方法基础上又得到一种改进方法,即CCM-MCVM的方法。该方法基于前后向矢量重构理论,利用阵列互相关矩阵不含噪声的特点,把互相关矩阵的每一个列矢量作为前向矢量,通过对其前向矢量元素进行复共轭变换并颠倒顺序得到后向矢量。利用所有的前后向矢量来重构信号的协方差矩阵并提取信号的子空间,该方法相较于CCM-ESPRIT算法具有损失阵列孔径小,估计精度高的特点。理论分析和仿真结果表明了该方法在低信噪比和小快拍数条件下相较于对比算法具更好的估计性能。     

叶绿素含量BP反演模型的光谱信息输入因子构建研究

植被叶绿素含量的高光谱反演是当今研究的热点,传统后向传播(BP)神经网络是其常用的一种反演模型。高光谱数据虽然具有精细光谱分辨率,但也造成了大量的信息冗余与噪声;而小波包变换(WPT)可以有效地抑制高光谱数据噪声和压缩信号,同时主成分分析(PCA)能够很好地降低模型输入因子的维数并可简化网络结构。以盆栽玉米为研究对象,在玉米叶片光谱数据对数变换并一阶微分处理的基础上,针对叶绿素含量的BP反演模型,提出了基于相关系数(CC)、WPT和WPTPCA的输入因子构建方法,并形成了叶绿素含量的CC-BP、PCA-BP及WPT-PCA-BP三种反演模型。通过比较玉米叶片叶绿素含量的实测值与三种BP模型反演结果,表明基于WPT-PCA构建BP模型的输入因子数量虽仅有6个却并不影响其反演精度,也能包含原始光谱的92%信息,且优于基于PCA和传统CC所构建输入因子的BP模型反演能力。