高速铁路日常客运量的EMD-Informer组合预测方法

铁路客流需求的科学预测是进行运输组织方案决策的重要依据。以高速铁路历史客票数据为基础,结合经验模态分解(empirical mode decomposition,EMD)与机器深度学习中的注意力机制,提出高速铁路日客流量的EMD-Informer组合预测方法。首先采用EMD方法分解高速铁路客流量序列,获得具有周期特征和线路客流内在特征的模态分量,再利用Informer模型分别训练和预测各模态分解分量,并通过多头注意力机制高效挖掘客流数据内在规律和捕捉数据序列中的关键特征,在此基础上重组各分量预测值,从而得到高速铁路日常客流的整体高精度预测值。同时,根据结合问题特征的大量实验,明确可供实际运用参考的超参数设置规则。基于京沪高速铁路全线的实例计算分析表明,相对对比预测方法,EMD-Informer组合预测方法在高速铁路客流的单步预测及超前预测上均具有明显更小的预测误差。

Information Entropy Squeezing for a Atom in Mode-Mode Competition System

The entropy squeezing properties for a two-level atom interacting with a two-mode field via two differentcompeting transitions are investigated from a quantum information point of view.The influences of the initial state of thesystem and the relative coupling strength between the atom and the field on the atomic information entropy squeezingare discussed.Our results show that the squeezed direction and the frequency of the information entropy squeezing canbe controlled by choosing the phase of the atom dipole and the relative competing strength of atom-field,respectively.We find that,under the same condition,no atomic variance squeezing is predicted from the HUR while optimal entropysqueezing is obtained from the EUR,so the quantum information entropy is a remarkable precision measure for theatomic squeezing in the considered system.

Fast Mode Decision Method Based on Historical Information for H.264/AVC

Mode decision based on rate-distortion optimization algorithm brings high compression efficiency to H.264/AVC. However, heavy computation load is also added to the encoder at the same time. In order to reduce the computation burden of mode decision, this paper presented a fast mode decision method based on mode information of the previously coded frame. Moreover, all coding modes were activated when scene change occurs and a scheme to detect scene change was proposed. The simulation results show that compared to the original encoder, the proposed method achieves a reduction of over 38% encoding time and keeps almost the same PSNR and bitrate for a wide range of quantization parameter.

Research on the Relationship Between Information Fusion Method and Information Failure Mode in Integrated Navigation System

On the basis of the basic principles of weighted fusion, Kalman filtering and BP neural networks, the basic principles of information fusion methods used in integrated navigation systems are expounded. Through the analysis of the basic principles, the as-sociation of information fusion methods commonly used in integrated navigation systems and information failure modes is obtained: the information fault mode of weighted fusion method The model is closely related to the specific weight allocation method, which depends on the fault mode of the sensor or sub-system in which the weight is dominant;the information fault mode of the Kalman filtering information fusion method is a continuous mutation fault corresponding to the nonlinear time interval of the system;the in-formation fault mode of the BP neural network method is gradual with time. The information failure mode of the BP neural network method is a slowly varying fault that gradually accumulates over time. Starting from the complexity associated with the information fusion method and the information failure mode, it is pointed out that in order to systematically express the relationship between the information fusion method and the information failure mode, further research can be carried out.

Research on the Multi-Interactive Intelligent Teaching Model of College English in the Information Technology Environment

The multi-interactive teaching mode based on the information technology is a new teaching mode, which promotes students' autonomous learning and research-oriented learning through the online and offline effective interaction in and out of class, fully embodying the principle of the learning-orientation. Based on this, this paper focuses on the connotation and the theoretical basis o f the multi-interactive teaching model, as well as its implementation practices and characteristics, with a view o f providing help to improve the quality of the college English teaching.

Construction project investment control model based on instant information

Change of construction conditions always influences project investment by causing the loss of construction work time and extending the duration. To resolve such problem as difficult dynamic control in work construction plan, this article presents a concept of instant optimization by ways of adjustment operation time of each working procedure to minimize investment change. Based on this concept, its mathematical model is established and a strict mathematical justification is performed. An instant optimization model takes advantage of instant information in the construction process to duly complete adjustment of construction; thus we maximize cost efficiency of project investment.

Exploration and Practice of "One Three One" Talent Training Mode Focuses on the Intelligent Hardware Industry of Electronic Information Science and Technology Major

China's intelligent hardware industry is gradually mature,the upstream and downstream resources of the industrial chain are increasingly rich,and the software and hardware system is gradually improved,forming a complete industrial chain of producing,supplying and selling intelligent hardware.The smart hardware market consists of six segmented areas,such as intelligent medical and health equipment,intelligent vehicle-mounted equipment,intelligent service robot,and intelligent industry,which has become the intersection of information technology and traditional industries.Intelligent hardware engineering and technical personnel are the important support for Chongqing to implement the"innovation-driven development strategy action plan led by big data intelligence".Taking Electronic Information Science and Technology Major of Chongqing University of Arts and Sciences as an example to focus on intelligent hardware industry"One Three One"talent training mode,namely,institute of intelligent hardware platform construction,teaching staff,teaching resources construction and practical conditions,intelligent hardware training mode to explore and practice,for the similar construction and reform of colleges and universities related to provide reference and reference.

基于MIC-EEMD-改进Informer的含高比例清洁能源与储能的电力市场短期电价多步预测

随着电力现货市场的开展,短期电价预测对于各市场主体的决策有着重要意义,而高比例清洁能源与储能的不断接入给短期电价预测带来很大挑战。提出一种基于最大信息系数法(maximum information coefficient,MIC)、集成经验模态分解(ensemble empirical mode decomposition,EEMD)和改进Informer的短期电价多步预测模型。首先,采用MIC分析出与电价相关性较高的几类因素作为模型原始输入序列;然后,将上述原始序列进行EEMD分解后得到多条本征模函数(intrinsic mode function,IMF)和一个残余项后输入改进Informer分别得到翌日24点多步预测结果,再对预测结果进行滤波;最后,将滤波后序列的预测结果叠加得到最终的预测值。以西班牙电力市场数据进行验证,实验结果证明该模型可以有效提高电力市场短期电价多步预测精度。

基于STAGCN-Informer时空组合模型的风电功率预测方法

针对风电功率预测中,空间信息受时空波动性和随机性影响无法有效提取,导致预测精度不足的问题,提出一种基于变分模态分解(Variational Mode Decomposition,VMD)、融合时空注意力图卷积网络(Spatiotemporal Attention Graph Convolutional Network,STAGCN)和改进Informer的组合模型(STAGCN-Informer-DCP)。首先运用VMD对原始特征进行模态分解,提取出不同时间尺度上的特征信息。同时利用北方苍鹰算法(Northern Goshawk Optimization,NGO)优化VMD的核心参数(惩罚因子和K值)选择。其次,利用融合时空注意力的STAGCN模块动态捕捉目标风机与近邻相似风机的时空特征,并将其与原始的信号分量融合获得携带空间尺度信息的特征向量。最后使用改进的Informer模型提取时序上下文的长期依赖关系,并实现多步输出预测。实验结果表明,该组合模型能较好地捕捉动态时空依赖,并有效提高了中长期风电预测的准确度。

The Discussion of English Course Teaching Mode under the Information Circumstance

This article mostly discusses the classifications, characteristics and functions of the information teaching circumstance. Compared with the characteristics and disadvantages of traditional English teaching mode, taking the modern education technology and constructivism theory as the guide, discusses how to construct the new English course teaching mode under the information circumstance.

基于Information Commons的图书馆资源整合与服务创新

介绍了Information Commons的由来与发展,并从宏观上分析它的具体实践活动,包括创造共用、软件共有、研究共有、学习共有等。接着以IC在图书馆的应用为基础,探讨IC的出现给图书馆资源整合、服务理念及服务模式带来的创新。

Information Commons：高校图书馆读者服务的创新模式

Information Commons(信息共享空间)是20世纪90年代末,国外高校图书馆为适应用户需求的变化和研究的需要而建立起来的一种基础设施和新的服务模式,本文先介绍了IC的概念及特点,然后从几个方面介绍了国外图书馆IC的基本情况,最后论述了我国高校图书馆构建IC的策略及注意事项。

二次分解策略组合Informer的短期电力负荷预测方法

针对电力负荷数据存在的波动性、非平稳性而导致预测精度低的问题,提出一种具有二次分解重构策略的深度学习电力负荷预测模型。首先,对负荷数据进行基于局部加权回归的周期趋势分解(STL)-改进的自适应噪声完备集合经验模态分解(ICEEMDAN)二次分解,通过计算样本熵和最大信息数对分量进行重构;然后在Informer模型中引入非平稳性机制,并融合卷积神经网络对重构分量进行预测;最后,将各分量的预测结果线性相加,得到最终预测结果。实验结果表明,所提方法在3个评价指标上的预测误差均低于所对比模型,证明该预测方法可以有效降低数据的非平稳性并提高预测精度。

Fuzzy BC-k-modes:一种分类矩阵对象数据的聚类算法

传统的聚类算法主要对具有单值属性的数据进行聚类研究,针对矩阵对象数据的研究较少,提出一种新的fuzzy between-cluster k-modes(简称Fuzzy BC-k-modes)聚类算法。在Fuzzy BC-k-modes算法中,采用增加簇间信息(不同类中的对象到其他类中心的距离)去修正目标函数,在对修正的目标函数寻求局部最优解时,提出隶属度矩阵的更新公式。最后,在四个真实数据集上验证了Fuzzy BC-k-modes算法的有效性,并且分析了模糊因子与隶属度间的关系。

Integration technology research on marine geographical information system

It is urgent and necessary to integrate a marine geographical information system （MGIS） with marine remote sensing detection modules. On the basis of the current technology and features of applications, an open three-layer integration framework is designed. At the data layer, a two-level three-base integration mechanism based on the plug-in technology is applied; At the function layer, an integration mode based on API, DLL, EXE and COM is discussed; and at the application layer, a sharing mechanism based on the clients/service is adopted. As an example, the remote sensing integrated application information system of China＇s coastal zone and offshore （MaXplorer1. 0） with muhiecology remote sensing fusion and assimilation module, surge detection module as well as eight other thematic application modules is integrated, and the key technology of integration is discussed at different layers and in different modules. The result shows that it is possible to realize the conformity of technology and resources and to provide the incorporate technology platform for marine information operational functioning after applying the integration framework.

基于VMD-Informer-BiLSTM模型的超短期光伏功率预测

由于光伏出力的波动性和随机性给电网的规划和运行带来了挑战,因此提高光伏功率预测的精度对提升新能源系统的稳定性具有重要意义。为此提出了一种结合模态分解、多维特征建模、Informer、双向长短期神经网络(bidirectional long short-term memory network,BiLSTM)的超短期光伏组合预测模型。首先通过变分模态分解将光伏功率序列分解成不同频率的本征模态函数(intrinsic mode function,IMF),降低光伏功率信号的非平稳性与复杂度;随后使用离散小波变换提取天气因素中的细节分量,实现不同分解算法的优势互补,并用随机森林算法为每个IMF筛选冗余特征,然后将特征矩阵送入Informer进行建模,提取不同时间步中关键时刻的信息,提高对长时间序列的预测效率;最后为进一步提高模型预测精度,分析误差序列特性,利用BiLSTM进行误差校正。采用实际光伏数据进行算例分析,结果表明所提方法提高了超短期光伏功率预测精度。