Data-driven optimal operation of the industrial methanol to olefin process based on relevance vector machine

Methanol to olefin(MTO)technology provides the opportunity to produce olefins from nonpetroleum sources such as coal,biomass and natural gas.More than 20 commercial MTO plants have been put into operation.Till now,contributions on optimal operation of industrial MTO plants from a process systems engineering perspective are rare.Based on relevance vector machine(RVM),a data-driven framework for optimal operation of the industrial MTO process is established to fully utilize the plentiful industrial data sets.RVM correlates the yield distribution prediction of main products and the operation conditions.These correlations then serve as the constraints for the multi-objective optimization model to pursue the optimal operation of the plant.Nondominated sorting genetic algorithmⅡis used to solve the optimization problem.Comprehensive tests demonstrate that the ethylene yield is effectively improved based on the proposed framework.Since RVM does provide the distribution prediction instead of point estimation,the established model is expected to provide guidance for actual production operations under uncertainty.

Monotonic Vector Space Model (Ⅰ): Concepts and Operations

Monotonic vector space (MVS), as a novel model in which there exist some monotonic mappings, is proposed. MVS model is an abstract of many practical problems (such as image processing, system capability engineering, etc.) and includes many useful important operations. This paper, as the first one of series papers, discusses the MVS framework, relative important concepts and important operations including partition, synthesis, screening, sampling, etc. And algorithms for these operations are the focus of this paper. The application of these operations in system capability engineering will be dealt with in the second part of this series of papers.

HORIZONTAL LAPLACE OPERATOR IN REAL FINSLER VECTOR BUNDLES

A vector bundle F over the tangent bundle TM of a manifold M is said to be a Finsler vector bundle if it is isomorphic to the pull-back π^＊E of a vector bundle E over M（[1]）. In this article the authors study the h-Laplace operator in Finsler vector bundles. An h-Laplace operator is defined, first for functions and then for horizontal Finsler forms on E. Using the h-Laplace operator, the authors define the h-harmonic function and ho harmonic horizontal Finsler vector fields, and furthermore prove some integral formulas for the h-Laplace operator, horizontal Finsler vector fields, and scalar fields on E.

On the completeness of eigen and root vector systems for fourth-order operator matrices and their applications

In this paper, we consider the eigenvalue problem of a class of fourth-order operator matrices appearing in mechan- ics, including the geometric multiplicity, algebraic index, and algebraic multiplicity of the eigenvalue, the symplectic orthogonality, and completeness of eigen and root vector systems. The obtained results are applied to the plate bending problem.

Completeness of system of root vectors of upper triangular infinitedimensional Hamiltonian operators appearing in elasticity theory

This paper deals with a class of upper triangular infinite-dimensional Hamilto- nian operators appearing in the elasticity theory. The geometric multiplicity and algebraic index of the eigenvalue are investigated. Furthermore, the algebraic multiplicity of the eigenvalue is obtained. Based on these properties, the concrete completeness formulation of the system of eigenvectors or root vectors of the Hamiltonian operator is proposed. It is shown that the completeness is determined by the system of eigenvectors of the operator entries. Finally, the applications of the results to some problems in the elasticity theory are presented.

Composition Operators with Linear Fractional Symbols on Vector-Valued Bergman Spaces

Let φ and ψ be linear fractional self\|maps of the unit disk D and X a separable Hilbert space. In this paper we completely characterize the weak compactness of the product operators of a composition operation C φ with another one's adjoint C * ψ on the vector\|valued Bergman space B 1(X) for forms C φC * ψ and C * ψC φ.

SE-DEA-SVM evaluation method of ECM operational disposition scheme

Operational disposition of electronic countermeasures(ECM)is a hot topic in modern warfare research.Through fully analyzing the characteristics and shortcomings of the traditional operational disposition scheme,a super-efficient data envelopment analysis support vector machine(SE-DEA-SVM)method for evaluating the operational configuration scheme of ECM is proposed.Firstly,considering the subjective and objective factors affecting the operational disposition of ECM,the index system of operational disposition scheme is established,and we explain the solution method of terminal indexs.Secondly,the evaluation and algorithm process of SE-DEA-SVM evaluation method are introduced.In this method,the super-efficient data envelopment analysis(SE-DEA)model is used to calculate the weight of index system,and the support vector machine(SVM)method combined with the training samples of evaluation index is used to obtain the input-output model of evaluation value of combat configuration.Finally,by an example(obtaining five schemes),we verify the SE-DEA-SVM evaluation method and analyze the results.The efficiency analysis,comparison analysis,and error analysis of this method are carried out.The results show that this method is more suitable for military evaluation with small samples,and it has high efficiency,applicability,and popularization value.

Norms of Dual Complex Vectors and Dual Complex Matrices

In this paper,we investigate some properties of dual complex numbers,dual complex vectors,and dual complex matrices.First,based on the magnitude of the dual complex number,we study the Young inequality,the Hölder inequality,and the Minkowski inequality in the setting of dual complex numbers.Second,we define the p-norm of a dual complex vector,which is a nonnegative dual number,and show some related properties.Third,we study the properties of eigenvalues of unitary matrices and unitary triangulation of arbitrary dual complex matrices.In particular,we introduce the operator norm of dual complex matrices induced by the p-norm of dual complex vectors,and give expressions of three important operator norms of dual complex matrices.

COMPOSITION OPERATORS ON ANALYTIC VECTOR-VALUED NEVANLINNA CLASSES

Let φ be an analytic self-map of the complex unit disk and X a Banach space. This paper studies the action of composition operator Cφ： f→foφ on the vector-valued Nevanlinna classes N（X） and Na（X）. Certain criteria for such operators to be weakly compact are given. As a consequence, this paper shows that the composition operator Cφ is weakly compact on N（X） and Na（X） if and only if it is weakly compact on the vector-valued Hardy space H^1 （X） and Bergman space B1（X） respectively.

基于多变量模式分析的飞行学员脑功能连接的识别研究

目的基于多变量模式分析(multivariate pattern analysis,MVPA)对飞行学员和健康的普通人的大脑功能连接进行有效识别。材料与方法采集了40名已经取得执照的飞行专业在校学生与39名地面专业在校学生的功能磁共振数据。通过网络功能连接分析得到功能连接矩阵作为特征,分别通过最小绝对收缩选择算子(least absolute shrinkage and selection operator,LASSO)算法与独立样本t检验方法对特征降维。使用不同核函数的支持向量机(support vector machine,SVM)进行训练和预测,使用留一交叉验证法进行模型性能评估,最终根据训练后SVM模型中的权重定位对应脑区之间的功能连接。结果使用LASSO特征筛选的线性(linear)核SVM模型准确率为81.82%,敏感度82.05%,特异度81.58%,曲线下面积(area under the curve,AUC)为0.88。核函数对模型准确率的影响不大。模型中右侧中央旁小叶、双侧中央后回、双侧顶下缘角回、右侧梭状回、左侧眶部额中回、左侧顶上回、右侧眶部额下回有较高的权重,模型中的权重集中在感觉运动网络(somatomotor network,SMN)与默认模式网络(default mode network,DMN),分别占用所有权重的25.62%和25.27%。结论结合LASSO算法进行特征筛选的SVM可以对飞行学员大脑进行有效识别,并且有更好的可解释性和更小的过拟合。模型权重信息反映了飞行学员主要在运动能力和感知能力有别于普通人。

低功耗矢量水声信号采集处理系统设计与实现

为进一步降低矢量水听器信号采集处理系统对水下滑翔机平台的续航影响,设计了一种以ARM和DSP为核心控制器的低功耗信号采集处理系统。利用ARM控制器实现系统的多种工作模式,以达到不同状态下的最佳功耗控制。此外,在控制软件中加入必要的优先级设计和流水线设计,以提高系统硬件控制效率,最后利用功率测试仪对系统不同工作模式进行测试。结果表明,低功耗模式下平均功耗约为1.27 W,实时处理模式下平均功耗约为4.29 W。在长时间连续工作情况下,可大幅度降低实时处理模式的工作时间占比,功耗控制效果显著,有效证明了该方案的有效性和优越性。

基于矩阵半张量积解四元数广义Sylvester矩阵方程组

该文利用矩阵半张量积求解四元数广义Sylvester矩阵方程组.首先将实矩阵半张量积运算推广到四元数矩阵,进而利用四元数矩阵半张量积提出四元数矩阵在向量算子下的一些新结论,利用这些结论将四元数矩阵方程组转化为四元数线性方程组,最后转化为实线性方程组,从而得到四元数广义Sylvester矩阵方程组有解的充要条件及通解表达式,并给出其极小范数解.最后通过数值算例说明该方法的有效性.

三芯电缆中间接头压接缺陷多物理场及温度梯度场分析

为解决传统电缆中间接头压接缺陷识别温度敏感性差、环境影响因素多、设备检测精度低等问题,提出了一种基于温度梯度矢量分析的电缆中间接头缺陷识别方法。通过对10 kV三芯电缆开展多场耦合计算,采用矢量分析方法获得了电缆中间接头的温度梯度场和热流密度分布矢量图,形成了电缆中间接头温度梯度场分布规律。结果表明:电缆中间接头表面温度梯度主要受x方向(电缆径向)温度梯度的影响;正常接头表面温度变化率不超过16.6%,其表面温度梯度变化率达49.8%;缺陷接头表面温度变化率不超过21.8%,其表面温度梯度变化率达57.8%,温度梯度场相较温度场有更高的检测灵敏度。开展电缆中间接头温度梯度场分析研究,对电缆接头致热型缺陷运维效率提升有积极意义。

向量值次线性算子的交换子在变指数Herz-Morrey空间上的加权估计

利用变指数A(p(·))权理论及广义BMO范数性质,我们证明了一类向量值次线性算子的交换子在加权变指数Herz-Morrey空间MK^(α)(·)λ_(q)(·)ω上的有界性,其中α(·),p(·)和q(·)均为变指数.

基于AES的改进视频加密测试

为了使得视频加密技术具有更加良好的加密效果,在原本AES加密算法基础上进行了创新,通过加密元素的选取、运动矢量的加密方案设计、DCT变化系数的加密方案设计三部分进行了展开分析与实验;经过后续仿真验证,密钥敏感性得到了很好的提升;视频质量方面,改进后的AES算法的RGB直方图可知其加密效果和解密还原效果良好,经过对比分析,改进后的AES算法加密的SSIM值最小、PSNR值更低,即加密效果相对更好;经过计算可知,编码时间百分比增加了0.23%,解码时间百分比增加了8.56%,对NPCR、UACI以及加密前后视频帧像素对的相关系数等参数分别测试可知,相比前人的几种加密方式,改进后视频加密安全性能最好;综上分析可知,改进后视频加密效果最好。

向量值多线性奇异积分算子的加权Sharp不等式

利用不等式、原子分解性质研究了向量值多线性算子,证明了某些向量值多线性奇异积分算子的一个加权Sharp不等式,并利用此不等式得到了该向量值多线性算子的加权L^(p)型不等式和L log L型不等式。

基于BERT-TextCNN的汽车评论情感分析

通过基于预训练转换器(Transformer)双向编码器表征的文本卷积神经网络(text convolutional neural network model based on pre-training bidirectional encoder representations from transformer,BERT-TextCNN)模型实现汽车网站评论的情感分析,其目的在于研究用户对汽车产品和汽车服务的情感态度。首先采用基于Transformer的双向编码器表征(bidirectional encoder representations from transformer,BERT)模型能有效解决汽车评论中存在的一词多义问题,并产生包含丰富信息的动态词向量,然后结合文本卷积神经网络(text convolutional neural network,TextCNN)模型中的卷积运算和池化运算提取关键特征,最后通过softmax函数计算评论文本情感的概率分布。试验结果表明,BERT-TextCNN模型在情感分类中相比几种常见的神经网络模型的精度、召回率和F1值均有所提升。

基于小样本数据的风电备品备件需求预测方法研究

全球风电装机容量不断攀升,我国风电产业也出现跨越式发展,风电行业的蓬勃发展给运维管理工作带来巨大挑战。风电运维管理中的备品备件需求预测是决定运维成本的重要因素,尤其是复杂工况下的备品备件预测是风电运维管理面临的亟待解决的问题。本文围绕风电行业运维数据呈现的小样本特征,对风电行业的小样本备品备件需求预测方法开展研究。根据风电行业备品备件运维需求特征,选择小波变换提取备品备件的历史需求数据的趋势与细节特征,选择合适的时间窗将时间序列转化为监督学习形式的时间序列对,提出混合遗传算法和支持向量回归的预测算法GA-dwtSVR,最后开展了验证实验,实验结果表明:相较于传统预测方法,GA-dwtSVR在风电备品备件需求预测中有较好的性能,可以满足风电行业备品备件基于小样本数据的预测需求。

信息病毒查杀的电网主机操作安全性保护平台

以保障电力系统主机操作网络安全运行为目的,构建基于电力信息病毒查杀的电网主机操作安全性保护平台.该平台利用站内联网交换机和防火墙连接保护子站和操作子站内若干电网主机后,连接电网主机操作单元,用户通过该单元执行数据下载、数据分析、Zookeeper服务等电网主机操作,同时使用集成第三方软件的病毒查杀单元内Mcafee、Avira和Kav杀毒软件,查杀用户执行电网主机操作过程中软件内存在的病毒.以支持向量机算法作为基础,建立电网主机操作安全等级审计模型,然后电网主机操作单元认证电网主机操作软件,并对电网主机操作安全等级进行审计后,执行相关电网主机操作,同时连接外网防火墙,并利用Web发布服务、物联网APN专线以及电信专用服务器等,将电网主机操作结果呈现在移动终端APP上,完成电网主机操作安全性保护过程.