关键词： Temperature  

摘要： Abstract: This article proposes an algorithm for solving the modeling problem with numerous variables in the MATLAB programming language. According to the developed algorithm, the MATLAB model calculates the values of solar insolation and ambient temperature of a particular location depending on five variables simultaneously: (1) time of day (τ), (2) day of the year (n), (3) tilt angle (β), (4) azimuth angle (γ) and (5) albedo (ρ). Meanwhile, the calculation results form a multidimensional database of solar insolation and ambient temperature of a given location for any values of the above variables and their *** example of using the created database is given for calculating the annual total photovoltaic panel productivity and productivity dynamics on days of the year for latitude 41°2995′, which corresponds to a location in the city of Tashkent. The developed algorithm can be used to analyze/forecast the efficiency of solar plants using parameters from the database and other variables, such as temperature coefficient, dust content, nominal efficiency, wind speed, humidity, etc., as input data. The availability of data makes it possible to use data for both one-way and two-way reception of radiation as for bifacial photovoltaic panels. The database created based on the proposed program model can serve as a basis for comparison with other calculations or measurement databases and for studying or determining other parameters that affect possible differences in the compared data and creating advanced modeling algorithms for the solar irradiance. © 2020, Allerton Press, Inc.

关键词： 价值转移   数学算法   互联网领域   参与方   架构   信任   链  

摘要： 区域链之所以能够做很多事情,是因为区域链具有价值转移和信用转移的优势.这是一种通过数学算法来实现的"去信任"架构."去信任"架构就是系统中的参与方无须信任对方就可以完成交易和协作.这是传统互联网领域中最薄弱的一个环节.

关键词： 数学建模   改进遗传算法   最小二乘法   拟合  

摘要： 研究计算机数学建模中改进遗传算法与最小二乘法的具体应用。针对基本遗传算法中存在的数据冗余问题,提出了一种改进遗传算法,结合最小二乘法建立计算机数学模型,以更好的应对数据变化。对具体的问题进行计算机数学建模,并进行验证,结果表明改进后的遗传算法具有一定的辨识能力,寻找最优,大大提高了工作效率。利用MATLAB软件求解方程组,确定参数的使用范围,解决估算参数范围和辨识问题。仿真结果证明改进后的遗传算法与最小二乘法结合后建立数学模型,大大增加了个体搜索空间,比基本遗传算法提高了90%的运算率。

关键词： 计算教学   对比   算理   算法  

摘要： 《义务教育数学课程标准(2011年版)》将"运算能力"作为十个核心词之一,足见计算学习在数学教学中的重要性。本文以苏教版数学一年级下册"两位数加整十数、一位数(不进位)"为例,着重通过三个层次的对比,让学生感受多种算法,既有效地沟通了算理与算法之间的联系,同时又使学生加深了对计算方法的认识。

关键词： 数学思想   计算机编程   算法优化  

摘要： 本文结合编程教学实践,立意于研究计算机编程中数学思想对算法的优化性,通过对比实验数据,验证数学思想在算法优化中的重要作用,总结出数学思想优化算法的策略,以期编程学习者能养成运用数学思想优化算法的意识,提高算法效率。

关键词： Radioactive   Nuclide   Decay   Modeling  

摘要： Radioactive decay processes, such as alpha decay, produce decay chains where the mass numbers of nuclides decrease as larger nuclides expel energetic particles to form smaller nuclides. Under these conditions, the coefficient matrix that describes the differential rate expressions for radioactive decay can be made lower triangular. With this special structure, formulating an algebraic solution to the decay chains can be done by first formulating the eigenvectors that make up the coefficient matrix, which can then be solved using forward substitution for a lower triangular matrix. This work details the derivation of algebraic solutions for decay chains of any number of stable and unstable nuclides with any number of branching based on this eigenvector analysis. A prototype computational code was developed to validate and compare this methodology against a number of other methods for solving similar systems. A two-phase sorting algorithm yielding the lower triangular matrix structure was established to apply the developed algebraic solutions for decay chains involving beta-emitting radionuclides transformed into daughter nuclides without change in their mass number. The methodologies produced in this work provide an efficient way to estimate nuclide fractions from natural decay processes. (C) 2019 Elsevier B.V. All rights reserved.

关键词： Container drayage scheduling   Linearization   Mathematical modeling   Multi-resource constraint   Large neighborhood search  

摘要： Multi-resource constraints in container drayage operations exist widely in real-life scenarios but have been seldom reported in literature. This study mainly addresses a container drayage problem with a limited number of empty containers at depots, with the goal of minimizing both the number of trucks in operation and the total working time of trucks. The problem is firstly formulated as a bi-objective mathematical model based on a so-called determined-activities-on-vertex graph. Three schemes are proposed to handle the developed mathematical model as: (a) a nonlinear constraint is linearized;(b) a parameter that is a large enough constant in the model is analyzed and tuned deeply;and (c) the bi-objective model is converted into a single-objective model. Furthermore, a large neighborhood search algorithm is designed to solve the problem. The two solution methods are validated and evaluated based on randomly generated instances as well as instances from literature. Numerical experimental results indicate that the methods can provide optimal or near-optimal solutions for medium- and large-scale instances in a short running time.

关键词： Flexible job-shop scheduling problem   Sequencing-flexibility   Bacterial foraging optimization algorithm   Simulated annealing   Weighted tardiness   Decision support system  

摘要： The flexible job-shop scheduling problem (FJSP) is an extension of the classical job-shop scheduling problem (JSP) in which operations can be performed by a set of candidate capable machines. An extended version of the FJSP, entitled sequencing flexibility, is studied in this work, which considers precedence between the operations in the form of a directed acyclic graph instead of a sequential order. In this work, a mixed integer linear programming (MILP) formulation is presented to minimize weighted tardiness for the FJSP with sequencing flexibility. Due to the NP-hardness of the problem, a novel biomimicry hybrid bacterial foraging optimization algorithm (HBFOA) is developed, which is inspired by the behavior of E. colt bacteria in its search for food. The developed HBFOA search method is hybridized with simulated annealing (SA). Additionally, the algorithm has been enhanced by a local search method based on the manipulation of critical operations. Classical dispatching rules have been employed to create the initial swarm of HBFOA, and a new dispatching rule named minimum number of operations has been devised. The developed approach has been packaged in the form of a decision support system (DSS) developed on top of Microsoft Excel-a tool most small and mid-range enterprises (SME) use heavily for planning. A case study with local industry is presented to validate the proposed HBFOA and MILP. Additional numerical experiments using literature benchmarks are further used for validation. The results demonstrate that the HBFOA outperformed the classical dispatching rules and the best integer solution of MILP when minimizing the weighted tardiness and offered comparable results for the makespan instances.

关键词： 公式识别   字符识别   结构识别   系统设计  

摘要： 随着社会信息技术的发展,电子档书籍已经成为重要的知识载体。数学公式是书籍重要的页面信息,其识别不仅要识别字符还要公式识别结构,一直以来都是实体书籍电子化的一个难点。本文设计了一种数学公式识别的方法,该方法应用卷积神经网络结构进行字符识别,利用运算符号的作用域和中心线来实现公式结构的识别。本文采取公式字符切割、字符识别、公式结构识别这三个步骤实现数学公式的识别。首先采用连通域与规则相结合的方法实现常见数学公式的有效切割。本文设计了卷积神经网络作为字符识别器,由3个卷积层和3个全连接层组成,卷积核大小为33,最后一层的全连接层的神经元个数为275。该字符识别器在Infty的数据集测试集的准确率可以达到98.90%。在公式结构识别方面,提出了一种基于运算符号和中心线的方法。首先,充分利用了公式中不同种类运算符作用域固定的特点将其分为三类,分别是分号、特殊符号和二元运算符号,其中特殊符号包括根号、求和符号、积分符号等大符号,这些运算符有多个作用域。先将这些作用域内的子式判别出来并进行识别,将识别结果直接送入运算符的Latex语言表达式,再进行识别上标和下标等位置关系。然后基于字符在书写时受四线格约束的情况,提出中心线的概念,将常见字符分类为向上类、向下类、全占类和中心类,依据字符的中心线类别不同计算阈值判断两个字符之间的位置关系。这样的方法能够全面地识别常见数学公式的结构。本文设计并实现了一个书籍内容识别系统。该系统应用客户端、服务器的架构实现。获取页面图片之后,在客户端进行页面预处理和页面图片分割,将分割结果发送到服务器。利用页面目标分类模型将页面信息分类后,对文本行和公式进行识别。然后服务器向客户端发送页面目标识别结果,最后在客户端文件生成可编辑文件。其中文本识别应用的是CRNN模型,公式识别中字符识别应用的是本文提出卷积神经网络结构,公式结构识别应用的是本文提出的方法。本文最终实现了书籍内容识别系统。

关键词： 自然语言处理   深度学习   词性标注   语料库  

摘要： 随着计算能力的大幅提高和互联网技术的飞速发展,人类对自然语言处理的研究不断深入.在此背景下,语料库语言学逐渐茁壮成长起来.尤其是近几年来,随着深度学习的崛起,语料库更是成为了神经网络算法得以有效运行的基础.在自然语言处理领域,词性标注是实现自然语言处理任务目标的一个基础环节,也是对文本数据的一个预处理过程,它的准确度将很大程度上影响到后续目标任务的性能.语料库的词性标注准确率越高,语料库的规模越大,神经网络模型的性能就越好.故词性标注语料库的构建与研究逐渐成为了国内外学者的研究热点.数学学科作为自然科学的基础学科,与各行各业的发展都有着密切的联系,它的发展对于我国科技的进步起着至关重要的作用.而目前国内外尚没有专门的带有词性标注的数学专业语料库,这严重影响了数学学科文献的机器翻译及其他自然语言任务的实现.因此,本文以词性标注为主,针对数学科技文献数据构建了一个具有一定规模的词性标注语料库.本文设计了构建数学专业词性标注语料库的算法.首先,我们将神经网络,条件随机场进行结合,构建神经网络框架;其次,我们借助新闻词性标注语料库数据,不断添加数学数据到训练集,测试集,验证集中,同时去掉相同句数的新闻数据,之后使用新闻和数学的混合数据去训练新的模型;最后,神经网络模型在多次迭代之后,我们得到一个对数学数据词性标注效率较高的模型以及一个数学专业词性标注语料库.语料库的准确率为98.36%,而现存的新闻词性标注语料库的准确率介于94%-98%之间,由此可见,我们构建的词性标注语料库的准确率是非常高的.以此语料库为基础,我们可以进行自然语言处理的其他任务研究.本文使用训练过程中产生的模型在纯数学数据上进行了测试实验,实验结果显示,随着模型不断优化,新产生的模型学习到的数据分布逐渐由新闻数据分布转变为数学专业数据的分布,在测试数据上解码效率越来越高,直到模型解码正确句子占全部数据比例不发生变化,则得到最优模型.最优模型在解码纯数学数据时,它的解码效率为69.85%(以句为单位),这比纯新闻数据训练的模型解码纯数学数据的解码效率(为12.82%)要高很多.通过对比分析可得,我们得到的最优模型学习到了数学数据的分布,使用它去标注数学专业文献原始语料,通过特定的阈值筛选,可以得到标准的词性标注语料库数据,从而避免了人工标注的高成本,低效率的缺点.除此之外,文章中我们设计的构建数学专业文献词性标注语料库的算法,也为其他学科科技文献语料库的构建提供了借鉴.