Robot’s Emotion Decision by Energy and Entropy Concepts

This study is to introduce concepts of energy and entropy to describe a robot＇s emoton decisien. It chooses the dimensional approach based on factors of pleasure and arousal for the merit of the interpolation between enotions. Especially, Circumplex model which has also two axes： pleasure and arousal is used. Besides, the model indicates how emotions are distributed in the two-dimensional plane. Then by the definition of psychodynamicsthe energy states （mental energy and physical energy） are matched to pleasure and arousal respectively that are the basis of Circumplex model. The mental energy is updated by the result of Prospect theory which measures the value of gain and loss as pleasure factor. And the physical energy is updated by the result of hedonic scaling which measures levels of arousal from pleasure computed by Prospect theory, and the result of intensity of stimuli. Then the energy states are fed by entropy. The feedback loop by entropy satisfies the 2nd law of thermodynamics. The energy states generated by stimuli and fed by entropy take a position in the plane of Circumplex model. Then distances between the current position and other emotions are cornputed to get a level of each emotion, proportional to the inverse of the distance.

Research on Energy Decision System based on Network Technology

This paper analyzes the C/S structure and B/S structure of their own characteristics, building energy decision support system based on network technology, elaborated Web GIS and database connections and other technical principles and applications. Mixed B/S structure mode development to achieve information management, energy consumption is forecast to show the query using the amomat of energy, energy supply and demand dynamic equilibrimn analysis and other functions, with simple, efficient, easy to operate and so on, and has good scalability and maintainability.

Performance Improvement of the DSRC System Using a Novel S and Π-Decision Demapper

Based on the constellation diagram of the different modulations, a novel S and Π-decision rule is designed for the analog demapper of the orthogonal frequency-division multiplexing (OFDM) systems. The dedicated short-range communications (DSRC) system is chosen as an OFDM platform to compare the performance among the proposed S and Π-decision decoder, hard-decision and soft-decision decoders. Simulation results demonstrate that both the complexity and performance of S and Π-decision demapper used for M-ary quad-rature amplitude modulation (QAM) OFDM system can be greatly improved. The number of decisions be-tween the received symbol and constellation points can be simplified to look up table times for M-ary QAM OFDM system.

Study of Product Quality Information and Construction of Its Decision-Making Support System and Measures to Improve People's Quality Concept

The 21th century is a "quality century". Facing the s evere challenge about quality, all the developed industry countries attach impor tance to the research and education about quality project. With the deepening of the reform of China and it’s joining to WTO, many industries and enterprises wi ll be confronted with severe challenge while being offered the opportunity to wa lk up to the world. The arrival of knowledge economy and web time has provided t he space for the innovation on enterprise management. It has also provided new m ethod and model for enterprise management, that is online management based on th e internet. The online management use an enterprise model including the goal of the enterprise and the scheme to carry out it to represent the whole operational of all the work. The online management is constructed on the integration of exp ert system, database, software and other technical models. How to scientifically manage the entire system of product, precisely master the condition of quantity and quality and analyze the rules of the change of produc t quality in order to carry out the overall control of quality, guarantee the go od condition of product as well as improve product quality and prolong it’s se rvice life was discussed in this thesis. Based on the above research the thesis also discussed how to develop the study on product quality information and const ruct it’s decision making support system. The concrete measures to improve peopl e’s concept of quality are also raised. Based on the type analysis, scientific management of product quality information should start from dynamic information of quality, instruction information as we ll as the information about the use, maintain, management and conservation of pr oduct, and induce them and further propose a design scheme of feedback system on product quality information, and then set up the blueprint of the information s ystem of product quality. This thesis included 7 chapters:1.the type of quality information;2.the content of quality information;3.the construction of the se nsitive web for transfering information, and collecting and transfering quality information rapidly and effectively;4.the way and procedure to transfer informa tion;5.the construction of quality information library for all phases; 6.the d isposal of quality information;7.the decision making support system of the mana gement of quality information;8.attach importance to three key factors on impro ving quality concept.

The Development of the Generation Expansion Planning System Using Multi-Criteria Decision Making Rule

The power expansion planning is large and capital intensive capacity planning. In the past, the expansion planning was established with the proper supply reliability in order to minimize social cost. However, such planning cannot be used in the power markets with many market participants. This paper proposed the power expansion planning process in the power markets. This system is composed of Regulator and Generation Company （GENCO）＇s model. Multi-criteria decision making rule is used for regulator model and several scenarios for GENCO model are applied.

Religious Figures on Dalai's Decision

ＲｅｌｉｇｉｏｕｓＦｉｇｕｒｅｓｏｎＤａｌａｉ＇ｓＤｅｃｉｓｉｏｎ￥ＴｈｅＤａｌａｉ＇ｓＭａｙａｎｎｏｕｎｃｅｍｅｎｔｔｈａｔａＴｉｂｅｔａｎｂｏｙｗａｓｔｈｅｒｅｉｎｃａｒｎａｔｅｄｃｈｉｌｄｏｆｔｈｅｌａｔｅ１０ｔｈＢｒｅｓｅｎｔｍｅｎｔｉｎＣｈ...

OPENING UP NEW HORIZONS IN OUR R&D WORK BY IMPLEMENTING THE PARTY'S DECISION TO SPEED UP CHINA'S S&T PROGRESS

The 1990s is a key historical decade in the process of China’s on-going modernization drive.When reviewing the overall picture of the socio-economic trend and S&T development around the world during the last two decades,we may reach the conclusion that a new technological revolution is taking shape.For both the developed and the developing countries,challenges and opportunities are close at hand.In the Decision to Speed up China’s S&T Progress Jointly Issued by the CPC Central Committee and the State Council.a developmen-

Decisive Stage of All-round Construction of Well-off Society Is a New Opportunity for China's Human Rights Cause

The p eriod of"The 13th Five-Year Plan"for national socioeconomic development(2016-2020)is a decisive stage of completion of all-round construction of a moderately prosperous society in China,and the year of 2016 is the first annum of this decisive stage.The Central Committee of the Communist Party of China(CPC)headed by General Secretary Xi Jinping has,with a new stance and new ideas,set strategic tasks

基于曲线决策融合的SDN饱和攻击检测方法

针对软件定义网络(SDN)交换机和控制器的饱和攻击是SDN中的主要安全问题。在使用集成学习方法检测饱和攻击时,现有方法通常使用距离或熵值等简单的信息计算方法修正证据,可能存在信息丢失问题,降低饱和攻击检测精度。为解决上述问题,提出一种基于曲线决策融合的饱和攻击检测方法(SACOIN)。SACOIN首先计算多分类器概率矩阵的混乱程度修正多分类器内证据;随后将多分类器概率矩阵转换为曲线并去除噪声,提取重构小波的信号特征组成特征矩阵;然后计算特征矩阵行内互信息,基于上述互信息修正多分类器间证据;最后使用D-S证据理论融合修正证据,得到最终检测结果。实验结果表明,SACOIN在检测针对SDN交换机和控制器的饱和攻击时的准确率、精确率、召回率、F1值分别为92.3%、93%、92.1%、91.3%。

基于B/S架构的多用户在线程序评判系统设计与实现

针对现有在线程序评判系统存在的判题效率和安全性问题,本文在分析多用户在线评判系统的功能需求基础上,提出了一种基于B/S架构采用Struts+Spring+Hibernate(SSH)组合框架的多用户在线程序评判系统,介绍了系统的程序评判原理及采用的安全机制。该系统不但可以用于ACM程序设计等比赛,而且可以为程序设计课程教学提供很好的辅助实验平台支持。

基于D-S证据理论的决策级多特征融合苹果分级方法

针对单特征苹果分级的不确定性和低正确率,通过图像处理提取大小、形状、颜色和缺陷4类能反映苹果外观品质的主要特征,引入信息融合的思想,以单特征初步分级的结果作为证据,用D-S证据理论的方法进行决策级融合处理,实现苹果的多特征综合分级,进一步提高可靠性和分级正确率。80个测试样本的分级试验表明,苹果分级正确率达92.5%,与单特征分级相比,此方法正确识别率高、稳定性好、效果显著。

基于S型曲线的指标非线性标准化研究

目前,数据标准化处理通常采用的是线性变换方法,然而在处理非均匀分布的指标数据集合时,尤其是对局部集中分布数据的处理存在一定局限性,例如无法有效地将数据划分层级,缺乏辨识性等。利用Logistic曲线函数的特性构建S型曲线模型,可对指标数据进行非线性标准化处理。该方法能够在不改变数据序列及整体分布的前提下对各数据点的取值进行非线性放缩,在将数据稠密的区间进行扩大的同时将数据稀疏的区间进行压缩;同时,针对包含正负值的指标数据集合可以有效地将负值数据突显出来,并且是对特定领域指标数据非线性标准化的补充。

基于SVM和D-S证据理论的多特征融合杂草识别方法

针对单一特征识别杂草的低准确率和低稳定性,提出一种支持向量机(SVM)和D-S证据理论相结合的多特征融合杂草识别方法。在对田间植物图像处理的基础上,提取植物叶片的颜色、形状和纹理等3类视觉特征,分别以3类单特征的SVM分类结果作为独立证据构造基本概率指派(BPA),运用D-S证据组合规则进行决策级融合,根据分类判决门限给出最终的识别结果。试验结果表明,多特征决策融合识别方法正确识别率达到97%以上。

多属性决策及D-S证据理论在底板突水决策中的应用

针对煤矿底板突水包含很多不确定因素和时空随机性的特点,探索多属性决策理论在底板突水预测中应用研究。采用案例统计与专家分析相结合的方法,赋予影响底板突水各种因素的权重,建立突水概率指数法的融合规则,提出突水概率指数法预测采场工作面底板突水的新方法。在获得底板突水概率指数的基础上,基于D-S证据理论,建立由突水、临界、不突水、不确定组成的识别框架,构建以富水性指数、构造指数、隔水层指数、水压指数、矿压指数为证据体的突水融合决策模型。通过在肥城煤田实例应用,说明经证据体的多次融合,能够提高突水概率的可信度,降低不确定性,证明基于多属性决策和D-S证据理论建立的工作面底板突水两级融合决策模型的有效性和可行性。

D-S证据理论中的基本概率赋值转换概率方法研究

D-S证据理论的基本概率赋值(BPA)能够有效地表示和处理不确定信息,在不确定信息处理中得到了越来越多的应用。但是如何基于BPA做出决策依然是一个有待解决的问题,将BPA转化为概率函数进行决策是一个简单可行的方法。文章基于证据理论中命题的信度函数和似真函数提供的信息,提出了一种新的基本概率赋值转换概率方法,该方法相比于现有经典TBM模型中Pignistic概率转换和基于似真函数的转换,可以更有效地利用系统已知的信息,实现基本概率赋值到概率分布的合理转换。算例表明了所提出方法的有效性。

基于D-S证据理论的群决策专家意见集结方法

将D S证据理论应用于群决策,指出其优势在于应用基本可信度分配函数描述专家意见,并能区分对所描述对象不知道和否定的差异。在专家意见集结上,Dempster证据合成规则有一定的局限性,当证据发生冲突时,会得到不合理的结果。定义了证据间的分歧度,提出一种专家意见集结方法,使得专家意见发生冲突时亦可得到合理的集结结果。

基于距离测度的D-S证据融合决策方法

针对多源信息融合结果并依据D-S证据结构进行决策的问题,提出一种基于距离测度的D-S证据决策方法·该方法结合决策基元和非决策基元的属性进行决策,将依据D-S证据的决策问题分解成两个层面:属性层面、证据层面·属性层面上,给出候选决策从证据焦元获得支持度的方法;证据层面上,基于辨识框架幂集的元素,构造一个证据焦元向量空间,引入候选决策的理想状态向量,定义距离测度,构建决策模型·最后,对多源水质监测信息融合结果进行决策分析,结果表明该方法是合理、有效的,且具有处理冲突或非冲突证据的优点·

基于改进不完全S变换与决策树的实时电能质量扰动分类

提出了基于一种改进不完全S变换(Improved Incomplete S-transform)与决策树的实时电能质量分类方法,其主要关注分类准确性与运算时间。根据主要频点所在频段,采用独立的高斯窗来处理不同的信号成分以减小海森堡测不准(Heisenberg's uncertainty)带来的时频分辨率限制,增强了对扰动的抗噪能力同时减小了响应时间。然后通过动态测度对改进不完全S变换结果进行特征提取。通过5个区分度强的特征量,采用优化决策树对电能质量扰动进行分类。通过一个基于DSP-FPGA的硬件平台来验证该方法。仿真与实验证明了该方法具有良好的应用前景。

基于D-S证据理论的态势评估方法

正确的战场态势评估是指挥决策的基础。本文对态势评估问题进行了分析,提出用D-S证据理论构造态势评估模型,根据不断到来的战场信息做出合理推断。实验结果证明该方法是有效的。

D-S证据理论在目标识别中的应用

D-S证据理论的应用过程与事件密切相关,针对目标识别这一特定领域,研究了证据理论中基本概率赋值的获取、组合规则及决策规则等问题。识别实例中,利用灰关联分析法来处理基本概率赋值的问题,对多传感器基本概率赋值进行组合后,最终通过决策完成对2种车型的类型识别,经计算仿真表明,D-S证据理论在目标识别中具有一定的有效性和优越性。