基于MAS的制导雷达网络化抗干扰作战体系结构研究

通过对地空导弹网络化抗干扰作战的分析,结合DAI(Distributed Artificial Intelligence,分布式人工智能)技术中的MAS(Multi-Agent System,多智能体系统)技术,建立了基于MAS的制导雷达网络化抗干扰作战系统,并给出了该系统的Agent组成框架,定义并说明了各Agent的功能及相互关系.     

Identifying crash propensity using specific traffic speed conditions

INTRODUCTION: In spite of recent advances in traffic surveillance technology and ever-growing concern over traffic safety, there have been very few research efforts establishing links between real-time traffic flow parameters and crash occurrence. This study aims at identifying patterns in the freeway loop detector data that potentially precede traffic crashes. METHOD: The proposed solution essentially involves classification of traffic speed patterns emerging from the loop detector data. Historical crash and loop detector data from the Interstate-4 corridor in the Orlando metropolitan area were used for this study. Traffic speed data from sensors embedded in the pavement (i.e., loop detector stations) to measure characteristics of the traffic flow were collected for both crash and non-crash conditions. Bayesian classifier based methodology, probabilistic neural network (PNN), was then used to classify these data as belonging to either crashes or non-crashes. PNN is a neural network implementation of well-known Bayesian-Parzen classifier. With its superb mathematical credentials, the PNN trains much faster than multilayer feed forward networks. The inputs to final classification model, selected from various candidate models, were logarithms of the coefficient of variation in speed obtained from three stations, namely, station of the crash (i.e., station nearest to the crash location) and two stations immediately preceding it in the upstream direction (measured in 5 minute time slices of 10-15 minutes prior to the crash time). RESULTS: The results showed that at least 70% of the crashes on the evaluation dataset could be identified using the classifiers developed in this paper.     

基于文本挖掘的煤矿安全隐患分析

为充分有效利用煤矿安全隐患数据,揭示安全隐患数据中隐藏的重要信息和潜在规律,用于指导安全管理实践,在对安全隐患数据进行预处理的基础上,利用词云和词频统计方法展现安全隐患概况,运用文本挖掘技术和社会网络分析方法研究煤矿安全隐患信息的关联关系和分布规律。结果表明,将文本挖掘技术和社会网络分析方法应用于文本型安全隐患数据分析具有可行性,有助于煤矿企业提高对安全隐患数据的认知并制定合理的安全隐患治理措施。     

中国农业生态效率测度与驱动因素

为适应我国进入高质量发展阶段,实现农业资源的高效率利用,本文以中国30个省、直辖市、自治区（不包括西藏、香港、澳门和台湾）的2003~2017年度面板数据,运用考虑生态非期望产出和期望产出两阶段网络EBM模型,构建资源环境投入—环境治理两阶段模型,测度各省（直辖市、自治区）的农业生态效率,并结合Malmquist指数进行动态分析.以测度的农业生态效率为被解释变量,在考虑农业劳动力素质的门限效应下对其驱动因素进行分析.结果显示：2007年以前中国各地区农业生态效率,呈西部到东部逐渐收敛趋势.2007年以后,呈东西部高,中部塌陷,东西部与中部地区发展失衡.农业劳动力素质存在门限效应,门限值为1012.62.当门限变量值≥1012.62时,农作物种植结构对农业生态效率的影响存在促进作用;反之,则有抑制作用.财政支农水平对提高农业生态效率有正向影响,农用机械投入强度、农业生态环境防护投资额对提高农业生态效率有负面影响.     

Bacterial succession in epiphytic biofilms and deciduous layer sediments during Hydrilla verticillata decay: A field investigation

Submersed macrophytes decay is an important natural process and has important role in mass and energy flow in aquatic ecosystems. However, little is known about the dynamical changes in nutrients release and bacterial community during submersed macrophyte decay in natural environment. In this study, a field observation was conducted in a wetland dominated with Hydrilla verticillata for 36 days. Increase of H₂O₂ and malondialdehyde (MDA) content and decrease of soluble proteins concentration were detected in leaves during H. verticillata decay. Meanwhile, ammonium-N, soluble microbial products (SMP) and TOC concentration increased in overlying water. According to bacterial 16S rRNA Illumina sequencing analysis, the Shannon values were lower in epiphytic biofilms than deciduous layer sediments. The relative abundances of Proteobacteria, Cyanobacteria and Actinobacteria were higher in epiphytic biofilms than in deciduous layer sediments (P < 0.05). Co-occurrence network analyses showed that a total of 578 and 845 pairs of correlations (|r| > 0.6) were identified from 122 and 112 genera in epiphytic biofilms and deciduous layer sediments, respectively. According to co-occurrence patterns, eight hubs were mainly from phyla Proteobacteria, Acidobacteria and Parcubacteria in epiphytic biofilms; while 37 hubs from the 14 phyla (Proteobacteria, Bacteroidetes, Acidobacteria, Chloroflexi, et al.) were detected in deciduous layer sediments. Our results indicate that bacterial community in deciduous layer sediments was more susceptible than in epiphytic biofilms during decay process. These data highlight the role of microbial community in deciduous layer sediments on nutrients removal during H. verticillata decay and will provide useful information for wetland management.     

达里诺尔湖溶解性有机质荧光特征与人工神经网络非线性响应研究

研究湖泊水体DOM（溶解性有机质）特征及其与复合环境要素的响应关系,对水生态系统保护及生物地球化学循环研究具有重要意义.以内蒙古达里诺尔湖（简称“达里湖”）为研究区,利用紫外-可见光谱技术（UV-vis）和人工神经网络模型（ANN）方法,开展达里湖水体DOM特征识别及其水质响应关系的研究.结果表明:①达里湖水环境污染程度较为严重,水体呈富营养化趋势.DOM吸收系数〔α（355）〕表明,夏季湖体DOM浓度较高.②建立了α（355）与达里湖水体ρ（TN）、ρ（TP）、pH和ρ（Chla）的人工神经网络模型,该模型验证期和测试期决定系数（R2）分别为0.78和0.84,该模型具有较好的DOM模拟和预测能力.③人工神经网络模型参数敏感性分析结果表明,α（355）对ρ（Chla）变化最敏感,敏感性水平为31.95%;其次为pH和ρ（TN）,α（355）对二者变化的敏感性水平分别为28.53%和25.54%;α（355）对ρ（TP）变化敏感性最低,敏感性水平为8.16%.研究显示,达里湖DOM对富营养化的发生具有较显著的表征影响,人工神经网络模型可为富营养化预测提供科学参考.      

窖水中微生物降解污染物的关键细菌

为探究以氮、磷及有机物污染为主要特征的、窖水中参与污染物降解的关键细菌及它们之间潜在的相关关系,基于16S rRNA的微生物组学截面数据,分析了窖水中细菌群落结构与功能及其与水质因子间的相关性,并通过微生物物种的同现或相关种间作用推断模型,构建了7个微生物菌属间的共现性关联网络.结果表明,窖水中存在具有相对特定生态功能的细菌,进行着诸多活跃的与新陈代谢功能基因相关的代谢活动;窖水微生物类群共现性关联网络中大部分节点的菌属营互利共生类型的生态关系;Lacibacter、Arthrobacter、Candidatus Protochlamydia、Methylocaldum、Sulfuritalea、Mycobacterium、Aquirestis、Rhodobacter、Methylotenera等菌属拥有较高的点度中心度;较强的互作关系发生在Sulfuritalea-Rhodobacter、Azospirillum-Rhodobacter、Methylocaldum-Rhodobacter、Arthrobacter-Rhodobacter、Rhodoplanes-Rhodobacter、Candidatus Protochlamydia-Rhodobacter、Methylotenera-Rhodobacter、Rhodobacter-Aquirestis、Mycobacterium-Rhodobacter、PlanctomycesCandidatus Solibacter、Planctomyces-Legionella、Hymenobacter-Adhaeribacter、Luteolibacter-Crenothrix之间.综合分析节点微生物相关性、点度中心度及菌属间的互作强度,认为Rhodobacter、Methylocaldum、Methylotenera、Acinetobacter、Novosphingobium、Planctomyces、Hymenobacter、Luteolibacter为参与窖水污染物微生物降解的关键细菌,Rhodobacter为关键细菌的代表属.研究结果加深了对窖水中污染物微生物降解机制的认识.     

基于CMAQ与前馈神经网络的区域大气污染物浓度快速响应模型

建立大气污染可控源排放-复合污染水平的函数关系,实现给定排放情景下环境污染物浓度的实时响应,是大气污染物浓度预测与减排效果评估等的重要技术前提.本研究对污染物浓度影响因子变化空间进行拉丁超立方采样,使用CMAQ区域多尺度空气质量模型的预测值作为输入数据,通过前馈神经网络模型构建基于统计机器学习的长三角区域污染物浓度快速响应模型.结果表明经过模型结构选择与参数调整,基于前馈神经网络的快速响应模型能够快速准确还原出不同减排情景下长三角区域PM_(2.5)浓度的预测值.外部验证情景下相关系数CORR达到0.999以上,MB与ME均值达到了-0.046μg·m~(-3)和0.6162μg·m~(-3),实现了比RSM更加快速与准确的预测,不同时段与不同污染物浓度的准确预测则验证了其普适性.     

双基球扁药工艺优化仿真研究

目的优化双基球扁药成球工艺参数,解决由于目前双基球扁药理论研究不充分、控制模型不明确、生产工艺参数调控依靠人工经验所导致的药品成球后直径、弧厚偏差大的问题。方法利用BP神经网络在处理复杂非线性映射问题上的强大的能力,对成球关键工艺参数与成球质量指标进行建模,并应用成球工艺过程仿真数据对其进行训练,将训练得到的BP神经网络模型用于优化成球工艺参数。同时利用仿真数据进行检验模型的可靠性。结果训练后BP神经网络均方误差为0.001,成球直径误差率为1.27%,成球弧厚误差率为2.08%,成球质量参数误差均很小,可以满足工艺要求。结论该BP神经网络模型具有较高精度,适用于含能材料工艺优化,提出的成球工艺优化方法能有效降低成球试制成本,缩短生产周期。     

基于遥感的洪水灾害承灾体神经网络的提取方法探讨

洪水灾害承文体信息的快速获取是洪水灾害评估的基础．卫星遥感所获得的遥感图象包含了有关该承灾体的空间分布信息．常规目视解译难以快速并准确地从遥感图象中提取这些信息．神经网络技术的发展为解决这一问题提供了新的工具．本文阐这了将神经网络用于从遥感图象中自动提取洪水灾害承文体的空间分布信息的基本原理和方法,并结合实例说明了这种方法的有效性