BP网络及其应用

基函数选取和系数求解是函数逼近中的两个关键问题,目前广泛应用的回归分析或其它映射方法均有局限性。采用人工神经网络中的BP网较好地解决了这两个问题,对理论和实际的数据进行了比较与验证,结果表明.BP网络有助于提高数据处理的可靠性与预测的精度.     

沼气在果菜粮贮藏业上的应用

文中介绍沼气这种新型绿色能源在果菜粮贮藏中的应用情况,详尽描述了其在贮藏果品、蔬菜中的应用过程,文中对其作用的原理和益处也给予了适当介绍。     

Expression and Characterization of HGV E2 cDNA in Pichia pastoris

A 559 base pair fragment of cDNA locating at the putative E2 region of GBV-C/HGV was in-serted into Pichia pastoris expression vector pPICgK in the reading frame of α-factor secreting signal pep-tide. The recombinant expression plasmid pPIC9K-E2 was introduced into P. pastoris GSll5 with electro-poration and recombined with the host genome by homological recombination. The His+Mut+ recombinantyeasts were selected and cultivated in the BMMY medium. After 3 days induction with 0. 5% methanol,the target protein(E2) accumulated up to 30% of total proteins in the supernatant. The expressed E2 pro-tein was proved possessing antigenicity and high specificity with Western blot and ELISA probed with serafrom the immunized rabbits and the patients infected by GBV-C/HGV.     

AC发泡剂新工艺研究—Ⅲ电解氧化法

用钌钛金属阳极在含有Ｃｌ—和Ｂｒ－离子的酸性溶液中进行电解氧化联二脲制备偶氮二甲酰胺，获得产品得率为９３—９８％、纯度为９９．６％、电流效率为８８－９２％。母液可循环使用，基本不产生废液．     

森林面积分布系统的最优控制存在性问题

本文研究森林面积分布系统的最优控制问题。本文将林木总和长消率作为控制变量，以“范数最小”来衡量其最优性，利用空间Ｌ＾２（０，Ｔ）Ｔ〉０的自反、光滑和严格凸性，给出了森林面积分布系统最优控制的一个存在性定理。     

可压缩流体网络的建模方法

本文运用网络分析手段,首次提出了任意拓扑结构可压缩流体网络动态特性的建模方法.该方法突破了目前在流体网络分析中普遍采用的“小扰动”线性化处理方法的局限。     

铁素体-贝氏体球墨铸铁的初步研究

本文研究了不同比率的铁素体-贝氏体两相组织对球墨铸铁拉伸特性和冲击韧性的影响。当铁素体球墨铸铁在900℃奥氏体化,400℃等温淬火,贝氏体体积分数为54％和93％时,延伸率及断面收缩率均出现两个峰值。实验证明,当铁素体和贝氏体的比率在最佳范围时,能明显提高其强度和韧性。X—射线衍射证实,贝氏体为无碳化物上贝氏体。能获得良好的韧性,主要由于在石墨球周围形成环状组织的无碳化物上贝氏体,可能是这种高强度、高韧性的贝氏体在石墨-贝氏体界面阻止了裂纹萌生的结果。     

Learning attention-based strategies to cooperate for multi-agent path finding

Multi-agent path finding (MAPF) is a challenging multi-agent systems problem where all agents are required to effectively reach their goals concurrently with not colliding with each other and avoiding obstacles. In MAPF, it is a challenge to effectively express the observation of agents, utilize historical information, and effectively communicate with neighbor agents. To tackle these issues, in this work, we proposed a well-designed model that utilizes the local states of nearby agents and outputs an optimal action for each agent to execute. We build the local observation encoder by using residual attention CNN to extract local observations and use the Transformer architecture to build an interaction layer to combine local observations of agents. With the purpose of overcoming the deficiency of success rate, we also designed a new evaluation index, namely extra time rate (ETR). The experimental results show that our model is superior to most previous models in terms of success rate and ETR. In addition, we also completed the ablation study on the model, and the effectiveness of each component of the model was proved.     

土茯苓种苗质量分级标准研究

探究土茯苓(Smilax glabra)种苗质量分级标准,为土茯苓人工种植规范化发展奠定基础。本研究随机选取50株土茯苓种苗进行调查,首先应用相关分析确定苗木分级指标,然后通过逐步聚类分析法修正分级,最终计算出土茯苓种苗的苗木分级标准。结果表明,根数和球茎数为土茯苓种苗的质量分级指标。土茯苓种苗分级标准为Ⅰ级种苗：根数≥37.4,球茎数≥6.8;Ⅱ级种苗：28.3≤根数<37.4,4.2≤球茎数<6.8;Ⅲ级种苗：根数<28.3或球茎数<4.2。所抽取的土茯苓样苗中,Ⅰ、Ⅱ级种苗(合格苗)分别占34%和20%,Ⅲ级种苗(不合格苗)占46%。本研究通过逐步聚类分析法制定的土茯苓种苗质量分级标准结果可靠,说明此方法可行,能为土茯苓人工种植繁育研究以及生产优质的土茯苓种苗提供理论资料。     

上海市集成电路及机电一体化产品的比较优势研究

本文用一种“节汇成本比较”的方法,对上海市集成电路及机电一体化产品的比较优势进行调查研究,提出了上海市在这两个产品领域发展重点的建议。