基于卷积神经网络的拖拉机工况识别

农机工况识别在细化农机作业状态和帮助掌握区域污染物排放趋势方面有着重要的研究价值。基于拖拉机不同运行状态下的行驶速度、发动机转速以及实时油耗等时间序列,首次提出将图像识别方法引入到拖拉机工况识别中的思路,并分别应用参数优化的支持向量机与卷积神经网络对实际作业拖拉机工况进行研究。结果表明:(1)基于参数优化的支持向量机可以较好地实现样本点的工况识别且识别准确度达到99.851 9%,但无法实现农机工况的连续性识别,同时无法对农机工况转换阶段进行有效识别。(2)以拖拉机运行速度与发动机转速等信息构建样本图像来描述农机工况变化的数据表达,并在此基础上应用卷积神经网络可以有效实现农机工况的连续性识别,且识别准确率可以达到93.3%。本研究在农机工况识别方面具有一定参考价值,并为后续农机不同工况下区域污染物排放研究提供技术支持。     

Use of netting and whitewash spray to protect papaya plants against Nivun Haamir (NH)-dieback disease

A disease causing the decline of papaya (Carica papaya) plants was noticed in the Jordan Valley in 1982. The disease caused severe yellowing of the upper leaves and dieback of the apex. The disease was not transmissible mechanically and continued efforts to associate viruses and viroids have so far been unsuccessful. Epidemiological observations suggested that the disease is airborne and probably caused by a mollicute. In order to test this hypothesis, plots were covered by an insect-proof net and with white nets providing 15, 30 and 50% shade. The 30%, 50% and insect-proof nettings provided complete protection, and the 15% netting reduced disease incidence to <2%, compared with>37% in the uncovered control plot. Spraying plants at weekly intervals with a commercial whitewash solution was also found to be effective in reducing disease incidence. The advantages of using a range of netting field treatments as a simple means for obtaining information on the epidemiological nature of a new disease are described.     

三种草坪草净光合速率和蒸腾速率的日变化特点研究

研究了在晴天的条件下，假俭草、细叶结缕草和地毯草净光合速率和蒸腾整编的日变化特点。结果表明：（1）假俭草和细叶绍缕草的净光合速率的日变化曲线均为双峰型，而地毯草为三峰型，3种草出现峰值的时间不同，均有明显的“午睡”现象，日平均净光全速率的大小顺序为俭草＞地毯草＞细叶结缕草；（2）假俭草和细叶结缕草的蒸腾速率的日变化曲线均为双峰型,而地毯草为单峰型，3种草出现峰值的时间有所判差别，“午睡”现象均不明显，日平均蒸腾速率的大小顺序为细叶绍缕草＞地毯草＞假俭草；（3）不同生态因子与3种草坪草净光合速率、蒸腾速率的相关程度均不相等。同种态因子与不同草坪草的净光合速率、蒸腾速率的相关程度也不一样。     

人工神经网络在电力系统短期负荷预测中的应用

根据电力系统短期负荷变化的特性，提出BP模型在实际负荷预测应用中的方法和步骤．对BP网络结构、样本空间、收敛性等作了有针对性的研究．结果表明：多层神经网络应用于电力系统短期负荷预测是可行和有效的．其预报结果比传统的负荷预测方法更准确、经济、效果更好．     

用单层分化法进行胚胎干细胞神经分化的研究

用单层分化法研究了 ES细胞从 0～ 316 h(第 14天 )的神经分化过程。无饲养层培养的 ES细胞在无血清的N2 B2 7培养基中开始神经分化。从 72 h开始 ,神经外胚层的特异性转录因子 Sox1表达 ,在克隆的边缘逐渐出现向外迁移的细胞 ,形状呈圆形或椭圆形 ,有 1～ 2个细胞突起。在 191h(第 8天 )之后 ,细胞可迁移至克隆边缘的 2 0 0 μm之外 ,这些有细长突起的细胞表达神经元特异性蛋白 Tau。以后 ,迁移细胞的突起开始交叠成网状。 316 h(第 14天 )时 ,细长的纤维可跨越在不同的克隆之间 ,长达上千微米。此时 ,免疫染色的结果表明 ,在克隆的中央有较多的圆形 Nestin阳性细胞 ,而克隆的边缘有大量的 β- tubulin 阳性细胞 ,它们交织成复杂的网状图像。阶段性表达基因的表达时序研究表明 ,ES细胞的特异性转录因子 Oct4开始减弱时 ,Sox1表达开始 ,此时细胞由全能状态进入早期神经分化状态 ;Tau的表达是 Sox1表达的下游事件。这一时序与胚胎发育中神经发生的时序相符。单层分化法能直观地展示 ES细胞的神经分化过程 ,可作为研究细胞分化、胚胎发育及药物筛选的体外模型     

电流模式神经网络

把电流模式电路引进了神经网络理论，提出了一种用电流模式电路构成的神经网络阈值处理单元，并由此构成了电流模式Ｈｏｐｆｉｅｌｄ神经网络，此网络中信号都是以电流的形式出现的。     

在DELPHI5.0平台上利用NETBIOS设计网络通讯软件的探讨

在组建六道湾煤矿企业信息网的过程中,我们利用DELPHI5.0平台所提供的NET-BIOS接口单元,解决了从企业原有的专用网络上获取广播数据包的异构网络间通讯问题,本文介绍了DELPHI5.0平台所提供的NETBIOS接口单元的组成及主要功能,并详细说明了网络广播数据包接收器的编程方法.     

Analysis of Growth and Productivity of Indian Mustard (Brassica juncea) in Relation to FYM, Nitrogen and Source of Fertilizer

A field experiment was conducted during the winter seasons of 1992–93 and 1993–94 at Anand to study the effect of FYM, nitrogen and source of fertilizer on growth and yield of mustard [ Brassica juncea (L.) Czernj & Cosson]. The results showed significant variation in leaf area index (LAI), crop growth rate (CGR), dry matter production and seed yield. The direct effect of farmyard manure (FYM) was conspicuous in improving the growth of mustard. FYM application at 10 tonnes ha⁻¹ significantly increased the LAI, CGR and dry matter accumulation per plant at almost all the stages during first year study (1992–93) and in pooled analysis. Similarly, nitrogen application registered maximum LAI, CGR at 75 kg level and RGR and NAR at 50 kg level at almost all the during both years. Sulphur carrying source (Ammonium sulphate plus single super phosphate) increased all stages growth characters. Maximum dry matter accumulation per plant and seed yield were recorded with highest levels of FYM (20 tonnes ha⁻¹), N (75 kg ha⁻¹) and source having S. Seed yield was strongly associated with LAI and dry matter accumulation per plant at all the stages.     

晚稻稻瘟病BP神经网络分区预报

应用相关分析方法分析了浙江省19个县1988～1999年晚稻稻瘟病发病与有关环境因子的关系,筛选了8个气象因子用于晚稻稻瘟病发生程度长期预报。根据各预报因子与稻瘟病发病程度相关性,采用邻接二维图论聚类分析法,将19个点(县)划分为4个生态区。每个生态区内运用BP神经网络技术建立模型,并进行拟合和试报。1997～1999年试报验证,在划分稻瘟病生态区的基础上,应用BP神经网络模型对稻瘟病进行长期预测预报是可行的,3年试报成功率分别是78.95％、84.21％和78.95％。文中还对该方法与过去常用的预报方法的试报结果作了比较。     

Restoration of tuber production by exposure to unshaded daylight in Eleocharis kuroguwai Ohwi shaded during the early tuber formation period

The loss of final tuber weight of Eleocharis kuroguwai Ohwi by shading during the early tuber formation period (TFP) is overcome by exposure to unshaded daylight thereafter (late TFP). In the present study, the growth parameters that contribute to the dry matter increase (DMI) per day of tubers in the late TFP were examined. DMI of the tuber during the late TFP was determined by that of the whole plant and the ratio of the DMI of the tuber to that of the whole plant during this period. The ratio of the DMI of the tuber to that of the whole plant during the late TFP was significantly correlated with the DMI of the whole plant during the first 14 days of the late TFP. During the late TFP after the exposure to unshaded daylight, DMI of the whole plant correlated with the surface area of the stem (SAS) and net assimilation ratio (NAR), and the SAS correlated with the stem dry weight (DW) and specific stem-surface area (SSA). SSA negatively influenced NAR, but NAR was increased by unshading. During the late TFP after shading, the effect of the decrease of the stem DW due to shading on the DMI of the whole plant was mitigated by the large SAS and high NAR. These results indicate that the growth parameters that contribute to the DMI of tuber during the late TFP after exposure to unshaded daylight are SAS and NAR just after unshading, and SSA during this period.