利用改进BP神经网络预测初产奶牛产奶量的方法

针对传统BP算法的缺陷,提出了一种采用L-M训练法的BP神经网络。在此基础上建立了基于改进BP神经网络非线性系统的奶牛305d产奶量预测模型,此模型可以提前215d预测初产奶牛305d产奶量,从而实现提早进行选择,加速奶牛育种工作进程。并通过具体的实验验证了改进BP神经网络预测模型的有效性。     

基于GF-6/WFV卫星遥感的大田冬小麦叶片氮素含量估测

为对大田冬小麦叶片氮素含量（LNC）进行快速、准确及无损监测,通过在江苏省泰州泰兴市、盐城大丰区和南通如皋市布设冬小麦遥感监测大田试验,在获取试验样点冬小麦冠层红光波段反射率（RED_ref）、近红外波段反射率（NIR_ref）和计算的十个光谱指数（RVI、NDVI、DVI、SAVI、OSAVI、MSR、RDVI、EVI2、NLI和SVI）基础上,将12个遥感光谱指标与冬小麦LNC进行相关分析,选出与LNC相关性较好的作为模型输入变量,构建基于BP神经网络的冬小麦LNC估测模型, 并利用GF-6/WFV卫星遥感影像对县域冬小麦LNC的空间分布开展监测。结果表明,12个遥感光谱指标与冬小麦LNC之间存在不同程度的相关性,其中NDVI、RVI、MSR、OSAVI和NLI与冬小麦LNC的相关性较好（相关系数不低于0.65）。将优选的5个遥感光谱指标作为模型输入变量,构建基于BP神经网络的冬小麦LNC估测模型（LNC-BPEM）,模型的估测精度r²=0.866,RMSE=0.246%,ARE=12.9%。将冬小麦LNC-BPEM估测模型和GF-6/WFV影像结合对县域冬小麦LNC的空间信息监测,获得了如皋县域冬小麦LNC的空间分布特征,该区域冬小麦LNC范围在0.9%~2.0%（长势正常）的种植面积为29 693.3 hm²,占冬小麦总种植面积的74%。这说明利用GF-6/WFV卫星的多个遥感光谱指标与神经网络结合建模可有效估测县域大田冬小麦叶片氮素含量。     

基于校园网环境下机房总管Ⅱ系统的设计与实现

本文介绍了一种基于校园网实现机房自动化管理系统的设计思路与实现方法.该系统用助于提高机房的管理水平、使用效率与管理质量.     

采摘机器人位置校正方法研究——基于无线传感网络和超带宽调频方法

超带宽雷达是新近出现的一种高分辨率雷达,当信号带宽和其发射中心频率的比大于0.25时,就被称作超带宽,带宽越大,雷达对目标的分辨能力越高。超带宽雷达具有高分辩能力,如果将其使用在采摘机器人定位导航系统中,可以有效地提高采摘机器人的定位导航精度和效率。为此,提出了一种基于无线传感网络和超带宽调频技术的采摘机器人位置校正方法,并以实际采摘环境作为研究对象,对采摘机器人的位置校正过程进行了测试。测试结果表明:采用超带宽调频无线传感网络方法,定位校正精度要高于传统的定位算法,且定位效率高,对于新型自动化采摘作业机器人的研究具有重要的指导意义。     

A multi-analysis approach for space–time and economic evaluation of risks related with livestock diseases: The example of FMD in Peru

This study presents a multi-disciplinary decision-support tool, which integrates geo-statistics, social network analysis (SNA), spatial-stochastic spread model, economic analysis and mapping/visualization capabilities for the evaluation of the sanitary and socio-economic impact of livestock diseases under diverse epidemiologic scenarios. We illustrate the applicability of this tool using foot-and-mouth disease (FMD) in Peru as an example. The approach consisted on a flexible, multistep process that may be easily adapted based on data availability. The first module (mI) uses a geo-statistical approach for the estimation (if needed) of the distribution and abundance of susceptible population (in the example here, cattle, swine, sheep, goats, and camelids) at farm-level in the region or country of interest (Peru). The second module (mII) applies SNA for evaluating the farm-to-farm contact patterns and for exploring the structure and frequency of between-farm animal movements as a proxy for potential disease introduction or spread. The third module (mIII) integrates mI–II outputs into a spatial-stochastic model that simulates within- and between-farm FMD-transmission. The economic module (mIV) connects outputs from mI–III to provide an estimate of associated direct and indirect costs. A visualization module (mV) is also implemented to graph and map the outputs of module I–IV. After 1000 simulated epidemics, the mean (95% probability interval) number of outbreaks, infected animals, epidemic duration, and direct costs were 37 (1, 1164), 2152 (1, 13, 250), 63 days (0, 442), and US$ 1.2 million (1072, 9.5 million), respectively. Spread of disease was primarily local (<4.5 km), but geolocation and type of index farm strongly influenced the extent and spatial patterns of an epidemic. The approach is intended to support decisions in the last phase of the FMD eradication program in Peru, in particular to inform and support the implementation of risk-based surveillance and livestock insurance systems that may help to prevent and control potential FMD virus incursions into Peru.     

基于电子商务的网络财务管理探悉

电子商务的蓬勃发展给企业管理带来巨大冲击，财务管理作为企业管理链条上的重要一环，其管理模式也因此而发生了重大变革。文中首先介绍了网络财务管理产生的背景及其概念，讨论了网络财务管理的优势所在，明确了网络财务管理产生和发展的技术基础，重点对网络财务管理的开发与实施展开了探讨。     

牛奶中抗生素残留检测研究进展

随着人们对食品安全的重视,牛奶中各种有害物质的残留受到了消费者的极大关注。虽然中国已经建立牛奶抗生素残留的标准方法,但操作复杂、应用不便,且难以解决目前某些抗生素残留的新情况,有待进一步更新。文章综述了牛奶中抗生素残留检测方法进展,旨在为进一步开展牛奶抗生素残留检测提供参考。     

Egg production forecasting: Determining efficient modeling approaches

Several mathematical or statistical and artificial intelligence models were developed to compare egg production forecasts in commercial layers. Initial data for these models were collected from a comparative layer trial on commercial strains conducted at the Poultry Research Farms, Auburn University. Simulated data were produced to represent new scenarios by using means and SD of egg production of the 22 commercial strains. From the simulated data, random examples were generated for neural network training and testing for the weekly egg production prediction from wk 22 to 36. Three neural network architectures-back-propagation-3, Ward-5, and the general regression neural network-were compared for their efficiency to forecast egg production, along with other traditional models. The general regression neural network gave the best-fitting line, which almost overlapped with the commercial egg production data, with an R(2) of 0.71. The general regression neural network-predicted curve was compared with original egg production data, the average curves of white-shelled and brown-shelled strains, linear regression predictions, and the Gompertz nonlinear model. The general regression neural network was superior in all these comparisons and may be the model of choice if the initial overprediction is managed efficiently. In general, neural network models are efficient, are easy to use, require fewer data, and are practical under farm management conditions to forecast egg production.