基于时间序列分解方法的太湖未来特征水位预测

基于太湖沿湖5个水文台站1956—2006年的逐日平均水位数据,提取得到了逐年特征(平均、最高和最低)水位,运用时间序列分解方法模拟和预测了太湖未来15a的特征水位,并探讨了太湖未来洪灾情势。将水位序列分解为趋势成分、周期成分和随机成分,得到时间序列分解模型,并进行了水位序列模拟和预测。分析结果表明,该模型的模拟精度比较理想,可以实现未来较长期的特征水位预测。预测得到未来15a后的最高水位可能达到4.05m,显著超过太湖警戒水位,因此需引起相关部门的重视。     

基于BP神经网络的离心泵关死点功率预测

离心泵关死点功率至今还不能通过理论计算求得。该文介绍了BP神经网络的结构和特点及其在离心泵性能预测领域的应用现状。基于BP神经网络建立了离心泵关死点功率的预测模型。给出了预测模型的输入模式,并应用试凑法确定了BP神经网络中间隐含层的数目。用46组数据该预测模型进行了训练并给出了神经网络权值和阈值,用3组数据该预测模型进行了仿真并对仿真结果进行了线性回归分析。研究结果表明:建立的离心泵关死点功率预测模型具有比较高的预测精度,其预测平均偏差为4%,可以应用于工程实践中离心泵关死点功率的理论求解。     

北京市水产养殖病害测报系统设计与实现

研究开发了北京市水产养殖病害测报系统,通过B/S架构实现水产养殖病害测报的电子化、网络化以及智能化,从而提高了对数据的采集、分析处理和应用能力,为政府部门决策提供可靠及时的参考,为北京市水产养殖病害的防治工作提供有力的信息化工具。     

基于信息融合的灌区水情预测方法研究

引入信息融合理论,通过在系统输入变量上叠加"干扰流量",基于对灌区用水模糊特征分析估计其不确定性,实现分散且孤立的实时监测数据之间的关联,使系统保持必要的信息冗余,进而实现灌区水情信息融合。该方法可用于建立具有跟踪和预测能力的灌区实时仿真系统,克服了信息融合理论用于非线性时变系统的诸多限制,扩展了信息融合理论在系统关系确定的民用领域的应用。     

The development of mature plant resistance in four potato cultivars against aphid-inoculated potato virus YO and YN in four potato cultivars

Summary Field-grown potato plants of cvs King Edward, Record, Maris Piper and Désirée were inoculated on seven different dates during the growing season of 1987 and 1988 with either potato virus Y^O (PVY^O) or PVY^N, using three viruliferous peach-potato aphids (Myzus persicae) per plant. In each cultivar, the proportion of progeny tubers infected with PVY^O or PVY^N was high in plants inoculated during the four weeks following emergence, the proportion declining to zero or close to zero in the subsequent 4–6 wks.     

基于核函数支持向量回归机的耕地面积预测

科学预测耕地保有量是耕地保护的基础,对缓解用地矛盾、保证粮食安全具有重要指导意义。为探讨不同核函数支持向量回归机(support vector regression,SVR)对耕地面积预测的影响,该文以惠州市为例,分别采用多元回归、BP神经网络及3种不同核函数SVR建立耕地面积预测模型并进行对比试验。预测结果精度分析显示,RBF核函数SVR预测结果平均相对误差为0.54%,均方根误差为0.007,精度最高;Sigmoid核函数SVR预测结果对应误差分别为1.12%及0.012,精度次之;多项式核函数SVR预测结果对应误差为分别为2.71%及0.032,高于BP神经网络模型,但低于多元回归模型。研究表明,在现有3种常用核函数SVR耕地面积预测模型中,基于RBF核函数SVR模型预测能力最强,其次是sigmoid核函数;而多项式核函数则效果较差。     

Tracking the potato late blight pathogen in the atmosphere using unmanned aerial vehicles and Lagrangian modeling

A means for determining the aerial concentration, C (sporangia m⁻³), of plant pathogenic spores at various distances from a source of inoculum is needed to quantify the potential spread of a plant disease. Values of C for Phytophthora infestans sporangia released from an area source of diseased plants in a potato canopy was quantified in three ways: (1) by using Rotorods to sample the air just above the source, (2) by using unmanned aerial vehicles to sample the air at altitudes up to 90 m above the source and at downwind distances up to 500 m from the source, and (3) by using a Lagrangian stochastic simulation of sporangia flight trajectories to tie these two measurements together. Experiments were conducted using three potato crops over two years. Model predictions of time-average, crosswind-integrated concentrations were highly correlated (r = 0.9) with values of C measured using the unmanned aerial vehicles. The model describes the release and dispersal of sporangia from a potato canopy to a downwind distance of 500 m. Thus, it may have utility as a part of an area-wide decision support system by helping to predict risk of disease spread between neighboring or distant potato fields.     

河北北部二代粘虫发生的预测

根据丰南县1981～1990年粘虫田间发生的调查资料,经回报检验,10a 粘虫发生情况的回报准确率均为90%,明显高于经验预报,而且能在粘虫实际发生前1个月做出定量预报,具有较好的实用价值。     

区域水资源可持续利用预警指标体系构建的探讨

 对建立区域水资源可持续利用预警指标体系进行了初步的探讨。在对构建区域水资源可持续利用预警指标体系的指导思想、设计思路和设计原则进行了论述后,设计了目标层、准则层、指标层3个层次的区域水资源可持续利用预警指标体系框架。并简要预示了区域水资源可持续利用预警指标体系的测评方法。     

Prognosis of the nutritional status of apple trees based on prebloom leaves and flowers

Abstract

The utilization of prebloom tissues to predict nutrition of apple (Malus domestica Borkh.) trees is limited. Therefore, the aim of this study was to assess the possible use of mineral analyses of the prebloom spur leaves and flowers to forecast the nutrition of ‘Idared’ apple trees in the mid-summer (based on the leaves from current-season shoots). The prebloom leaf concentrations of phosphorus, potassium, magnesium, boron and manganese were positively correlated with those in the summer leaves. The significant relationships between nitrogen, calcium, boron, and manganese in the flowers and those in the mid-summer leaves were also found. However, only the relationship between boron in the prebloom spur leaves and in the summer leaves was sufficiently strong. The calculated optimal range of boron in the prebloom spur leaves was 20–25?mg kg^?1.