基于最小二乘支持向量机模型预测田面糙率

田面糙率是影响地面灌溉质量的重要参数。基于最小二乘支持向量机建立了两类4个田面糙率预测模型,并进行了验证。结果表明第一类模型预测值(即作物地采用LSSVM-N-I3、裸地采用LSSVM-N-I1,翻耕地采用LSSVM-N-I2)相对误差最大值为9.7%;第二类模型预测值(即LSSVM-N-II模型)相对误差最大值为10.5%,由此可见两类模型都具有较高的预测精度,可以用于田面糙率的预测。     

优化运行理论在泵站综合自动化系统中的应用

利用改进的神经元网络构建了泵站流量以及效率预测模型，并利用改进的遗传算法构建了泵站优化模型。将这两个模型应用到泵站综合自动化系统中，可以完成对泵站流量及效率的准确预测，并在此基础上实现泵站的优化运行。讨论了泵站优化运行理论在泵站综合自动化系统中应用的步骤和方法。     

灰色动态模型群法在河流水质预测中的应用初探

在对原始数据序列对数变换的基础上，依据灰色系统理论，构造了由6个GM（1，1）模型组成的灰色动态模型群，并用于淮河干流枯水期氨氮浓度变化趋势预测。研究表明，灰色动态模型群法能够充分利用近期水质资料信息预测水质变化趋势；相对单个GM（1，1）模型，灰色动态模型群法能有效改善随机波动数据序列的拟合效果，提高预测精度。     

灌溉预报与节水灌溉决策专家系统研究

从节水灌溉的基本原理入手，分析灌溉预报中的各个参量，提出了节水灌溉决策的依据、方法及评价标准。介绍了灌溉预报决策专家系统中由框架库、规则组库和黑板3部分组成的综合知识体结构的知识表示技术。     

黑龙江省水稻单产增长潜力预测方法研究

水稻是黑龙江省的主要粮食作物。加入WTO后,水稻生产、贸易、进出口等方面都面临新的形势,同时水稻产量的稳定和增长对于国家粮食安全、人民生活水平提高以及农民奔小康都有着重要意义。因此,探讨了应用Peal曲线函数模型对黑龙江省水稻单产进行预测的方法。     

立木生物量方程的优度评价和精度分析

以建立东北落叶松和南方马尾松立木生物量模型为例,对模型的优度评价和精度分析进行专题研究。首先,在综合分析各种常用模型评价指标的基础上,提出6项基本评价指标,并通过对2个树种的生物量模型进行评价,证明是可行和有效的;然后,基于线性回归估计的基本假设及其置信区间计算公式,提出适合立木生物量模型预估的条件均值和单一预估值置信区间的估计方法,并确定2个树种生物量模型的置信区间;最后,通过蒙特卡罗模拟进行随机再抽样试验,结果表明采用检验样本进行适用性检验的做法不可取,应该利用全部样本(不分建模样本和检验样本)来建立模型,它能充分利用样本信息,使模型的预估误差最小。     

Implications of the number of years assessment on recommendation of common bean cultivars

In the common bean crop in Brazil, the requirement of the value for cultivation and use trials is that these experiments must be conducted over two years in three locations per region. Information in regard to the necessary number of years to ensure precision in cultivar recommendation and the influence of evaluated years in the estimation of the GE interaction are still scarce. Using grain yield of five check varieties assessed over 11 years in three sowing seasons per year, the aims of this study are as follows: to measure the relative contribution of the GE interaction, evaluating the implication of the number of years in the estimates of the GE interaction, and infer how many years are needed to ensure precision in cultivar recommendation. For instances, analysis of variance was carried out involving all environments and also combinations of years. The results showed that the GE interaction was greater than all other cross‐effects involving lines. The use of at least two years allows good coincidence in cultivar recommendation compared to the whole period. Increasing the evaluation time is a good strategy, especially when it is difficult to grow three different sowing seasons.     

Junction stiffness analysis and vibration noise prediction of wind power speed-increase gearbox

In order to study the vibration characteristics and radiation noise of wind power speed-increase gearbox, a torsional vibration model of wind power speed-increase gearbox is established based on the analysis of supporting stiffness of bearing and contact stiffness of gear pairs. By solving the vibration differential equation with the help of Matlab, the frequency and corresponding vibration mode are obtained. With taking stiffness excitation, error exaction and meshing impact exaction into account, the dynamic finite element model of speed-increase gearbox is set up, and the dynamic response simulation is carried out. Regarding vibration displacement of the nodes on gearbox surface as boundary condition, an acoustic boundary element model of speed-increase gearbox is built. The surface acoustic pressure of gearbox and the radiation noise of field points are solved by the direct boundary element method. The results show that there is a great difference between torsion frequency and excitation frequency of wind power speed-increase gearbox, and so the resonance doesn't occur. The maximum structural noise and radiation noise mainly appear near the double octave of the gears meshing frequency of high-speed grade.     

Prediction of the Economic Development for the Circumjacent Reservoir Region of the Longtan Hydropower Project Using Grey Quantitative Analog

In order to overcome some technical difficulties,we adopt ingenously the grey models combined with the quantitative analog method,and select the Dan Jiangkou Project which has accumulated practical operation data for a long time as the analog project to predict the economic influence for the circumjacent reservoir region of the Longtan hydropower project.On the basis of analog we have built various industrial economic grey models,and compare them with the corresponding predictions for the unconstructed reservoir case.The predictions of the former are greatly larger than the later ones.and hence provides a certain scientific basis for the central leaders to make a decision for constructing the huge hydropower project ultimateiy.