组合预测方法及其在粮食产量预测中的应用

提出了一种组合预测方法，以“预测偏差绝对值的和最小”为标准，用线性规划确定组合权系数的最优加权算法，对青海省的粮食总产量进行了预测分析。结果表明，此模型具有较高的精度。     

分析阶梯流量井流试验数据的线性回归法

文中应用线性回归法，对一次阶梯流量井流试验数据进行适当交换后，可得回归直线方程，与目前采用的多次阶梯流量井流试验数据的分析相比较，在同样可以得到井流实验欲求的井损系数C，综合参数和含水层导水系数T三个参数的情况下，至少可以使抽水量的变化次数减少两次。实际算例表明，一次阶梯流量井流与多次阶梯流量法的计算结果非常吻合。     

Conversion to organic arable farming in The Netherlands: A dynamic linear programming analysis

Several studies show that organic farming is more profitable than conventional farming. However, in reality not many farmers convert to organic farming. Policy makers and farmers do not have clear insight into factors which hamper or stimulate the conversion to organic farming. The objective of this paper is to develop a dynamic linear programming model to analyse the effects of different limiting factors on the conversion process of farms over time. The model is developed for a typical arable farm in The Netherlands central clay region, and is based on two static liner programming models (conventional and organic). The objective of the model is to maximise the net present value over a 10-year planning horizon. The results of the analysis of a basic scenario show that conversion to organic farming is more profitable than staying conventional. In order to arrive at the actual profitable phase of organic farming, the farmer has to pass through the economically difficult 2-year conversion period. Sensitivity analysis shows that if depreciation is 25% higher than conventional fixed costs due to machinery made superfluous by conversion, conversion is less profitable than staying conventional. Also the availability of hired labour, which can be constrained in peak periods, has a strong effect on the cropping plan and the amount of area converted. Further analysis shows that a slight drop (2%) in organic prices lowers the labour income of the farmer and makes conversion less profitable than conventional farming. For farmers, a minimum labour income can be required to ‘survive’. The analysis shows that constraint on minimum labour income makes stepwise conversion the best way for farmers to overcome economic difficulties during conversion.     

多元线性回归和神经网络在水库调度中的应用比较研究

采用多元线性回归和改进人工神经网络算法分别建立水库多目标调度函数，并进行实例计算验证。通过比较分析，表明人工神经网络的非线性映射能力能够更好地反映水库调度中多个自变量和因变量之间的复杂关系，具有较高的模拟精度和较好的可行性，且应用简便。     

电磁振动排种器振动特性的理论分析

排种器是播种机械的核心,是决定播种机性能的主要因素。在设计排种器时,首先要对其进行振动特性分析,找出影响排种器排种性能的因素,并作为后续设计的理论依据。为此,介绍了电磁振动排种器的工作原理,并对其振动特性进行了理论分析,从而为电磁振动排种器的设计提供了理论依据。     

泌乳期中国荷斯坦牛转化孕酮浓度的影响因素分析及其表型预测

【目的】本研究基于试验测定的中国荷斯坦牛血液和乳汁中的孕酮浓度表型，探究了影响中国荷斯坦牛体内孕酮分泌的环境因素、孕酮浓度的变化趋势、孕酮浓度与乳成分的关联程度以及血乳中孕酮浓度的预测方法。【方法】试验于2021年8月在北京、山东两个牧场采集不同胎次、泌乳阶段和妊娠状态的中国荷斯坦泌乳牛的奶样、尾根血样，测定孕酮浓度，最终获得402条乳汁孕酮浓度和298条血液孕酮浓度表型用于数据分析。对孕酮浓度进行数据转化使其近似服从正态分布后，采用固定效应模型探究胎次、泌乳阶段、妊娠状态、牧场等固定效应对奶牛孕酮表型的影响，运用R语言cor函数计算孕酮与各乳成分间的关联，并利用偏最小二乘法和个体及乳成分信息对孕酮浓度进行预测，以建立孕酮浓度表型高通量获取手段。【结果】妊娠状态对转化乳汁孕酮浓度存在极显著影响（P<0.01），胎次、场效应对转化乳汁孕酮浓度均有显著影响（P<0.05），而转化血液孕酮浓度只受到妊娠状态影响（P<0.01）；全乳固体、乳脂率、乳蛋白率、脂蛋比与转化乳血孕酮浓度均存在显著或极显著的正相关关系（r=0.14~0.37，P<0.05；P<0.01）；基于本试验数据，乳成分与个体信息对转化乳血孕酮浓度的预测准确性不高（R²=0.030~0.17），但如果增加血液或乳汁的转化孕酮浓度对乳汁或血液的转化孕酮浓度进行预测，预测效果则有大幅提升（R²=0.40）。【结论】影响泌乳期中国荷斯坦牛转化孕酮浓度的因素除妊娠状态外，可能还包括饲养条件与胎次。此外转化孕酮浓度与乳脂率、乳蛋白率等乳成分呈极显著相关。基于乳成分信息与转化孕酮的关系，获得了对中国荷斯坦牛乳血转化孕酮浓度预测的可用策略，为今后牧场的繁殖辅助管理、奶牛育种新性状研发以及孕酮浓度的高通量获取等提供了新思路。     

Discrimination between Japanese and Chinese Products using trace elements concentration of kernel in pickled Japanese apricot

pp. 875–880

The trace-element composition of kernel in pickled Japanese apricot (Prunus mume Sieb. et Zucc.) was determined using an inductively coupled plasma optical emission spectrometer in order to distinguish between Japanese products and Chinese products.

Strontium and barium concentrations in the kernels of Chinese products were 10 or more times those of the Japanese ones. When based on 8.0 mg kg^?1 of strontium concentration in kernel, 93.2% of sample was distinguished as Japanese products or Chinese ones.

Applying principal component analysis using 9 elements (Mn, Zn, Fe, Ni, Ba, Sr, Cu, Co, Cr), the pickled Japanese apricots tend to separate into two countries. Linear discriminant analysis (LDA) using 9 elements allowed a reasonable classification of pickled Japanese apricots according to the country of production.

The result of the analysis of K-nearest neighbors (KNN) was better than that of LDA.     

基于PCA-ARIMA模型的高血压发病率预测

目的高血压发病率是政府和相关医学工作者预防和监测高血压的重要依据之一。方法利用主成分分析(principal component analysis,PCA)对因子进行线性筛选,获得保留因子后利用ARIMA进行建模预测,即为PCA-ARIMA多维时间序列组合预测模型。结果高血压发病率的拟合与独立预测结果表明,PCA-ARIMA优于PCA-MLR、ARIMA等参比模型。结论本文提出的基于主成分分析和ARIMA模型(PCA-ARIMA模型)的建模有助于提高模型的预测精度。     

Spatial optimization by simulated annealing and linear programming

Forest management planning comprises selection among treatment alternatives in management units. A traditional linear programming (LP) approach may effectively account for a profit maximization objective combined with sustainability constraints, e.g. on the temporal distribution of harvest volume flows, cash‐flow, and net present value development, but it fails to account for spatial constraints, especially those associated with final felling. By applying a simulated annealing adjacency model based on net present value maximization and combined with an LP consequence computation model, it is possible to delineate optimal strategies of final felling scheduling. Evaluation is made of the trade‐off between (1) the incremental cost (determined by use of the LP model) of an optimal adjacency model solution, and (2) the potential damage cost resulting from adjacency characteristics such as windthrow and bark injuries. The decision support system may contribute significantly to reduce damage costs and may improve the reliability of forest management planning.