烟台地区水面蒸发BP神经网络模型研究

运用BP神经网络模型对水面蒸发进行研究,并与多元线性回归和主成分回归2种方法的拟合结果进行比较。结果显示,多元线性回归各参数均通过t检验,拟合较好;主成分回归中,参数b2未通过t检验,拟合效果不如多元线性回归好。BP神经网络模型、多元线性回归、主成分回归建立的水面蒸发量观测值与拟合值的回归方程中决定系数分别为0.8986、0.7993、0.7984。应用BP神经网络进行分析,相对误差小于10%的样本个数超过总样本个数的40%,相对误差不超过30%的样本个数接近80%;而其它2种方法相对误差大于10%的样本个数超过总样本数的80%,相对误差大于50%的接近总样本个数的30%。可见,应用BP神经网络模型进行水面蒸发量计算,远优于其它2种方法,应用此方法进行水面蒸发量的预测是非常理想的。     

Ratios of Nutrients in Lowland Rice Grown on Two Iron Toxic Soils in Nigeria

ABSTRACT

Iron (Fe) toxicity is a widespread nutritional soil constraint affecting rice production in the wetland soils of West Africa. Critical levels of total iron in plant causing toxicity is difficult to determine as different rice cultivars respond to excessive Fe^{2 +} in various ways in what is called “bronzing” or “yellowing” symptoms (VBS). An investigation was conducted to evaluate the relationship between plant growth and nutrient ratios at four iron levels (1000, 3000, 4000 μ g L^?1) and control. This involved two rice cultivars (‘ITA 212’ and ‘Suakoko 8’), and two soil types (Aeric Fluvaquent and Aeric Tropaquept). The experimental design was a 2 × 2 × 4 factorial in a completely randomized fashion with four replications. The results showed that nutrient ratios [phosphorus (P)/Fe, potassium (K)/Fe, calcium (Ca)/Fe, magnesium (Mg)/Fe, and manganese (Mn)/Fe), Fe content, and Fe uptake vary widely with the iron levels as well as with the age of the cultivars. The iron toxicity scores expressed as VBS increased with increasing Fe^{2 +} in the soils, resulting in simultaneous reduction of the following variables: plant height, tiller numbers/pot, relationships grain yield (GY) and dry matter yield (DMY). There were no significant difference between nutrient ratios, Fe contents, Fe uptake, the GY and DMY of both rice cultivars on both soil types. Multiple stepwise regression analysis showed that Fe uptake and Fe contents contributed 42% and 17% respectively to the variation in the grain yield of ‘ITA 212’ on Aeric Tropaquept. On both soil types and cultivars, Fe uptake and Fe content contributed between 26 and 68% to the variation in the DMY, while the nutrient ratios (P/Fe, K/Fe, Ca/Fe, and Mn/Fe) contributed between 3% and 13% DMY. Thus, it could be concluded that iron toxicity in rice is more a function of a single nutrient (Fe) rather than nutrient ratios.     

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。     

Relationship of morphological variables to rubber production in guayule

Summary The objective of this study was to find measurable characters to aid the plant breeder in predicting both rubber percentage and rubber yield of guayule (Parthenium argentatum Gray). A field of 234 guayule breeding lines was harvested by clipping in February 1986. Forty-two of these lines were selected for further evaluation on the basis of superior regrowth, rubber content and yield. These 42 lines were reharvested in February 1987. Nineteen variables were evaluated including percent rubber and resin, rubber and resin yield (g/plant), plant height (cm), width (cm), and volume (m³), fresh and dry weight (kg/plant), percent dry weight, stem number, mean stem diameter (cm), total and mean stem circumference (cm), total and mean stem area (cm²), total stem circumference/total stem area (cm^-1), and regeneration at two dates. Dry weight (kg/plant) had the best correlation to rubber yield (r=0.85), and plant volume best predicted dry weight (r=0.83). These variables did not correlate as highly with percent rubber. The best statistical regression model for percent rubber included four variables, fresh and dry weight, percent dry weight, and plant volume. This model accounted for 51% of the variation. In comparison, the best regression of rubber yield had an R²=0.85 for plant height and width, volume, and dry weight.     

黄淮冬麦区气象因子与小麦晚霜冻害关系研究——以商丘市为例

基于1956-2015年商丘市气象数据和霜冻害资料,研究黄淮冬麦区商丘市气象因子与霜冻害的关系,以及气温、地温、相对湿度、风速与草面温度的偏相关和多元线性回归关系。结果表明:随着全球气温变暖,商丘市冬春季积温以4.35℃·d·a~(-1)的速率逐年增加;晚霜冻害除受冷空气活动影响外,与前期积温和降水量密切相关,冬春季积温偏高或降水量偏少的情况下易发生晚霜冻害;随着小麦幼穗的发育,其对低温的敏感度增加,且可引起霜冻害的最低温度有逐渐升高的趋势;气温、地表温度、平均相对湿度、平均风速与草面温度呈极显著相关(P0.01),偏相关系数的大小表现为地表温度气温平均相对湿度平均风速,可见,除气温、地表温度外,平均相对湿度、平均风速也对霜冻害的发生及轻重程度起关键作用;各因子与草面温度可用模型表述为Y=0.558ST+0.482AT+0.087RH+1.304WS-12.704,经检验,线性回归方程成立,可通过该模型对草面温度进行监测,并为未安装草面温度传感器的地区提供可靠的冻害评估结果。     

Selection proposal of cultivars of spring barley in the years from 2001 to 2004, using Joint Regression Analysis

Joint Regression Analysis is a widely used technique in the analysis of series of experiments for cultivar comparison. We show how this technique may be used in the selection of recommended cultivars. We apply this technique to a series of experiments on spring barley cultivars carried out from 2001 to 2004. The data in these designs were kindly made available by the Central Institute for Supervising and Testing in Agriculture in Brno, Czech Republic.     

Selection of locations of knots for linear splines in random regression test‐day models

Using spline functions (segmented polynomials) in regression models requires the knowledge of the location of the knots. Knots are the points at which independent linear segments are connected. Optimal positions of knots for linear splines of different orders were determined in this study for different scenarios, using existing estimates of covariance functions and an optimization algorithm. The traits considered were test‐day milk, fat and protein yields, and somatic cell score (SCS) in the first three lactations of Canadian Holsteins. Two ranges of days in milk (from 5 to 305 and from 5 to 365) were taken into account. In addition, four different populations of Holstein cows, from Australia, Canada, Italy and New Zealand, were examined with respect to first lactation (305 days) milk only. The estimates of genetic and permanent environmental covariance functions were based on single‐ and multiple‐trait test‐day models, with Legendre polynomials of order 4 as random regressions. A differential evolution algorithm was applied to find the best location of knots for splines of orders 4 to 7 and the criterion for optimization was the goodness‐of‐fit of the spline covariance function. Results indicated that the optimal position of knots for linear splines differed between genetic and permanent environmental effects, as well as between traits and lactations. Different populations also exhibited different patterns of optimal knot locations. With linear splines, different positions of knots should therefore be used for different effects and traits in random regression test‐day models when analysing milk production traits.     

Pacific sardine (Sardinops sagax, Jenyns 1842) landings prediction. A neural network ecosystemic approach

In this study the performances of computational neural networks (CNNs), multiple linear regressions (MLRs) and generalised additive models (GAMs) to predict Pacific sardine (Sardinops sagax) landings and to analyse their relationships with environmental factors in the north area off Chile were studied. For this purpose several local and global environmental variables and indexes (sea surface temperature, sea level and Ekman transport index in the Chilean coast and, sea surface temperature in the area Niño 3 + 4 and Niño 1 + 2, and the south oscillation index) were considered as inputs or independent variables. Additionally, several CNNs were calibrated and validated adding the anchovy (Engraulis ringens) landings in the same area as model inputs. The time lags of the variables considered were selected through analysis of the non-linear cross-correlation functions and an alternative form of sensitivity analysis based on the approach of the missing value problem. The analysis of error measures with validation data set showed that the best results were obtained when local and global variables were used separately and combined with anchovy landings. Globally, the best result was given by a CNN with 18 input variables (model CNN 6(II) which only considered global variables and anchovy landings) and 10 neurons in a hidden layer. For this configuration the explained variance was slightly higher to 86% which supposed a standard error of prediction of 7.66%. These results were significantly better than those obtained with MLRs and GAMs. The strong correlation between predicted and observed sardine landings suggests that CNNs captured the trend of the historical data. Also, the generalisation capacity together the sensitivity analysis allowed us to identify the variables with a high weight in the model and partially to interpret the statistical functional relationships between these environmental variables and sardine landings.     

黄烷酮类衍生物的抗菌活性QSAR研究

为了研究黄烷酮衍生物分子结构与杀菌活性的关系,本文以34种黄烷酮衍生物的杀菌活性为指标,以量子化学计算软件PCLIENT提取化合物的初始分子描述符,经高维特征筛选以及多轮末尾淘汰进行特征选择,获得了14个重要分子描述符,并基于保留描述符构了建多元线性回归、支持向量回归模型,获得了较高的预测精度。最后基于支持向量机非线性解释体系,对保留分子描述符的作用进行解析,各描述符对影响黄烷酮衍生物的杀菌活性的重要性依次为SssO > MATS6e > SeaC2C2aa > EEig04d > EEig03x > C-028 > R7v+ > L2e > X4A > Mor25m > G1m > Mor17m >BEHv1 > RDF045m,为指导黄烷酮类杀菌剂的合成提供了重要指导。     

植烟土壤磷素与烟叶化学成份及品质的相关性研究

采用相关以及多元线性回归分析法，对植烟土壤中的全磷、速效磷、缓效磷与烟叶化学成份及评吸质量的关系进行研究，找到其相关性，得出相应的定量关系式，并评价了磷形态与烟叶化学成份及评吸质量作用的显著水平。