应用近红外光谱分析技术快速测量蔗汁锤度

本文论述了应用 ASD- LabSpec近红外光谱仪实现甘蔗糖汁锤度的快速测量     

蔗汁更正锤度与甘蔗蔗糖份的相关分析

本文收集了本所从１９８７－１９９５年的全部糖份化验分析数据，对甘蔗的蔗汁更正锤度匚甘蔗蔗糖份进行了不同品种、不同月份的相关分析，得出蔗汁更正锤度与甘蔗蔗糖份呈极显著正相关关系，并建立不同品种、不同时期甘蔗蔗糖份的数学模型：ｙ↑＾＝－２．２８８３＋０．８７３６ｘ（式中的ｙ↑＾为甘蔗蔗糖份，ｘ为改正后的蔗汁锤度）。绝对误差为０－０．２７。     

近红外光谱法快速测定甘蔗蔗汁蔗糖分

本试验应用傅立叶变换近红外光谱透射技术快速定量分析甘蔗蔗汁蔗糖分。利用内部交叉证实和自动优化功能和对预测数学模型进行优化,得到蔗汁蔗糖分预测数学模型的决定系数为99．41％，均方差为0．214；蔗汁蔗糖分近红外分析结果的准确度是可接受的。对近红外光谱技术在甘蔗科研和生产中的应用前景进行了讨论。     

甘蔗品系苗期螟害率与田间蔗汁锤度的关系

应用数理统计方法从多角度对168个甘蔗品系的蔗苗螟害枯心率(2.47%～46.13%)与甘蔗成熟期健康蔗茎田间蔗汁锤度(18.2%～25.6%)的关系进行分析研究。结果表明:甘蔗苗期螟害枯心率的高低与甘蔗品系田间蔗汁锤度的高低没有显著关系。     

混合汁中蔗糠含量对沉降和过滤影响的探讨

该文用实验和动态生产试验的方法探讨了蔗糠含量对沉降速度、过滤速度和干滤泥转光度的影响；指出，过量或过少蔗糠含量对这三者会产生急剧干扰；认为，亚法澄清的过滤尚需适量蔗糠作助滤剂等。     

近红外法甘蔗榨季分析

本文叙述了用近红外(NIR)光谱法分析甘蔗竹一种新系统。甘蔗是取自钻／压化验室预碎机、切得很细的全蔗样品。试验是在一间大型甘蔗糖厂(12000吨/日)中于1994／1995整个榨季内每天进行。与化验室数值相较，在准确度与精密度两方面，用近红外分析全蔗以确定可收回总糖分是成功的，而且在劳力、设备和人员的费用上都有相当大的节省。     

甘蔗蔗汁直接旋光度检测的简化方法

以桂糖93/102等57个甘蔗品系(种)样本为材料.时蔗汁直接旋光度采用常规方法和直接上机检测的简化方法进行两年实验,并对实验数据进行变异分析、差异分析及简单直线回归分析.结果表明:两种方法检测的蔗汁直接旋光度和相应的蔗汁蔗糖分,其间的平均值、标准差、变异系数、最小值及最大值极为相近,与常规方法比较,简化方法的蔗汁蔗糖分偏差远小于5%;简化方法的结果变异系数更小;但简化方法的蔗汁蔗糖分高于常规方法且其均数差异经检验达极显著水平;简化方法与常规方法检测的蔗汁直接旋光度的相关系数r=0.9998**,建立了对简化方法检测结果进行纠正的一元直线回归方程y=0.9992x-0.06094.     

蔗汁生产啤酒发酵特性的研究

论述了在50L啤酒生产设备上,分别用大米与蔗汁为辅料生产啤酒的对照试验,通过考察两种辅料所得麦汁的主要成分,对两种麦汁的外观发酵度、可发酵浸出物、酵母利用还原糖和α-氨基氮的情况、酒精生成量及双乙酰等发酵参数进行研究.结果表明,酵母在添加蔗汁辅料的麦汁中能正常发酵,所得啤酒的各项理化指标均符合国家标准GB4927.     

基于神经网络的甘蔗产量预测系统

首次将人工神经网络应用到甘蔗产量的预测中,介绍了ＢＰ网络模型及其算法,讨论了系统的开发方法,并对系统进行验证。对广西忻城糖厂蔗区８８／８９－９７／９８榨季的甘蔗单产和相应的气象条件运行结果表明模型具有较高的精度,复测误差在－５．３－１０．２％。     

Estimating regional productivity based on demographic structure with artificial neural network

This paper explores the impact of the age structure on regional productivity. An estimation model based on artificial neural network (ANN) was developed on the assumption that demographic change, due to aging and migration has a significant effect on the regional productivity, especially in rural regions. A multilayer perceptron ANN model was applied to consider the composition of demographic structure rather than ratio between two population groups such as aged-child ratio. Regional productivity was estimated by applying the estimation model developed in this research study to population and aggregate product data of sixteen South Korean cities and counties, from 2000 to 2011. Developed model is trained with data of sixteen cities and counties, from 2000 to 2009, and verified with observation data and estimation results of 2010 and 2011. The results revealed that gross regional domestic product per capita, which represents regional productivity, is significantly related to demographic structure and can be estimated by age structure.     

Relationships between nutrients and sucrose concentrations in sugarcane juice and use of juice analysis for nutrient diagnosis in Japan

Sugarcane is an important economic crop in southwest Japan, but its production is decreasing. To increase sugar production, both sugarcane yield and quality should be improved. Fertilizer management is one of the factors that influence sugarcane quality. We accordingly focused on nutrients present in sugarcane juice and attempted to identify the key factors affecting sugarcane quality. We collected sugarcane samples from 2013 to 2015 from all of the sugar mills in Japan and examined the relationships between juice nutrients and sucrose concentration. Juice analysis over 3 year showed that potassium (K⁺) and chloride (Cl^?) were the most abundant cation and anion in the juice and that both negatively correlated with the sucrose concentration. K⁺ and Cl^? concentrations significantly varied depending on production areas and those with higher K⁺ and Cl^? concentrations had a low sucrose concentration. This finding suggests that sugarcane in those areas may have been supplied with these two ions in excess. Electrical conductivity (EC) in the juice always positively correlated with K⁺ and Cl^? concentrations. EC may thus be a reliable indicator of K⁺ and Cl^? concentrations and could be used for nutrient diagnosis because of its ease of measurement. For improving sugarcane quality, we recommend that potassium chloride, which supplies both K⁺ and Cl^? and is a commonly used potassium fertilizer for sugarcane production in Japan, should be used in lower quantities in a year following one in which the EC of sugarcane juice at harvest is found to be high.     

Simulation for response of crop yield to soil moisture and salinity with artificial neural network

In saline fields, irrigation management often requires understanding crop responses to soil moisture and salt content. Developing models for evaluating the effects of soil moisture and salinity on crop yield is important to the application of irrigation practices in saline soil. Artificial neural network (ANN) and multi-linear regression (MLR) models respectively with 10 (ANN-10, MLR-10) and 6 (ANN-6, MLR-6) input variables, including soil moisture and salinity at crop different growth stages, were developed to simulate the response of sunflower yield to soil moisture and salinity. A connection weight method is used to understand crop sensitivity to soil moisture and salt stress of different growth stages. Compared with MLRs, both ANN models have higher precision with RMSEs of 1.1 and 1.6 t ha⁻¹, REs of 12.0% and 17.3%, and R² of 0.84 and 0.80, for ANN-10 and ANN-6, respectively. The sunflower sensitivity to soil salinity varied with the different soil salinity ranges. For low and medium saline soils, sunflower yield was more sensitive at crop squaring stage, but for high saline soil at seedling stage. High soil moisture content could compensate the yield decrease resulting from salt stress regardless of salt levels at the crop sowing stage. The response of sunflower yield to soil moisture at different stages in saline soils can be understood through the simulated results of ANN-6. Overall, the ANN models are useful for investigating and understanding the relationship between crop yield and soil moisture and salinity at different crop growth stages.     

Changes in juice quality and sugarcane yield with recurrent selection for sucrose

Sugarcane (Saccharum spp. hybrids) breeding programs in Louisiana have made improving sucrose content a top priority because a short growing season limits cane yield. Using a recurrent selection strategy, the cultivars with the highest sucrose content are crossed, and a new generation of cultivars is selected from the progeny. This study was designed to determine how selection primarily for sucrose content has modified physiological characters, and impacted sucrose content and yield. Five cultivars were randomly selected from each of seven generations of recurrent selection in Louisiana and planted in two experiments. The plant and first stubble crops were harvested late in the harvest season from each experiment. Cane yield and juice quality were determined. Cultivars from the last three generations were superior to cultivars from the first three generations for Brix % cane, sucrose % cane, purity, theoretically recoverable sugar (TRS), cane yield and sugar yield. Fiber % cane was not different among the generations. Selection primarily for sucrose has increased Brix % cane from 14% to 16%, sucrose % cane from 12% to 14%, purity from 82.5% to 87.3%, and TRS from 98 to 122 kg Mg⁻¹. A plateau in juice quality and sucrose yield in the last three generations may indicate that: (1) Louisiana's short growing season may restrict sucrose accumulation; (2) the genetic potential for late season juice quality has been reached with currently available germplasm; or (3) the inclusion of lower juice quality Saccharum spontaneum germplasm into the breeding program in order to increase disease tolerance, cold tolerance, and ratooning ability has diluted the effect of recurrent selection for sucrose.     

The use of k-means and artificial neural network to classify cotton lint

The use of High Volume Instrument (HVI) to measure cotton lint characteristics produces high dimensional data. A model which utilized Kohonen Self Organizing Maps (SOM) to visualize cotton lint HVI data, k-means clustering technique to cluster the data and Probabilistic Neural Network (PNN) for data classification was designed and tested using Kenyan cotton lint. According to the model the Kenyan cotton lint can be grouped into four clusters, which were successfully classified by using PNN with a correlation coefficient (R-value) of 1.     

Predicting properties of single jersey fabrics using regression and artificial neural network models

In our previous works, we had predicted cotton ring yarn properties from the fiber properties successfully by regression and ANN models. In this study both regression and artificial neural network has been applied for the prediction of the bursting strength and air permeability of single jersey knitted fabrics. Fiber properties measured by HVI instrument and yarn properties were selected as independent variables together with wales’ and courses’ number per square centimeter. Firstly conventional ring yarns were produced from six different types of cotton in four different yarn counts (Ne 20, Ne 25, Ne 30, and Ne 35) and three different twist multipliers (α _e 3.8, α _e 4.2, and α _e 4.6). All the yarns were knitted by laboratory circular knitting machine. Regression and ANN models were developed to predict the fabric properties. It was found that all models can be used to predict the single jersey fabric properties successfully. However, ANN models exhibit higher predictive power than the regression models.     

甘蔗家系配合力研究与应用

采用100个甘蔗亲本配制了144个杂交组合,分析其有性世代主要农艺性状、田间锤度、蔗糖分配合力,次年参照常规选育方法选择宿根杂种优良株系。结果表明：早熟组合前7名均为以桂糖与CP为亲本的组合,以ROC22作为父本易得到植株较高和较高糖分的F1;在小区含糖量组合配合力前30名的组合中,小区含糖量组合配合力与小区含糖量的特殊配合力和父本的一般配合力呈正相关,相关系数分别为0．67和0．44。     

Assessment of gastric cancer survival: using an artificial hierarchical neural network

This study is designed to assess the application of neural networks in comparison to the Kaplan-Meier and Cox proportional hazards model in the survival analysis. Three hundred thirty gastric cancer patients admitted to and surgically treated were assessed and their post-surgical survival was determined. The observed baseline survival was determined with the three methods of Kaplan-Meier product limit estimator, Cox and the neural network and results were compared. Then the binary independent variables were entered into the model. Data were randomly divided into two groups of 165 each to test the models and assess the reproducibility. The Chi-square test and the multiple logistic model were used to ensure the groups were similar and the data was divided randomly. To compare subgroups, we used the log-rank test. In the next step, the probability of survival in different periods was computed based on the training group data using the Cox proportional hazards and a neural network and estimating Cox coefficient values and neural network weights (with 3 nodes in hidden layer). Results were used for predictions in the test group data and these predictions were compared using the Kaplan-Meier product limit estimator as the gold standard. Friedman and Kruskal-Wallis tests were used for comparisons as well. All statistical analyses were performed using SPSS version 11.5, Matlab version 7.2, Statistica version 6.0 and S_PLUS 2000. The significance level was considered 5% (alpha = 0.05). The three methods used showed no significance difference in base survival probabilities. Overall, there was no significant difference among the survival probabilities or the trend of changes in survival probabilities calculated with the three methods, but the 4 year (48th month) and 4.5 year (54th month) survival rates were significantly different with Cox compared to standard and estimated probabilities in the neural network (p < 0.05). Kaplan-Meier and Cox showed almost similar results for the baseline survival probabilities, but results with the neural network were different: higher probabilities up to the 4th year, then comparable with the other two methods. Estimates from Cox proportional hazards and the neural network with three nodes in hidden layer were compared with the estimate from the Kaplan-Meier estimator as the gold standard. Neither comparison showed statistically significant differences. The standard error ratio of the two estimate groups by Cox and the neural network to Kaplan-Meier were no significant differences, it indicated that the neural network was more accurate. Although we do not suggest neural network methods to estimate the baseline survival probability, it seems these models is more accurately estimated as compared with the Cox proportional hazards, especially with today's advanced computer sciences that allow complex calculations. These methods are preferable because they lack the limitations of conventional models and obviate the need for unnecessary assumptions including those related to the proportionality of hazards and linearity.