Increasing productivity and protein content using early-maturing rices and efficient nitrogen management

Germplasm with shorter duration than that of the currently grown varieties is being generated to maximize productivity of irrigated rice. However, short-duration varieties often produce yields lower than the medium- and long-duration varieties. Experiments were conducted during the 1980–82 dry and wet seasons to increase productivity through the use of very early-maturing rices and the improved management of nitrogen (N) fertilizers.Results over three years showed that IR58 and IR9729-67-3 (growth duration 100 ± 5 days) yield as well as or higher than IR36 although earlier maturing. They generally had a higher productivity (kg ha^–1 day^–1) than IR36 (110 ± 5 days).Three years' data suggest that the improved timing of broadcast applications of urea in split doses increased grain yield comparable with the basal incorporation of slow-release sulfur-coated urea (SCU) or deep point-placement of urea supergranules (USG).Results on elite breeding lines showed that the early-maturing IR9729-67-3 produced higher protein yield ha^–1 than longer duration varieties such as IR8 and IR42 in the dry season. Furthermore, contrary to earlier results, single basal incorporation of slow-release SCU increased the protein yield of rice by 53 kg ha^–1 and deep point-placement of USG by 43 kg ha^–1 over split application of prilled urea.     

THE ESTIMATION AND PREDICTION OF SUBSET BILINEAR TIME SERIES MODELS WITH APPLICATIONS

Abstract. In this paper we define subset bilinear time series models, and then describe an algorithm for the estimation of these models. It is also pointed out that for this class of non-linear time series models, it is possible to obtain optimal several step predictors. The estimation technique of these models is illustrated with respect to three time series, and the optimal several steps ahead forecasts of these time series models are calculated. A comparison of these forecasts is made with the forecasts obtained by the best linear autoregressive and threshold autoregressive models. The residuals obtained from the models are tested for independence and Gaussianity using higher order moments.     

Polymer/surfactant interaction—Its relevance to detergent systems

The field of polymer/surfactant interaction is reviewed in this work. Results from two investigative methods,viz., dialysis and surface tension, are discussed, illustrating the main behavioral patterns and outlining the principles of the interactions. Next, aspects of the interaction phenomena that appear to have relevance to detergent formulation are presented. These include solution rheology, solubility control and surface conditioning. Lastly, the importance of surface activity of the polymer itself is stressed, culminating in a discussion of the properties of hydrophobically modified water-soluble polymers (“polymeric surfactants”), both alone and in the presence of conventional surfactants. Based on the Samuel Rosen Memorial Award lecture, given at the AOCS Annual Meeting, Anaheim, CA, April 1993.     

贝叶斯网络分类器近似学习算法

贝叶斯网络在很多领域应用广泛,作为分类器更是一种有效的常用分类方法,它有着很高复杂度,这使得贝叶斯网络分类器在应用中受到诸多限制。通过对贝叶斯网络分类器算法的近似处理,可以有效减少计算量,并且得到令人满意的分类准确率。通过分析一种将判别式算法变为产生式算法的近似方法,介绍了这种算法的近似过程,并将其应用在了贝叶斯网分类算法中。接着对该算法进行分析,利用该算法的稳定性特点,提出Bagging-aCLL 集成分类算法,它进一步提高了该近似算法的分类精度。最后通过实验确定了该算法在分类准确率上确有不错的表现。     

Piezo- and pyroelectricity in polymer blends of poly(vinylidene fluoride)/poly(methyl methacrylate)

Compatible polymer blends of poly(vinylidene fluoride) (PVDF) and poly(methyl methacrylate) (PMMA) can be used as suitable model systems for investigating the relationship between the physico-chemical structure of polymers and their piezo- and pyroelectric activity. The structure of PVDF/PMMA blends can be varied over a very wide range which can lead to a strong influence on the piezo- and pyroelectric activity and the corresponding coefficients d₃₁ and g₃. The values of d₃₁ and g₃ were found to vary over nearly five decades whereas the normalized coefficients $\text{d}{}_{31}\text{P}$ and $\text{g}{}_3\text{P}$ remain largely unaffected. This emphasizes the importance of the molecular processes causing the macroscopic polarization P during the poling procedure. For a given polarization P and a given temperature T the properties of the polymer matrix, however, are far less important for the values obtained for d₃₁ and g₃. The experimental results were compared with theoretical predictions based on models which were recently developed by Tashiro et al., Broadhurst et al. and by Mopsik et al.. Considering the appropriate scope of each model a good agreement between theory and experiment is observed and general contradictions have not been found.     

深基坑工程安全评价的尖点突变模型

为评价深基坑工程失稳的突变特征,采用突变理论建立了深基坑工程尖点突变安全评价模型.根据工程安全监控理论,引入突变理论,以基坑水平变形为例,考虑深基坑土体的流变特性,建立了基于时效影响因子的深基坑工程尖点突变安全评价模型.实例分析表明,尖点突变安全评价模型克服了采用常规设置监控值评价时参数选择困难,并能反映基坑失稳突变性的特点,可以量化评价基坑安全程度,正确评价安全状态,具有适应性强、精度高的优点.     

Zonal disintegration phenomenon in enclosing rock mass surrounding deep tunnels -- Elasto-plastic analysis of stress field of enclosing rock mass

The zonal disintegration phenomenon (ZDP) is a typical phenomenon in deep block rock masses. In order to investigate the mechanism of ZDP, an improved non-linear Hock-Brown strength criterion and a bi-linear constitutive model of rock mass were used to analyze the elasto-plastic stress field of the enclosing rock mass around a deep round tunnel. The radius of the plastic region and stress of the enclosing rock mass were obtained by introducing dimensionless parameters of radial distance. The results show that tunneling in deep rock mass causes a maximum stress zone to appear in the vicinity of the boundary of the elastic and the plas-tic zone in the surrounding rock mass. Under the compression of a large tangential force and a small radial force, the rock mass in the maximum stress zone was in an approximate uniaxial loading state, which could lead to a split failure in the rock mass.     

朴素贝叶斯与全贝叶斯在DFI的应用比较研究

P2P应用软件检测技术主要是深度包检测法和深度流量检测法.深度包检测法不能检测出加密的和未知的对等网络流应用,深度流量检测方法可以克服深度包检测法这个缺点.将模糊识别领域里比较成熟的贝叶斯分类技术应用到对等网络深度流检测中.结合实际项目,对贝叶斯的两个分类器-朴素贝叶斯和全贝叶斯的算法、训练结果、运行结果进行研究,实验研究表明朴素贝叶斯和全贝叶斯分类器能够快速准确地找到P2P流应用,朴素贝叶斯分类器准确度占据优势,全贝叶斯运行时间占据优势.     

EVALUATION OF SURFACE FORCE-PORE FLOW AND MODIFIED SURFACE FORCE-PORE FLOW MODELS FOR REVERSE OSMOSIS TRANSPORT

The surface force-pore flow (SF-PF) model of reverse osmosis transport and the extended and modified form (the MD-SF-PF model) have been employed to predict the performance of four aromatic polyamide (FilmTec, FT30)reverse osmosis membranes. The evaluation is based on a comparison of model predictions with experimental data. Dilute sodium chloride-water solution experimental data were used to estimate model parameters. The models are then used to predict flux and separation at various operating pressures and concentrations. Membrane performance (i.e., separation and permeate flux) can be well predicted by the MD-SF-PF model while the SF-PF model predicts the performance for the sodium chloride system less satisfactorily.