基于BP神经网络的人机联合率定SWMM研究

为解决雨洪管理模型(SWMM)在率定过程中参数复杂、过程繁琐等问题。以西宁市某地块为例建立SWMM模型,利用Morris筛选法进行参数灵敏度分析,并根据灵敏度分析的结果进行人工率定;另外利用BP神经网络对模型进行率定,并结合参数灵敏度对其进行优化。对3种率定方案进行分析,结果表明:水文水力模块参数的相对灵敏度基本一致,其中灵敏度较大的参数为子汇水区面积(Area)、不透水率(Imperv)和不透水区洼地蓄积量(Destore-Imperv),并且不同降雨条件下模型参数的灵敏度存在差异。经过优化后的BP神经网络参数率定方法的模型模拟效果最好,纳什系数最大。结合灵敏度优化BP神经网络的人机联合率定方法一方面能提高BP神经网络率定的准确性,另一方面又能提高传统人工率定的效率。     

Prediction of baking results from farinograph measurements by using stepwise linear regression and artificial neuronal networks

In an effort to extract additional data from farinograph experiments a model was developed to simulate the measurements and correlate the parameters of the model with results from baking tests. This additional information can be used in bakeries to predict the baking properties of the flours and adjust the recipes to maintain a constant product quality. For this eight different flours were characterized with a farinograph and 13 different results from baking experiments. An approach with five nonlinear differential equations was able to model the farinograph measurements very well (average R² = 0.995 ± 0.005). While a stepwise multilinear regression only showed weak correlations in cross validation between a single parameter of the model and the baking volume (R² = 0.745) and the volume yield (R² = 0.796) respectively, the artificial neuronal network was more successful. For the baking weight (R² = 0.926), the dough yield gross (R² = 0.909) and net (R² = 0.913) strong correlations were found. A good correlation for the baking volume (R² = 0.853) was also determined, while the volume yield showed comparable results to the linear regression (R² = 0.792).     

Construction and evaluation of introgression lines and fine mapping of ehd8 from Jinghong common wild rice (Oryza rufipogon)

Heading date is one of the most important traits in rice and regulated by multiple genes. Common wild rice is the ancestor of Asian cultivated rice and harbours abundant genetic diversity. To use wild rice resource in rice breeding, a set of 154 introgression lines (ILs) covering 93% of the genome of Jinghong common wild rice was constructed in the background 'Teqing', using 208 simple sequence repeat markers evenly distributed on 12 chromosomes. Among the ILs, the line JIL64 displayed late heading independent of photoperiod. Genetic analysis using the two F₂ populations crossed ''Teqing'/JIL64 and JIL64/'Teqing' revealed that late flowering was controlled by a recessive gene on chromosome 8 (designated early heading date 8, ehd8), and ehd8 was fine mapped to the 50‐kb region flanked by markers RM22221 and 64Indel4. Sequencing and qRT‐PCR demonstrated that LOC_Os08g01410 and LOC_Os08g01420 were deleted in JIL64 and may be associated with the late heading of Jinghong common wild rice. These findings lay a practical foundation for characterizing ehd8, and the ILs help to mine genes from Jinghong common wild rice.     

人工神经网络在电力系统短期负荷预测中的应用

根据电力系统短期负荷变化的特性，提出BP模型在实际负荷预测应用中的方法和步骤．对BP网络结构、样本空间、收敛性等作了有针对性的研究．结果表明：多层神经网络应用于电力系统短期负荷预测是可行和有效的．其预报结果比传统的负荷预测方法更准确、经济、效果更好．     

常见混凝土质量通病及防治措施

介绍了几种常见混凝土质量通病产生的原因,并提出了具体的防治措施。     

四种杀菌剂对小麦根病病原菌的室内药效试验

室内药效试验表明：麦根宁1号、敌力脱对小麦根腐病菌的生长均有良好的抑制作用。麦根宁Ⅰ号稀释300倍时，平均抑菌圈直径最大；敌力脱稀释800倍时，平均抑菌圈直径最大。同时，麦根宁Ⅰ号、敌力脱Ⅱ号，粉锈宁对小麦全蚀病菌生长也有明显的抑制作用。粉锈宁，麦根宁Ⅰ，Ⅱ号在稀释10000倍时，抑菌率均达100％；敌力脱在稀释15000倍时，平均抑菌率仍达89．7％。     

电流模式神经网络

把电流模式电路引进了神经网络理论，提出了一种用电流模式电路构成的神经网络阈值处理单元，并由此构成了电流模式Ｈｏｐｆｉｅｌｄ神经网络，此网络中信号都是以电流的形式出现的。     

Imperfect Competition and Common Resource Exploitation—An Analysis of Renewable Resource Management with Non—use Values

This article analyzes the importance of non-use values in the management of natural resources with a model of imperfect competition.By constucting a framework of duopolistic exploiters.the incorporation of non-use values based on self-interest and altruistic motives appears to be significant for determining the degree of inefficiency caused by the problem of common property at the non-cooperative Nash equilibrium.It is found that when the non-use value placed by the altruist exploiter is bounded by that placed by the pure selfinterest exploiter,the effect of market power is dominated by the effect of common exploitation.In this case,the exploiters‘ harvesting strategy will response in the same direction to the change of each other‘s harvesting at the non-cooperative Nash equilibrium.In contrast,when the non-use value placed by the altruist is substantially larger or smaller than that placed by the self-interest exploiter,one exploiter‘s increase in harvesting will lead to a decrease in harvesting of the other exploiter at the non-cooperative Nash equilibrium.     

Identification of blackeye cowpea mosaic virus from germplasm of yard-long bean and from soybean,and the relationships between blackeye cowpea mosaic virus and cowpea aphid-borne mosaic virus

Two potyvirus isolates, one from germplasm of yard-long bean (Vigna unguiculata ssp.sesquipedalis) introduced into the Netherlands, and another one from soybean plants (Glycine max) in Indonesia, were compared with two virus isolates of blackeye cowpea mosaic virus (BICMV) from the USA and a Moroccan isolate of cowpea aphid-borne mosaic virus (CAMV). It is proposed that all five isolates be now considered BICMV on the basis of host ranges, symptoms and serology. From our results, and a reassessment of the literature it is suggested to drop the name CAMV in favour of BICMV.Samenvatting Twee potyvirussen, de een in Nederland ingevoerd met genenmateriaal vanVigna unguiculata ssp.sesquipedalis en de ander uit planten van sojaboon (Glycine max) in Indonesië, werden vergeleken met twee isolaten van blackeye cowpea mosiac virus (BICMV) en een Marokkaans isolaat van cowpea aphid-borne mosaic virus (CAMV). Op grond van waardplantenreeksen, symptomen en serologie stellen de auteurs voor om alle vijf isolaten te beschouwen als BICMV. Gebaseerd op de verkregen resultaten en een kritische beschouwing van de literatuur wordt de aanbeveling gedaan om de naam CAMV te laten vallen ten gunste van BICMV.     

Identification of Resistance to Common Bacterial Blight Disease on Bean Genotypes Grown in Turkey

Common bacterial blight (CBB) in edible beans (Phaseolus vulgaris), incited Xanthomonas campestris pv. phaseoli, reduces bean yields and seed quality. The main objective of this study was to determine resistance to common bacterial blight in bean genotypes. Twenty-two bean genotypes grown in Turkey including common and snap bean cultivars/lines were collected from different parts of Turkey and tested for resistance against to Xanthomonas campestris pv. phaseoli strain MFD-11. All the common and snap bean lines/cultivars tested were moderately susceptible, susceptible or highly susceptible, except AG-7117 which was found resistant to Xanthomonas campestris pv. phaseoli. This is the first report of a resistance source in a common bean line (AG-7117) against Xanthomonas campestris pv. phaseoli.