基于产品族的核心模块演进分析与评价方法

针对产品族设计中核心模块演进问题,研究了产品族核心模块演进规律与演进过程评价,提出了一种模块演进规律分析与评价方法。该方法研究了客户需求与产品核心模块设计之间的关系,并在现有产品核心模块基础上,进一步对模块部件的标准化、继承性等指标进行了分析;在需求分析过程中,建立了一种基于贝叶斯网络的核心模块演进分析模型,利用历史数据对该模型进行了训练,将客户需求作为证据对该模型进行推理,进而得出核心模块在设计与工艺方面的改变规律;在演进规律分析基础上,对现有核心模块部件设计参数、标准部件重用度进行了分析,获得了核心模块演进过程综合评价值。最后,用小型轮式装载机后车架模块的演进过程验证了该方法的可行性和正确性。     

Design of nondestructive test system for plasma clots

One of the most important measures that are used to guarantee blood transfusion safety is to detect clots in the plasma before transfusion. To overcome the disadvantages of manual detection method, this research designs a nondestructive testing (NDT) system for plasma clots inspection based on machine vision technique and artificial neural networks. The key technology for system design are studied and presented. Image acquisition is performed by custom-designed software based on MATLAB platform, and the methods of image cut, reverse color, median filter as well as gray cutting are adopted to preprocess image. The use of fisher discrimination method, combined with iterative threshold segmentation method and the selection of connected domain, can successfully eliminate the interference of air bubble and correctly extract the image of plasma clots. Plasma clots are discriminated by a recognition model based on artificial neural network BP algorithms. The results of clinical contrast experiment shows that the system can effectively detect whether plasma contains plasma clots and the new system shows a much higher degree of repeatability and stability. From the image acquisition and processing to the recognition of plasma clots, the detecting time of a sample is no more than 1 min.     

基于复杂网络从地区间经济贸易角度对我国贫富差距成因的分析

近年来,复杂性科学特别是复杂网络成为研究热点,它提供了一种全新的视角,有助于我们从整体上把握社会贫富问题的复杂性。本文基于复杂网络理论,通过建立动态的地区间经济贸易网络,从地区间经济贸易角度对我国贫富差距成因进行了分析,并最终给出了基于复杂网络对贫富差距成因的分析法较以往分析方法的优势及本文的相关结论。     

Spatially targeting the distribution of agricultural input stockists in Malawi

Developing the rural agricultural input markets in sub-Saharan Africa can improve the current low productivity of smallholder farmers. Malawi has seen significant efforts in addressing the availability of agricultural inputs at village level in the last few years; for example, the improvement of rural agro-dealer networks. Nevertheless inputs are still difficult to obtain for many remote smallholder farmers. Spatial analysis can help in the expansion of input stockists, especially agro-dealer networks, by assessing the coverage of existing input outlets and deriving optimum locations for village-level input stockists.We address three research questions. First, what is the locational efficiency of the current village-level stockists of inputs (Citizens Network for Foreign Affairs – Rural Agricultural Market Development Trust trained network of agro dealers and public sector)? Secondly, how many village-level stockists of markets are needed to reach 60% of the population in the central region of Malawi within one hour? Finally we address the potential spatial components of the sustainability of input stockists relating to the potential demand from smallholder farmers and the access to bulk supplies. The problem of finding the optimum location for village-level stockists of markets is addressed in two stages, using spatial analysis in conjunction with location–allocation models. First, the locational efficiency of the existing network of stockists of inputs is determined, followed by the establishment of a set of optimal sites for village-level stockists of inputs. A final step explores the viability of stockists and calculates the population surrounding the stockists taking into account competition from other sources of inputs and the accessibility of the selected stockists to potential wholesalers who are bulk distributors of farm inputs.Our results show that locational efficiency can be assessed in terms of the differential access of households to resources and transport. Often, these differences are not considered in covering problems and can have a large effect on the physical access to inputs. The results can be used to define which areas are inherently difficult to serve with agricultural inputs and could inform efforts to provide incentives to remote areas. Further implications for input policies in Malawi are that improvements in road infrastructure might not directly benefit the poorest farmers (if they are walking) but could serve to reduce the wholesale prices and therefore the retail price. In addition, the improvement in roads might increase the number of potential customers of any particular stockist, with economies of scale allowing the reduction of prices while ensuring a satisfactory profit margin for the stockist. The results of our models imply that Citizens Network for Foreign Affairs may need to train stockists over a wider area to increase the access to inputs of those smallholder farmers with least resources.     

基于人工神经网络的金属切削刀具磨损的识别与控制

根据金属切削刀具磨损与切削力的关系，建立了基于人工神经网络的切削刀具磨损的识别与控制系统，初步实现了用人工神经网络对金属切削刀具磨损的动态识别与控制。     

Plant‐frugivore interactions in an intact tropical forest in north‐east Thailand

Fleshy‐fruited plants in tropical forests largely rely on vertebrate frugivores to disperse their seeds. Although this plant–animal interaction is typically considered a diffuse mutualism, it is fundamental as it provides the template on which tropical forest communities are structured. We applied a mutualistic network approach to investigate the relationship between small‐fruited fleshy plant species and the fruit‐eating bird community in an intact evergreen forest in northeast Thailand. A minimum of 53 bird species consumed fruits of 136 plant species. Plant‐avian frugivore networks were highly asymmetrical, with observed networks filling 30% of all potential links. Whereas some of the missing links in the present study might be due to undersampling, forbidden links can be attributed to size constraints, accessibility and phenological uncoupling, and although the majority of missing links were unknown (58.2%), many were probably due to a given bird species being either rare or only a very occasional fruit eater. The most common frugivores were bulbuls, barbets and fairy‐bluebirds, which were responsible for the majority of fruit removal from small fleshy fruited species in our system. Migratory birds seemed to be a minor component of the plant‐frugivore networks, accounting for only 3% of feeding visits to fruiting trees; they filled 2% of the overall potential networks. The majority of interactions were generalized unspecific; however, Saurauia roxburghii Wall. appeared to be dependent on flowerpeckers for dispersal, while Thick‐billed Pigeons were only seen to eat figs.     

基于全卷积神经网络的猪背膘厚快速准确测定

为了有效地弥补猪B超图像人工手绘背膘厚的不足,为生猪育种工作提供更精准和稳定的背膘厚测定新方法。该研究将全卷积神经网络（Fully Convolutional Networks,FCN）模型应用于猪B超图像的背膘分割和背膘厚测定上,开发出一套使用Python调用FCN模型对猪B超图像背膘厚进行自动测定的系统。通过开展验证集验证试验、屠宰比对试验和人员比对试验,发现模型测定结果和标注结果之间差异不显著（P>0.05）,两者相关系数达0.92（P<0.01）;B超标准测定背膘厚和FCN分割测定背膘厚的相关系数达到0.97（P<0.01）;专家组组内标准差为0.17 mm（最小）,行业外组组内标准差为1.67 mm（最大）,而FCN分割结果稳定性强,不受人员因素的影响。因此,该方法可以实现对外种猪B超背膘厚的精准、快速、稳定测量,减少猪场对专业人员的依赖,降低测定人员培训成本,减少工作人员工作量。     

变时滞随机Cohen-Grossberg神经网络的指数稳定性的新判据

通过建立よ算子不等式,利用随机分析技巧得到了变时滞随机Cohen Grossberg神经网络的p-阶矩指数稳定的一个充分条件,并给出实例进行了说明.     

Neural network prediction of performance parameters of an inclined plate seed metering device and its reverse mapping for the determination of optimum design and operational parameters

Seed planting equipment with inclined plate seed metering devices is the most commonly used equipment for planting of peanut crop in India. For obtaining the high yield, it is very essential to drop the peanut seeds in rows maintaining accurate seed rate and seed spacing with minimum damage to seeds during metering. This mainly depends on forward speed of the planting equipment, rotary speed of the metering plate and area of cells on the plate. The relationship between these factors and the performance parameters viz., seed rate, seed spacing and percent seed damage can be established using regression analysis. But they may not be very accurate and may pose difficulty in the determination of inputs for a set of desired outputs (reverse mapping). Hence, an attempt has been made in this paper to develop the feed forward artificial neural network (ANN) models for the prediction of the performance parameters of an inclined plate seed metering device. The data were generated in the laboratory by conducting experiments on a sticky belt test stand provided with a seed metering device and an opto-electronic seed counter. The generated data was used to develop both statistical and neural network models. The performance of the developed models was compared among themselves for 4 randomly generated test cases. The results show that the ANN model predicted the performance parameters of the seed metering device better than the statistical models. In order to determine the optimum forward speed of the planting equipment, peripheral speed of the metering plate and the area of cells on the plate to obtain the recommended seed rate of 33.33 seeds/m², seed spacing of 100 mm and percent seed damage of 0.2% with 100% fill of the cells, a novel technique of reverse mapping using ANN model was followed. It was observed that the optimum forward speed of the planting equipment and optimum area of cells on the metering plate had good correlation with size of seed. Linear regression equations were developed to predict the optimum forward speed of the planting equipment and optimum area of cells on the metering plate using the size of seeds as independent parameter. The peripheral speed of the metering plate of 0.237 m/s was found to be optimum for the size of seeds in the range of 95.42-123.01 mm². However, the results need to be verified by conducting planting operation under actual field conditions.     

基于BP神经网络的数字识别

介绍了字符识别的几种方法及神经网络的基本原理,并将BP神经网络应用于数字识别,选取最佳的隐含层节点数及训练样本个数,实现了基于BP神经网络的数字识别。仿真试验结果表明,BP神经网络可以对阿拉伯数字进行快速、准确的识别,具有广泛的应用前景。