适用于智能传感器系统的SVM集成研究

以支持向量机(SVM)为代表的人工智能技术在智能传感器系统中得到了广泛的应用,但传统的SVM有"灾难性遗忘"现象,即会遗忘以前学过的知识,并且不能增量学习新的数据,这已无法满足智能传感器系统实时性的要求。而Learn++算法能够增量地学习新来的数据,即使新来数据属于新的类,也不会遗忘已经学习到的旧知识。为了解决上述问题,提出了一种基于壳向量算法的Learn++集成方法。实验结果表明:该算法不但具有增量学习的能力,而且在保证分类精度的同时,提高了训练速度,减小了存储规模,可以满足当下智能传感器系统在线学习的需求。     

高精密恒温系统在NDIR气体分析仪中的应用

当前基于非分光红外（NDIR）技术的气体分析仪设计过程中存在很多干扰源,这些干扰源对检测精度影响的轻重程度各有不同,其中,红外池中,温度的稳定性对气体的吸收影响大,此外,对于特定的吸收气体,恒温点同样重要。设计了一种基于增量式PID算法的高精度温控系统,选择对气体吸收影响最小的恒温点,并控温范围为±0.1℃,相比现有同类温控系统,精度提高近10倍。通过试验比较不控温情况下的结果和应用在本恒温系统中的测试结果表明：高精度恒温系统下,NDIR气体分析仪的信噪比得到了明显改善。该成果将直接应用到国家＂十二·五＂计划项目中。     

Ammonia volatilization from urea as affected by the addition of iron pyrites and method of application

Laboratory incubations were conducted to determine the ammonia (NH₃) loss from urea as affected by the addition of coarse and ground (fine) pyrites at 1:1, 1:2 and 1:5 urea: pyrite (w/w) ratios and methods of application (surfaceapplication, incorporation and placement). Coarse pyrites (>-2mm) were not effective in reducing NH₃ loss from urea when surface applied even at the highest ratio of pyrite (15.9% vs 18.7% without pyrite). Ground pyrites (0.1–0.25 mm), in 1:1 ratio, had about 5% less NH₃ loss than the urea alone treatment. Higher ratios of pyrites reduced NH₃ loss much more. Ammonia losses were the most with surface-applied urea (18.9%) and the least (13.5%) when placed (2.5 cm) below the soil surface. Addition of ground pyrite to surface-applied urea (1:1 ratio) decreased the loss to 13.2%. Urea+pyrite placed below the soil surface had the least loss (9.8%). Results indicate that combined application of urea and fine pyrite could reduce NH₃ loss.     

基于资源分配网络的小数据集并行集成学习方法

机器学习领域中,如何在小规模的训练数据集上获得一个具有稳定的高计算精度的算法模型,一直以来都是一个棘手而富有挑战的问题。从算法模型出发,提出了一种基于扩展卡尔曼滤波器的资源分配网络并行集成学习方法。该集成系统由多个带有扩展卡尔曼滤波器的资源分配网络（RANEKF）组成,并且每个RANEKF子网的输入由原始数据集中的输入经过随机权值的修正得到。通过和其他神经网络构成的集成学习算法的实验对比,发现提出的方法在小训练集上拥有更高的计算精度和稳定性。     

Effect of granule size and the placement geometry on the efficiency of urea supergranules for wetland rice grown on a permeable soil

In a laboratory experiment 5 cm depth of water was allowed to percolate daily down through a 15 cm thick soil (Typic Ustipsamment) layer. It was observed that leaching losses of urea supergranules (USG)-N could be decreased by about 20% by the placement of four 0.25 g granules at four points instead of one 1 g granule at one point. In field microplots, the placement of approximately 30 granules of 0.30 g size instead of 9 granules of 1.00 g size resulted in reduced leaching of USG-N and, in turn, increased rice yield. In a follow-up field study, the advantage of more frequently placed USG was confirmed. As compared with 1 g USG placed in the usual manner in the center of four rice hills, increasing the density of placement in soil produced 15% more rice grain. Further increase in rice yield could be obtained by increasing the number of USG placed in the soil and decreasing the size of the granule from 1.00 g to 0.70 or 0.35 g. With USG of 0.35 and 0.70 g yields were equal or sometimes even slightly higher than with split application of prilled urea on a heavily percolating, low-CEC, light-textured soil.     

Tool path generation framework for accurate manufacture of complex 3D sheet metal parts using single point incremental forming

The prototyping of complex sheet metal parts using single point incremental forming (SPIF) requires the generation of optimal tool paths and/or tool path sequences that ensure that the formed part is within geometric design specifications. The presence of a multitude of features on complex parts leads to multiple inaccuracy inducing phenomena occurring simultaneously due to interactions between the features. This paper proposes a network analysis methodology using topological conceptual graphs to capture the effects of different phenomena on the final accuracy of a sheet metal part manufactured by SPIF. Using this framework optimized tool paths can be generated that compensate for the inaccuracy inducing behavior. Tool path generation algorithms to create partial tool paths that account for the accuracy of specific features in the part based on the proposed framework are also presented. Finally, the creation of integrated tool paths maintaining complementarity between tool paths and desired continuity behavior using non-uniform cubic B-splines is illustrated. A number of case studies demonstrating the applicability of the integrated framework are discussed, where the maximum deviations in the part are significantly reduced and the average absolute deviations for the complete part are brought down to less than 0.5 mm.     

Nitrogen transformations during hydrolysis and nitrification of urea

Nitrogen transformations occurring in ten soils fertilized with urea were determined during incubation in the laboratory for four weeks. Urea was applied at one rate, but with different placement methods. Urea was applied in solution, as prills with a 1 cm grid spacing and as prills with no spacing. Unfertilized soils and soils amended with KNO₃ solution were included as controls.Nitrite accumulated in the majority of soils treated with urea, and the maximum nitrite concentration measured was directly related to initial soil pH. Cumulative gaseous N losses as percentages of applied N were: NH₃, 0-59.6; N₂, 0-4.9; N₂O, 0-9.9; KMnO₄-N, 0-1.3; CH₃ONO, 0-<0.1. No gaseous N evolution was detected in control treatments. Gaseous N losses were correlated with soil pH (NH₃) maximum NO ₂ ^- concentration (N₂, N₂O, KMnO₄-N) and organic C content (N₂, N₂O). Fertilizer placement effects were generally not significant and were small in comparison with differences between soils.

---

Resumo Foram determinadas, durante incubação em laboratório por quatro semanas, as transformações do nitrogênio ocorridas em dez solos fertilizados com uréia. A uréia foi aplicada a um só nivel, mas com diferentes métodos de aplicação: em solução e em grânulos com 1 cm de espaçamento de grade e em grânulos sem espaçamento. Solos não fertilizados e solos corrigidos com solução de KNO₃ foram incluidos como controles.O nitrito acumulado na maioria dos solos tratados com uréia e a concentração máxima de nitrito medido foram diretamente relacionados ao pH inicial do solo. As perdas cumulativas de N gasoso tomadas em percentagens de N aplicado foram: NH₃, 0-59.6; N₂, 0-4.9; N₂O; 0-9.9; KMnO₄-N, 0-1.3; CH₃ONO, 0-<0.1. Não foi detectada liberação de N gasoso nos tratamentos de controle. As perdas de N gasoso foram relacionadas com o pH do solo (NH₃), concentração máxima de NO ₂ ^- (N₂, N₂O, KMnO₄-N) e teor de C orgânico (N₂, N₂O). Efeitos da aplicação de fertilizante não foram de um modo geral significativos e foram pequenos em comparação com as diferenças entre os solos.

     

基于VPR的FPGA布局算法时延改进

基于模拟退火的现场可编程门阵列(FPGA)布局算法在计算关键度时存在一定的偏差。为此,提出一种FPGA布局时延改进算法。利用不同的模拟退火温度和交换接收率,以及前后2次布局的时延代价差,对FPGA布局的时延代价进行补偿。通过增加时延补偿模块来调整布局的代价函数,达到重新寻找布局过程中被遗弃的较优解的目的。实验结果表明,在MCNC基准电路上使用改进算法,布局的时延代价和线网代价分别比改进前的算法减少19.2%和0.5%。此外,电路的关键路径时延也得到了不同程度的改善,使得布局质量在各个方面都明显优于优化前的通用布局布线算法。     

基于动态有噪网络的增量社团发现算法

在动态社会网络中,诸如垃圾邮件之类的噪声会影响网络的稳定性,导致其社团结构难以被准确发现。针对该问题,提出一种采用增量结构的社团发现算法。利用相对熵处理噪声,通过改进的增量算法发现社团结构。实验结果表明,该算法针对不同动态社会网络的发现性能均优于传统动态社团发现算法,其模块度可达到0.8左右,互信息值变化也较平稳,可有效避免噪声对算法性能的影响。     

一种基于Hadoop的动态树增量更新方法

为适应真实环境中数据量大、流程复杂、计算密集的数据挖掘需求,提高传统树增量更新挖掘效率,改变已有算法的串行执行方式,提出一种基于Hadoop的动态树增量更新方法。介绍云计算、模型与执行流程等基本概念,针对现有Hadoop平台中任务调度的随机分配策略,设计一种动态云平台中的资源调度与分配算法,以期达到成本消耗的最小化,给出树增量更新挖掘算法以及2个并行算法（DeleteFreqTree和FindNewTree）,完成树数据的增量挖掘工作。实验结果表明,该并行算法有效可行,具有高效性与良好的扩展率,能够对海量树数据进行更新挖掘。