基于能量优化的LEACH路由协议改进

针对LEACH路由算法中簇头选举随机性和簇头与基站直接通信导致能量消耗过快且不平衡的特点,提出新的改进算法,以达到降低能耗目的。在改进算法中,簇头剩余能量高于网络平均能量。根据簇头节点与基站的相对位置划分不同区域,簇头节点发送数据采用多跳方式,避免簇头节点能量消耗过快,达到平衡网络能量消耗的目的。仿真表明,通过改进簇头选举条件和采用多跳路由的方式,即使在数据通信量增加的情况下,依然能够延长网络通信时间。     

基于马尔可夫链的自适应性神经网络训练算法

提出一种基于马尔可夫链的自适应性神经网络训练方法,对传统的S型激励函数进行了改进,建立了自适应性的神经网络分类器。在假设样本中噪声服从于正态分布的情况下建立最大似然估计,通过后验概率建立马尔科夫链对样本进行训练,提高了神经网络训练速度。在轴承故障诊断中的测试结果表明,该算法可以迅速稳定地训练出神经网络,有效提高诊断的分类结果。     

低(等)压合成氨的最佳H2/N2与催化活性的关系

采用高压催化剂性能评价实验装置，在压力分别为7.0,10.0和15.0MPa，温度分别为350,400和450℃条件下，在H2／N2为1.6-3.0范围内研究了H2／N2对A301，ZA－5和A110－2型催化剂的活性和合成塔出口氨浓度的影响。在压力和空速一定的条件下，最佳H2／N2随反应温度而异。在350，400和450℃下，最佳H2／N2分别为1.8-2.2,2.2-2.5和2.5-3.0。由此可见，合成氨反应的速率达到最大值时的最佳H2／N2值与反应的进程有关。为此提出了催化剂效率K（Catalysis efficiency)的概念来表征在催化剂作用下反应器出口氨浓度趋近平衡的程度，即K＝CNH3／C^*NH3。根据实验结果，得到了最佳H2／N2与催化剂效率的定量关系：（H2／N2）m=1.5(1 CNH3/C^*NH3)=1.5(1 K)。由此可以根据催化剂在不同反应条件下的催化剂效率来确定最佳H2／N2。凡是会降低催化剂效率的因素，都会使最佳H2／N2降低。各种影响因素对最佳H2／N2的影响中，反应温度的影响最大，其次是空速和催化剂的活性，而压力和惰性气体含量的影响相对较小。在低温（低压）下合成氨，宜采用较低的H2／N2。     

基于浏览器 /服务器的橡胶加工网络测控系统

通过对客户机／服务器模式和浏览器／服务器模式的比较，选择浏览器／服务器模式组建了基于Internt/Intranet的远程橡胶加工网络测控系统，并介绍了系统的开发模式、方法和注意事项。     

Balanced pasture fertilization in the Basque Country: I. Phosphorus and potassium budgets on dairy farms

Dairy farming is the main agricultural activity of the Basque Country. A dairy farm is characterized as a system with soils and crops, forage, cattle and manure as main components, and in such a system, nutrient cycling is very important to maintain soil fertility and optimize forage production. To quantify nutrient transfers in the cycle, a simple system was developed and has been applied to seventeen farms to examine its ability to achieve a balanced P and K fertilization. These farms have provided data on inputs (fertilizer, feeds, concentrates), pasture and manure management, and outputs (milk production), and soil samples have been taken from farm pastures. Phosphorus and K in excreta and uneaten pasture is used with a relatively high efficiency as suggested by the relatively high efficiency of P and K utilization by the pasture that usually ranges from 70 to 90%. Concentrate feeding (3000 kg cow^–1 yr^–1) represents one of the main P and K inputs in Basque Country dairy farms, averaging 26 and 66 kg ha^–1, respectively. Besides, release of K in the soil through slow liberation from non-exchangeable sites was estimated as 30 kg ha^–1. Thus, a high efficiency in excreta recycling would diminish substantially P and K mineral fertilizer needs. Farm nutrient budgets appear to be a convenient tool for determining nutrient shortages and surpluses at farm level, and thus they are considered as a first step to support a better management of maintenance fertilization of permanent pastures.     

Mussoorrie rock phosphate-pyrite mixture as phosphate fertilizer

Mussoorrie rock phosphate (MRP), MRP + pyrite (25% by weight), diammonium phosphate (DAP), ammonium polyphosphate (APP) and nitrophosphate (NP) were compared in a field experiment as fertilizers for wheat. At 20 kg P ha^–1, MRP was only 6 per cent as effective as DAP. However, when it was mixed with pyrite, the efficiency of MRP increased to 64 per cent at 20 kg P ha^–1 compared with 97 per cent at 40 kg P ha^–1. The P requirement for a targeted yield for 4.5 t ha^–1 decreased from 39.4 kg P ha^–1 as MRP to 23.7 kg P ha^–1 as MRP + pyrite. Of the other P fertilizers studied, NP was as effective as DAP, whereas APP was 9 to 37 per cent more effective than DAP. However, the P requirement as DAP, NP and APP for a targeted yield of 4.5 t ha^–1 was similar (11 ± 0.5 kg P ha^–1).     

PRIVACY LEAKAGE IN HEALTH SOCIAL NETWORKS

Members of health social networks may be susceptible to privacy leaks by the amount of information they leave behind. The threat to privacy increases when members of these networks reuse their pseudonyms in other social networks. The risk of re‐identifying users from such networks requires quantitative estimates to evaluate its magnitude. The estimates will enable managers and members of health social communities to take corrective measures. We introduce a new re‐identification attack, the social network attack, that takes advantage of the fact that users reuse their pseudonyms. To demonstrate the attack, we establish links between MedHelp and Twitter (two popular social networks) based on matching pseudonyms. We used Bayesian networks to model the re‐identification risk and used stylometric techniques to identify the strength of the links. On the basis of our model 7‐11. 8% of the MedHelp members in the sample population who reused their pseudonyms in Twitter were re‐identifiable compared with 1% who did not. The risk estimates were measured at the 5% risk threshold. Our model was able to re‐identify users with a sensitivity of 41% and specificity of 96%. The potential for re‐identification increases as more data is accumulated from these profiles, which makes the threat of re‐identification more serious.     

浅谈燃气轮机的清洗方法

由于燃气轮机吸收含有大量灰尘等杂质的空气,会在压气机叶片上形成积垢,从而导致机组出力下降,该文介绍在实际工作中,根据不同的情况使用不同的设备、采取不同的清洗方法来除去压气机叶片上的污垢,以提高整台机组的工作效率.     

Increasing urea-N efficiency for transplanted lowland rice by pneumatic injection: Yield and economics at the farm level

To increase the fertilizer-N efficiency in lowland rice (Oryza sativa L.) cultivation, new management practices are needed. Main cause of the present low efficiency is the low N recovery by plants, as a considerable part of the N applied is lost; deep placement techniques improve the recovery. A pneumatic injector, with which urea prills can be point-placed at a depth of 5–10 cm in paddy soils, was tested in 38 on-farm trials in 1989/90, mostly during the wet season. The experiments, located in Africa and Asia, focussed on differences in grain yield between conventional methods of broadcasting urea and injection by the pneumatic injector, at recommended N-rates. The study shows that the pneumatic injector is effective as a tool to improve the N fertilizer efficiency. The average yield increases per region, resulting from the use of the injector, ranged from about 250 to 1300 kg grain ha^–1. The value of the yield increase would allow most farmers to recover the costs of the injector within one season, even if labour was hired to carry out the injections. The average labour requirement of the injector was 40 hours ha^–1. In Indonesia, injection of prilled urea gave yields similar to those obtained with urea briquettes.     

水泥回转窑的智能控制

文章介绍了PIT Navigator对水泥熟料生产自动化控制的实现,基于神经网络的自学习、自适应控制元件,将相关的过程数据记录下来,随后借助数据关联性进行分析,在预测和自学习过程模型(模型预测控制)中处理这些数据,并计算过程参数的实时预测,使用这些工具,可以实现更稳定的窑况,同时提高产量和降低能耗。