Kohonen网络与BP网络的集成应用研究

本文介绍了 Kohonen神经网络对输入数据进行聚类方法在卷烟配方中的应用 ,提出了从核心样本动态搜索 BP网络训练样本的新探索 ,摒弃了过去 BP算法中训练样本固定不变 ,互不相交的方法 ,实现了 BP网络和 Kohonen网络动态无缝集成。     

Fuzzy Optimum Model of Semi-Structural Decision for Lectotype Optimization of Offshore Platforms

In the process of concept design of offshore platforms, it is necessary to select the best from feasible alternatives through comparison and filter. The criterion set, used to evaluate and select the satisfying alternative, consists of many qualitative and quantitative factors. Therefore, the selection is a problem of multicriteria and semi- structural decision- making. Different from traditional methods in semi- structural decision- making, a new framework and methodology is presented in this paper for evaluation of offshore platform alternatives. First, the criterion set is established for the evaluation of alternatives. Next, the approach is studied to construct the relative membership degree matrix, in which both qualitative and quantitative factors are consistent with the uniform calculating standard. And then a new weight-assessing method is developed for calculation of the weights based on the relative membership degree matrix. Finally, a multi-hierarchy fuzzy optimum model is adopted to select t     

LabVIEW设计中压力传感器的RBF神经网络温度补偿

在石油平台注水压力监测系统设计中 ,采用LabVIEW虚拟仪器平台 ,嵌入逼近能力强和收敛速度快的RBF神经网络 ,以人工环境实验数据为样本进行训练 ,实现了压力传感器的智能网络温度补偿。结果显示 ,此方法能够在压力、温度变化较大的恶劣环境下 ,获得很高的补偿精度。     

Optimization Mathematical Model of Pile Forces for Offshore Piled Breasting Dolphins

An optimization mathematical model of the pile forces for piled breasting dolphins in the open sea under various loading conditions is presented. The optimum layout with the well distributed pile forces and the least number of piles is achieved by the multiplier penalty function method. Several engineering cases have been calculated and compared with the result of the conventional design method. It is shown that the number of piles can be reduced at least by 10%—20% and the piles‘ bearing state is improved greatly.     

基于网络药理学探究补阳还五汤治疗颈椎病的作用机制

目的：基于网络药理学与分子对接技术分析补阳还五汤治疗颈椎病的作用机制。方法：使用TCMSP、化学专业数据库获取补阳还五汤的活性成分,并对潜在靶点进行预测及规范。分别从PharmGKB、DisGeNET、OMIM、GeneCards数据库中得到颈椎病疾病靶点,利用韦恩图获取补阳还五汤与颈椎病的交集靶点。通过CytoScape软件构建中药-活性成分-疾病靶点和蛋白质-蛋白质相互作用（PPI）网络,获得核心有效成分与关键靶点,利用David数据库对潜在靶点进行富集分析。最后运用AutoDock Vina软件对补阳还五汤核心有效成分与关键靶点进行分子对接验证。结果：共获得补阳还五汤治疗颈椎病的有效活性成分97个,包括槲皮素、山柰酚、黄芩素、木犀草素等;交集靶点64个,关键靶点有白细胞介素-6（IL-6）、肿瘤坏死因子（TNF）、丝氨酸/苏氨酸蛋白激酶1（AKT1）、白细胞介素-1B（IL-1B）等;主要涉及肿瘤坏死因子（TNF）、白细胞介素-17（IL-17）、磷脂酰肌醇3-激酶（PI3K）/蛋白激酶B（Akt）等信号通路,分子对接结果显示核心有效成分与关键靶点之间结合紧密,为补阳还五汤治疗颈椎病提供相应条件。结论：该研究在总体上预测了补阳还五汤治疗颈椎病的活性成分、靶点和信号通路,作用途径广泛,为下一步的临床应用提供参考及思路。     

人工神经网络在地质学判别分析中的应用

将现代人工神经网络方法应用于地质学的判别分析，通过实例计算表明，无论是对于两组判别分析还是多组判别分析，人工神经网络方法都是切实可行的。     

A Linear Programming Model for Seabed Oil-Gas Pipe Network

This paper analyzes the pipe network system of oil-gas collection and transportation foroffshore oilfield development.A"0-1"integer linear programming model is constructed to optimize theinvestment of seabed pipe network.The mathematical model is solved by the spanning tree method ofgraph theory and network analysis.All spanning trees of a network graph compose all the feasible solu-tions of the mathematical model.The optimal solution of the model is the spanning tree with the minimumcost among all spanning trees.This method can be used to optimize the seabed pipe network system andgive a minimum cost plan for the development of offshore marginal oilfield groups.     

基于模拟退火算法的饲料配方设计

本文在饲料配方设计中首次引入新颖、高效的组合优化算法--模拟退火算法，并获得理想的数值结果，说明了在传统的饲料配方设计中高技术的应用可以获得更高的效益.     

应用遗传神经网格方法分析赤潮监测数据

简要介绍了遗传神经网络方法的原理 ,探讨了应用遗传神经网络方法研究辽东湾海域丹麦细柱藻 (Leptocylindrusdanicus)赤潮与其环境因子之间的关系 ,计算了各环境因子对丹麦细柱藻赤潮的贡献。结果表明 ,温度、盐度、DIN的变化对研究海域丹麦细柱藻种群密度的增长有比较重要的影响 ,DIN是营养限制因子。遗传神经网络是分析赤潮监测数据的有效方法 ,值得进一步探索     

Neural network for wave forecasting among multi-stations

Unlike in the open sea, the use of wind information for forecasting waves may encounter more ambiguous uncertainties in the coastal or harbor area due to the influence of complicated geometric configurations. Thus this paper attempts to forecast the waves based on learning the characteristics of observed waves, rather than the use of the wind information. This is reported in this paper by the application of the artificial neural network (ANN), in which the back-propagation algorithm is employed in the learning process for obtaining the desired results. This model evaluated the interconnection weights among multi-stations based on the previous short-term data, from which a time series of waves at a station can be generated for forecasting or data supplement based on using the neighbor stations data. Field data are used for testing the applicability of the ANN model. The results show that the ANN model performs well for both wave forecasting and data supplement when using a short-term observed wave data.