基于结构继承的贝叶斯网结构学习优化设计

采用遗传算法建立贝叶斯网络的优化学习结构,一直是贝叶斯网络研究倍受关注的课题.传统遗传算法的个体设计存在需要反复进行无环性检验的问题,降低了进化效率.针对这个问题,提出一种新的个体编码方式.考虑到进化过程中家族得分的可继承性,提出基于家族继承的结构评分改进算法,进而设计相应的改进遗传算法.实验结果表明,改进算法在BN建网精度与效率上都得到明显提升.     

基于遗传算法的关联议题并发谈判

基于遗传算法的并发谈判在电子商务应用中具有独特的优势,但已有的研究尚未考虑议题之间的相关性、动态权重的变化规则依赖于对手信息的获取,极大地限制了其使用价值.对此,提出议题分组的方法以解决议题的相关性问题,而议题权重的变化则采取从历史资源中发掘知识的方法进行动态调整.构建了基于遗传算法的关联性议题并发谈判模型,给出了模型的形式化描述、谈判算法设计和动态权重更新方案.通过对模型的实验和比较分析,证实了该方案能够更方便地满足用户谈判的多样性需求,解决谈判中议题关联性的问题,而且能够快速、有效地得出最优谈判结果.     

一种新型ICA算法在RFID系统中的应用

针对无线射频识别系统在通信过程中的防碰撞问题,从信号处理的角度,提出一种基于盲源信号并行分离技术的防碰撞处理算法,其中包括信号白化技术,以及一种改进的独立元分析(ICA)信号并行分离方法.功能仿真结果证明,该算法的吞吐率最高达到69％,与吞吐率只有18％的传统Aloha算法相比,该算法能提高防碰撞能力.     

点到平面距离公式的一种新证法

该文根据球面上一点切平面的方程和平移变换,对空间任一点P(x0,y0,z0),分P点为坐标原点和P点不是坐标原点两种情形,给出了到平面π:Ax By Cz D=0的距离公式d=|Ax0 By0 Cz0 D|/√A2 B2 C2 的一种新证法.     

大坝监测分析中的时间序列叠合模型

叙述了用时间序列叠合模型建立大坝监测预报模型的2种基本方法:统计-时间序列模型和确定性时序叠合模型。通过统计软件SPSS实现2种时序叠合模型的实例表明，在环境量记录完整的情况下，统计-时间序列模型比单纯的统计回归模型精度要高;在缺少环境量监测或与环境量的因果关系不明显的情况下，用确定性叠合模型分析和预报大坝监测效应量的测值，精度也很高，并能够分离出趋势性分量、周期分量和随机平稳分量，分析的结果简单明了，是大坝监测分析中一种比较好的方法。     

UPS并机数字锁相误差分析

针对不间断电源(UPS)并联系统同步问题,进行锁相误差分析,并探讨了减少锁相误差的措施和方案。通过在两台30kVA样机上进行并联实验,验证了方案的正确性。     

基于Matlab的0-1背包问题的动态规划方法求解

背包问题是经典的NP-hard组合优化问题之一,在经济管理、资源分配、投资决策、装载设计等领域有着重要的应用价值。文中用动态规划方法解决0-1背包问题,通过在Matlab6.5环境下对其算法进行测试和与其他方法对比分析,表明应用该方法可节省大量的计算时间,因而具有更高运行效率。     

The generalized Riemann problem method for the shallow water equations with bottom topography

This paper extends the generalized Riemann problem method (GRP) to the system of shallow water equations with bottom topography. The main contribution is that the generalized Riemann problem method (J. Comput. Phys. 1984; 55 (1):1–32) is used to evaluate the midpoint values of solutions at each cell interface so that the bottom topography effect is included in numerical fluxes, and at the same step the source term is discretized with an interface method in which only mid‐point values are plugged in. This scheme is well balanced between the flux gradient and bottom topography when incorporating the surface gradient method (SGM) (J. Comput. Phys. 2001; 168 (1):1–25) into data reconstruction step, and it is also suitable for both steady and unsteady flow simulations. We illustrate the accuracy of this scheme by several 1‐D and 2‐D numerical experiments. Copyright © 2005 John Wiley & Sons, Ltd.     

A finite element method for poro mechanical modelling of geotechnical problems using local second gradient models

In this paper, a new finite element method is described and applied. It is based on a theory developed to model poromechanical problems where the mechanical part is obeying a second gradient theory. The aim of such a work is to properly model the post localized behaviour of soils and rocks saturated with a pore fluid. Beside the development of this new coupled theory, a corresponding finite element method has been developed. The elements used are based on a weak form of the relation between the deformation gradient and the second gradient, using a field of Lagrange multipliers. The global problem is solved by a system of equations where the kinematic variables are fully coupled with the pore pressure. Some numerical experiments showing the effectiveness of the method ends the paper. Copyright © 2005 John Wiley & Sons, Ltd.     

Using immune genetic algorithm based hybrid techniques for short-term unit commitment problem

This paper presents a hybrid chaos search (CS), immune algorithm (IA)/genetic algorithm (GA), and fuzzy system (FS) method (CIGAFS) for solving short-term thermal generating unit commitment (UC) problems. The UC problem involves determining the start-up and shut-down schedules for generating units to meet the forecasted demand at the minimum cost. The commitment schedule must satisfy other constraints such as the generating limits per unit, reserve, and individual units. First, we combined the IA and GA, then we added the CS and the FS approach. This hybrid system was then used to solve the UC problems. Numerical simulations were carried out using three cases: 10, 20, and 30 thermal unit power systems over a 24 h period. The produced schedule was compared with several other methods, such as dynamic programming (DP), Lagrangian relaxation (LR), standard genetic algorithm (SGA), traditional simulated annealing (TSA), and traditional Tabu search (TTS). A comparison with an immune genetic algorithm (IGA) combined with the CS and FS was carried out. The results show that the CS and FS all make substantial contributions to the IGA. The result demonstrated the accuracy of the proposed CIGAFS approach.