基于粒子群算法的GM(1,1)在经济发展预测中的应用

采用粒子群算法对GM(1,1)模型中的参数进行优化求解,然后将得到的模型应用于四川省经济发展预测,并和传统采用最小二乘法求解参数的GM(1,1)的预测结果进行了对照.结果表明,采用粒子群算法优化参数的GM(1,1)模型预测效果明显优于传统采用最小二乘法求解参数的GM(1,1)的模型预测效果.     

A model for analyzing the transshipment competition relationship between the port of Hong Kong and the port of Kaohsiung

The transshipment is an important issue in the current marine transportation. This paper constructed a mathematical programming model to elaborate the transshipment port selection for the shipping company. This model is tested by the data collected from the ports of Hong Kong and Kaohsiung. The results show that this model can be used to explain the transshipment competition relationship between the ports of Hong Kong and Kaohsiung well. A sensitivity analysis was also executed. The sensitivity analysis results showed that both the port of Hong Kong and Kaohsiung should decrease the charges of port and increase the efficiency of loading and discharging. Based on the sensitivity analysis results, some interesting conclusions and helpful suggestions were obtained for the managers of the ports of Hong Kong and Kaohsiung to improve their port management.     

基于GA优化的灾情评估普适公式

为建立适用于多种灾情的评估指数公式,运用遗传算法,优化公式中的参数.该公式对于分级指标项数没有限制,使用范围广泛.应用该公式对1990年我国部分省市发生的9次自然灾害进行等级划分,并与其它方法的评估结果进行比较,验证了公式的正确性.     

基于遗传算法的免疫网络设计

从人体免疫网络抵御抗原与计算机网络抵御入侵的相似性,提出一个基于遗传算法的免疫网络并应用于分布式入侵检测系统.将一个基于每秒网络数据量为参数的时间序列作为检测子的抗体表达形式,在单个网段检测网络数据,通过各个网段之间的抗体交流,实现抗体的优化.在抗体的进化过程中,提出了较好地评价抗体适应度的评估函数,为优化抗体的遗传算法的实现打下了基础.     

EVOLUTION OF URBAN SYSTEM IN NEW ECONOMIC CIRCUMSTANCES AND PLANNING COUNTERMEASURES

As one kind of organizational relation of regional space, in essence, urban system is the spatial reflection of the deep social economic movement.(HERZOG,1991). Since the 1980s, by means of the globalization of economy and high technology and the particular spatial effect, the evolution of urban system has taken on one brand-new landscape different from the traditional one, which also raise new concepts and requirements for urban system planning. Meanwhile, it′s no doubt that active plannin…     

A new method to determine the tropopause

The tropopause has a complex structure and some interference information may exist in high-resolution global positioning system （GPS）/low earth-orbiting （LEO） radio occultation （RO） data. The position of the tropopause cannot be accurately determined using traditional cold point tropopause （CPT） and lapse rate tropopause （LRT） algorithms. In this paper, an integrative algorithm is developed to determinate tropopause parameters. The algorithm is applied to GPS/COSMIC RO data to obtain a global distribution of the height and temperature of the tropopause. This algorithm improves the utilization rate of GPS/LEO RO data by 30% compared with that from the traditional CPT method. The rationality and reliability of GPS/LEO RO data in probing the Earth＇s atmosphere are verified by our study of the tropopause using COSMIC data.     

基于满意优化原理的网络QoS路由研究

Qos（Quality of Service）路由选择问题是网络多媒体信息传输的关键技术.随着新兴的分布式多媒体应用大量出现和网络多媒体应用技术的高速发展,高效的QoS支持将变得越来越重要,更是彰显了QoS路由选择问题的重要性.提出了一种基于满意优化原理和遗传算法的QoS路由求解算法来解决QoS路由问题,该算法能极大地缩短路由求解时间,提高路由求解的成功率,并能在最大程度上避免出现拥塞,满足了网络流量工程的需要.     

免疫遗传算法及其在VRP中的应用

物流配送车辆路径问题（VRP）是一类典型的NP问题。在基本遗传算法的基础上,根据生物的免疫系统原理,提出一种改进的算法——免疫遗传算法。在算法中构造一种基于抗体浓度的群体多样性保持策略,引入免疫算子和免疫记忆库。将该算法应用于求解VRP问题,实验结果表明算法可以实现解的多样性,避免出现早熟收敛,可以有效防止进化过程中最优解退化的可能,是求解车辆路径问题的一种有效的算法。     

基于模糊C均值的进化神经网络分类模型

将模糊C均值方法与进化算法相结合,提出了一种新的进化神经网络分类模型。实验结果表明,模型与其它标准的分类方法：C4.5、CART、BP神经网络、模糊ARTMAP比较,有更高的分类精度。     

���������Ż�RBF�����ڶ�̬����Ԥ���е�Ӧ��

?????????????????????????????????????????????????RBF?????????????????????CIOA????RBF??????????????????????????????CIOA?????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????RBF???????????????????????У???Ч????????????????????     

洪灾应急疏散路径规划算法优化

应急疏散是救灾工作的重要环节,合理的路径规划能有效缩短疏散时间,减少人员伤亡.本文以疏散总时间最短为目标,考虑需求控制,容量限制,交通延误,公平分配和资源节约等约束条件,对经典Dijkstra算法进行改进;并采用混合拆分疏散方法,构建洪灾避难应急疏散路径规划模型.运用C#语言编写算法,求解最佳路径,基于自主开发的应急疏散分析工具MiniGIS,对规划路径进行动态模拟,依据反馈逐次优化算法.结果表明:理想算法,延时-改进算法,逆行-改进算法均能为县域尺度的洪灾避难疏散路径选择提供参考,但理想算法适用于组织简单,高度有序的疏散情景,延时-改进算法考虑了除交通拥堵之外的延误,与真实情况更为接近,逆行-改进算法避免了因中途路径调整而出现的"回头路",在时间最短次优的条件下,更有利于疏散过程管理与资源节约,其结果被认为是此次应急疏散路径规划的最优解.