| 引用本文: | 任凯,李营,罗雯军.大型船舶流行病传播控制过程多智能体模型.[J].国防科技大学学报,2021,43(6):153-162.[点击复制] |
| REN Kai,LI Ying,LUO Wenjun.Multi-agent model of epidemic control process for large cruise ships[J].Journal of National University of Defense Technology,2021,43(6):153-162[点击复制] |
|
| |
|
|
| 本文已被：浏览 5631次 下载 4535次 |
| 大型船舶流行病传播控制过程多智能体模型 |
| 任凯1,李营1,罗雯军2 |
|
(1. 海军工程大学 动力工程学院, 湖北 武汉 430033;2. 中国舰船研究设计中心, 上海 201108)
|
| 摘要: |
| 基于复杂网络和传播动力学相关原理,构建了符合大型船舶人员交互特征的社交网络,定义了传播网络层级结构,分析了交互网络拓扑生成的5项规则,给出了大型船舶流行病传播社交网络构成方法。基于多智能体技术,研究了交互网络节点成员属性和流行病传播特点,给出了智能体成员状态空间的基本构成形式,融合防护治疗、管控隔离、信息交互等因素的物理特征,构建了智能体状态迁移和行为交互过程算法,并分析论证了衰减函数的结构及区间特性。对比距离恒定模型和随机游走模型,分别对典型大型船舶流行病传播过程分4种工况12种状态进行了仿真计算,结果表明:随机游走模型更适合信息量丰富条件下的大型船舶流行病早期传播过程模拟仿真和疫情防控策略分析。 |
| 关键词: 复杂网络 多智能体 社交网络 流行病传播 |
| DOI：10.11887/j.cn.202106018 |
| 投稿日期：2020-04-24 |
| 基金项目:国家部委基金资助项目(1020201021001) |
|
| Multi-agent model of epidemic control process for large cruise ships |
| REN Kai1, LI Ying1, LUO Wenjun2 |
|
(1. College of Power Engineering, Naval University of Engineering, Wuhan 430033, China;2. China Ship Research & Design Center, Shanghai 201108, China)
|
| Abstract: |
| Based on the theory of complex network and propagation dynamics, a social network was constructed, which conforms to the interaction characteristics of large cruise ships. The hierarchical structure of the communication network was defined, five rules of topology generation of the interaction network were analyzed, and the social networks construction method of the large cruise ships epidemic communication were given. Based on the multi-agent technology, the properties of the members of the interaction network nodes and the epidemic propagation characteristics were studied, the basic form of agent member state-space was given, the physical characteristics of protection and treatment, management and control isolation, information interaction and other factors were integrated, the algorithm of agent state transfer and behavior interaction process were constructed, and the structure and interval characteristics of attenuation function were analyzed and demonstrated. Compared with the constant-distance model and the random-walk model, the simulation calculations of the epidemic transmission process of typical large cruise ships were carried out under 4 working conditions and 12 states, respectively. Results show that the random-walk model is more suitable for the simulation of the early epidemic transmission process of large cruise ships and the analysis of epidemic prevention and control strategies with abundant information. |
| Keywords: complex network multi-agent social network epidemic propagation |
|
|
|
|
|
