Delayed Dynamics of SIR Model for COVID-19

This paper presents a new modified SIR model which incorporates appropriate delay parameters leading to a more precise prediction of COVID-19 real time data. The efficacy of the newly developed SIR model is proven by comparing its predictions to real data obtained from four counties namely Germany, Italy, Kuwait, and Oman. Two included delay periods for incubation and recovery within the SIR model produce a sensible and more accurate representation of the real time data. In the absence of the two-delay period () the dynamical behavior of the model will not correspond to today’s picture and lag the detection of the epidemic peak. The reproductive number R0 is defined for the model for values of recovery time delay of the infective case. The effect of recovery time may produce second wave, and/or an oscillation which could destabilize the behavior of the system and a periodic oscillation can arise due to Hopf bifurcation phenomenon.

基于边划分理论的谣言传播模型

针对谣言传播过程中传播态节点恢复时会受其邻居节点状态影响的问题,提出了一种基于边划分理论的谣言传播模型。首先,使用改进的边划分理论建立起谣言传播的动力学方程组,推演出谣言在复杂网络上的传播范围值和爆发阈值;然后,通过数值仿真实验研究网络结构、传播概率和基础恢复概率等参数对谣言传播的影响;在此基础之上,提出了可以有效控制谣言传播范围和爆发阈值的免疫策略。理论分析和仿真结果表明,与经典的SIR模型相比,提出的谣言传播模型缩短了谣言传播的周期,传播态节点比例的峰值则有小幅提高。对比实验发现,与现有的随机免疫策略相比,当谣言的传播概率较大时,优先免疫连接小度节点的边能得到更小的谣言传播范围;反之,当谣言的传播概率较小时,优先免疫连接大度节点的边可以有更小的谣言传播范围。研究结果表明,提出的谣言传播模型符合谣言消退期的特征,为谣言传播的预测与控制提供了理论和数值上的支持。