基于改进SEIR模型的新冠肺炎疫情分析与预测

Analysis and Prediction of the COVID-19 Based on an Improved SEIR Model

针对新型冠状病毒的潜伏期较长,基于每日发布的新冠疫情数据集,在经典SEIR (Susceptible-Exposed-Infected-Recovered)模型的基础上,考虑了隐性传播人群,并且将确诊人群分为两类(一类感染者具有传染能力;一类感染者由于处于隔离期间,其感染能力可忽略不计),构建了基于改进SEIR的新冠肺炎传播动力学模型。以2021年12月15日到2022年1月13日的西安市疫情数据为依据,拟合得到了改进SEIR模型的动力学参数,对西安市COVID-19疫情进行预测和评估。结果表明,基于改进SEIR传染病动力学模型对疫情的理论估计与西安市疫情的实际情况较为符合,数据可视化和医学隔离等措施对抑制疫情大面积传播有重要作用。

Considering the lengthy incubation period of the novel coronavirus,based on the COVID-19 data set released daily,this paper improves the classic SEIR(Susceptible-Exposed-Infected-Recovered) model and establishes the transmission dynamics model of COVID-19,in which the inapparent transmission population is included,and the confirmed population is divided into two categories(a class of infected persons has the infectivity,while another class of infected persons has negligible infectivity because they are in the quarantine period).The kinetic parameters of the improved SEIR model are fitted by using the epidemic data in Xi'an from December 15,2021,to January 13,2022,and the COVID-19 epidemic in Xi'an is predicted and evaluated.The results show that the theoretical estimate of the epidemic based on the improved SEIR model is in line with the actual situation of the epidemic in Xi'an,and the measures such as data visualization and medical isolation have an important role in inhibiting the large spread of the epidemic.