Isolation Treatment Strategies For Emergency Control

As a kind of effective emergency control, isolation control has been widely employed in practice, such as epidemics control, Internet worm outbreak control, power system splitting and etc. But the theory of this strategy has not been discussed systemically. In this dissertation, based on optimal control theory, the design of dynamic optimal isolation control laws is proposed. And the applications in SARS epidemics control and Internet worm control are discussed. Furthermore, based on graph theory, the islanding problems of large system under emergency are also investigated. This dissertation gives a pilot study on isolation control theory and the major works are as follows:Firstly, in the absence of valid vaccine or medicines, quarantine and isolation strategies are the most important and effective measures against the outbreaks of SARS. Based on a SEQIJR transmission model, a control pair representing the quarantine and isolation strategies is introduced and incorporated in this model. And the application of the dynamic quarantine and isolation control for SARS epidemics control is discussed. The design of dynamic optimal and sub-optimal isolation control laws is proposed via Pontryagin's Maximum Principle and genetic algorithm, respectively. The simulation results illustrate the effectiveness of the optimal and sub-optimal strategies for outbreak control. And the results also demonstrate that the maximum implementations of quarantining and isolation strategies in the early stage of the epidemic are of very critical impacts in the both cases of optimal and sub-optimal control. This gives a theoretical interpretation to the practical experiences that the early quarantine and isolation strategies are critically important to control the outbreaks of epidemics. Furthermore, our results also show that the proposed sub-optimal control can lead to performances close to the optimal control, but with much simpler strategies for epidemics control in practical use.Secondly, differed from the research on signal isolated region in first part, this part concerns the spread of disease from region to region by means of traveling population and related dynamic quarantine and isolation strategies. A multigroup SARS transmission model is constructed and pairs of control variables in terms of the quarantine and isolation strategies are introduced in this model. The optimal control laws in each region are obtained via the Pontryagin's Maximum Principle, The simulation results well illustrate how the disease spreads from region to region by means of traveling population. Furthermore, the results not only demonstrate the importance of the early quarantine and isolation strategies but also the necessity of the observation and quarantine of travelers between regions to control the outbreaks of epidemics. This gives theoretical interpretations to the practical experiences that the early quarantine and isolation strategies, as well as the observation and quarantine of travelers between regions, are critically important to contain the epidemic.Thirdly, based on the Two-Factor transmission model, the application of the dynamic isolation control for network worm control is discussed. To this end, one control variable representing the isolation strategy is incorporated in the model. The optimal and sub-optimal isolation control laws are obtained via Pontryagin's Maximum Principle and genetic algorithm, respectively. The simulation results illustrate the effectiveness of the optimal and sub-optimal isolation strategies for worm outbreak control. And the results also demonstrate that the maximum implementation of isolation strategy in the early stage during the outbreak is critically important in worm control.Finally, based on system islanding, the concept of isolation control is introduced. By graph theory, the islanding problems of large system under emergency are discussed. And necessary conditions of optional isolation strategy and feasible isolation strategy are proposed. Besides, two types of isolation schemes, optimal cost scheme and time primary scheme are proposed and employed in network worm control and power system splitting, respectively.