| Resiliency refers to the ability of a system to recover from a disruptive or traumatic event and return to previous or normal work.Resilience is considered to be an inherent capability and inherent property of the system.Resilience systems reduce the possibility of damage or failure by improving the ability to absorb damage propagation during interruptions.To reduce consequences such as social and economic impacts,adaptation to disruptions should be strengthened for timely recovery of the system.With the continuous application and expansion of resilience in various fields such as ecology,engineering and economy,resilience has been more widely understood in the academic field in recent years.Resilience engineering can be regarded as an extension of system security engineering.Resilience engineering system has been advocated as a new safety management paradigm,and it has also received more and more attention and application.In general,the research on resilience has the following shortcomings.First,the research depth on resilience is still insufficient.Most of them stay on the surface analysis,such as the measurement of resilience.Most literature uses qualitative evaluation methods to evaluate system resilience.The evaluation method is relatively abstract and more difficult to understand than the quantitative evaluation method.Second,there are also some difficulties in network modeling and analysis of resilience systems,and there is very little literature that incorporates the consideration of maintenance strategies into the system’s elastic recovery and resilience assessment.It also lacks the formulation and analysis of corresponding maintenance strategies and the corresponding example applications.Third,there is also a lack of classified discussion of general and wartime situations in the formulation of maintenance strategies involving elastic recovery.Fourth,in the existing system recovery strategies,the setting of the adaptive function of the algorithm rarely involves the shortcoming of task importance information,and there are relatively few constraints on time and spare parts.In this way,it is difficult to simulate the actual situation of rush repair and wartime repair in a limited time,so it is also difficult to obtain corresponding rush repair and wartime system resilient recovery strategies.In view of the above,this article summarizes the definition and measurement standards of resilience firstly,including qualitative and quantitative measurement methods.Corresponding basic information is added to the modeling of elastic system networks,andthe formulation of resilient recovery strategies is divided into general situations and wartime situations for classification and discussion.Improve the generation of ancestral populations in genetic algorithms,according to the priority model and roulette strategy to generate half of the initial individuals.Use improved genetic algorithm to formulate maintenance strategy.In general,the economy is taken as the optimization target,the group maintenance mode is considered,and the genetic algorithm is used to solve the resilient recovery strategy with the lowest economic cost.In wartime situations,mission importance is used as an optimization goal,and improved genetic algorithms are used to solve the system’s resilient recovery strategy that considers group parallel maintenance and priority recovery of key important functions in a limited time.Group and parallel maintenance factors are included in the scope of system resilience research,and an in-depth system resilience analysis study is performed.Finally,a quantitative evaluation of system resilience is performed,and a comparative analysis of this resilience recovery strategy with a normal recovery strategy shows the excellent maintenance strategy and the resilience of the system. |
