Research On Resilience Optimization Of Consecutive-k-out-of-n Systems Based On Reconfiguration Modes

Consecutive k-out-of-n systems consist of n components arranged in a line or a circle,which are widely used in a variety of major engineering systems,such as vacuum accelerator systems,microwave tower systems,spacecraft relay station systems,quality control batch sampling systems,nuclear accelerator photography systems.This kind of system has always been the focus of the design of redundant systems and the configuration of resources.Considering the reconfigurability of consecutive k-out-of-n system,its system resilience can be improved by the reconfigurable modes integrating the rigid method with high performance(adding redundant spare parts and replacing components)and the flexible soft method(changing positions of components and selecting maintenance modes).Under the constraints of risk events and maintenance resources,how to design the configuration of the consecutive k-out-of-n system with maximum system resilience in the defense phase and recovery phase reasonably is a key issue that needs to be solved urgently for the design of redundant systems and the configuration of resources.The innovations and main research content are listed as follows.(1)Based on the improved system reliability of consecutive k-out-of-n systems,we discussed the resilience evaluation method of consecutive k-out-of-n systems in the defense phase,recovery phase,and defense-recovery oriented two phases.The resilience importance measures of consecutive k-out-of-n systems are developed in the defense phase and recovery phase,respectively.(2)This paper analyzes the influences of the component configuration(component redundancy and arrangement)and external risks(risk types and risk modes)on the system defense ability of consecutive k-out-of-n systems in detail.Secondly,the optimization model of external risks-based system defense ability is constructed.The defense importance-based genetic algorithm(DIGA)is proposed to solve the optimization model.Simulation experiments the distribution rules of component redundancy in the optimal system defense configuration under different risks.The simulation results are shown as follows:(1)In the cases of consecutive risks,F systems add redundant components at intervals,and the component module at the k-th position must add the redundant components;while G systems add redundant components continuously at the positions of occurring risk events.(2)If the risks are at intervals,prioritizing the addition of redundant components where risk events occur,and then adding redundancy to the positions where the risks do not occur.The larger the value of k,the easier F system is to add redundant spare parts in intervals.while G systems add redundant components to consecutive k positions.(3)This paper analyzes the influences of the component configuration(maintenance sequence and maintenance mode)and the different number of workers on the system recovery ability.The optimization model of maintenance resource-based system recovery ability is constructed,and the recovery importance-based genetic algorithm(RIGA)is proposed to solve the optimization model.The performance of RIGA with the classical genetic algorithm is compared in simulation experiments,and the distribution rules of maintenance sequence and maintenance modes in the optimal system recovery configuration are analyzed.The simulation results are shown as follows: The component modules with maintenance priority should use the accelerated repair mode,which can be selected according to the recovery importance.(4)A defense-recovery phase-based resilience joint optimization model of consecutive k-out-of-n systems is constructed.Then,a resilience importance-based joint optimization algorithm is developed,which is a two-stage algorithm combining DIGA with RIGA to solve the joint optimization model.The simulation experiments analyze the changes in the optimal system resilience under different numbers of workers and cost allocation ratios.Taking the production monitoring system as an example,the optimal system resilience configurations under different risks are analyzed,which can verify the redundancy allocation rules in defense configurations and the distribution rules of maintenance modes and maintenance sequence in recovery configurations.