TY - JOUR
T1 - Restoration of interdependent network against cascading overload failure
AU - Zhong, Jilong
AU - Zhang, Feng Ming
AU - Yang, Shunkun
AU - Li, Daqing
N1 - Publisher Copyright:
© 2018 Elsevier B.V.
PY - 2019/1/15
Y1 - 2019/1/15
N2 - Many networks are physically or logically interdependent with each other, such as smart power grid, city traffic network and communication systems, where cascading overload failure becomes a major threat. Based on a load-dependent cascading model, we investigate the restoration characteristics in the consideration of repair resource, timing and load tolerance, for different coupling strength and network topologies in interdependent networks. We find that the restoration on the network with different coupling strength may lead to two extreme system effects with early repair: full recovery or completely collapse. Furthermore, SF–SF network is sensitive to repair resources, while repair effect of ER–ER network increases sharply when load tolerance is increased. When overloads are triggered in an ER network coupled with a SF network, the restoration effect can be obviously worse than other topology combinations. Our findings may help to design restoration strategy for interdependent networks and improve the system resilience.
AB - Many networks are physically or logically interdependent with each other, such as smart power grid, city traffic network and communication systems, where cascading overload failure becomes a major threat. Based on a load-dependent cascading model, we investigate the restoration characteristics in the consideration of repair resource, timing and load tolerance, for different coupling strength and network topologies in interdependent networks. We find that the restoration on the network with different coupling strength may lead to two extreme system effects with early repair: full recovery or completely collapse. Furthermore, SF–SF network is sensitive to repair resources, while repair effect of ER–ER network increases sharply when load tolerance is increased. When overloads are triggered in an ER network coupled with a SF network, the restoration effect can be obviously worse than other topology combinations. Our findings may help to design restoration strategy for interdependent networks and improve the system resilience.
KW - Cascading failure
KW - Interdependent network
KW - Load-dependent
KW - Restoration
UR - https://www.scopus.com/pages/publications/85054232234
U2 - 10.1016/j.physa.2018.09.130
DO - 10.1016/j.physa.2018.09.130
M3 - 文章
AN - SCOPUS:85054232234
SN - 0378-4371
VL - 514
SP - 884
EP - 891
JO - Physica A: Statistical Mechanics and its Applications
JF - Physica A: Statistical Mechanics and its Applications
ER -