This paper provides an unified framework for resilience hierarchical-multi-level system-of-systems (SoS), from hypergraph modeling. To actively detect physical faults that may affect SoS and adapt the overall operating modes, a resilience approach is proposed, composed of both bottom-up monitoring and top-down reconfiguration strategies. Based on the fault detection and isolation (FDI) algorithm that is applied to cyber-physical component systems (CPCS), located at the bottom level, a bottom-up monitoring is enabled detecting faulty behaviors and informing the supervisor at a higher level of the fault information. Meanwhile, upon the received fault information, the supervisor is able sending instructions to component systems at lower levels to optimize the performance of the whole SoS, which is named as top-down reconfiguration. Compared with previous works mainly performing the reconfiguration from only one aspect, this work develops a comprehensive reconfiguration framework from communication, operational independence and mission perspectives based on hypergraph drawn from related viewpoints. A case study on a SoS composed of intelligent autonomous vehicles (IAV) and unmanned aerial vehicles (UAV) are given finally to validate the effectiveness of theoretical results.