When power systems encounter outages and largescale blackouts, system restoration is critical and should be carried out with dedicated schemes. In this past, most studies divided the power system restoration into three stages (i.e., black-start zone partitioning, network reconfiguration, and load restoration) and deal with them separately. After that, few studies considering the three stages together were emerging while the support of multiple flexible resources, i.e., renewable energy source (RES), electric vehicle system (EVS) and energy storage system (ESS), were not considered comprehensively. Therefore, a bi-level coordinated power system restoration (BiCPSR) model is proposed in this work considering the support of multiple flexible resources. In the upper level, two network topology indices that describe the “reachability” and “shortest reachable distance” of buses in power systems, and the restoration characteristics of generators and loads are utilized for optimizing the start-up sequence of generators and network reconfiguration. In the lower level, the uncertainties of RES and EVS are considered by various scenarios and the support of multiple flexible resources is utilized cooperatively for accelerating the restoration process and maximizing the restorable load. Case studies on the revised IEEE 39-bus, WECC 179-bus and the actual Zhejiang power systems are performed to illustrate the basic features of the proposed model and its availability in bulk power systems. The comparisons between the proposed model and other models are also performed to illustrate the strengths of the proposed model.