|
| Step number | Step name | Main operation |
|
| 0 | Initialization | Compute the initial load and capacity of all stations in the interdependent PTN (see the detailed introduction in the Table 5, station initial load definition, and station capacity definition) |
| 1 | Trigger the cascading failures | The cascading failures are triggered by attacking a single station that has a certain characteristic (see the detailed introduction in the Table 5, attack strategy). |
| 2 | Search adjacent stations and interdependent stations of the attacked stations | On the one hand, there are some edges between the attacked station and its adjacent stations. These edges will be used to transit the redistribution loads from the attacked station. On the other hand, the interdependent station of the attacked station is also influenced by the failure of the attacked station. |
| 3 | Redistribute the load to the adjacent stations and impose coupled effect to the interdependent stations | For the adjacent stations, the redistribution loads from the attacked station are redistributed to the adjacent stations according to certain rule (see the detailed introduction in the Table 5, rule of failure load dynamic redistribution).
For the interdependent stations, the coupled effect is imposed by considering the combined influences of the type of interdependent network and the interdependent type between interdependent stations (see the detailed introduction in the Table 5, type of interdependent network, and interdependent type between interdependent stations). |
| 4 | Compare and estimate the station state | Compare updated station load with station capacity (this operation is conducted to all receiver stations of redistribution loads and all influenced stations by coupled effect), if the updated station load is larger than the station capacity, the station will be failed (failure state, see the detailed introduction in the Table 5, station state). Then, go on to carry on the step 2 and step3 successively. Furthermore, carry on this process repeatedly, until no stations occur cascading failures any more, and at this moment, the whole process of cascading failures ends. |
| 5 | Compute the resilience measurement indicators | Compute various resilience measurement indicators before and after cascading failures occurring (see the detailed introduction in the Table 5, resilience measurement indicator). In particular, if resilience measurement indicator is a local indicator or an indicator measuring each time step, this resilience measurement indicator should be computed in every time step (a typical example can be seen in the work of Zhang et al.[32]). |
| 6 | Change the value of the control parameters of cascading failures based dynamic resilience model | The cascading failures based dynamic resilience model of interdependent PTN has various control parameters for better controlling the model to be more realistic. Thus, change the value of various control parameters and repeat the steps above, so that a more complete and comprehensive numerical simulation analysis is conducted (a typical example can be seen in the work of Zhang et al.[71]). |
|
