Relaxation or decomposition of the MIQCP normalization constraint

Develop relaxation or decomposition approaches that eliminate or otherwise resolve the nonconvex normalization constraint in the mixed-integer quadratically constrained attacker-selection formulation, thereby simplifying the optimization problem and improving computational time.

Background

The proposed attacker-selection problem is reformulated as a mixed-integer quadratically constrained program (MIQCP). Although this formulation can be solved by commercial solvers, its target-state normalization constraint remains nonconvex. The paper identifies relaxation or decomposition methods as a future direction because removing this nonconvexity could substantially reduce computational effort and make the formulation more tractable for larger power systems.