Recovery-Oriented Optimal Investment Planning for Power Grid: A Two-Stage Risk-Averse Optimization Approach

The occurrence of high-impact, low-probability natural disasters poses a threat to the power grid, leading to potential widespread disruptions and raising stringent requirements for power grid restoration capability. Moreover, the indeterminate magnitude of damage resulting from such catastrophes introduces additional risks. In order to develop a far-sighted plan and enhance the restoration capability of power grid, this paper proposes a stochastic recovery-oriented power grid investment planning model which integrates preventive hardening measures and responsive backup generator (BG) support to mitigate the impact of disasters and facilitate efficient rehabilitation. The recovery process entails a sequential restoration of impaired transmission lines, offline buses, offline generators, and offline renewable power generations (RPGs). Then, a risk-averse planning strategy using two-stage stochastic information gap decision theory (IGDT) is adopted in the proposed baseline model to counter the uncertainty surrounding the severity of natural disasters. The effectiveness of the proposed method is validated through numerical tests conducted on the New England 10-machine 39-bus system, which demonstrate its potential in reducing recovery time and minimizing losses during power grid restoration by allocating additional resources.