An Improved Time-domain Decomposition Approach for Chance-Constrained Fast Production Cost Minimization Simulation

Chance-constrained production cost minimization (PCM) simulation has been employed for long-term power system simulations and assessments considering uncertainties. However, solving such a problem is time-consuming for its numerous binary variables. Traditional time-domain decomposition (TDD) is widely used for acceleration, but plenty of temporal constraint violations make the resulting solution significantly less accurate. To address this drawback, this paper proposes an improved TDD approach based on optimal initial decomposition segment (IDS) to generate sub-problems, thus reducing the temporal constraint violations across sub-problems. Simulation results show that the proposed approach provides the near-optimal solution with less than one violation in 100,000 constraints, and improves the computational efficiency by 3 orders of magnitude.