Comparison of an Improved Metaheuristic and Mathematical Optimization Based Methods to Solve the Static AC TNEP Problem

The complexity of current and future electricity networks demands the use of more accurate models to solve the Transmission Network Expansion Planning (TNEP) problem. To deal with this issue, formulations based on AC network equations have been proposed by the research community. Although the AC formulations exist, they do not work with problems with a large number of candidate transmission paths, different planning scenarios due to convergence issues or infeasible solutions. Also, it has been difficult for the power system community to be aware of the real advantages and disadvantages of the existing approaches due to the lack of rigorous and fair comparisons among them. In this research work, a full non-convex AC formulation to solve the TNEP problem is proposed. It considers in an integrated fashion reactive power expansion, the (N-1) contingency criterion and operational costs. The formulation is solved by an improved non-convex optimization algorithm in a two-stage approach. Also, a fair and rigorous quantitative and qualitative comparison among the proposed approach and other state-of-art metaheuristics and mathematical programming methods has been performed. Simulation results show that the proposed formulation and solution method are superior to other approaches with respect to reliability and suitability for cases with large search spaces and different scenarios. Results are shown for four test systems, namely the Garver 6-bus system, the IEEE 24-bus system, the IEEE 118-bus system, and a modified version of the IEEE 300-bus system.