Multi-objective transmission expansion planning based on Pareto dominance and neural networks

This paper presents an algorithm to solve the multi-objective transmission expansion planning (TEP) problem including the investment and reliability criteria. The reliability is considered by using the Expected energy not supplied (EENS) index. The main contribution consists on handling the reliability criterion in the optimization process, which tends to provide solutions with better trade-off between the mentioned criteria. For that purpose, a novel probabilistic algorithm called non-dominated Monte Carlo simulation (ND-MCS) is proposed to allow solving the multi-objective TEP problem with suitable computational effort and efficacy even considering the probabilistic feature of reliability in the optimization. In addition, a Support Vector Machine (SVM) network is applied embedded within the ND-MCS. The proposed methodology integrates the Pareto dominance method as a convergence criterion to MCS and a fuzzy criterion to support the decision making. The effectiveness of the proposed approach is tested in three systems, including a practical Brazilian network.