Go top
Ph. D. Thesis information

Decision support methods for large-scale flexible transmission expansion planning

Sara Lumbreras Sancho

Supervised by Andrés Ramos Galán

The long permitting processes involved in transmission expansion mean that transmission decisions must anticipate the evolution of the system, particularly in what respects to new generation developments. Consequently, it is necessary to develop methods to propose robust, flexible plans that are able to adapt to the future as it unfolds. Moreover, the integration of markets results in planning increasingly large areas. Then, it is no longer possible to rely on the experience of a TSO (Transmission System Operator) to understand the structure of the network and provide functions such as the proposal of potentially interesting candidate transmission lines. This adds to the fact that Transmission Expansion Planning (TEP) is already a complex, computationally demanding problem. - This thesis develops formulation and algorithmic enhancements to make accessible the large problem sizes currently involved in TEP studies. Working within a Benders’ decomposition framework, we propose Semi-Relaxed cuts to deal with large numbers of discrete decision variables. In addition, we develop a Progressive Contingency Incorporation algorithm to evaluate the effect of component failures efficiently. - We develop methods to propose robust, flexible plans. We apply Real Options Valuation to identify the key high-potential transmission lines and analyze their main value drivers. In addition, we propose a regularization approach for Stochastic Optimization which produces solutions that are able to perform relatively well despite modifications in the definition of scenarios. - We propose automatic mechanisms for candidate discovery. In addition, a candidate analysis method disentangles the synergistic relationships among transmission lines. Finally, we combine Ordinal Optimization, a metaheuristic that identifies relatively good solutions efficiently, with Mixed Integer Programming (MIP). The proposed technique extracts valuable information about the structure of the solution, which is then exploited to reduce computation time.

Universidad Pontificia Comillas. Madrid (España)

09 June 2014

Request Request thesis to the author

Aviso legal  |  Política de cookies |  Política de Privacidad

© Universidad Pontificia Comillas, Escuela Técnica Superior de Ingeniería - ICAI, Instituto de Investigación Tecnológica

Calle de Santa Cruz de Marcenado, 26 - 28015 Madrid, España - Tel: (+34) 91 5422 800