19 de octubre de 2018
Resumen:
Transmission Expansion Planning (TEP) has the purpose of identifying new equipments to be inserted on a transmission grid to optimize some pre-defined objective function while ensuring the secure and economic supply of the demand forecasted along an extended planning horizon. Its non-linear and non-convex natures turn TEP a challenging problem. In addition, the phenomenon of combinatorial explosion of investment alternatives typically requires a huge computational effort to solve it.
In recent years TEP has been approached with relaxed mathematical models to overcome these challenges, even though there is no guarantee that the optimal solution of the relaxed problem is even feasible if tested in the original real problem. Moreover, in the last years, power systems have been changing towards a more active load pattern, where consumers also produce electricity and now act as ”prosumers”, mainly due to advances related with the dissemination of Distributed Energy Resources (DERs) as Distributed Generation (DG), electric vehicles, energy storage, smart grids, microgrids and demand response programs. To further increase the challenges of TEP problems, the share of renewable energy sources with intermittent, reduced predictability and controllability characteristics is increasing, and the mathematical models should take this issue into account. Besides, the unbundling of the electricity sector in several activities, some of them provided in a regulated way and some others under competition, poses a number of challenging problems namely because in several areas there are conflicting objectives associated to different stakeholders.
In order to overcome these issues, this Thesis compares different mathematical models regarding the accuracy and time required to solve the TEP problem. Besides, new algorithms and techniques are proposed in order to improve the computational performance in solving the problem. In this Thesis it is also evaluated the impact under several scenarios of solar DGs and electric vehicle charging policies in the planning task. The uncertainties in renewable energy, electric demand and equipment availability are also addressed as well as the impact of the most common decision-making processes adopted in the TEP literature and on the total system cost. A new worst-case parameter is proposed in order to ensure that the system is sufficiently robust to overcome conditions with high electricity demand and low renewable energy generation. Finally, a novel efficient method is proposed to handle with the proposed multiobjective optimization TEP formulation required to give a trade-off between the specified objectives.
The results indicate that although relaxed mathematical models can solve the TEP problem faster, they may display unreliable and in some cases even unfeasible solution plans.
Besides, other techniques as reduction of the search space and parallel computing can be allied to real mathematical models in order to get reliable solutions in a timely manner.
The results also indicate that distributed generation can enable the reduction of operation costs, transmission losses and emissions, although it is only possible to reduce investment costs in new equipment if the peak load is also reduced by the DGs. Regarding the impact of electric vehicles, the investments in new equipment could be postponed if it is adopted an efficient management approach to control the charging that includes the “valley-filling effect” and/or the “peak-shaving effect”. Besides, numerical simulations proved that the TEP conducted considering only the peak load to quantify investment requirements is not sufficient to ensure the safe operation of the system in normal conditions for any other off-peak load scenarios. Finally, the results obtained allow concluding that multiobjective approach gives the decision maker a higher flexibility to decide as well as more information about the solutions presented, which in turn are associated to trade-offs between the specified objectives.
Cita:
P. Vilaça (2018), Transmission expansion planning - a dynamic multiobjective approach considering uncertainties. Oporto (Portugal).
