| Summary: | Renewable energy is beginning to play a major role in the production of clean and inexhaustible energy to supply electricity demand and to hedge against the price volatility of natural gas and oil. However, renewables are expected to increase existing net-demand variability and unpredictability. This will mean that in order to maintain the balance between demand and supply, flexible generation or demand will have to modify their production or consumption at higher rates and frequencies for larger shares of renewables in order to accommodate such extensive and rapid changes as effectively and economically as possible. This thesis addresses the relevance of generation and demand flexibility for the future expansion of power systems that expect large contributions from renewable resources and high levels of vehicle electrification. A novel fully integrated large-scale mixed-integer linear generation expansion planning model was developed, which incorporates detailed modelling of generation and demand (specifically electric vehicles) flexibility characteristics. Computational tractability and efficiency of the model are achieved by clustering generation and flexible demand resources, which allows using integer instead of binary decision variables. The use of integer variables allows reducing the model size in terms of both decision variables and constraints, and also avoids non-linearities in the model formulation. Case studies on conventional generation flexibility show that total system costs are underestimated by up to 24% with a traditional generation expansion planning model when compared to the results obtained with the proposed model. In addition, the optimal generation mix calculated by the traditional model is not only infeasible in terms of security, but also inefficient for absorbing available renewable energy and much more expensive of operating. The case studies also show that reductions in the minimum stable generation, and improvements in ramping capability, reduce the curtailment of renewable energy by up to 73%, as well as the total system costs by up to 20%. The case studies on EV flexibility and its impact on generation expansion planning, on the other hand, show that if the flexibility potential of flexible EV is not utilized, the installed capacity can increase by up to 50%, the total system costs can rise in by to 18%, the level of renewables curtailment can become up to 69x bigger, and the average energy prices can climb by up to 18%, with respect to the case in which EV flexibility is fully utilized. Finally, the developed model is able to produce useful indicative energy planning results that can help regulators, system planners and analysts to design and assess the proper market conditions, energy policies, and incentives required to deliver secure, affordable, sustainable and less polluting power systems in the future.
|
|---|
