Kris Poncelet

Selecting Representative Days for Capturing the Implications of Integrating Intermittent Renewables in Generation Expansion Planning Problems

Due to computational restrictions, energy-system optimization models (ESOMs) and generation expansion planning models (GEPMs) frequently represent intraannual variations in demand and supply by using the data of a limited number of representative historical days. The vast majority of the current approaches to select a representative set of days relies on either simple heuristics or clustering algorithms and comparison of different approaches is restricted to different clustering algorithms. This paper contributes by: i) proposing criteria and metrics for evaluating representativeness, ii) providing a novel optimization-based approach to select a representative set of days, and iii) evaluating and comparing the developed approach to multiple approaches available from the literature. The developed optimization-based approach is shown to achieve more accurate results than the approaches available from the literature. As a consequence, by applying this approach to select a representative set of days, the accuracy of ESOMs/GEPMs can be improved without increasing the computational cost. The main disadvantage is that the approach is computationally costly and requires an implementation effort.

Applicability of a Clustered Unit Commitment Model in Power System Modeling

Clustered unit commitment (CUC) formulations have been proposed to provide accurate and fast approximations to the unit commitment (UC) problem. In these formulations, identical or similar plants are grouped into clusters. This way, the binary commitment variables of all the plants within a cluster can be replaced by a single integer variable. This approach has recently been mainly used for tractably integrating flexibility constraints in generation expansion planning problems. However, a thorough general validation is still missing. In addition, these formulations do not provide commitment schedules on a plant-by-plant level and hence cannot be used directly for operating actual systems or markets. A first contribution of this paper is to show that errors can be introduced both due to the problem formulation and the grouping of nonidentical units. A case study is presented in which these errors are quantified under different conditions. Overall, the error in approximating the total cost does not exceed 0.06%. A second contribution of this paper is the development of a hybrid approach, which sequentially uses a CUC and a traditional UC model. This approach allows to reduce the computational cost of solving the UC problem while providing a guaranteed feasible and near-optimal solution.

Myopic optimization models for simulation of investment decisions in the electric power sector

Generation expansion planning models optimize investment and operational decisions over a time horizon of multiple decades, thereby typically assuming perfect foresight (PF). Recently, myopic optimization models, in which the foresight is restricted to a certain period, have been suggested to more realistically simulate the short-sightedness of investment decision makers. The literature has shown that the modeled level of foresight can have a significant impact on the results obtained. However, the literature does not contain an in-depth analysis of the investment decision making process in myopic optimization models. As a result, the implications of using myopic optimization models to simulate the decision making of private agents in liberalized electricity markets are unclear. This paper provides fundamental methodological insights into the decision making in both PF and myopic optimization models. The projections, at the time the investment decision is made, of the short-run profits that can be obtained by investing in a generation asset are analyzed in this regard. This analysis reveals a major limitation of the decision making process in myopic optimization models, i.e., the approach does not extrapolate trends, in terms of changes in the projected SR profits, expected within the window of foresight to later periods. This leads to decision making which cannot be considered to reflect reality.