Abstract
The price evolution in electricity market with large share of renewables often exhibits a deep volatility, triggered by external factors such as wind and water availability, load level and also by business strategies of market agents. Consequently, in many real applications, the performance of electricity price is not appropriate. The goal of this article is to analyze the available market data and characterize circumstances that affect the evolution of prices, in order to allow the identification of states that promote price instability and to confirm that class segmentation allows increasing forecast performance. A regression technique (based on Artificial Neural Networks) was applied first to the whole set and then to each class individually. Performances results showed a clear advantage (above 20%) of the second approach when compared to the first one.

Abstract

Abstract
In the last decades, the weight of renewable energies sources (RES-E) in the electricity generation mix of most European countries has considerably increased, constituting an important contribution to the transition towards a low-carbon economy. Until very recently, RES-E were supported by favorable investment mechanisms specially designed to endorse investment in RES-E. More recently, as RES-E are becoming increasingly more competitive (especially wind and solar photovoltaic), RES-E are starting to be remunerated according to market mechanisms. This has generated a lively debate on the economic pros and cons of dispatching RES-E in the market. This paper contributes to this debate by developing a game theoretical model in the context of which we analyze how the inclusion of RES-E in the electricity wholesale market affects equilibrium outcomes under demand and supply uncertainty. Then, we examine how the inclusion of RES-E in the electricity wholesale market impacts firms' incentives to invest in conventional energy sources, characterizing the optimal investment under demand and supply uncertainty. We find that, when RES-E capacity and asymmetry in firms' marginal production costs are sufficiently high, RES-E producers may strategically reduce the market price, in order to evict the less efficient conventional source in that period. Although, in the short-run, this strategy may actually favor energy consumers (since prices are lower), the expectations of inactivity periods (regardless of whether they arise for strategic or market reasons) may negatively affect investment in back-up capacity, possibly leading to an increase in future prices (since less back-up capacity is available). Finally, we provide an analytical characterization of optimal investment levels in conventional energy sources under demand and supply uncertainty.

Abstract

Abstract
The penetration of distributed renewable energy sources in the distribution grid is increasing considerably in the last years. This is one of the main causes that contributed to the growth of technical problems in both transmission and distribution systems. An effective solution to improve system security is to exploit the flexibility that can be provided by distributed energy resources (DER), which are mostly located at the distribution grids. Their location combined with the lack of power flow coordination at the system operators interface creates difficulties in taking advantage of these flexible resources. This paper presents a methodology based on the solution of a set of optimization problems that estimate the flexibility ranges at the distribution and transmission system operators (TSO-DSO) boundary nodes. The estimation is performed while considering the grid technical constraints and a maximum cost that the user is willing to pay. The novelty behind this approach comes from the development of flexibility cost maps, which allow the visualization of the impact of DER flexibility on the operating point at the TSO-DSO interface. The results are compared with a sampling method and suggest that a higher accuracy in the TSO-DSO information exchange process can be achieved through this approach.

Abstract
Several important real world problems of predictive analytics involve handling different costs of the predictions of the learned models. The research community has developed multiple techniques to deal with these tasks. The utility-based learning framework is a generalization of cost-sensitive tasks that takes into account both costs of errors and benefits of accurate predictions. This framework has important advantages such as allowing to represent more complex settings reflecting the domain knowledge in a more complete and precise way. Most existing work addresses classification tasks with only a few proposals tackling regression problems. In this paper we propose a new method, MetaUtil, for solving utility-based regression problems. The MetaUtil algorithm is versatile allowing the conversion of any out-of-the-box regression algorithm into a utility-based method. We show the advantage of our proposal in a large set of experiments on a diverse set of domains.