Abstract
This paper proposes an adaptation of the Q-Learning reinforcement learning algorithm, for the identification of the most probable scenario that a player may face, under different contexts, when negotiating bilateral contracts. For that purpose, the proposed methodology is integrated in a Decision Support System that is capable to generate several different scenarios for each negotiation context. With this complement, the tool can also identify the most probable scenario for the identified negotiation context. A realistic case study is conducted, based on real contracts data, which confirms the learning capabilities of the proposed methodology. It is possible to identify the most probable scenario for each context over the learned period. Nonetheless, the identified scenario might not always be the real negotiation scenario, given the variable nature of such negotiations. However, this work greatly reduces the frequency of such unexpected scenarios, contributing to a greater success of the supported player over time. © 2019, Springer Nature Switzerland AG.

Abstract
This research proposes a novel methodology for adaptive learning in electricity markets negotiations, based on the principles of the determinism theory. The determinism theory states that all events are predetermined due to the cause-effect rule. At the same time, it is unmanageable to consider all causes to a certain effect, making it impossible to predict future events. However, in a controlled simulation environment, it is possible to access and analyze all involved variables; thus, making the application of this theory promising in such environments. This research applies the principles of the determinism theory to a new learning methodology, which optimizes players' actions, considering the predicted behavior of all involved players, with the objective of maximizing market gains. A case-based reasoning approach is used, providing adaptive context-aware decision support. Results show that the proposed approach is able to achieve better market results than all reference market strategies.

Abstract

Abstract
Energy consumption forecasting is crucial in current and future power and energy systems. With the increasing penetration of renewable energy sources, with high associated uncertainty due to the dependence on natural conditions (such as wind speed or solar intensity), the need to balance the fluctuation of generation with the flexibility from the consumer side increases considerably. In this way, significant work has been done on the development of energy consumption forecasting methods, able to deal with different forecasting circumstances, e.g., the prediction time horizon, the available data, the frequency of data, or even the quality of data measurements. The main conclusion is that different methods are more suitable for different prediction circumstances, and no method can outperform all others in all situations (no-free-lunch theorem). This paper proposes a novel application, developed in the scope of the SIMOCE project (ANI vertical bar P2020 17690), which brings together several of the most relevant forecasting methods in this domain, namely artificial neural networks, support vector machines, and several methods based on fuzzy rule-based systems, with the objective of providing decision support for energy consumption forecasting, regardless of the prediction conditions. For this, the application also includes several data management strategies that enable training of the forecasting methods depending on the available data. Results show that by this application, users are endowed with the means to automatically refine and train different forecasting methods for energy consumption prediction. These methods show different performance levels depending on the prediction conditions, hence, using the proposed approach, users always have access to the most adequate methods in each situation.

Abstract
Futures contracts are a valuable market option for electricity negotiating players, as they enable reducing the risk associated to the day-ahead market volatility. The price defined in these contracts is, however, itself subject to a degree of uncertainty; thereby turning price forecasting models into attractive assets for the involved players. This paper proposes a model for futures contracts price forecasting, using artificial neural networks. The proposed model is based on the results of a data analysis using the spearman rank correlation coefficient. From this analysis, the most relevant variables to be considered in the training process are identified. Results show that the proposed model for monthly average electricity price forecast is able to achieve very low forecasting errors.

Abstract
Electricity markets are complex and dynamic environments, mostly due to the large scale integration of renewable energy sources in the system. Negotiation in these markets is a significant challenge, especially when considering negotiations at the local level (e.g., between buildings and distributed energy resources). It is essential for a negotiator to be able to identify the negotiation profile of the players with whom he is negotiating. If a negotiator knows these profiles, it is possible to adapt the negotiation strategy and get better results in a negotiation. In order to identify and define such negotiation profiles, a clustering process is proposed in this paper. The clustering process is performed using the kml-k-means algorithm, in which several negotiation approaches are evaluated in order to identify and define players' negotiation profiles. A case study is presented, using as input data, information from proposals made during a set of negotiations. Results show that the proposed approach is able to identify players' negotiation profiles used in bilateral negotiations in electricity markets.