Negotiation context analysis in electricity markets. (1st June 2015)

• Record Type: Journal Article
• Authors: Pinto, Tiago; Vale, Zita; Sousa, Tiago M.; Praça, Isabel
• Is Part Of: Energy. Volume 85(2015)
• Page Start: 78
• Page End: 93
• Publication Date: 2015-06-01
• Subjects: Adaptive learning -- Artificial intelligence -- Context awareness -- Electricity markets -- Multiagent simulation; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2015.03.017
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Contextualization is critical in every decision making process. Adequate responses to problems depend not only on the variables with direct influence on the outcomes, but also on a correct contextualization of the problem regarding the surrounding environment. Electricity markets are dynamic environments with increasing complexity, potentiated by the last decades' restructuring process. Dealing with the growing complexity and competitiveness in this sector brought the need for using decision support tools. A solid example is MASCEM (Multi-Agent Simulator of Competitive Electricity Markets), whose players' decisions are supported by another multiagent system – ALBidS (Adaptive Learning strategic Bidding System). ALBidS uses artificial intelligence techniques to endow market players with adaptive learning capabilities that allow them to achieve the best possible results in market negotiations. This paper studies the influence of context awareness in the decision making process of agents acting in electricity markets. A context analysis mechanism is proposed, considering important characteristics of each negotiation period, so that negotiating agents can adapt their acting strategies to different contexts. The main conclusion is that context-dependant responses improve the decision making process. Suiting actions to different contexts allows adapting the behaviour of negotiating entities to different circumstances, resulting in profitable outcomes. Highlights: Context analysis methodology considering factors that influence the market price. Integration of context analysis in ALBidS (adaptive learning strategic bidding system) to provide adaptation of agents' actions. Simulation of MIBEL, EPEX and Nord Pool electricity markets based on real data. Testing the performance of context analysis in players' decision making process. Context-dependant adaptive responses result in profitable outcomes.