Bidding Strategy in pay as bid markets by Multi- Agent Reinforcement Learning

Publish Year: 1392
نوع سند: مقاله کنفرانسی
زبان: English
View: 1,076
  • Certificate
  • من نویسنده این مقاله هستم

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این Paper:

شناسه ملی سند علمی:

PSC28_238

تاریخ نمایه سازی: 25 اردیبهشت 1393

Abstract:

In a deregulated electricity market, market participants use optimal bidding strategies in order to maximize their individual profits, due to the uncertainty and dynamic of electricity market, players’ optimal bidding strategy is not determined easily. Power markets have two objectives. The first which is the Maximization of their profit and the second is their utilization rate. It is essential for players to identify complex behavior by learning through a continuous exploiting and exploring process. Reinforcement learning is a good way to make a decision in these incomplete information markets. In this paper agent-based simulation is employed to study the incomplete power market operation under pay-as-bid pricing market. The market is considered SFE model that players proposed both bids and powers in their suppliers’ functions. Power suppliers are modeled as adaptive agents capable of learning through the interaction with their environment, following a Reinforcement Learning algorithm. The SA-Q-learning algorithm, a slightly changed version of the popular Q-Learning, is used in this Paper. In this paper new state-action definition are proposed. The results of proposed state-action method on five bus power systemare compared with different state-action definitions and the superiority of this definition in two different cases is shown

Authors

Mohammad Amin Tajeddini

Department of Electrical and Computer Engineering University of Tehran, Tehran, Iran

Mahdi Imani

Department of Electrical and Computer Engineering University of Tehran, Tehran, Iran

Hamed Kebriaei

Department of Electrical and Computer Engineering University of Tehran, Tehran, Iran