Modeling of Suppliers' Learning Behaviors in an Electricity Market Environment

Energy Management System for Smart Grid Consumers with Advanced Usage Information

“…The pricing elasticity can be utilized to achieve the load shaving when the generation capacity reaches the limits. Markov Decision Process [21] and game theories [22] can be employed for the optimization of the expected value of the profit. Mote Carlo simulation is an effective tool which can be used to validate and verify the results derived from analytical methods.…”

Agent-based modeling of the demand-side system reserve provision

“…the demand-side) on the other, is appropriate for ABM approach because all players seek to adjust their behavior to market outcomes in order to improve their profits and benefits. However, most of the agent-based models were developed to investigate behavior of GenCos [20][21][22][23][24] although few also included analysis of the demand-side response [25][26][27].…”

“…In most cases it is used to evaluate how GenCos can raise their profits in the process of producing the electricity and offering it on the electricity market [21,24,[28][29][30][31].…”

“…While various learning algorithms were applied to in different ABM electricity market simulators, a number of them argued for the application of the Q-learning [21,26,32]. In addition, several modifications of the Q-learning algorithm are being investigated, such as balancing between exploration and exploitation of an individual agent to reach the optimal policies for individual states, using the Metropolis criterion and the SA-Q-learning algorithm [29,30].…”

Evaluation of wholesale electric power market rules and financial risk management by agent-based simulations

“…Yu et al [24] modeled generation companies as Q-Learning agents. The results demonstrated that Q-Learning facilitates the GenCo agent exploiting the market in the absence of a MPM process.…”