Group:Network Science Group
Title: Optimal Wind Power Integration in Microgrid: A Dynamic Game Approach, Path-Loss
Speaker: Chenye Wu, Zhaoquan Gu University
Time: 2011-11-18 12:30-2011-11-18 14:00
Venue: FIT 1-222

Abstract:

Chenye Wu:

Title: Optimal Wind Power Integration in Microgrid: A Dynamic Game Approach

Abstract:

Renewable energy resources, such as wind and solar power, are rapidly becoming major generation technologies in the United States and around the world. However, large scale integration of renewable is still very challenging due to their intermittency and fluctuations. With the growing interest in Smart Grid and Smart Metering Systems, one promising option to tackle these challenges is to design demand side management (DSM) algorithms to shape the load to follow the amount of renewable energy generated, such as the output of wind turbine which is the focus in this paper. Note that such coordination of load with wind power generation requires users' cooperation and their participation in DSM programs. Hence, in this paper, we propose two DSM mechanisms using Dynamic Game Theory to encourage user participation in the integration of wind power. The first approach is to formulate a potential game for efficient cost sharing among users. The second approach is to introduce a smart pricing model that aligns users behavior with the needs of the microgrid using a simple and myopic energy consumption scheduling strategy. We analyze the two designed dynamic games over a long period of time and investigate the efficiency of the constructed game model at equilibrium. Finally, the system performance is assessed using computer simulation. Simulation results show that the simple myopic strategy can reduce generation cost by 35% compared to the case where no demand side management is performed. The cost allocation approach can further reduce the generation cost by 21%.

Zhaoquan Gu:

Path-Loss Fluctuations: Towards Robust Scheduling Algorithms in the SINR Model

Abstract:

The SINR model has attracted much attention recently in the field of wireless networks. The path loss exponent in this model, usually denoted by \alpha, is generally treated as constant that lies between two and six. However, in real scenarios, the exact value of \alpha is hard to find; in addition the attenuation of signal powers transmitted through different areas vary, which may cause the value of \alpha to ebb and flow among all wireless requests.

We initiate the study about the impact of the \alpha that fluctuates on the physical (SINR) model. We prove that for any given \alpha and its fluctuation \delta, i.e. a path loss exponent that may vary between \alpha-\delta and \alpha+\delta, a specific topology can always be constructed which is extremely vulnerable to the small change of \alpha. We call algorithms that still perform well despite a fluctuating \alpha \alpha-Robust'' algorithms and give an idea how to design \alpha-Robust algorithm for the MIS problem.