3 Presentations

• 
  10:30 AM - 10:55 AM

  Risk Management in Energy Storage Planning

  • Somayeh Moazeni, presenter, Princeton University
  • Warren Powell, Princeton University
  • Belgacem Bouzaiene-Ayari, Princeton University

  An optimal policy for a finite planning horizon, among some energy source, energy carrier electricity, and a battery storage is sought to satisfy the electricity load and minimize the expected cost (and risk) of the system. We present a stochastic optimization framework for the problem, and develop a novel tractable parallel computational stochastic dynamic programming technique based on the direct policy search method and a parallel multi-start pattern search derivative free optimization. Impacts of the underlying non-stationary stochastic processes of the electricity price and load on the minimum expected cost energy flow structure, its associated expected cost and Value-at-Risk are then extensively discussed.

• 
  10:55 AM - 11:20 AM

  Decision-Support in the Hydro Energy Market: A Stochastic Approach with Risk Aversion Measures

  • Michel Gendreau, presenter, Polytechnique Montréal
  • Raphael E.C. Gonçalves, Polytechnique Montréal

  We present a stochastic approach for decision-support in the hydro energy market. For that, we consider risk aversion metrics based on the Conditional Value-at-Risk (CVaR), which can provide more coherent decisions and allow the decision-maker to add its risk perception into the model formulation. We present a case study based on an academic example of the Hydro-Québec (HQ) trading market environment.

• 
  11:20 AM - 11:45 AM

  ANNULÉE/CANCELLED: SMART-ISO: Modeling Uncertainty in Renewable Sources of Energy

  • Marcos Leone-Filho, presenter, Princeton University
  • Warren Powell, Princeton University
  • Hugo Simao, Princeton University

  We present a stochastic modeling framework that allows for the analysis of the integration of high levels of renewable sources of energy (e.g. wind and solar) into a large transmission system. It is comprised of a multi-scale optimization and simulation system, including a day-ahead and an hour-ahead unit commitment models and a real-time simulation model. We will illustrate the application of the system through some preliminary results of an off-shore wind integration study.