Risk and unit commitment decisions in scenarios of wind power uncertainty


A challenge to the stochastic programming paradigm in systems with high wind power.

A decision framework matching risk against a stochastic evaluation of costs.

A stochastic modeling of risk criteria related to the aversion to worst events.

Pareto front cost vs. risk is not convex, precluding simplistic trade-off analysis.

Decisions based only on a stochastic paradigm may put the system at unacceptable risk.


This paper addresses the problem of decision making in Unit Commitment in systems with a significant penetration of wind power. Traditional approaches to Unit Commitment are inadequate to fully deal with the uncertainties associated to wind, represented by scenarios of forecasted wind power qualified by probabilities. Departing from a critique of planning paradigms, the paper argues that a stochastic programming approach, while a step in the good direction, is insufficient to model all aspects of the decision process and therefore proposes the adoption of models based on a Risk Analysis paradigm. A case study is worked out reinforcing this perspective. In a multi-objective context, the properties of the cost vs. risk Pareto-optimal fronts are analyzed, where risk may be represented by aversion to a worst scenario or a worst event. It is shown that the Pareto-optimal front may not be convex, which precludes a simplistic use of tradeoff concepts. It is also shown that decisions based on stochastic programming may in fact put the system at risk. An evaluation of risk levels and cost of hedging against undesired events is proposed as the paradigm to be followed in Unit Commitment decision making.


  • Wind power;
  • Stochastic programming;
  • Risk analysis;
  • Unit commitment

Be the first to comment

Leave a Reply

Your email address will not be published.