Monte Carlo computation of power generation production cost under unit commitment constraints

A highly efficient Monte Carlo procedure for estimating the mean and variance of electric power production cost under unit commitment constraints is proposed. Such estimates are useful in making near-term operational decisions. The authors show that, for this purpose, it is essential to consider the stochastic processes associated with generation unit outages. When unit commitment constraints are taken into account in the production-costing model, a combined combinatorial and continuous optimization problem needs to be solved at each hour for each Monte Carlo run. This tends to make the computations quite long. They propose a method to reduce the required number of Monte Carlo runs by using a control variate technique on batches. They present the results for a small electric power generating system where they achieved a reduction of sample size by a factor of 10 for the variance and by a factor of 42 for the mean