A Markov model for evaluation of electric power generation system production costs.

摘要

The cost of producing electric power is an important economic measure to both utilities and regulators. Power companies use cost estimates in capacity planning, fuel management, and operational planning. Regulatory agencies use them in setting the rates charged to the customers. To estimate the cost, power companies routinely run production costing software packages based on probabilistic models that account for load variations over time and the uncertainty in the utilization of the generating units due to failures and repairs.; This research uses two different approaches to evaluate the production cost Monte Carlo simulation and an analytical approach using a Markov chain model for the available capacities of the generating units and the system load. The Monte Carlo simulation method accounts for the up and down processes of the generating units to provide better estimate for the mean and the variance of production cost as well as marginal cost. Sensitivity analyses based on the simulation method have been used to provide insight on how the statistics of production cost are affected by the change of system parameters.; In the analytical approach, the asymptotic mean of production and marginal costs is calculated based on the steady-state probabilities of the Markov chain. The "fundamental matrix" underlying the Markov chain is used to calculate the asymptotic variance of the production and marginal costs. A recursive scheme is developed so that these formulations can be applied to a real life power system. Two different Markov chain load models have also been developed to represent the stochastic loads. These models are suitable for use in the Fundamental Matrix Method (FMM). Numerical results show that when the load is periodic, the FMM provides accurate estimate of the variance of production and marginal costs for long time horizons with less computation time as compared to the Monte Carlo method.