Self-power generation scheduling model under uncertainty in Energy Intensive Enterprises

Scheduling of the self-generation plants is very helpful to both the energy saving and emission reduction for Energy Intensive Enterprises. Uncertainty exists, so how to reduce the cost considering the uncertainty is an important problem to be studied. In this paper, the objective function is to minimize the overall cost of electricity and gas, and two different models are presented: the deterministic model and the stochastic model. Uncertainties of electricity load and gas supply are captured and analyzed by the scenario tree method. Numerical experiments are performed under a system with four units. Results show that during off-peak periods, most of the enterprises load is afforded by the grid; and that during peak periods, most of load is afforded by the self-generation plants. In addition, the results also indicate that the cost is highly dependent on forecast errors. The more accurate prediction is, the less cost the enterprises need to pay.