Optimal Scheduling In Smart Grids With Prediction Error Probability Limitation

We consider demand response, supply management and power scheduling in a smart grid in the presence of renewable energy sources. We maximize the total social welfare while constraining the probability of renewable forecast error, and solve the ensuing optimization problem to derive the best conventional power production and user consumption schedules. The resulting optimal policies demonstrate their superiority over a competing policy in terms of social welfare, particularly when the constraint is tight.