Economic management and planning based on a probabilistic model in a multi-energy market in the presence of renewable energy sources with a demand-side management program

Abstract

A key issue in the optimal operation of power systems is the economically efficient use of microgrids while considering demand-side management. Implementation of demand-side management programs reduces the cost of power system operation. It also requires financial incentive policies. Therefore, in this paper, the probabilistic modeling of energy for large-scale consumers in the presence of different energy sources such as energy storage systems, renewable energy sources and a microturbine relying on power exchange-based bilateral contracts is performed while considering a demand-side management program. Since renewable energy sources such as wind and solar resources have uncertainties, an autoregressive moving-average based scenario generation has been applied to model their behavior. To reduce the cost of purchasing the required energy, storage and demand-side management systems will directly aid big industries. The market price uncertainty model, load and output power of renewable energy sources are also included in the problem formulation. Market price, load, temperature and radiation forecast error of photovoltaic systems is modeled using a normal distribution to generate the scenarios. The Weibull distribution is used to generate variable wind speed scenarios for the wind power output uncertainty model. In uncertain situation of decision making, the decision maker has to evaluate the optimal decisions during a decision horizon by the uncertainty environment. Optimal energy management in microgrids is usually formulated as a nonlinear optimization problem. Due to the nonlinear and discrete nature of the problem, solving it in a centralized manner requires a large volume of computation in the central microgrid controller. To solve it, therefore, a new seagull-based algorithm is proposed. In the developed model, it is combined with the genetic algorithm to strengthen its local and global search capability because the genetic algorithm has proper perfor-mance in binary search due to cross-over and feature selection operators. Finally, the effect of energy storage systems and demand response program on suggested microgrids are examined, and four test cases are considered to prove the capability of the suggested stochastic energy procurement problem. Obtained numerical analysis prove the efficiency of the suggested stochastic program.