ABSTRACT
Dynamic pricing is one of the most important management policies to help stabilize imbalanced energy in a smart grid network. The power demands of each microgrid will interact with each other through the price policy, which is greatly influenced by the mean (collective) behavior and random fluctuation of the smart grid network. Therefore, a novel system of a mean field stochastic smart grid network is modeled to describe the interaction with each microgrid by mean field stochastic theory. For the market operation of a smart grid system, managers of the smart grid network expect a dynamic pricing policy to achieve their desired stored energy working level not only with parsimonious price but also by attenuating the effect of external disturbance due to unpredictable intermittent renewable energy sources (RESs). In this chapter, a multi-objective optimal H2 /H∞ dynamic pricing policy is proposed to achieve H2 optimal desired reference energy storage tracking and H∞ optimal attenuation of the effect of external disturbance by integration-based state feedback control in a mean field stochastic smart grid network system. Since it is difficult to solve the multi-objective optimization problem directly, an indirect method is proposed to transform the MO H2 /H∞ dynamic pricing policy problem in the mean field stochastic smart grid network system into a linear matrix inequality–constrained MOP. The LMI-constrained MOP is still not easily solved directly by the conventional multi-objective evolution algorithm. Therefore, a reverse-order LMI-constrained MOEA is proposed to solve the LMI-constrained MOP of the dynamic pricing policy in the management of the mean field stochastic smart grid network efficiently. A simulation example of the mean field stochastic smart grid network system is also provided to illustrate and validate the proposed multi-objective H2 /H∞ dynamic pricing management.
