State-based potential game approach for distributed economic dispatch problem in smart grid

A modified state-based potential game approach is proposed for the distributed economic dispatch problem in smart grid. Its basic idea is to formulate the objective function of each agent in the designed game as a local estimated Lagrange-like function, thus the stationary state Nash equilibrium incorporates exactly the global optimum of the original centralized economic dispatch problem. This approach can handle complicated equality and inequality coupled constraints, guarantee the asymptotic and quick convergence to the global optimum, entails low communication and information among agents, and possesses adequate robustness to the environment uncertainties. Simulations performed on the IEEE 9 bus system demonstrate the performance of the proposed approach.