Study Of Real-Time Pricing Algorithm Based On Gradient Projection Method For Smart Grid

Smart grid which is built on the basis of an integrated, high-speed bidirectional communication network with a blend of many advanced technologies, including:advanced sensing and measurement technology, advanced equipment technology, advanced control methods and advanced decision support systems technology and so on, in order to achieve the goals of reliability, security, economic, efficient, environmental friendly and use safety of power grid. Since the advent of the smart grid, relevant research work has made significant development. Traditional power marketing model does not consider the role of demand-side. The power grid is just for the simple grid power transmission. The smart grid is an interactive network of energy supply and demand, which makes real-time pricing to become a possibility. Real-time pricing is the demand response measure based on price, and it is a dynamic pricing mechanism to balance the network load, to ensure the operation of grid securely and stably, to optimize resource allocation and to improve energy efficiency and users’ satisfaction with electricity. Therefore, study of real-time price algorithm will have profound theoretical and practical significance.This paper is based on the development of smart grid to introduce the current construction situation and research status of smart grid, which explores the implementation of real-time pricing in the smart grid and discusses the advantages and disadvantages of real-time pricing algorithm under DSM. In addition, to the utility of the users and electricity supplier as the starting point this paper analysis the performance and the applicable conditions of the real-time pricing algorithm for the smart grid system with a single electricity supplier and depth studies the real-time pricing algorithm based on the gradient projection.Taking into account the aggregate utility with users and electricity supplier, the real-time pricing algorithm based on utility maximum establishes the relevant optimization model, via solving the algorithm, which can obtain the best price and higher aggregate utility. However, this algorithm applies only to the case of small-scale users, and the convergence rate of each time slot is not optimal.In response to these problems, this paper proposes a real-time pricing algorithm based on the gradient projection method, which chooses decreasing risk aversion utility function as user’s utility function and uses the gradient projection method to solve the optimization model. This algorithm can quickly obtain the optimal price to ensure the load balancing of supply and demand, to achieve full utilization of energy, to improve the effectiveness of the user and the electricity supplier and make the grid run more stable, more reliable. And when the user scale increases, the algorithm also is applicable. The simulation results show that compared with the real-time pricing algorithm based on utility maximum, the proposed algorithm has better performance in this paper.