A Study On Demand Response And Energy Dispatch Strategy For Smart Grids Based On Real-time Price

Smart grid is a large-scale cyber-physical system, where the market participants do not have the global information of the market. Therefore, implementing distributed market management is a key problem to be solved. This thesis studies distributed demand response and energy dispatch strategy using convex optimization and game theory for large commercial consumers, industrial consumers, and small residential consumers. The main contributions are as below.Firstly, the distributed load control algorithm and the real-time pricing strategy are developed for large commercial consumers and industrial consumers. The adjustment algorithms of the pricing parameters are designed for the pricing strategy that is to match supply with demand. The global asymptotic stability is proved based on the Lyapunov stability theory, and the distributed implementation of the algorithms is given.Secondly, the energy consumption control and real-time pricing algorithm is proposed for small residential consumers, and the global asymptotic stability of the algorithm is proved. Further, the input-to-state stability of the algorithm under disturbances is proved based on the nonlinear control theory.Thirdly, the joint demand response and generation dispatch strategy is proposed based on the hierarchical game theory for large commercial consumers and industrial consumers. The demand response strategy is obtained from the Nash equilibrium of a noncooperative game, and the generation dispatch strategy is obtained by the backward induction method. The energy consumption control algorithm and generation dispatch algorithm are developed based on the subgradient method, and the convergence is proved.Finally, the joint demand response, energy purchase, and generation dispatch strategy is proposed based on the Stackelberg game theory for small residential consumers,and the demand response strategies of the consumers, the energy purchase strategies of the retailers, and the generation dispatch strategies of the generators are obtained based on the duality principle and the KKT condition.The main findings in this thesis are as follows. The distributed load control and realtime pricing algorithm shows rapid convergence speed, matches supply with demand, and shifts the peak load. The joint demand response and generation dispatch strategy can also match supply with demand and shift the peak load. Comparing with the fixed pricing strategy, the real-time pricing strategy can e?ectively reduce the peak-to-average ratio, the daily average price, and the daily load in electricity markets. The joint demand response,energy purchase, and generation dispatch strategy can guarantee the retailers to obtain profits from the two-stage electricity trading.