Search

The problem of optimal dispatching of a power system containing a high proportion of renewable energy is of great significance for the realisation of new energy consumption and the economic and reliable operation of the power system. For the solution of non‐linear, non‐convex, multi‐objective problems for the optimal operation design of a power system with wind and photovoltaic access, traditional methods have difficulties in terms of computational real‐time and iterative convergence. To address this issue, a deep reinforcement learning‐based optimal scheduling method for the hybrid power system is proposed, which enables continuous action control to obtain an optimal scheduling strategy through the interaction between the agent and the hybrid power system. Firstly, a mathematical description of the optimal scheduling problem containing wind power and photovoltaic power system is presented, and the state space, action space, and reward function of the agent are designed. Secondly, the basic framework of the deep reinforcement learning optimal scheduling model is constructed, and the basic principles of the twin delayed deep deterministic policy gradient algorithm are introduced. Finally, the effectiveness of the deep reinforcement learning model for day‐ahead optimal scheduling of the hybrid power system is verified by means of an arithmetic analysis of the modified New England 39‐bus system. [ABSTRACT FROM AUTHOR]

With the incorporation of renewable energy, load frequency control (LFC) becomes more challenging due to uncertain power generation and changeable load demands. The electric vehicle (EV) has been a popular transportation and can also provide flexible options to play a role in frequency regulation. In this paper, a novel adaptive composite controller is designed to solve the LFC problem for the interconnected power system with electric vehicles and wind turbine. EVs are used as regulation resources to effectively compensate the power mismatch. First, the sliding mode controller is developed to reduce the random influences caused by the wind turbine generation system. Second, an auxiliary controller with reinforcement learning is proposed to produce adaptive control signals, which will be attached to the primary proportion‐integration‐differentiation control signal in a real‐time manner. Finally, by considering random wind power, load disturbances and output constraints, the proposed scheme is verified on a two‐area power system under four different cases. Simulation results demonstrate that the proposed adaptive composite frequency control scheme has a competitive performance with regard to dynamic performance. [ABSTRACT FROM AUTHOR]