More and more renewable energy sources are integrated into novel power systems. The randomness and fluctuation of such renewable energy sources bring challenges to the static stability and safety analysis of novel power systems. In this work, a multilayer deep deterministic policy gradient is proposed to address the fluctuation of renewable energy sources. The proposed method is stacked with multilayer deep reinforcement learning methods that can be continuously updated online. The proposed multilayer deep deterministic policy gradient is compared with other deep learning algorithms. The feasibility, effectiveness, and superiority of the proposed method are verified by numerical simulations.
Affected by environmental factors, equipment aging, operating status, etc., the parameters of photovoltaic (PV) models will deviate from the original setting parameters. In order to accurately identify the dynamic parameters of photovoltaics under the general simulation model, traditional parameter identification methods mainly use heuristic intelligent optimization algorithms for direct solution. Due to the limited data collected and the strong randomness of the algorithm, it is easy to make the identification accuracy and stability of photovoltaic parameters difficult to meet the requirements. To this end, this paper proposes an optimal identification method for PV dynamic parameters driven by data expansion. Firstly, the PV external characteristic data is fitted and generalized, which used the generalized regression neural network (GRNN). Then, the extended high-quality data can be used for dynamic parameter identification for PV cell. To confirm the performance of the proposed algorithm in this paper, this paper expands based on the actual external characteristic data of different proportions and uses the general PV simulation model to conduct comparative tests on various commonly used algorithms. The case studies under different scenarios show that the proposed algorithm can provide a more reliable and well-represented fitness function to the metaheuristic algorithms. Therefore, the optimization accuracy and stability of the proposed algorithm for dynamic PV cell parameter identification can be significantly improved simultaneously.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.