Accurate wind speed prediction is extremely critical to the stable operation of power systems. To enhance the prediction accuracy, we propose a new approach that integrates bidirectional long short-term memory (BiLSTM) with fully adaptive noise ensemble empirical modal decomposition (CEEMDAN), the RIME optimization algorithm (RIME), and a multi-head self-attention mechanism (MHSA). First, the historical data of wind farms are decomposed via CEEMDAN to extract the change patterns and features on different time scales, and different subsequences are obtained. Then, the parameters of the BiLSTM model are optimized using the frost ice optimization algorithm, and each subsequence is input into the neural network model containing the MHSA for prediction. Finally, the predicted values of each component are weighted and reconstructed to obtain the predicted values of wind speed time series. According to the experimental results, the method can predict the short-term wind speeds of wind farms more accurately. We verified the effectiveness of the method by comparing it with different models.
The flywheel energy storage (FES) array system plays an important role in smoothing the power output of wind farms. Therefore, how to allocate the total charging and discharging power of wind farms to individual FES system (FESS) units has long been a research hotspot. We propose a hierarchical coordinated control strategy applying the concept of the state of energy (SOE) of FESS units. Energy allocation is achieved by investigating the consensus algorithm of the SOE for both an undirected graph and unbalanced directed graph. Then, we propose another approach to convert the adjacency matrix into a column stochastic matrix using the idea of the PageRank algorithm for an unbalanced directed graph and achieve an SOE consensus of FESS units using the power iteration method. Simulation results demonstrate that the proposed algorithm is effective.
1 July 2023 in Journal of Renewable and Sustainable Energy
Due to the volatility and intermittency of renewable energy, injecting large amounts of renewable energy into the grid will have a tremendous impact on the stability and security of the network. In this paper, we propose the hierarchical energy optimization of flywheel energy storage array system (FESAS) applied to smooth the power output of wind farms to realize source-grid-storage intelligent dispatching. The energy dispatching problem of the FESAS is described as a Markov decision process by the actor-critic (AC) algorithm. In order to solve the problems of stability and low sampling efficiency of the AC algorithm, the soft actor-critic (SAC) algorithm, a deep reinforcement learning (DRL) algorithm based on the model-free off-policy method of the maximum entropy framework, is adopted. Furthermore, SAC and prioritized experience replay (PER) are utilized to greatly improve learning efficiency and sample utilization. The experimental results show that SAC-PER has better performance and stability in energy optimization of the FESAS.
Owing to the volatility and intermittency of renewable energy generation units in microgrids, effective energy scheduling methods are essential for efficient renewable energy utilization and stable microgrid operation. In recent years, microgrid energy optimization scheduling based on deep reinforcement learning (DRL) has made significant progress. With the development of the microgrid, the drawbacks of the traditional DRL agent, such as long training time and poor convergence effect, are gradually revealed. This paper proposes a generative adversarial imitation learning method with Wasserstein distance for optimal energy scheduling in the microgrid. This method combines a proximal policy optimization algorithm to optimize energy scheduling and reduce microgrid operating costs. First, the agent adaptively learns the action exploration process by imitating expert trajectories. Second, based on the generative adversarial theory, a discriminator network is added, and the Wasserstein distance is introduced into the discriminator network to distinguish between the generative and expert strategies. This feedback assists in updating the neural network parameters. Finally, the effectiveness of the proposed method is verified through an arithmetic example analysis.