The virtual synchronous generator (VSG) exhibits high-dimensional complexity, necessitating a tradeoff between accuracy and complexity when constructing small-signal models. To address this issue, this article proposes a data-driven modeling approach based on long short-term memory (LSTM) neural networks. The focus is on mapping relationships between electrical quantities, considering the influence of irrational factors on model accuracy. A detailed data-driven modeling approach for VSG is proposed and verified in this article. Due to the time-series correlation in the electrical data generated during VSG operation, the LSTM algorithm, known for its excellent time-series prediction capabilities, is chosen to construct the VSG data-driven model. Several complex VSG application scenarios are used to validate the effectiveness and accuracy of the proposed modeling approach. In conclusion, the LSTM-based VSG data-driven modeling outperforms small-signal models and recurrent neural network data-driven models in terms of accuracy and stability.

Partially rated dc interlinking converters are recognized for their high-gain power regulation capabilities, which effectively synergize active power across dc microgrids (DCMGs). Integrating energy storage units (ESUs) to address the intermittent nature of renewables in DCMGs has become an enhanced requirement for these converters. This article proposes a partially rated multiport interlinking converter (PMIC) that incorporates a distributed ESU. The PMIC controls a floating voltage and a bidirectional shunt current on the dc line, ensuring full galvanic isolation for the ESU while operating at a low dc-link voltage. It regulates multidirectional power flow and balances power during peak and off-peak periods. A decentralized droop-based power flow control strategy is proposed for the PMIC, which distributes renewable energy generation, load consumption, and ESU utilization proportionally across the system. The control strategy includes two tailored continuously differentiable activation functions, Sigmoid and hyperbolic tangent, to facilitate autonomous global power-sharing and seamless ESU engagement. Simulation and experimental case studies confirm the PMIC's capability to smooth renewable energy fluctuations and enhance the power and voltage profiles of DCMGs.

Aiming at the problems of promoting new energy consumption, reducing carbon emissions, load fluctuations, and multi-agent conflict of interests in the networked microgrid system, this article proposes a microgrid optimization operation strategy based on demand response and reward-penalty ladder-type carbon trading mechanism. First, in order to determine the electricity sales price of the system, an optimal scheduling model for microgrid operators is established, including gas cost, electricity profit for users, and surplus power supply network profit. Second, a demand response strategy on electricity price and low-carbon compensation incentives is proposed on the user side. The transaction model is embedded between microgrid operators and users into the master–slave game framework, and a multi-slave game collaborative optimization model is established with microgrid operators as leaders and users as followers. The existence and uniqueness of Stackelberg game are proved, and the differential evolution algorithm and CPLEX solver are used to solve the proposed model. Finally, an example of a microgrid system including three community users is provided to show the effectiveness of the proposed model and strategy.

The mathematical mechanism model of modular multilevel converters (MMC) exhibits high-dimensional complexity, necessitating a trade-off between accuracy and efficiency during the modeling process. To address this issue, this article proposes a nonmechanism modeling approach for MMC, focusing on the mapping relationships between electrical quantities. The proposed method utilizes the convolutional neural network - long short-term memory - attention mechanism (CNN-LSTM-AM) neural network algorithm to drive the data-driven modeling of MMC. Operational data from the MMC system is utilized as the training set. The CNN layer selects input features with strong correlations, while the LSTM layer captures temporal dependencies among the features. The AM layer performs weighted aggregation on the LSTM output to enhance prediction accuracy. In addressing the challenge of fitting frequency, which cannot be directly achieved using electrical quantities, the frequency model incorporates a Hilbert–Huang transform (HHT) decomposition of voltage and current signals. The extracted instantaneous frequencies and amplitudes serve as additional input features to improve the fitting of output features. This approach effectively resolves the issue of inadequate prediction accuracy in the frequency model. Additionally, data-driven models for MMC inverters and rectifiers are trained using simulation data and evaluated for accuracy using a separate test set.

Both efficiency and operation cost are the key to offshore microgrids operation. At the same time, due to the lack of external energy support, the paralleled converters in offshore microgrids often needs to handle the operation in a wide load range. In that case, this article proposes a dual-level optimal control framework to improve overall operation performance of offshore microgrid with paralleled converters within a wide load range. First, a normalized nonlinear relationship between power loss and operation cost of paralleled converters is established. Based on it, a multiobjective optimal function is established. Then, the optimal operation condition is derived by Lagrange multiply method with the established converter performance index. Furthermore, optimal power sharing considering both efficiency and operation cost is proposed at first level. Then, second level control is proposed to improve system performance in a wide load range. The proposed decentralized dispatch strategy is realized by a consensus protocol mechanism with the mealy machine based on the established performance index. Experiment results in a scaled-down prototype are given to validate the effectiveness of the proposed dual-level control strategy. The proposed strategy is able to optimize performance of paralleled converters under different power profiles.

Offshore microgrids, due to their remote location and lack of external energy support, face significant challenges in wide-range load operation and maintenance. Consequently, efficiency and reliability are critical concerns for converters in offshore dc microgrids. This article presents an optimal control strategy aimed at enhancing both efficiency and reliability. A normalized nonlinear relationship between power loss and thermal stress of a paralleled converter is first established. Based on this, a dual-objective optimization function with an active weight function as well as a system overall performance index is established. The active weight function dynamically adjusts the control priority based on converter efficiency and switching device thermal stress. Then, the optimal power-sharing strategy is derived by the Lagrange multiplier method with the proposed optimal function. Additionally, to accommodate a wide load range, an optimal selection strategy for operating converter combinations is proposed, requiring only low-bandwidth communication. Experiment verification is given to validate the effectiveness of the proposed control strategy. The experiment results demonstrate that the proposed control strategy can improve the overall performance of offshore microgrids by optimizing efficiency and reliability.