In the direct current (DC) microgrid composed of multiple distributed generations, due to the different distances between various converters and the DC bus in the system, the difference of the line resistance will reduce the current sharing accuracy of the system. The droop control was widely used in the operation control of the DC microgrid. It was necessary to select a large droop coefficient to improve the current sharing accuracy, but a too large droop coefficient will lead to a serious bus voltage drop and affect the power quality. In view of the contradiction between the voltage regulation and load current sharing in the traditional droop control, a hierarchical control algorithm based on the improved droop control of the fuzzy logic was proposed in this paper. By improving the droop curve, the problems of voltage regulation and current sharing were solved simultaneously. The effectiveness of the algorithm was verified by simulation.

Battery cell voltage is an important evaluation index for electric vehicle condition estimation and one of the main monitoring parameters of the battery management system, and accurate voltage prediction is crucial for electric vehicle battery failure warning. Therefore, this paper proposes a novel hybrid gated recurrent unit and long short-term memory (GRU-LSTM) neural network to predict electric vehicle lithium-ion battery cell voltage. Firstly, Pearson coefficient correlation analysis is carried out to determine the input parameters of the neural network by analyzing the influence factors of the voltage parameters, and the hyperparameters of the neural network are determined through cross-validation to construct the lithium-ion battery single-unit voltage prediction model based on GRU-LSTM. Secondly, the voltage prediction accuracy and robustness of the GRU-LSTM model are verified by training the historical data of real vehicles in spring, summer, fall, and winter, combined with four different error indicators. Finally, the feasibility of the proposed method is verified by designing hierarchical warning rules based on the prediction data to realize the accurate warning of multiple voltage anomalies.

Low-voltage direct current (DC) microgrid based on distributed generation (DG), the problems of load mutation affecting the DC bus under island mode, and the security problems that may arise when the DC microgrid is switched from island mode to grid-connected mode are considered. Firstly, a DC bus control algorithm based on disturbance observer (DOB) was proposed to suppress the impact of system load mutation on DC bus in island mode. Then, in a grid-connected mode, a pre-synchronization control algorithm based on a neural network adaptive control was proposed, and the droop controller was improved to ensure better control accuracy. Through this pre-synchronization control, the microgrid inverters output voltage could quickly track the power grid’s voltage and achieve an accurate grid-connected operation. The effectiveness of the algorithms was verified by simulation.

In the context of weak grids, vector-controlled modular multilevel converters (MMC) suffer from issues such as low inertia, low damping, and poor system stability. To address these challenges, this paper proposes a control strategy for virtual synchronous generators (VSGs) based on a fuzzy logic control algorithm. The conFtrol strategy leverages the capability of fuzzy algorithms to handle the fuzziness and uncertainty of input signals, enabling adaptive control of the virtual inertia and damping coefficient of the VSG, thus empowering the system with autonomous frequency and voltage regulation capabilities. When the system deviates from or approaches the stable operating point, increasing or decreasing the virtual inertia allows for dynamic adjustment of the virtual inertia and damping coefficient in response to load fluctuations during MMC operation. Through simulation verification, it is demonstrated that the proposed control method provides inertia support to the system during sudden changes in active load during MMC grid-connected operation. This control method achieves adaptive adjustment of the virtual inertia and damping coefficient, effectively enhancing system stability. The simulation results validate the effectiveness and correctness of the proposed control strategy.

Accurate estimation of the state of charge (SOC) of lithium-ion batteries is critical for assessing the safety and remaining range of electric vehicles. However, due to the complex and variable operating environment of batteries and their highly nonlinear internal mechanisms, achieving high-precision SOC prediction remains a central challenge in current research. To this end, this paper proposes a nonlinear Hammerstein model based on the Hippopotamus Optimization Algorithm (HO) to optimize the backpropagation neural network, thereby enhancing the accuracy of SOC prediction. The HO-BP-Hammerstein model optimizes the BP neural network architecture using the Hippopotamus Algorithm and conducts SOC prediction accuracy tests on real-world data. Experimental results demonstrate the superiority of the proposed method through comparative accuracy analysis of various SOC prediction approaches under different operating conditions, confirming its significant engineering application value.

The remaining useful life (RUL) prediction of lithium-ion batteries remains challenging for hybrid models due to high computational redundancy and hyperparameter sensitivity under complex operating conditions. To address these issues, this paper proposes a novel hybrid framework that integrates a CNN–BiLSTM–Attention network with a Crested Porcupine Optimizer (CPO). The key innovation lies in the simultaneous co-optimization of model structure and parameters: a group-level structured pruning strategy eliminates redundant convolutional kernels to reduce complexity, while the CPO algorithm dynamically optimizes critical hyperparameters (e.g., BiLSTM hidden nodes, attention dimensions) using RMSE as the fitness function. This dual optimization achieves a balance between model lightweighting and predictive accuracy. Experimental results on real-world electric vehicle datasets demonstrate that the optimized model reduces Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) by 60.7% and 55.2%, respectively, compared to the baseline CNN–BiLSTM–Attention model. Furthermore, the model maintains high robustness across multiple datasets (R² > 0.996), validating its strong generalization capability in small-sample and high-noise scenarios.