In DC microgrids, a large-capacity hybrid energy storage system (HESS) is introduced to eliminate variable fluctuations of distributed source powers and load powers. Aiming at improving disturbance immunity and decreasing adjustment time, this paper proposes active disturbance rejection control (ADRC) combined with improved MPC for n + 1 parallel converters of large-capacity hybrid energy storage systems. ADRC is utilized in outer voltage control loops, and improved MPC is employed in inner current control loops of n battery converters. Droop control is adopted to obtain power distribution between n battery converters, and a DC bus voltage compensator is used to compensate voltage deviations and maintain constant DC bus voltage. The low-pass filter (LPF) is adopted to obtain high-frequency power as the reference for the supercapacitor converter, ADRC is also utilized in the outer power control loop, and MPC is employed in the inner current control loop. Compared with traditional observers, the voltage expansion state observer of the proposed ADRC control is independent of the system model and parameters and consequently has strong disturbance immunity, and significantly reduces voltage overshoots during power fluctuations. The MPC-based inner current control loops of n + 1 converters accelerate current response speed and significantly decrease switching losses. Simulation and experimental results indicate that utilizing the proposed control strategies, large-capacity HESS has stronger anti-interference ability, shorter regulation time, smaller switching loss, and simultaneously maintains the stability of the DC bus voltage.

Mixed potential theory is frequently employed for analysing large‐disturbance stability. The precision of the current stability criterion for DC microgrids controlled by virtual DC motors (VDCM), which relies on mixed potential theory, is inadequate. This is primarily because the criterion does not account for the control parameters of the DC bus voltage control link and the virtual DC motor link. To address these issues, the paper initially formulates the control aspect of the VDCM using a controlled current source. Subsequently, it derives the system's mixed potential function from the model, enabling the development of a stability criterion that includes the control parameters of the DC bus voltage control link, the virtual DC motor link, and the current tracking link. It also clarifies how parameters not directly included in the criterion affect system stability. Through simulations and experimental validations, it is demonstrated that the proposed stability criterion effectively captures the impact of control parameters on system stability and precisely delineates the system's power stability boundary, offering insights for system parameter optimization.

This paper presents a loop analysis and angle recovery (LAAR) based reactive power optimization for three-phase unbalanced and weakly-meshed active distribution networks (ADNs). For each loop, the width first search (WFS) method is used to find the breakpoint bus of each tie line, and then all loops are broken up at these breakpoint buses by disconnecting tie lines, placing added buses, and adding compensation powers, so that the original weakly-meshed ADN can be precisely converted into an equivalent radial ADN. Furthermore, the traditional second-order cone programming can be employed for the equivalent radial networks to find the global optimal solution. However, the compensation powers are not constant values but related to the bus voltages, which are unknown before opening the loops. To address this problem, we design an iterative method to dynamically update the values of compensation powers until the convergence criterion is met. Moreover, the voltage angles are recovered for all loops at each iteration. The effectiveness of the proposed LAAR method is demonstrated by 9 cases, and the results show that the LAAR method can achieve a better convergence performance than the traditional method.

In isolated microgrids, virtual synchronous generators (VSGs) are utilized to enhance frequency stability. However, the presence of these generators can lead to undesirable levels of voltage or current during transient angle instability. The current Lyapunov transient stability analysis method is challenging to utilize due to the complexity of acquiring the energy function. Furthermore, existing data‐driven microgrid transient stability methods do not easily adapt to changes in topology. Inaccuracies may also be present in existing stability analysis methods, as they fail to account for nonlinearities such as saturation and dead zone. To address these issues, we propose a microgrid transient stability analysis method based on a message‐passing graph neural network (MPNN). When the topology changes, the information bus and transmission line are aggregated and updated through the message‐passing mechanism in the graph neural network, and we enhance the approach by introducing current saturation as an additional feature to the transient stability analysis model. We present simulation results for a microgrid test case to validate the effectiveness of the proposed method, and the evaluation accuracy is maintained above 98%.

Existing liner power flow (LPF) models do not consider the non-smooth constraint characteristics of the voltage source converter (VSC) and on-load tap changer (OLTC) that contain operation limit and dead zones, which limits its application. Thus, a three-phase LPF model for distribution networks is proposed, which considers non-smooth constraint characteristics. Firstly, smoothing functions are used to effectively fit the non-smooth constraint characteristics, resulting in control functions that are continuous and differentiable. Then, based on the first-order Taylor series expansion, the three-phase power flow equations are physically linearized, and the terms to compensate errors are obtained through the partial least squares (PLS) method. Compared with the model without considering the non-smooth constraints, the three-phase LPF model considering the non-smooth constraints can accurately represent the operating characteristics of the VSC and the OLTC in the actual distribution network, thus providing more reliable power flow calculation results. Based on a typical 42-node distribution network, the proposed model is compared and analyzed against other LPF models. The results indicate that the proposed model has the ability to handle non-smooth constraints, with higher computational accuracy compared to existing data-physical hybrid driven models.