Amer Mohammad Yusuf Mohammad Ghias

State-of-charge (SoC) balancing is commonly utilized for modular battery energy storage systems (MBESSs). However, the existing methods assume that the MBESS power demand can always be met without violating constraints (for example, battery voltage limits). Hence, the existing methods may violate battery constraints, which can jeopardize the MBESS and trigger its protections. Moreover, the literature has largely overlooked a viable methodology for predicting the maximum power that a constrained MBESS can process over a given interval, a quantity referred to as the state-of-power, which is essential for effective MBESS scheduling. Addressing these gaps, this article presents a constraint-oriented SoC balancing method that guarantees the safe operation of MBESSs. In particular, the proposed method minimizes the MBESS SoC imbalance as permitted by constraints, and curtails the MBESS output power minimally if the power demand cannot be met without violating constraints. Furthermore, a computationally affordable state-of-the-power prediction algorithm for MBESSs operating under the proposed method is developed. The proposed method and algorithm are validated via simulations and experiments.

Conventional solid-state transformer (SST) architectures based on the cascaded H-bridge converter typically require large high-voltage side capacitors to suppress low-frequency voltage ripples. These capacitors are further stressed during load transients, complicating efforts to reduce their capacitance. Large capacitor banks increase system footprint and reduce reliability, especially with high-power-density electrolytic capacitors. The capacitors absorb transient power imbalance between the input and output ports of an SST. Conventional control strategies often lead to uneven distribution of transient energy, placing disproportionate stress on certain capacitors and resulting in oversized capacitance requirements. To address these challenges, this article proposes a reserve-energy-aided control strategy for effective energy distribution across DC-link capacitors in low-capacitance SSTs. The proposed solution is validated using a 1-kVA, 230-V/250-V, 50-kHz single-phase SST prototype, demonstrating improved transient response and operational stability under varying load conditions.

Model inaccuracies arising from operational and environmental uncertainties often compromise the performance of controllers for power electronic systems. This paper develops an end-to-end adaptive control framework to address this challenge by combining physical principles with expert knowledge within a unified Physics-Informed Neural Network architecture. To this end, expert control policies acquired via imitation learning are formulated as differentiable constraints within the loss function. By co-optimizing these expert-knowledge constraints with fundamental physical laws and safety-guided stability regularization, the neural network is trained to simultaneously identify real-time system dynamics and directly map them to an optimal control action. The resulting controller generates adaptive control signals to effectively counteract system uncertainties. Its effectiveness is comprehensively validated across multiple topologies, including DC/DC converters and a DC/AC inverter. Under parameter variations and external disturbances, the proposed method achieves faster transient response and robust stability, exhibiting improved performance over both well-tuned linear controllers and advanced data-driven under adverse conditions.

Variable speed pumped storage (VSPS) possesses significant potential for power system frequency regulation. This paper proposes a novel optimal trajectory control for VSPS to enhance power system frequency response (FR) by fully leveraging its regulation potential, while maintaining its rotor speed stability. First, the speed stability of VSPS during FR process in both turbine and pump modes is investigated. Following this, an analytical expression for the rotor speed is derived, specifically for the turbine mode, given its susceptibility to speed instability. This expression enables not only an efficient assessment of rotor overspeed risks but also a precise characterization of critical points along the operational trajectory. Then, the design principles of additional power control for VSPS are proposed to achieve a smooth switch of operation stages. Based on these principles and derived expressions, the optimal operation trajectory of VSPS is determined by system disturbances and the unit's operating conditions. This adaptability of the proposed strategy optimizes power response and strengthens FR capability. Furthermore, an overspeed control is developed to ensure rotor speed stability. Finally, the IEEE 39-bus system is utilized to test the proposed strategy under various operating conditions, demonstrating its superior performance in improving power system FR.

The integration of reconfigurable battery storage (RBS) systems into dc shipboard microgrids (SMGs) holds considerable potential by improving cell-level flexibility and system adaptability compared to conventional battery storage systems. This article proposes an online-trained deep reinforcement learning (DRL) control framework for an RBS system interfaced with a bidirectional dual-active-bridge (DAB) converter, aiming to manage its dynamic operation while enabling adaptive current regulation and cell balancing. Considering the nonlinear dynamics and real-time uncertainties inherent in practical SMGs, an automated online training framework is developed to bridge the sim-to-real gap that often hinders the generalization of offline-trained DRL methods. To this end, the proposed online training framework is implemented on a dSPACE real-time platform, enabling the controller to continuously interact with the physical system and refine its policy with real-world experiences. Experimental results validate that the proposed online-trained DRL approach achieves convergence 20.1% faster, in terms of training episodes, compared to simulation-based training. Comparative studies demonstrate that the proposed control strategy achieves superior dynamic performance, reducing settling time by 43.6% compared to the simulation-trained controller and by 53.7% compared to the benchmark predictive controller. It also enables faster charge-discharge transitions, completing mode switches within just 9.1 ms. Furthermore, the approach improves battery balancing and transient current handling, thereby improving its suitability for SMG applications.

Traditional current reference generators (CRGs) used in low-voltage ride-through (LVRT) control schemes often optimize a single objective, such as minimizing power oscillations, at the expense of other performance metrics such as total harmonic distortion (THD). This article proposes a many-objective optimized current reference generation strategy (MaO-OCRGS) that yields a set of tradeoff solutions, each defined by proposing two decision variables and their associated power quality performance metrics, i.e., power oscillations ($\Delta p$, $\Delta q$), $\mathit {THD}$, and the unbalance index ($\mathit {UI}$). A hybrid physics-based and data-driven (HPD) approach is proposed to construct the performance objective functions of a many-objective optimization (MaOO) problem, where the power oscillations are theoretically derived, and an artificial neural network (ANN) is proposed to model the relationships between the decision variables and current-related metrics ($\mathit {THD}$ and $\mathit {UI}$). A nondominated sorting genetic algorithm III (NSGA-III) is employed to solve the MaOO problem, yielding a set of optimal decision variables. A subsequent postoptimization selection then identifies feasible solutions that prioritize one performance metric while keeping the others within acceptable limits, thereby avoiding undesirable tradeoffs and resulting in three optimized CRGs. Experimental validation demonstrates that the optimized CRGs achieved through the proposed MaO-OCRGS effectively enhance one performance objective without unacceptable degradation in others, as the other metrics are maintained within their predefined limits, thus offering flexibility while enhancing the overall LVRT performance under unbalanced voltage conditions.

The cascaded H-bridge (CHB) converter is an attractive topology to interface energy sources, such as battery energy storage or photovoltaic systems, with the medium-voltage grid. The submodules in CHB converter-based energy systems unavoidably process different active power during operation. However, there exist inherent limits that prohibit an arbitrarily imbalanced intra-phase active power distribution. Failing to consider such limits can result in output current distortion and capacitor voltage deviation, which can jeopardize the CHB converter. Crucially, these limits vary depending on the modulation method utilized, whose selection is, thus, of utmost relevance. Accordingly, this paper derives and compares the feasible intra-phase active power imbalance range of existing modulation methods. The analysis is experimentally corroborated in a 1-kVA single-phase CHB converter with six submodules per phase. Based on the comparison, suggestions on which modulation methods are preferred for different CHB applications are provided.