State of health (SOH) estimation is foundational for effective health prognosis and management of lithium-ion batteries. However, conventional data-driven approaches struggle to ensure robust generalizability when training data lacks sufficient SOH labels. In this work, we explore the potential utility of ageing information, gathered from routine operational data beyond the standard input collection region, as proxy labels (PLs) to establish nonlinear mappings between a short charging segment and SOH. To achieve this, we develop a multitask learning (MTL) framework for battery health monitoring, which comprises two critical components: 1) a deep MTL model, and 2) a multivariate Gaussian process regression (GPR) model. The deep MTL model is designed with four distinct tasks, each with a specific goal to exploit data under different labeling conditions, facilitating the extraction of domain-invariant and target-specific features for PL estimation. Using the estimated PLs, the GPR model is employed to provide precise mappings to SOH. Validations are conducted on batteries with diverse chemistries and cycling temperatures, particularly under severe label scarcity below 5%. The results confirm the effectiveness of the proposed method, outperforming baseline methods and achieving a mean absolute error of less than 1.3% in all case studies.

Solving alternating current optimal power flow (AC-OPF) problems for large-scale power systems is essential for operations integrating renewable energy. However, conventional numerical methods for AC-OPF face challenges such as high computational costs and convergence difficulties as system size increases. To address these challenges, existing research utilizes direct current OPF (DC-OPF) or data-driven methods. DC-OPF linearizes the AC-OPF problem by considering inherent physical characteristics of the power system, such as voltage variations, providing an approximate solution. Meanwhile, data-driven methods leverage its strong end-to-end learning capabilities to solve AC-OPF effectively. Although both methods are sufficiently fast, DC-OPF can suffer from accuracy limitations due to its simplifying assumptions, which neglect reactive power and voltage deviations. On the other hand, data-driven approaches may exhibit degraded performance when trained on limited data and are prone to constraint violations. To address these problems, this study proposes PrOPF, a physics-informed framework that refines DC solutions to solve AC-OPF. Instead of learning from scratch, PrOPF treats DC-OPF results as physical priors and learns the mapping from DC to AC solutions using a graph neural network. A physics-informed loss with soft constraints is introduced during training, and hard constraints are enforced during inference to improve physical consistency. Extensive simulations are conducted on various IEEE test systems with renewable energy. The results demonstrate that our PrOPF achieves higher prediction accuracy and lower constraint violations compared to other data-driven OPF methods, and attains up to 30x acceleration over the traditional interior point method.

The direct current (DC) ultrafast electric vehicle charging hub (UFEVCH) represents a cutting-edge infrastructure designed to meet the growing high-power charging demands of electric vehicles (EVs) and support the goal of carbon neutrality. The way that UFEVCH realizes ultrafast charging is its ability to gather power by allocating power modules (PMs) to charging dispensers. Intelligent and real-time dispatching of PMs can enhance the overall utilization rate of UFEVCH and reduce losses associated with frequent reallocation. However, the PM dispatch has yet to be comprehensively studied. This study introduces a novel UFEVCH model that simulates the operational conditions of PMs and formulates a PM dispatch problem to achieve a higher utilization rate and modular cost-effectiveness. A real-time PM dispatch method, termed reinforcement learning and secondary dispatch (RLSD), is proposed to accommodate the uncertain behavior of EV charging. This method employs reinforcement learning (RL) to optimize PM allocation in real-time. In addition, the secondary dispatch (SD) ensures the reliability of UFEVCH by refining the RL solutions. The proposed UFEVCH model and RLSD method are validated using data from over 100 practical UFEVCH in Shenzhen, China, demonstrating nearly doubled service quality compared with numerous model-free and model-based methods. In addition, a theoretical analysis is provided to illustrate the convergence of the RLSD method.

Behind-the-meter (BTM) battery energy storage systems (BESS) are becoming increasingly important in the power system with the proliferation of intermittent distributed renewable energy sources. Stacked revenue can be achieved by providing multi-services to the power grid, justifying the substantial upfront cost of BTM BESS and promoting their future adoption. This paper focuses on optimizing the operation strategy of BTM BESS to maximize the time average stacked revenue obtained from multiple service markets, including energy arbitrage, frequency regulation, photovoltaic (PV) power smoothing and reactive power compensation. Challenges arise from the coupling of operation decisions both among multiple services and over the temporal dimension, considering the different decision timescales of service markets as well as the battery dynamics. The uncertainty of stochastic parameters further complicates the optimization process. To address these challenges, this paper proposes a novel two timescale online optimization scheme based on the Lyapunov optimization framework. The uncertainties are tackled by making decisions online, and the computation complexity is highly relieved by relaxing the temporal coupling with a drift-plus-penalty technique. Theoretic analyses are conducted to prove that the solution of this relaxed online decision problem is always feasible for the original one, and it can achieve near-optimum with a constant optimality gap. Extensive simulations utilizing the energy and frequency regulation data from the real market validate the effectiveness of our proposed scheme.