ABSTRACT State‐of‐charge (SOC) inconsistency among units in vanadium redox flow battery (VRFB) stations causes voltage‐limit violations and premature charge/discharge cutoffs, thereby degrading station‐level energy utilization and dispatch performance. This paper proposes a station‐level multi‐objective coordinated model predictive control (MPC) strategy using a mixed‐integer‐representable ReLU neural surrogate and hybrid bound tightening. The main innovation is to integrate OCV‐reference selection, SOC‐equalising station‐level MPC and feasibility‐based/optimization‐based bound tightening (FBBT + OBBT) into a unified control‐oriented framework that preserves terminal‐voltage safety whilst reducing online computational burden. First, an open‐circuit‐voltage (OCV) reference trajectory is selected at the single‐stack level to operate in a high flow‐rate sensitivity region and mitigate concentration polarization. Second, a station‐level MPC formulation jointly tracks the OCV reference, equalises inter‐cabin SOC and satisfies dispatch requirements under voltage, concentration, current and power constraints. Third, a ReLU‐based surrogate replaces the nonlinear electrochemical model and is embedded into the MPC through Big‐M constraints, whilst the bilinear power term is handled by McCormick envelopes. The hybrid FBBT + OBBT scheme tightens neuron bounds by 28.64%–59.24% and reduces preprocessing time by 14% compared with pure OBBT. Simulation results show that the proposed strategy shortens charging time by about 10%, improves voltage efficiency from 75.59% to 79.95%, improves energy efficiency from 71.20% to 74.64% and reduces station‐level charge/discharge actions from 31 to 29 under the same 100 MWh dispatch target, demonstrating improved safety, efficiency, SOC consistency and operational economy.
ABSTRACT Indian distribution utilities face persistent challenges such as billing inefficiencies, high aggregate technical and commercial (AT&C) losses, power theft and operational constraints. Although smart meters provide advanced monitoring capabilities, interoperability issues persist in Advanced Metering Infrastructure (AMI) due to multivendor deployments. India has adopted DLMS/COSEM under IS 15959 to enable standardised and interoperable metering systems. This paper presents a structured compliance assessment of Indian smart meters against IEC 62056 and IS 15959, identifying key gaps in communication profiles, security mechanisms, and interface class implementation with respect to international practices. Based on the identified gaps, the paper proposes an IEC 62056‐aligned framework to improve interoperability, data reliability and system integration in multivendor AMI environments. In addition, a conceptual Emergency Load Profile is introduced to support controlled load management by enabling limited power supply to critical loads during contingencies or nonpayment conditions. The proposed approach enhances grid reliability, operational efficiency and utility‐level flexibility, while also supporting large‐scale smart meter deployments aligned with national initiatives such as Make in India and Atmanirbhar Bharat.
Transient stability in Dispatchable Virtual Oscillator Control (dVOC) during interconnection operations in microgrid systems remains a significant challenge. This paper proposes a Synergistic Phase-Damping Strategy (SPDS) that integrates phase compensation and virtual damping within the dVOC control framework to improve transient stability. The proposed method is validated on a 20-kV medium-voltage microgrid comprising three 5-MW dVOC-based grid-forming inverters under three-phase-to-ground fault conditions. The phase compensator mitigates switching-induced delays in active and reactive power regulation, while the virtual damping mechanism suppresses oscillations through high-pass-filtered current feedback. Simulation results demonstrate that the proposed SPDS reduces the post-fault oscillation frequency from 48 to 17 Hz, shortens the oscillation duration from 200 to 95 ms, and decreases voltage/current recovery times from approximately 89-24 ms compared with conventional dVOC. Furthermore, the proposed strategy improves active and reactive power stabilisation while maintaining stable voltage regulation throughout fault recovery. These results confirm that SPDS significantly enhances the robustness and transient performance of dVOC-based inverters in medium-voltage microgrid applications.
Renewable energy bases in Desert-Gobi-Wasteland (DGW) areas connected with high-voltage direct current (HVDC) transmission systems show notable weak-grid features, including low inertia, weak damping and a low short-circuit ratio. This results in strong active power-voltage (P-V) coupling effects, making existing real-time active power control methods ineffective. To tackle this problem, this paper proposes a real-time active power control method that simultaneously considers static voltage security constraints and the effective regulation space of various resources. The method ensures safe and stable system operation while efficiently using regulation resources. First, the causes of the strong P-V coupling effects in weak HVDC sending-end AC power systems integrated with renewable energy sources are analysed. Sensitivity indices are used to measure the degree of coupling under different operating conditions. Next, considering the impact of uncertainties in renewable energy output on identifying system voltage weak points, combined with the impedance modulus margin indicator (IMMI), a voltage weak point identification approach based on probabilistic power flow is proposed. The effects of active power regulation from resources such as wind power, photovoltaic and HVDC links on these weak points' voltages are analysed and a quantitative method for active power regulation space of multiple resource types, considering static voltage security constraints, is presented. Then, by integrating voltage-active power sensitivity and the effective regulation capacity of various resources, a real-time active power control strategy is designed. This strategy incorporates resource prioritisation and differentiated power dispatch, aiming to meet active power regulation needs while reducing negative impacts on voltages at weak points. Finally, the effectiveness of the proposed approach is validated using a modified IEEE 30-bus system and a modified IEEE 57-bus system.
ABSTRACT Currently, rural energy consumption is characterised by extensive usage with low conversion efficiency and limited renewable energy integration. Moreover, optimisation strategies focused solely on economic objectives often fall short in addressing real‐world needs. To tackle these challenges, this study introduces a dual‐layer optimisation approach for shared energy storage (SES), grounded in thermodynamic economics theory and emphasising energy efficiency–economy coupling in multi‐regional rural integrated energy systems (RIES). We first suggest an AC/DC hybrid architecture with SES to bolster the power grid's flexible adjustment capacity. Subsequently, we present an exergy economic flow (EEF)‐based optimisation model that intricately links the quality and cost attributes of energy flow. This model also incorporates a dual‐layer optimisation configuration for SES, considering both long‐term and short‐term costs across various typical scenarios. To ensure efficient and rapid resolution, the model undergoes linearisation, relaxation and scaling, ultimately being converted into a single‐layer mixed‐integer linear programming (MILP) problem. Simulation outcomes indicate that our proposed method enhances economic metrics by 41.9% and energy efficiency metrics by 14.9% compared to conventional optimisation strategies centred on economic goals. These improvements hold substantial significance for the sustainable low‐carbon development of rural regions.
The power control loop in a virtual synchronous generator (VSG) exhibits inherently insufficient damping, making it prone to low-frequency oscillation (LFO) under disturbances. Moreover, the dynamic coupling between active and reactive power can induce unnecessary reactive power fluctuations during active power transfer. This not only affects voltage stability but may also further exacerbate LFO. Leveraging the static VAR compensator (SVC)'s capability for fast reactive power support and supplementary damping, this study proposes an LFO suppression strategy based on the NPOD-SVC-VSG grid-connected system. First, this study develops a small-signal model and state-space representation of the SVC-VSG grid-connected system. Eigenvalue analysis is then employed to investigate the stability influence mechanisms in the SVC-VSG grid-connected system under weak interactions. Furthermore, a Phillips-Heffron model of the SVC-VSG system is developed for the mechanism analysis of LFO. To enhance system damping, nonlinear power oscillation damping (NPOD) is proposed that adaptively adjusts gain based on oscillation amplitude while considering the impact of communication delay between the SVC and VSG. NPOD is incorporated into the voltage control loop of the SVC, and its parameters are designed using the phase compensation method that accounts for communication delay. Finally, MATLAB/Simulink simulations demonstrate that the proposed NPOD-SVC-VSG strategy effectively suppresses LFO, increasing the system damping ratio by 10.93% compared to the VSG strategy. The strategy also rapidly compensates for reactive power deficits during transients, thereby enhancing system voltage stability.
Accurate multi-energy load forecasting plays a vital role in the optimal scheduling of integrated energy systems (IES). However, existing forecasting methods face two major challenges: insufficient extraction of complex features within load data, which limits prediction accuracy, and ever-increasing model complexity, which leads to a sharp rise in computational resource demands. To tackle the above-mentioned challenges, this research introduces a load forecasting scheme derived from multiscale lightweight deep learning. First, the dynamic coupling characteristics of load data are analysed using correlation coefficients to determine model-specific input configurations. Then, an improved variant channel multiscale bottleneck residual network (VC-MBResNet) is proposed to obtain high-dimensional load feature data. Finally, by leveraging shared underlying parameters and an enhanced adaptive loss function for each subtask-using both soft and hard weight-sharing strategies-the extracted features are integrated into a multi-task learning (MTL) framework. A bidirectional long short-term memory (BiLSTM) network is employed as the shared layer, with an attention mechanism embedded in its hidden layers to enhance the focus on critical temporal features. Experimental results demonstrate that the proposed approach surpasses existing models in terms of both prediction accuracy and computational efficiency.
The solid oxide fuel cell (SOFC) is an innovation power generation device in the form of an electric stack. The temperature distribution is crucial for the stable and safe operation of the system. Considering the difficulty of accurately measuring the internal temperature of the device, this paper proposes a temperature estimation method based on the Luenberger-sliding mode observer (L+S) and the Kalman filter (KF) observer. First, a discrete-time state-space model of the SOFC stack is established using a one-dimensional (1D) theoretical model that considers system noise. Secondly, the input variables of the observer are filtered and combined with the condition number of the measurement matrix to produce the optimal combination. The estimation performance of the model is enhanced through decoupling of error systems and vibration suppression. The Kalman filter observer model is constructed by combining the stack model with the Kalman filter (KF) algorithm. Then, the Luenberger-sliding mode observer model is constructed based on the theories of the Luenberger observer (L) and the sliding mode observer (S). The estimation performance of this model is enhanced through the decoupling of the error system and vibration suppression. The Kalman filter observer model is constructed by integrating the electric stack model with the Kalman filter (KF) algorithm. This method solves the time difference between the observed state and the actual system state by filtering noise to improve the estimation accuracy of the model. Finally, a multidimensional performance evaluation index system was established, and the estimation performance of various observer models was compared through simulation experiments and data analysis to verify the effectiveness of the temperature estimation model proposed in this article.
Distributed photovoltaic (DPV) power forecasting is essential for grid stability but remains challenging due to strong intermittency and meteorological uncertainty. Existing data-driven models often lack physical interpretability and struggle to capture asymmetric dependence structures, leading to unreliable predictions during extreme weather. This paper proposes a copula-guided transformer (CGT) framework that integrates statistical dependence mining with physics-informed deep learning. Specifically, Gaussian copula is used for global feature screening, whereas Clayton and Gumbel copulas quantify asymmetric tail dependencies-revealing the conditional lower-tail inhibitory effect of rainfall and the upper-tail driving effect of irradiance on power output. These copula-derived parameters are embedded as physical priors into a guided encoder, where a temporal convolutional network (TCN) dynamically regulates attention weights to enhance physical consistency. Validated on real-world DPV data from China, the CGT model offers significant performance advantages. The results demonstrate superior robustness across clear-sky, rainy and high-volatility scenarios by effectively mitigating spurious overestimation and response lag.
As the penetration rate of renewable energy represented by photovoltaic (PV) power in power systems continues to increase, the synchronisation stability of grid-connected inverters under weak-grid conditions faces severe challenges. The double second-order generalised integrator-based phase-locked loop (DSOGI-PLL), a key technology for grid-connected control in weak grids, exhibits excellent harmonic suppression capability. However, when significant fluctuations occur in grid frequency or voltage, its dynamic response and phase-tracking accuracy are significantly affected, even threatening the stable operation of the system. Analysis indicates that one of the key issues leading to the performance degradation of the DSOGI-PLL in weak grids is the lack of a fast and accurate frequency estimation. To address this, after comparing three frequency estimation methods-namely, the output frequency of the PLL itself, the grid-voltage zero-crossing detection frequency and the adaptive notch filter frequency-this paper proposes a multisource frequency fusion and adaptive damping DSOGI-PLL (MSFFAD-DSOGI-PLL). By analysing the harmonic content and frequency deviation of the power grid, the proposed method adaptively weight-fuses the results of the above three frequency estimation approaches and corrects the resonant centre frequency of the DSOGI in real time. On this basis, to balance the dynamic response and steady-state accuracy of the system under different operating conditions, an adaptive damping adjustment mechanism based on fuzzy logic linearisation is designed to achieve stable operation over a wide voltage/frequency range. Finally, experimental results demonstrate that compared with the traditional DSOGI-PLL, the proposed method exhibits stronger robustness and higher tracking accuracy under complex grid conditions such as large-range fluctuations in grid frequency and voltage.
In view of three-phase power unbalanced caused by the output power difference of each phases microsource of the modular multilevel converter half-bridge series microgrid (MMC-MG) operating in the islanded mode, a balance control method of inter-phase power is proposed. The MMC-MG topology and three inter-phase power dispatching modes are introduced. The system three-phase output power mathematical model is established, and the mechanism of the dc circulating current to realize the inter-phase power flow is described. A dc circulating current controller based on arm virtual voltage is designed. Its output signal is superimposed on each phase modulation indices as power balance control variable. According to the output power of each microsource and the load power, the inter-phase power regulation is distributed, and the regulation is compensated by the state of charge of the energy storage device. Simulation and experiments show that the proposed control strategy can achieve inter-phase power balance control, which has no effect on the output voltage and frequency of the system. Compared with general three-phase microgrid, MMC-MG can realise inter-phase power balance by its own circulating current control, without additional investment of power balance equipment.
The increasing integration of inverter-based resources (IBRs) into microgrids (MGs) poses considerable challenges for dynamic stability and energy management, primarily due to their variable and uncertain inertia and damping characteristics. Unlike conventional synchronous generators (SGs), IBRs exhibit complex nonlinear behaviour, complicating both mathematical modelling and real-time control. Battery energy storage systems (BESS) are essential for ensuring grid stability, operational efficiency and flexibility. Nevertheless, dynamic-aware energy management of BESS remains insufficiently explored, with current approaches often lacking adaptability to uncertainty and real-time requirements. This paper proposes a two-step hybrid framework combining quantum particle swarm optimisation (Q-PSO) with deep reinforcement learning (DRL). In the first step, Q-PSO efficiently generates an initial solution, significantly reducing computational demands. Subsequently, DRL dynamically refines this solution, effectively managing real-time uncertainties linked to inertia and damping variations. The proposed method addresses the non-Markovian nature of MG dynamics by constraining the DRL action space using the Q-PSO-derived solution, thereby alleviating the curse of dimensionality and enhancing training stability. Furthermore, dynamic constraints on frequency deviations and the rate of change of frequency (RoCoF) are incorporated to maintain robust grid stability during transients. Extensive simulations demonstrate that the proposed dynamic-aware energy management system (EMS) achieves economic efficiency improvements of 52.2% compared to Q-PSO alone and 22.7% compared to DQN alone. Additionally, BESS charging efficiency improves by 39.6% and 22.7%, whereas discharging efficiency increases by 38.3% and 28.25%, respectively, against the same benchmarks.
When multiple grid-forming and grid-following converters operate within an offshore energy system (OES), dynamic interactions among them can lead to poorly damped oscillations and potential instability. This paper proposes a computationally efficient reduced-order state-space modelling framework for small-signal stability and parametric sensitivity analysis of large-scale OESs. The approach replaces complex full-order analytical modelling with a practical MATLAB/Simulink-based linearisation procedure, enabling tractable stability assessment of systems comprising multiple wind plants, electrolysers, HVDC links and network components. The reduced-order model preserves the dominant dynamics while significantly decreasing the number of states, thereby improving computational efficiency for eigenvalue and sensitivity analyses. Using the linearised model, the influence of key control parameters is systematically quantified to provide explicit guidance for controller tuning and damping improvement. The accuracy of the proposed model is validated through comparison with detailed electromagnetic transient (EMT) simulations in PSCAD/EMTDC, demonstrating close agreement in dynamic responses and stability characteristics.
The rise of large-scale renewables has exacerbated frequency instability, revealing the limits of conventional frequency regulation frameworks in controlling deviations and incentivising fast-acting units (FAUs). Although the mileage-based payment framework promotes FAUs' participation in automatic generation control (AGC) services, efficient instruction dispatch (ID) across an increasing number of heterogeneous AGC units remains challenging. The existing mileage-based payment framework lacks validation under intermittent generation or generation outages scenarios. This work proposes a data-driven ID framework with a modified payment scheme, adding a penalty term for refining mileage calculation to handle intermittent generation or generation outages during AGC operation. The framework uses a multihead attention-based encoder-decoder model, where the encoder extracts latent features and the decoder predicts unit-specific instructions. Attention mechanism improves accuracy by prioritising critical features, whereas L2 normalisation, dropout and k-fold cross-validation enhance models' robustness under unforeseen scenarios. The model aggregates the flexibility of multiple FAUs into a single entity, termed the FAU aggregator. Trained on a synthetic dataset generated from the evolutionary optimisation-based ID framework and validation on an interconnected system accounting for disturbance due to intermittent renewable energy sources' output, FAU variability and stochastic communication effects. The results demonstrate a reduction in both frequency deviation and area control error in comparison with other ID frameworks.
Systemwide traction power outages in mass rapid transit systems (MRTSs) can cause trains to stop between stations, leading to service disruption and urgent passenger evacuation requirements. Onboard emergency self-traction systems (ESTSs) can mitigate this risk by enabling a stranded train to reach the nearest station at low speed. However, installing and maintaining ESTS on every train can increase implementation and lifecycle costs, motivating approaches that can ensure evacuation with partial ESTS deployment. This article introduces a power-sharing framework for ESTS in an MRTS. The primary aim is to reduce the number of ESTS units needed by facilitating power sharing between trains equipped with ESTS and those without it, while guaranteeing that all trains can safely and quickly evacuate passengers to the nearest station during a systemwide power outage. This study examines the Bangkok Mass Transit System (BTS) Skytrain Silom Line (dark green line) in Thailand, through computer simulation parameterised by real-world line and operational data, to evaluate the viability of this concept. It analyses variables such as the quantity of trains outfitted with ESTS, the distance from the train to the station where it halted, particular train specifications and additional pertinent aspects. The results reveal that ESTS can efficiently provide backup power to equipped trains and share surplus energy with other trains within the sharing radius with a success rate of 72.41%. This underscores the viability of ESTS operations and aids in the advancement of a more sustainable and efficient MRTS, despite the inherent constraints of ESTS.
In the small-signal synchronisation stability analysis of grid-following converters (GFL), the detailed interactions among multi-loop control mechanisms are often oversimplified or neglected, leading to significant modelling inaccuracies and potential stability assessment failures. To address this issue, this paper presents a motion-equation-based model for GFL converter that comprehensively incorporates multi-loop control dynamics, including the current control loop, AC voltage control loop, DC-link voltage control loop and PLL. Based on the proposed model, the coupling characteristics of multi-loop control are explicitly analysed and visualised in terms of system damping. The analysis reveals that low-frequency oscillations in GFL converters can be attributed to unintended negative damping induced by multi-loop control couplings. To mitigate this issue, an improved additional damping control (IADC) strategy is proposed, which enhances the synchronisation stability via effectively compensating for the negative damping caused by multi-loop interactions. Finally, the simulation and experiment studies are presented to validate the proposed analysis results and the effectiveness of the IADC strategy.
Accurate long-sequence net load forecasting is essential for reliable grid operation and renewable integration, yet it remains challenging under quasi-periodicity, sharp weather-driven variability and long-range error accumulation. We propose HarmoNet, an end-to-end dual-domain architecture for long-horizon deterministic and interval forecasting. HarmoNet (i) encodes coupled low/high-frequency representations with multi-scale temporal signals, (ii) integrates local pattern modelling and global dependency learning via a hybrid convolutional-transformer block and (iii) aggregates horizon-wide representations to mitigate drift in long-range prediction. Uncertainty is estimated with quantile regression (10%-50%-90%). We evaluate HarmoNet on four-year hourly net-load datasets from Belgium, Bulgaria and Italy (2016-2019) derived from the Open Power System Data platform, using eight exogenous meteorological covariates, over horizons of 96/192/336/720 h. Relative to the strongest baseline per dataset-horizon setting, HarmoNet reduces MAE by 14.2% on average (up to 22.5% on Italy at 720 h) and achieves average reductions of 28.4% in the Winkler score and 13.0% in pinball loss. Under deployment-oriented stress tests spanning high-volatility, peak-spike and steep-ramp windows, HarmoNet attains the best deterministic accuracy in 27/36 windows and the best probabilistic performance in 32/36 windows, indicating robust and deployment-friendly long-horizon forecasting.
Large-scale electric vehicles (EVs) integration demand coordinated power-transportation systems to optimise charging loads and grid flexibility, whereas existing strategies inadequately address range anxiety, peak demands and photovoltaic (PV) operational uncertainties. To address these issues, this work proposes a power-transportation coordination model integrating flexible swapping. The model strategically guides EV users towards swapping stations by embedding bounded rationality parameters within a stochastic user equilibrium framework, mitigating anxiety while balancing charging-swapping loads. This work also develops a distributionally robust optimisation framework to address PV uncertainty, enhancing resilience against generation fluctuations. Numerical simulations demonstrate that battery swapping integration significantly reduces distribution network peak loads and shortens user waiting times compared to charging-only approaches. The proposed method outperforms deterministic models by reducing conservativeness and aligning more closely with real-world operational dynamics, validating its efficacy in harmonising user behaviour and grid constraints under uncertainty.
Hybrid AC/DC microgrids that integrate photovoltaic, wind, battery and hydrogen energy systems are prone to DC-bus voltage fluctuations because of their low inertia and converter-based operation. This paper proposes a robust backstepping nonsingular fast terminal integral sliding mode controller that incorporates a virtual capacitor and a fractional-power reaching law to emulate synthetic inertia and improve transient damping. The controller coordinates energy exchange among distributed generation units and ensures precise DC-bus voltage regulation while managing bidirectional power transfer between AC and DC subgrids. An adaptive neuro-fuzzy inference system automatically tunes the controller gains in real time, and system stability is rigorously established through control Lyapunov functions. A detailed MATLAB/Simulink model, comprising PV, PMSG-based wind turbine, battery storage, electrolyser and PEM fuel cell, implements ANN-based MPPT to maximise renewable energy harvesting. The BNFTISMC is evaluated in three case studies against two benchmark controllers: the enhanced integral terminal SMC and the enhanced nonsingular terminal SMC. Under severe disturbances and varying load conditions, the proposed controller cuts overshoot by 75%-100%, reduces rise time by 58%-80% and completely eliminates mean absolute and mean squared errors. Processor-in-the-loop testing confirms zero steady-state error, whereas converter efficiency reaches 95.78% compared with 69.21% for the reference designs, demonstrating improved DC-bus voltage regulation, enhanced microgrid reliability and efficient real-time operation.
The rapid integration of renewable energy generators (REGs) alongside flexibility resources into the grid introduces significant data complexities, alongside increased unpredictability and variability within the distribution network. Reasonable allocation of flexibility resources can improve the utilisation of REGs and the flexibility of the distribution network. Addressing these challenges, an innovative two-stage model for distribution network expansion planning is proposed in this paper. The model uniquely combines the construction of new substations, the extension of existing lines, the integration of soft open points (SOPs) and strategic siting and sizing of energy storage systems (ESSs). Specifically, considering that SOPs can quickly regulate node voltages, thereby proposing flexibility zones as novel metrics to gauge the adaptability of the distribution network. Furthermore, a multivariate copula function is employed to delineate the correlation between REGs outputs and loads. This approach, devoid of reliance on extensive historical datasets, leverages discrete scenario ambiguity sets alongside norm theory to enhance scenario generation. To navigate the intricacies of the proposed model, a structured solution methodology encapsulated in a three-level process is developed. Validation through numerical simulations on the modified Portugal 54-bus system underscores the robustness and practicality of the methodology and solution framework in facilitating informed decision-making.
