IET Renewable Power Generation最新文献

Very short-term wind power forecasting (WPF) is an indispensable component of wind energy supply in modern power systems. To improve forecasting accuracy, this paper presents a cyber-resilient online WPF method based on model selection, called online lifelong learning (LL). In this study, the WPF is carried out in two scenarios for four regions of Iran with different geographical and climatic characteristics. In each scenario, the results of each model are evaluated using different evaluation indicators. The first scenario advances the forecasting procedure under the situation in which that the data for each region is clean and without manipulation. The results presented in this scenario show that the online LL model is superior to conventional models, with the minimum mean squared error (MSE) values of 26.85, 32.25, 33.58, and 34.58, for the regions of Tabriz, Mahshahr, Lootek, and Bojnord, respectively. The second scenario advances the WPF process under cyber-attacks. The scaling attack is implemented as a smart variant of a false data injection attack (FDIA), targeting the input variables of wind speed and wind direction. In this scenario, as well, it can be seen that the online LL model advanced the WPF process with the least possible error and was superior to other models.

Protection of DC microgrids is essential to ensure the safe and reliable operation of the whole power system. This paper introduces a simple and effective methodology for fault detection and classification in DC microgrids. The proposed method transforms the measured current from only one end of the protected line into a novel modal current. Then, it employs the second-order derivative of the Savitzky-Golay filter to introduce a novel fault detection index for discriminating the faults in a wide range of DC microgrid conditions. Accordingly, the index is first compared with a threshold to diagnose the fault, enabling the fault classification unit to start, which checks the index value for three consecutive samples. The application of the proposed method provides a communication-less protective scheme for various operation modes of DC microgrids, which can be easily implemented in real-world scenarios due to its simplicity and speed. Different simulation studies have been conducted using MATLAB/Simulink to evaluate the performance of the approach, considering various sources and loads. The proposed technique accurately distinguishes faulty conditions across lots of challenging scenarios. Additionally, the technique is verified in a small-scale experimental lab test. The results robustly demonstrate the rapid response and accuracy of the proposed method.

This paper presents a scenario-based stochastic optimization framework for mobile hydrogen energy storage systems (HESS) integrated with renewable generation and demand response. The model captures relocation dynamics across multiple buses and incorporates financial risk using conditional value-at-risk (CVaR). Key features include scenario reduction for wind and solar uncertainties, downside risk constraints, and temporal coupling between relocation and energy states. The objective function balances expected operating cost and risk exposure, enabling robust dispatch under uncertainty. Simulations on the IEEE 33-bus system show that mobile HESS units outperform stationary hydrogen configurations, achieving up to 8.4% cost reduction despite relocation penalties. Benchmarking against mobile battery systems reveals that battery-based setups yield lower operating costs—approximately 9.9% less—due to higher round-trip efficiency. However, the spatial and long-duration flexibility of HESS remains a strategic advantage, especially under extended congestion. The proposed framework advances mobile energy storage modelling by integrating mobility, uncertainty management, and risk-aware dispatch. It offers a scalable solution for resilient energy planning and can be extended to include load-dependent efficiency, probabilistic demand response, and multi-carrier energy coordination.

Energy management systems (EMSs) play a pivotal role in modern power systems by orchestrating resource optimization, cost reduction, and resilience enhancement amid increasing renewable penetration and decentralized energy paradigms. This paper introduces a novel three-tiered hierarchical EMS framework for multi-microgrid (MMG) networks, designed to address the dual challenges of dynamic energy coordination and operational efficiency in heterogeneous grid environments. The architecture consists of a distributed control layer that autonomously regulates distributed energy resources and energy storage systems within individual MGs through adaptive droop control. The optimization planning layer incorporates advanced day-ahead stochastic scheduling algorithms and introduces a novel quantum-inspired adaptive balancing strategy, which dynamically adjusts the obtained day-ahead schedules to mitigate forecasting errors and enhance operational robustness under uncertainty within an individual MG. The integrated grid management layer features a decentralized, privacy-preserving peer-to-peer energy exchange mechanism that coordinates surplus and deficit energy profiles across MMGs without compromising data privacy. The layer only required the energy access/deficit information along with the buy/sell price to optimally manage the energy transfer within MGs. Comprehensive simulations under heterogeneous forecast uncertainty scenarios show the effectiveness of the proposed EMS to optimize energy flows and cost efficiency while maintaining grid stability under various type of forecast errors. Moreover, comparisons with existing EMSs validate the optimal performance of the proposed method.

The energy function method developed from Lyapunov theory is a typical direct method to evaluate the transient synchronous stability of traditional power grids. Different from grid-forming (GFM) converters mimicking characteristics of synchronous generators (SGs), grid-following (GFL) converters behave like current sources. The integration of multiple GFL converters poses challenges of constructing energy functions for transient synchronous stability assessment, which is analysed in this paper. In addition, the damping terms of GFL converters are highly dependent on phase angles. Due to the indefinite damping terms of GFL converters, assessment results might be optimistic or conservative, if neglecting damping terms to construct energy functions. Thus, based on manifold theory, an alternative without constructing energy functions is proposed for transient synchronous stability assessment of multi-converter-based power grids. Firstly, a hybrid modelling with equivalent voltage and current sources is built to accommodate different network topologies and types of converters. Then, benefiting from the first-order approximation of manifolds and the geometrical analysis of the region of attraction (ROA) boundary, a criterion designed by manifold theory is proposed and employed for transient synchronous stability evaluation. In addition, the transient synchronous stability margin k_relζ is defined to avoid optimistic results. Finally, modified IEEE 4-generator 11-bus and IEEE 10-generator 39-bus systems are built on the MATLAB/Simulink platform to validate the effectiveness of the method.

Estimating electricity load is paramount in the strategic planning and operation of power systems, ensuring the efficient and sustainable operation of contemporary electricity supply networks. This study proposes a Stacked Ensemble (SE) model to address these concerns and enhance short-term net load forecasting (STNLF) by combining bidirectional long short-term memory, support vector machine, and random forest (RF) models. An autoencoder is also used to optimise the input data, which extracts critical features from the data and enhances predictive accuracy. This SE model is designed to incorporate Explainable Artificial Intelligence (XAI) as an integral part, providing users detailed insights into input variable influences such as temperature, solar radiation, and wind speed, all interpretable and relevant, thus improving the transparency of the forecasting process. The proposed method is evaluated against real historical time-series data in Austria from 2018 and 2019 regarding hourly measurements for temperature, solar energy, wind turbine generation, and actual loads. Results indicate that the SE model can capture complex data patterns, whereby XAI features actionable insights and proves a reasonable concurrent behaviour prediction with boundary-specific realities. Remarkably, the proposed SE model surpassed traditional machine learning approaches on many performance metrics while showcasing robustness and reliability in increasingly improving STNLF accuracy.

Power systems with high renewable energy (RE) penetration are increasingly vulnerable to frequency instabilities caused by the stochastic nature of RE sources and load-generation imbalances. Additional challenges such as communication delays in phasor measurement units (PMUs), nonlinearities like governor deadband (GDB) and generation rate constraints (GRC) and cross-coupling between load frequency control (LFC) and automatic voltage regulation (AVR) loops further impair system stability. This paper proposes an efficient control framework that combines a fractional-order tilt-integral-derivative (TID) controller, optimised using the salp swarm algorithm (SSA), with a continuous-time disturbance observer (DO) for real-time estimation and rejection of aggregate disturbances. Unlike traditional approaches that rely solely on complex feedback controllers, the integration of the DO improves robustness against nonlinearities and disturbances while maintaining a low-order control structure. The controller parameters are optimised over a nonlinear time-delay model of a two-area hybrid power system, considering both inter-area coupling and dynamic uncertainties, using SSA, genetic algorithm (GA), particle swarm optimisation (PSO) and Archimedes optimisation algorithms (AOA). The proposed SSA-TID-DO controller achieved the lowest objective function value (IAE = 0.2367), clearly outperforming SSA-TID (0.4082), SSA-PID (0.6413), PSO-TID (0.4434), GA-TID (0.4515) and AOA-TID (0.4435) controllers. The results highlight the effectiveness and practicality of the proposed strategy for robust frequency regulation in RE-dominated power systems.

The growth of penetration of wind sources in the power system along with demand response programmes (DRPs) has led the electricity market and decision-making process in a complicated pathway. The inherent uncertainty of demand behaviour and wind turbine generation increases the complexity of decision-making as well. In this research, a new economic index is introduced to assess the feasibility of the mathematical modelling of the power system's economic performance from the consumers' and producers' social welfare perspective. The uncertainty is represented using the empirical cumulative distribution function along with the Monte Carlo method, and a probabilistic market analysis is conducted. The role of the spot market along with different demand response modelling and uncertainty in market clearing is addressed, and the functionality of the model in a comprehensive survey is demonstrated. Real data on an 8-bus test system is surveyed, and the efficiency of this approach to discovering the locational marginal prices is cleared to empower the independent system operator to monitor the market precisely. The results illustrate that the producer's and customer's costs reduced significantly, and the integration of participants in the spot market increases their profit and income.

The stable operation of DC microgrids requires robust secondary control strategies capable of ensuring accurate power sharing and bus voltage regulation, even under adverse conditions. Traditional approaches are generally based on proportional sharing according to the nominal capacities of the converters, which limits operational flexibility in scenarios requiring critical load prioritisation or maintenance strategies. This paper proposes a distributed secondary control strategy that enables individualised power allocation among converters without compromising system stability. The effectiveness of the proposed approach is validated through simulations and practical experiments under different operating conditions, including load variations, disturbances, and communication failures, demonstrating its accuracy, robustness, and applicability.

This paper presents an isolated resonant high step-up DC-DC converter with ripple-free input current (RFIC). At the low-voltage side (LVS), an interleaved current-fed bridge is employed while its switches are operated with a constant duty cycle of 0.5 to achieve RFIC. By active clamp circuits, firstly, the zero voltage switching (ZVS) is ensured for LVS switches; secondly, the stored energy of the leakage inductance is recycled; and thirdly, the voltage stresses across the switches are clamped at the value of 2V_IN that is much lower than the high output voltage. The last item enables the implementation of low voltage rated switches with low ON-state resistance. Through an active voltage quadrupler (AVQ) at the high-voltage side (HVS) and a transformer, the voltage gain is significantly extended. The switches of the AVQ are responsible for voltage gain regulation and experience half of the output voltage during the OFF-state. Due to the provided resonance by the auxiliary inductor and multiplier capacitors, the currents of the body diodes of the HVS switches smoothly reach zero, diminishing the reverse recovery losses. The steady-state analysis of the proposed converter is presented in detail and is validated by the fabrication of a 400W 40–400 V prototype with 96.1% peak efficiency.