Driven by global climate change and the ongoing energy transition, the coupling between power supply capabilities and meteorological factors has become increasingly significant. In the long term, accurately quantifying renewable energy generation under the influence of climate change is crucial for the development of sustainable power systems. However, due to interdisciplinary differences in data requirements, climate data often lacks the hourly resolution and fail to capture short-term uncertainty of renewable energy resources. To address this limitation, a super-resolution recurrent diffusion model (SRDM) has been developed to enhance the temporal resolution of climate data and model the short-term uncertainty. The SRDM integrates a pre-trained decoder and a denoising network within a recurrent coupling mechanism to generate long-term, high-resolution climate data. The high-resolution climate data is then converted into power generation using the mechanism model, enabling the simulation of wind and photovoltaic power generation on future long-term scales. Case studies were conducted in the Ejina region of Inner Mongolia, China, using coupled model intercomparison project (CMIP6) data to simulation three representative climate pathways. The results demonstrate that SRDM outperforms existing generative models in generating super-resolution climate data. Furthermore, the research highlights the estimation biases introduced when low-resolution climate data is used for power conversion.

With the growing integration of wind power and increasing frequency of extreme weather events, accurately quantifying uncertainty is essential for reliable power system operation. However, the complex spatiotemporal dependencies of wind power often cause mismatches in scenario generation, limiting the effectiveness of stochastic optimization. To address these challenges, this study proposes a dynamic spatiotemporal aware latent diffusion mode (DSTALDM) which integrates deterministic forecast-based pretrained embedding networks and a VAE-based latent space mapping network. An embedding-based dynamic graph-aware module is employed to extract the expected dynamic graph from numerical weather prediction (NWP), enhancing the accuracy and interpretability of spatiotemporal correlation modeling. The latent space mapping module transforms the scenario generation process into a latent space, thereby effectively reducing the computational complexity of the denoising model. Comparative studies with state-of-the-art models demonstrate that DSTALDM can generate high-quality joint scenarios while accurately capturing spatiotemporal correlations. Experimental results confirm that DSTALDM delivers superior performance in uncertainty quantification and computational complexity.

The increasing penetration of renewable energy has strengthened the coupling between power systems and weather conditions, making annual hourly joint wind-photovoltaic (PV) power scenario generation essential for adequacy assessment and flexibility planning. However, existing methods are constrained by limited training samples, weak physical consistency, and difficulties in modeling long sequences. To address these challenges, a hierarchical conditional flow matching framework (H-CFM) is proposed for annual hourly joint wind-PV power scenario generation. The framework formulates the task as a two-stage process, consisting of hierarchical meteorological sequence generation and deterministic physical mapping. A year-solar term-day-hour hierarchical structure is introduced, where solar ecliptic longitude defines a circular manifold prior to characterize solar-term phases and encode seasonal periodicity. An overlap conditioning mechanism with boundary constraints is developed to improve continuity across adjacent days. The proposed method is validated using ERA5 data from Inner Mongolia, China. Results show that H-CFM outperforms benchmark models in terms of probabilistic distribution, temporal characteristics, and extreme event representation, achieving improved reproduction of annual energy levels, seasonal resource patterns, and extreme scenarios. Ablation studies further demonstrate the effectiveness of the solar-term phase prior and the overlap conditioning mechanism in capturing seasonal structure and maintaining long-horizon continuity.

ABSTRACT Reliable hydro‐wind‐PV joint scenarios under extreme drought are essential for assessing renewable power adequacy and operational flexibility in renewable‐dominant power systems. However, existing studies on joint hydro‐wind‐PV modelling often lack detailed weather‐process representation, adequate temporal resolution and consideration of cumulative drought effects, limiting their applicability to monthly scenario generation at hourly resolution during extreme droughts. To address this issue, this study develops a multi‐resolution variable‐alignment conditional diffusion model (TXDM) for monthly hydro‐wind‐PV joint scenario generation under extreme drought conditions. Unlike conventional methods, the proposed framework incorporates multi‐resolution weather indices to characterise drought‐related impacts on monthly hourly renewable power generation. Case studies in Sichuan demonstrate that TXDM improves deterministic and probabilistic accuracy, preserves spatio‐temporal dependence and generates credible few‐shot extreme drought scenarios. The generated scenarios provide reliable high‐resolution inputs for renewable power adequacy assessment, reserve estimation, storage dispatch and hydro‐wind‐PV coordination during extreme droughts.

With the increasingly severe situation of global climate governance and the gradual progress of the “dual-carbon” goal, the promotion of low-carbon transformation of the energy structure will become an important direction of the future national development strategy, and electric-hydrogen-carbon synergistic development mode is an important program to solve the above problems. Aiming at the above problems, this paper proposes a coupled electric-hydrogen-carbon system planning model based on energy flow-material flow synergy. Firstly, the typical synergistic development mode of electric-hydrogen-carbon coupled system is analyzed, and based on this, the theoretical framework of the whole link flow of energy-material flow is constructed, which reveals the energy flow process and dynamic balance mechanism of the system; secondly, by combining with the operation mechanism of the key equipment units, the planning model of the system is proposed by comprehensively considering the energy-material balance of the system; Finally, with the help of PyOptInterface, the model is solved to obtain the results of system capacity optimization, and the simulation curves of typical daily production are drawn, based on which the analysis of system economy, operation characteristics and planning feasibility is carried out.

In recent years, extreme weather events have occurred more frequently. The resulting equipment failure, renewable energy extreme output, and other extreme operation scenarios affect the smooth operation of power grids. The occurrence probability of extreme operation scenarios is small, and the occurrence frequency in historical operation data is low, which affects the modeling accuracy for scenario generation. Meanwhile, extreme operation scenarios in the form of discrete temporal data lack corresponding modeling methods. Therefore, this paper proposes a definition and generation framework for extreme power grid operation scenarios triggered by extreme weather events. Extreme operation scenario expansion is realized based on the sequential Monte Carlo sampling method and the distribution shifting algorithm. To generate equipment failure scenarios in discrete temporal data form and extreme output scenarios in continuous temporal data form for renewable energy, a Gumbel-Softmax variational autoencoder and an extreme conditional generative adversarial network are respectively proposed. Numerical examples show that the proposed models can effectively overcome limitations related to insufficient historical extreme data and discrete extreme scenario training. Additionally, they can generate improved-quality equipment failure scenarios and renewable energy extreme output scenarios and provide scenario support for power grid planning and operation.

As the penetration rate of distributed photovoltaics (PVs) increases, behind-the-meter (PVs) amplify the impact of meteorological factors on net load predictions. Public weather forecasts (PWFs) can serve as an alternative meteorological covariate to ensure prediction accuracy when numerical weather prediction (NWP) data are unavailable owing to high costs or data delays. PWFs are freely accessible to the public and significantly influence resident behavior. However, they provide limited information and have inherent heterogeneity with historical net load data in terms of data structure and features, posing challenges in feature extraction and fusion. To address these challenges, this study proposes a day-ahead net load prediction model cross-modal contrastive learning network (CCL-Net) that incorporates PWFs. The CCL module employs contrastive learning to minimize the gap between the representations of temporal meteorological data and PWF features, enriching the semantic features of PWFs. Considering the heterogeneity in features and data structure between future weather forecasts and historical net load data, CCL-Net incorporates a cross-modal fusion block to fuse both features, enhancing prediction accuracy. The proposed model is compared with benchmark methods, demonstrating superior performance. In addition, the analysis and ablation experiments of cross-modal technologies validate the rationality and superiority of the proposed approach.

The frequent occurrence of extreme weather events attributed to global climate change presents challenges to the supply-demand balance in power systems with high renewable energy integration. Modeling renewable energy scenarios can effectively guide power system operation and planning. However, obtaining diverse extreme renewable energy scenarios is challenging due to the limited availability of renewable power dataset resources. To tackle this issue, the pattern-guided diffusion model (PGDM) is proposed for controllable renewable energy scenario generation. Initially, we define the scenario pattern features associated with wind and solar power generation. A contrastive pre-training model is used to learn representations of renewable energy scenarios, aiding the downstream model in understanding scenario pattern features. Subsequently, a perceptual variational autoencoder is used to map the high-dimensional scenario into a low-dimensional latent space, thus reducing the computational burden. This is combined with a conditional latent diffusion model to achieve controllable scenario generation. The renewable power dataset from Belgian transmission operator Elia was used for the case study. The proposed PGDM demonstrated lower errors in controllable scenario generation and exhibited excellent generalization performance in low probability (few-shot) and novel (zero-shot) pattern scenario generation.

As the proportion of new energy generation in the power system increases fastly, the impact of the source side on the power system continues to rise, resulting in complex spatiotemporal uncertainty of source and load, which brings difficulties to the operation and dispatch of the power system. Source-load scenario generation technology quantifies and models the uncertainties of both the generation side and the load side, providing scientific decision support for the planning, operation, and scheduling of power systems. This paper first proposes a method for generating controllable source-load scenarios based on the diffusion model, combining the Contrastive Language-Image Pre-training model and the Variational Auto-Encoder to enrich scenario information, enriching the randomness of the generated scenarios. Secondly, based on speculative sampling acceleration, a diffusion model acceleration method for denoising is designed to save computational costs in scenario generation. Verification through case studies shows that the proposed scenario generation method effectively achieves controllable generation of four different scenarios. The denoising process acceleration method based on speculative sampling can achieve a 27.76% acceleration with a 7.09% loss in quality.

Under the ‘dual-carbon’ objective, the large energy base in the west is dominated by renewable energy sources, which have a high degree of uncertainty compared with conventional thermal and hydropower. In the current system evaluation, it is mainly dominated by single-level value evaluation, and lacks the whole-link value evaluation that considers system flexibility, safety and other factors. This paper takes the Gobi Desert as the research object. Firstly, an evaluation method for quantifying the value of renewable systems is proposed from a multidimensional perspective, and then an optimisation model based on stochastic time series production simulation and an evaluation model for quantifying the value of the system are constructed with respect to the stochastic uncertainty of renewable output and the source-network-storage operating characteristics of the system under different scenarios. Finally, the results of system value quantification show that this method is able to adequately reflect the generation capacity and transmission scale of renewable bases under the consideration of renewable sources uncertainty, which verifies the feasibility of the method.

The form of wind power forecast results can take various, including points, quantiles, intervals, or probability. Deterministic forecasting provides decision-makers with an intuitive representation, while probabilistic forecasting incorporates more uncertainty features. Regardless of the chosen form, the fundamental goal is to forecast the future wind power. This study aims to clarify the relationship between different forms of forecast results and proposes the shared multi-task network (SMTN) for short-term wind power forecasting. The SMTN maps the meteorological features to a shared latent space. Unlike existing forecasting models that only provide a single form of forecast results, SMTN uses multiple forecasting task modules to transform the latent space features into various form results to meet different forecasting demands. The results of the case study demonstrate that SMTN can cooperatively optimize multiple forecasting tasks and exhibit superior performance across various forecast results.

Stochastic chronological operation simulation (S-COS) is essential for analysing long-term supply-demand balance in power systems with high penetration of renewable energy. However, conventional methods face significant computational challenges due to inter-temporal constraints and numerous binary variables in multi-scenario annual simulations. This paper presents a novel data-driven, surrogate-assisted approach to accelerate year-round, scenario-based operation simulations. The proposed approach employs a temporal decomposition method to decouple the annual stochastic optimization problem into an inter-day scheduling model and multiple intra-day power dispatch models, which are efficiently solved using a data-driven surrogate model. Case studies on modified six-bus and IEEE 118-bus systems demonstrate the approach's adaptability to various scenarios and its scalability across different network scales. Results show that this approach improves computational efficiency by at least 100 times compared to conventional methods, with even faster performance in larger systems. It also maintains high accuracy, achieving an average annual operating cost error of only 1.35% relative to benchmarks.

Flexible load resources on the demand side are characterized by diverse types, small capacity, and wide distribution. Since these load resources cannot be individually invoked at the system level, it is becoming increasingly important to consider how to establish heterogeneous flexible load aggregates and utilize their adjustable characteristics. In response to the above problems, this paper establishes mechanistic models for three specific heterogeneous load resources, uses the basic homothetic polytope to approximate the original feasible region of loads, and aggregates the feasible region of heterogeneous load resources using the Minkowski sum, while ensuring the aggregation speed and accuracy. Finally, for scenarios of system load shedding and high wind and solar penetration, a demand response strategy on the load side is introduced, and the role of this method in system power supply guarantee and new energy consumption is illustrated through a numerical example.

With the rapid development of large wind power and photovoltaic base in desert, Gobi and desert areas in the west of our country, wind power and photovoltaic energy delivery become an important problem to be solved urgently. In this paper, multi-stage transmission-storage cooperative planning model is carried out for the real power grid, and a comprehensive evaluation indicators system is proposed from four dimensions of economy, reliability, cleanliness and flexibility. The results show that developing renewable energy is helpful to improve the operating performance of the power grid and enhance the comprehensive benefits of the system.

To reduce the difficulty and enhance the enthusiasm of private-owned electric vehicles (EVs) to participate in frequency regulation ancillary service market (FRASM), a decision aid model (DAM) is proposed. This paper presents three options for EV participating in FRASM, i. e., the base mode (BM), unidirectional charging mode (UCM), and bidirectional charging/discharging mode (BCDM), based on a reasonable simplification of users' participating willingness. In BM, individual EVs will not be involved in FRASM, and DAM will assist users to set the optimal charging schemes based on travel plans under the time-of-use (TOU) price. UCM and BCDM are two modes in which EVs can take part in FRASM. DAM can assist EV users to create their quotation plan, which includes hourly upper and lower reserve capabilities and regulation market mileage prices. In UCM and BCDM, the difference is that only the charging rate can be adjusted in the UCM, and the EVs in BCDM can not only charge but also discharge if necessary. DAM can estimate the expected revenue of all three modes, and EV users can make the final decision based on their preferences. Simulation results indicate that all the three modes of DAM can reduce the cost, while BCDM can get the maximum expected revenue.

Power systems with high penetration of renewable energy contain various uncertainties.Scenario-based optimization problems need a large number of discrete scenarios to obtain a reliableapproximation for the probabilistic model. It is important to choose typicalscenarios and ease the computational burden. This paper presents a scenario reduction networkmodel based on Wasserstein distance. Entropy regularization is used totransform the scenario reduction problem into an unconstrained problem. Throughan explicit neural network structure design, the output of the scenarioreduction network corresponds to Sinkhorn distance function. The scenarioreduction network can generate the typical scenario set through unsupervisedlearning training. An efficient algorithm is proposed for continuous/discrete scenarioreduction. The superiority of thescenario reduction network model is verified through case studies. Thenumerical results highlight high accuracy and computational efficiency of theproposed model over state-of-the-art model making it an ideal candidate for large-scale scenarioreduction problems

As the world moves towards an increased penetration of renewable energy, the importance of photovoltaic systems (PV) cannot be denied as these networks are deployed at rooftops of active buildings and provide power to local loads. The performance of this system is dependent on the MPPT algorithm and step-up DC-DC converter for maximum power utilization at varying operating conditions. This paper proposes an ANN-based modified flyback converter scheme for linking PV panels to dc loads. In the proposed design, the secondary side diode is replaced by a MOSFET which offers several advantages such as reduced voltage stress, improved efficiency and positive output voltage. Moreover, the large voltage and current spikes at primary side switch is reduced which increases the lifetime of MOSFET. The system performance is analyzed in Simulink through variations in load, temperature and irradiance. A comparative analysis with P&O-PSO controller shows that the proposed control and topology gives an improved transient response, making it an ideal choice for future PV systems.

Electric vehicles (EVs) are high-quality flexible resources to provide frequency regulation services. However, the EV real-time participation in ancillary service market for frequency regulation (FRASM) under the existing aggregation framework faces data privacy and interest conflicts problems. To solve these issues, this paper proposes the fully-decentralized aggregator (FDA) as a trusted and non-profit agent to replace the traditional aggregator and help EVs interact with FRASM in a decentralized manner. Then a fully decentralized aggregation framework based on consortium blockchain is constructed to protect the interaction data security during frequency regulation process. Furthermore, the inference network is introduced so that EVs can learn better performance and optimize regulation mileage without sharing their private data. This paper proposes an inference network based multi-agent soft actor-critic (IN-MASAC) method that realizes the decentralized interaction between EVs and FRASM and considers multiple user preferences. The case studies demonstrate the efficiency and scalability of the fully decentralized aggregation framework and IN-MASAC method.

Driven by global climate change and the ongoing energy transition, the coupling between power supply capabilities and meteorological factors has become increasingly significant. Over the long term, accurately quantifying the power generation of renewable energy under the influence of climate change is essential for the development of sustainable power systems. However, due to interdisciplinary differences in data requirements, climate data often lacks the necessary hourly resolution to capture the short-term variability and uncertainties of renewable energy resources. To address this limitation, a super-resolution recurrent diffusion model (SRDM) has been developed to enhance the temporal resolution of climate data and model the short-term uncertainty. The SRDM incorporates a pre-trained decoder and a denoising network, that generates long-term, high-resolution climate data through a recurrent coupling mechanism. The high-resolution climate data is then converted into power value using the mechanism model, enabling the simulation of wind and photovoltaic (PV) power generation on future long-term scales. Case studies were conducted in the Ejina region of Inner Mongolia, China, using fifth-generation reanalysis (ERA5) and coupled model intercomparison project (CMIP6) data under two climate pathways: SSP126 and SSP585. The results demonstrate that the SRDM outperforms existing generative models in generating super-resolution climate data. Furthermore, the research highlights the estimation biases introduced when low-resolution climate data is used for power conversion.