Accurate prediction of air conditioning (AC) loads at the feeder level is imperative for the effective implementation of targeted demand response and effective peak load management. However, it is challenged by data heterogeneity and scarcity, especially for new feeders. Traditional methods often fail under such conditions. This paper proposes a meta-learning-based Dynamic Hierarchical Forecasting Framework (DHFF) explicitly designed for efficient few-shot load forecasting. A core gating network dynamically fuses predictions from a global unified model and a feeder specific model, adapting based on context. The framework’s effectiveness was validated through dual testing scenarios. Outstanding performance was achieved in data-rich environments, with an RMSE of 0.0135, improving upon a strong LSTM baseline by nearly 80%. Furthermore, rigorous few-shot experiments confirmed its primary design goal: under extreme data scarcity (e.g., 5% data), DHFF demonstrated superior accuracy, improving up to 7.01% over standard approaches by intelligently leveraging generalized knowledge. These results validate DHFF as an adaptive, high-performing solution across both data-scarce and data-rich feeder forecasting scenarios.

Multi-terminal voltage-source-converter-based HVDC (VSC-MTDC) systems are increasingly used to integrate renewable energy and interconnect asynchronous AC grids, but conventional fixed-coefficient droop control cannot simultaneously limit DC-voltage deviations, reduce operating losses, and preserve converter power margins during operating-point switching. This paper hypothesizes that a rule-based fuzzy adjustment of the droop slope can provide smooth multi-objective coordination without inter-station communication. A dual Mamdani fuzzy controller is developed: one controller adjusts the weighting between loss-oriented and power-margin-oriented droop coefficients according to converter power margin, while the other introduces a voltage-deviation correction according to DC-bus voltage. The controller is implemented and verified in a five-terminal MMC-based VSC-MTDC model built in PSCAD/EMTDC, where simulation data are generated under heavy-load, light-load, and power-reference switching scenarios using specified line and converter parameters. Compared with conventional droop control, the proposed strategy improves power-margin utilization, reduces operating-point discontinuities, and raises the minimum DC voltage from 370.2 kV to 381.4 kV in the severe switching case. The results confirm that fuzzy-slope droop control can achieve smoother operating-point switching and better coordinated optimization among voltage stability, operating loss, and converter reserve margin.

Accurate air conditioning (AC) load prediction under extreme weather conditions is critical for power grid stability and energy management. While tree-based ensemble methods such as XGBoost, LightGBM, and Gradient Boosting have become the dominant paradigm in short-term load forecasting, their effectiveness for weather-sensitive AC load prediction—particularly during extreme weather events—remains insufficiently examined. This study presents a systematic comparison of four machine learning models—Support Vector Regression with RBF kernel (SVR-RBF), LightGBM, XGBoost, and Gradient Boosting—for daily AC load estimation conditioned on measured same-day meteorological and calendar features. Based on five years of processed load and weather data from a major city in Southwest China, we construct 31 features and evaluate model performance across four scenarios: normal days, weather-extreme days, high-load P85, and high-load P90. Hyperparameters are selected within the first four years by expanding-window validation, and the fifth year is held out for testing. Our results reveal that SVR-RBF achieves an overall R2 of 0.9772, substantially outperforming LightGBM (0.9091), XGBoost (0.8972), and Gradient Boosting (0.9094); paired moving-block bootstrap intervals for the tree-minus-SVR MAE differences exclude zero. The advantage of SVR-RBF is most pronounced under extreme conditions: on weather-extreme days, SVR-RBF attains R2=0.8316 versus R2=0.39 for the best tree model. This pattern is consistent with the smooth U-shaped temperature–load relationship captured by the RBF kernel. Additional trend-sensitivity analysis shows that annual load growth and target-level extrapolation also explain a substantial part of the tree-model degradation. Furthermore, we quantify the performance limitation of the restricted feature set: while normal-day estimation achieves R2=0.984, high-load P90 days reach R2=0.843. These findings support SVR-RBF as a strong baseline for measured-weather conditional AC load estimation while emphasizing that temporal shift and training-domain coverage must be considered when interpreting model differences. For energy-system applications, the lower errors under high-load and weather-extreme conditions are relevant to peak-demand assessment, reserve planning, and demand-side management.

To address the issue of inaccurate power flow calculations in the asymmetric coupling system of a power grid and traction network, this paper proposes a dynamic power flow calculation method for the “renewable energy–power grid–transportation network” asymmetric coupled system. First, by utilizing the asymmetric characteristics of the traction transformer, the dynamic asymmetric nodal admittance matrix for the “renewable energy–power grid–transportation network” coupled system is established, which facilitates the construction of the mixed power flow equations for the coupling of the power grid and transportation network. Next, when analyzing the asymmetric coupling system of renewable energy, power grid, and transportation network in mountainous areas, it is necessary to allocate the power of electric multiple units (EMUs) to the three-phase (A, B, C) power distribution. To address this, a three-phase power balancing strategy is proposed, incorporating both the single-phase loads of EMUs and the output of renewable energy sources. Thus, a three-phase power balance strategy is proposed, incorporating the single-phase load of traction load units and renewable energy output. Finally, a simulation study is conducted using a real system of a regional power grid and traction network as a case example, demonstrating the suitability and effectiveness of the proposed model.

Hydropower is a high-quality, clean energy source with low operating costs and the ability to facilitate peak regulation. To achieve the national strategic goal of clean energy transition, it is of great practical significance to conduct supply-demand balance analysis in hydropower-rich regions. This paper proposes a new method for hydropower output modeling and optimization scheduling in production simulation calculations. First, the traditional three-stage output curve model is used to solve the output arrangement of hydropower units. On this basis, daily scheduled power generation of hydropower unit is taken as the optimization variables, and optimization is performed considering the reservoir capacity constraints. The output of hydropower units is then rearranged using the three-stage output curve model according to the results of daily power generation. The effectiveness of the proposed method lies in its ability to consider reservoir capacity constraints, ensuring the correctness of the calculation results while maintaining the output scheduling principles of the original three-stage output curve model, thus achieving similar peak regulation effect as the original model. Case studies demonstrate the feasibility and effectiveness of the proposed method.

Air conditioning loads in power systems exhibit spatiotemporal heterogeneity across geographical regions, complicating accurate load forecasting. This study proposes a framework that integrates Deep Reinforcement Learning-guided DBSCAN (DRL-DBSCAN) clustering with a Graph Attention Network (GAT)-based Graph Neural Network to model spatial dependencies and temporal dynamics. Using meteorological features like temperature and humidity, the framework clusters geographical grids and applies GAT to capture spatial patterns. On a Pecan Street dataset of 25 households in Austin, the GAT with DRL-DBSCAN achieves a Test MSE of 0.0216 and MAE of 0.0884, outperforming K-Means (MSE: 0.0523, MAE: 0.1456), Hierarchical clustering (MSE: 0.0478, MAE: 0.1321), no-clustering (MSE: 0.0631, MAE: 0.1678), LSTM (MSE: 0.3259, MAE: 0.3442), Transformer (MSE: 0.6415, MAE: 0.4835), and MLP (MSE: 0.7269, MAE: 0.5240) baselines. This approach enhances forecasting accuracy for real-time grid management and energy efficiency in smart grids, though further refinement is needed for standardizing predicted load ranges.

During the implementation of active voltage support in wind farms, coordinating the operation of multiple wind turbines presents significant challenges. The dynamic response of the entire wind farm becomes complex during grid faults, making it difficult to achieve coordinated voltage support across different wind turbines. To address this, a coordination control strategy for doubly fed wind farms is here proposed which is based on Q-learning informed by the sensitivity of voltage. First, a method for calculating the voltage sensitivity of DFIG-based wind farms is introduced, utilizing the arbitrary polynomial chaos approach. Additionally, the operational constraints of wind farms are defined based on the average short-circuit ratio of reactive power. The voltage support characteristics of multi-machine wind farms under grid fault conditions are then thoroughly explored. Subsequently, an improved Q-learning algorithm is developed, based on the sensitivity of voltage. This algorithm aids in optimizing the control commands, thus enhancing the effectiveness of the voltage support system. Finally, adopting this voltage sensitivity as the basis for the coordinated control commands and applying the improved Q-learning algorithm as the implementation mechanism, a coordinated control strategy for active voltage support in DFIG-based wind farms is proposed. Simulation results demonstrate that the proposed control strategy can provide effective active voltage support during grid faults.

The highly uncertain and uncontrollable power output of renewable energy sources (RES), when integrated into power systems at high penetration levels, reduces system inertia and introduces uncertain changes in system structure, parameters, and frequency response characteristics. This renders traditional frequency regulation analysis methods and frequency response models inapplicable, lacking a generalized model to describe renewable energy’s participation in frequency regulation. Thus, this paper proposes a method where RES utilize suitable means to reduce load, thereby contributing to frequency regulation. Furthermore, employing Virtual Synchronous Machine (VSM) technology, these renewable energy units emulate the inertia and droop characteristics of Synchronous Generators (SG), enabling their equivalent modeling alongside traditional generators within a single-machine aggregate model. An SFR (System Frequency Response) model integrating renewable energy’s frequency regulation has been established. This model enables the analysis of the relationships between the system’s equivalent droop coefficient and the frequency nadir, nadir time, and quasi-steady-state point. Furthermore, the required equivalent droop coefficients are proposed for various sending-end system capacities and operating conditions. Finally, the model’s validity and accuracy are confirmed through a modified WSCC 4-machine 10-bus system, offering theoretical underpinnings for stable system operation and optimized operational planning.

As the penetration rate of renewable energy in the power grid continues to rise, the reserve criteria for traditional power grids dominated by synchronous generators (SGs) have difficulty meeting system frequency security requirements. This study proposes a frequency security-constrained optimization approach for the allocation of reserve capacity in high-penetration renewable energy grids that utilize multitype reserve resources, including SGs and nonsynchronous units, to address the frequency security issue. First, strategies and models for expanding the sources of frequency regulation reserves are analyzed, including various types of renewable energy generation, such as wind turbine (WT) curtailment and the combination of photovoltaic (PV) cells and battery storage. A refined reserve criterion is then proposed that considers multidimensional evaluation indices from both operational economy and frequency security aspects. Finally, a bilevel optimization model for reserve capacity allocation on multiple timescales that considers frequency security is constructed. The rationality and effectiveness of the proposed reserve allocation scheme were verified using a practical power grid in Southwest China.

With the rapid development of the power industry, the safe and stable operation of the power system has become an important basis for safeguarding economic and social development. In order to adapt to the trend of rapid growth of new energy sources and the demand for transformation and upgrading of the power industry, the capacity allocation scheme of the power system needs to be constantly optimized and adjusted. To address the development planning of new power systems, a new power system development planning model based on two-layer planning is constructed, taking into account the operational output results of various types of units and the interaction between the configuration schemes. Through the analysis of the evolution of the power system in a region in the next 15 years, the development pattern of the new power system in different stages under the high, medium, and low scenarios of new energy development is optimally calculated to guide the development of the power system. The results show that the model can provide a quantitative reference for decision-making on energy policies and power strategies in the medium and long term.

Driven by the carbon peaking and carbon neutrality goals, Sichuan Province, with its abundant hydropower and renewable energy resources, possesses the feasibility to pioneer in constructing a provincial-level carbon-neutral power system in China. However, current grid planning lacks sufficient consideration for carbon emission related goals and constraints. To investigate the carbon emission evolution path of Sichuan Power Grid under the carbon neutrality goal, firstly, the distribution and development status of clean energy sources in Sichuan are clarified in this paper, and the low-carbon technology combination for Sichuan Power Grid's transition is determined. Then, load forecasting and other relevant factors are also combined to form boundary conditions for planning. Based on this, power source and grid planning research are carried out, exploring the two-stage evolutionary trajectory—transition period and formation period. Finally, employing the carbon emission flow theory, the carbon emissions at each phase are analysed, depicting the progressive realization of carbon neutrality in Sichuan Power Grid from the perspective of carbon emission flow.

In the new situation, mega cities have entered a new round of rapid development, with electricity demand showing characteristics of diversification and high growth rates. However, with the gradual construction of new power systems under the “dual carbon” goals, urban grid structures have become more complex. Meanwhile, the requirements for urban environmental policies and regulations are becoming stricter. The 220 kV power grid faces multiple demands to enhance power supply capacity, improve safe and stable operation levels, and achieve efficient planning implementation. This paper discusses the planning ideas of the 220 kV power grid for mega cities in China, which comprehensively consider the impact of electricity demand trends, power supply layout, topography, and power channel resources. From the aspects of substation capacity and channel planning, network structure construction, and partitioning, a systematic planning idea for the 220 kV power grid in mega cities is formed in this paper. This idea can guide the planning and development of the 220 kV power grid in mega cities of China and also has good expandability.

The analysis and calculation of hybrid energy flow is the basis of planning, operation and market transaction of multi-energy systems. An optimal energy flow model of multi-energy systems considering N-1 component failure is proposed in this paper. The compressor and gas storage devices are modeled in the hybrid optimal energy flow model. The particle swarm intelligent optimization algorithm is used to solve this non-convex and non-linear optimization model. Case studies demonstrate the coupled and complementary operation of multi-energy systems can reduce the total operation cost by 8 % as compared to independent operation model. The hybrid optimal energy flow (HOEF) model of multi-energy systems with N-1 component failure can improve the robustness of HOEF results.

The scale of generation supply and power grid continues to expand, the grid structure is subject to the double constraints of short-circuit current and network stability, which is supposed neither be too tight nor too loose. How to find a balance point among them has become the key to power transmission network planning. In response to the above problem, three types of grid construction modes is proposed in this paper. The advantages and applicable scenarios of each modes in terms of power flow distribution, short-circuit current control effect, long-term development adaptability, and operational flexibility is analyzed. The feasibility of the proposed method are verified by a case study of a practical provincial power grid.

With the increase of the scale of UHVDC in the near urban power grid load center, the reduction of main reactive power sources such as local generator, dynamic reactive compensation, poses a serious threat to the transient voltage security of the receiving power grid. This paper analysis the coupling mechanism between the transient voltage stability and UHVDC, and proposes several measures to improve voltage stability, including condenser, optimization of VDCOL parameters and DC power emergency control. Taking a large urban power grid in southwest China as an example, the simulation analysis is carried out. The results show that the measures proposed in this paper can effectively improve the transient voltage stability of the urban power grid.

Two AC systems that are asynchronously interconnected will have the problem of lack of mutual support ability due to blocking the sharing of frequency modulation resources. In order to improve the frequency response characteristics of the asynchronous interconnected system, an additional frequency control strategy (AFC) with line commutated converter based HVDC (LCC-HVDC) participation is proposed. The strategy improves the transient characteristics of the system frequency response by adding frequency-var additional control. And the strategy also improves the voltage stability of the grid frequency regulation process by adding voltage-var additional control and adjusting the switching mode of the AC filter in LCC-HVDC. Finally, an equivalent model is built in PSCAD/EMTDC. The adaptability of LCC-HVDC to the proposed additional frequency control strategy is analyzed by a case study.

With the goal of “double carbon,” wind power and PV are connected to the grid on a large scale, which poses a great challenge to the safe operation of the grid. The synergistic optimization of “hydro-wind-PV hybrid systems” can bring into play the ability of coordination and mutual assistance between different types of power sources, which is one of the important measures to increase the safe operation of the power grid and reduce abandoned wind and photovoltaic power. Based on the analysis of the characteristics of hydro-wind-PV power, the “hydro-wind-PV” integrated network-source cooperative optimization scheduling model is constructed to maximize the source-load matching and the consumption of clean energy. Finally, the results of a case study of a hydropower-wind power-photovoltaic cluster project in southwest China are used to verify the rationality and feasibility of the model.

The evolution of power system is a dynamic and complex process. Whether the evolution of power system can adapt to the macro environment and develop with all subsystems within the system depends on the matching and integration development among different evolutionary subjects. Based on the dissipation theory and Brusselator model, this paper constructs an evolutionary synergy model of the power system, and quantitatively analyzes the evolutionary synergy relationship between the evolving subjects of the power system and the overall evolutionary process of the power system. Then, from the socio-technical system transformation theory, we analyze the evolutionary elements of macro-environmental level, meso-institutional level and micro-beneficial level in the process of system transformation in China and the interaction among them. Finally, by comparing the entropy changes of the evolutionary participant subjects and the overall power system at each stage, the endogenous mechanism of the power system, the external elements and their interactions on the system development are studied in order to reveal the evolutionary synergistic law of China’s power system and make suggestions for the development of the new power system.

Recent years, power transmission and transformation projects in Sichuan have heavy construction tasks and many construction sites. State Grid Corporation of China (SGCC) has increasingly increased its requirements for low-carbon and digital construction management., and it is difficult to meet the requirements by relying on the on-site supervision of managers. Therefore., the application and research of remote online monitoring and diagnosis technology for power grid construction is of great significance for low-carbon., digital and visual construction management. The research of this paper has focused on the digital upgrading of The whole process integrated digital monitoring platform for smart low-carbon power grid construction.

Many repeated manual feature adjustments and much heuristic parameter tuning are required during the debugging of machine learning (ML)-based transient stability assessment (TSA) of power systems. Furthermore, the results produced by ML-based TSA are often not explainable. This paper handles both the automation and interpretability issues of ML-based TSA. An automated machine learning (AutoML) scheme is proposed which consists of auto-feature selection, CatBoost, Bayesian optimization, and performance evaluation. CatBoost, as a new ensemble ML method, is implemented to achieve fast, scalable, and high performance for online TSA. To enable faster deployment and reduce the heavy dependence on human expertise, auto-feature selection and Bayesian optimization, respectively, are introduced to automatically determine the best input features and optimal hyperparameters. Furthermore, to help operators understand the prediction of stable/unstable TSA, an interpretability analysis based on the Shapley additive explanation (SHAP), is embedded into both offline and online phases of the AutoML framework. Test results on IEEE 39-bus system, IEEE 118-bus system, and a practical large-scale power system, demonstrate that the proposed approach achieves more accurate and certain appropriate trust solutions while saving a substantial amount of time in comparison to other methods.