With the increase of building integrated energy systems(BIES) and its increasingly prominent role in improving energy utilization efficiency, it is necessary to study how to enhance the coupling degree of equipment within BIES and its coordination with the power grid, thereby reducing the daily comprehensive operating cost of BIES. Therefore, a BIES day-ahead optimization scheduling model based on feedforward neural network(FNN) is proposed. Firstly, the structure of BIES with rooftop photovoltaics is analyzed. Secondly, a day-ahead PV output prediction method and process based on FNN are proposed, providing reliable PV power generation data for BIES optimization scheduling. Finally, with the objective of minimizing the daily comprehensive operating cost composed of electricity purchase cost, gas purchase cost, and equipment operation and maintenance cost, a day-ahead scheduling model for BIES is proposed. The result of the calculation examples show that the proposed PV output prediction method can effectively improve the prediction accuracy of PV power generation output, and the propose day-ahead optimization scheduling model can effectively improve the operational economy of BIES.

When a three-phase asymmetric fault occurs, the capacitor voltages between phases of the star-connected static synchronous compensator(STATCOM) valve group tend to be unbalanced, which is a key difficulty in the fault ride-through (FRT). To address this issue, based on the star-connected STATCOM topology, a single-phase instantaneous power model of the STATCOM is established to analyze the three-phase AC voltage characteristics under diverse grid faults, along with inter-phase power transfer and capacitor voltage variation patterns in the STATCOM. According to the integrated control idea of positive and negative sequence power, a capacitor voltage balancing control strategy combining open-loop and closed-loop control is proposed. An electromagnetic transient(EMT) simulation model built on RTDS is tested against various typical grid faults. Results demonstrate that the proposed strategy effectively balances capacitor voltages in STATCOM commutation valve. The capacitor voltage balancing control strategy using integrated control of positive and negative sequence power resolves limitations of conventional independent positive and negative sequence control methods and enhances STATCOM’s AC fault ride-through capability and operational safety margin.

To improve the recognition accuracy and robustness of the adjustable capability of electric vehicles, a Bidirectional Gated Recurrent Unit(BiGRU) deep learning model that integrates Adaptive Boosting(AdaBoost) mechanism is proposed. Firstly, based on the characterization of charging behavior from three dimensions: user, region, and time, an input system covering 13 key variables such as charging period, electricity consumption, and external environment is constructed to cluster the electric vehicle population and lay the foundation for regulating charging load in the power grid. Secondly, the BiGR model is utilized to mine temporal features, and the AdaBoos ensemble learning mechanism is used to enhance the model's generalization ability, improve the accuracy and robustness of identifying the adjustable capabilities of electric vehicles, and further improve the efficiency of electric vehicle scheduling. Finally, experimental validation is conducted on a real historical dataset to evaluate the identification effect of the model in different scenarios. The results show that the proposed method can effectively improve the accuracy of identifying the adjustable capability of electric vehicles, enhance their schedulability, and has a good effect on identifying the adjustable capability of electric vehicles in different dates and scenarios.

Aiming at the problem that traditional fluid-structure interaction methods are difficult to quickly obtain the galloping response of transmission lines under icing conditions in scenarios such as daily operation and maintenance monitoring and engineering emergency assessment, a finite element-based equivalent loading method for icing galloping loads is proposed, which aims to improve the efficiency of galloping response prediction to meet engineering emergency needs. The catenary theory is used for conductor shape-finding analysis to determine the initial equilibrium state of the conductor under the action of initial tension and self-weight. Based on the three-component aerodynamic force principle, combined with the lift coefficient, drag coefficient and torque coefficient of the iced conductor, the lift force, drag force and torque are calculated. The dynamic aerodynamic loads are equivalently converted into concentrated nodal loads based on the finite element model. Finally, the simulation analysis of galloping displacement response is realized with the help of the ANSYS APDL platform. The example results verify the applicability of the method to single conductors, split conductors and different icing shapes, and it can also be applied to galloping analysis under low wind speed conditions. Compared with the traditional method, this method does not require complex parameter input and long-time integral iteration, and can quickly output the conductor displacement response results, with a significant improvement in analysis efficiency, which can provide technical support for the formulation of anti-galloping decisions for transmission lines under icing disasters.

To address the issue that traditional line protection algorithms are difficult to adapt to changes in system characteristics after large-scale wind power integration into the power system, a protection method for wind farm outgoing lines based on time-domain voltage calculation and comparison is proposed. First, a time-domain calculation model of line voltage is constructed based on the resistor inductor based model, and the voltage values of characteristic points in the line are respectively solved. Then, through the correlation analysis of the waveform of the voltage calculation results at two points, the fault can be accurately identified. Adopting the time-domain full-scale analysis approach, this method can effectively overcome interference caused by special operating conditions in wind power systems, such as crowbar activation, the influence of transition resistance, and weak feed-in from wind farms. To verify its performance, a doubly-fed induction generator(DFIG) wind farm grid-connected model is built on the RSCAD simulation platform for simulation analysis. The results show that the proposed protection method can accurately identify faults, verifying its effectiveness and adaptability.

To address the challenges of insufficient ultimate breaking capacity and difficulty in extinguishing DC arcs in low-voltage small DC circuit breakers, research is conducted on optimizing the arc extinguishing performance of low-voltage small DC circuit breakers. A DC arc magnetohydrodynamic model is constructed by meshing the arc extinguishing system of a low-voltage small DC circuit breaker and coupling external circuit characteristics into it. By applying corresponding boundary conditions, the distribution of temperature, pressure and other parameters in the arc extinguishing chamber during the arc ignition process is successfully obtained. On this basis, the motion characteristics of the arc during the breaking process of low-voltage small DC circuit breakers are deeply explored, further revealing the inherent mechanism of the DC arc evolution process. A comprehensive arc extinguishing optimization method based on increasing circuit resistance, increasing constant arc extinguishing magnetic field, and adopting U-shaped double breakpoint contact structure is proposed. The simulation analysis results show that the optimization design method proposed in this paper can effectively elongate the arc column, raise the arc voltage, reduce the arc current and arc column temperature, effectively shorten the arc ignition time, and greatly improve the arc extinguishing ability and breaking performance of low-voltage small DC circuit breakers.

This paper proposes a lightweight 3D engine design for power systems based on the Metaverse, with the core focuses on model lightweighting, high-performance rendering, and computing network optimization. The model data volume is reduced by applying geometric and texture compression with Level of Detail(LOD) techniques. Screen Space Ambient Occlusion(SSAO) and geometry shader optimization is introduced to enhance rendering performance, and a cloud-edge hierarchical elastic computing network is constructed to achieve resource collaboration and energy consumption optimization. The engine effectively supports applications such as the “One-Map Power Grid” and virtual dispatching command centers. This study provides a new pathway for the intelligence and metaverseization of power system visualization.

When high-impedance faults occur in distribution networks with arc suppression coil grounded systems, their fault distribution characteristics differ significantly from those in non-grounded systems, and existing fault detection methods face challenges in effective detection. On this basis, a line amplitude comparison method based on instantaneous zero-sequence current is proposed. By constructing feeder zero-sequence current models and comprehensive zero-sequence current models for each feeder, and comparing these models while considering the direction of zero-sequence current, the method can effectively achieve accurate detection of high-impedance faults. This method is validated on a simulation platform for distribution networks with arc suppression coil grounded systems. Simulation results show that the proposed method can effectively identify faulted feeders. This research provides valuable reference for promoting the safe operation of distribution networks.

To address the bottlenecks in single-phase grounding fault disposal, including the low accuracy of traditional station-end fault line selection devices, prolonged average disposal time, and difficulties in detecting high-resistance/intermittent faults, a precise disposal method based on primary-secondary coordination and multi-source verification is proposed. This method innovatively establishes a three-tier technical framework: Firstly, a differentiated dynamic setting model is established. Based on the neutral point grounding method(ungrounded, arc suppression coil grounded) and line parameters(cable, overhead line length), the zero-sequence current reference value is derived via Kirchhoff's Current Law. By combining transient process analysis, protection setting values are dynamically generated, thereby resolving maloperation issues caused by empirical setting values. Secondly, the master station functionality is enhanced to support rapid decision-making. The four-remote point list is optimized by adding zero-sequence voltage and zero-sequence current remote measurements, and a "Single Command-Multiple Response" waveform transmission mechanism is established with the time reduced from 20 minutes to 1 minutes. An online setting values management module is developed to ensure consistency in multi-source verification. Finally, multi-source spatiotemporal topology verification is achieved by integrating the substation fault line selection device and distribution network terminal, while incorporating traveling wave positioning technology and wide-area synchronous intelligent sensors. Combined with the zero-sequence mutation quantity ratio criterion, this enables spatial topology verification between in-station line selection and out-station segment selection. The 200 faults simulation verification shows that under the transition resistance condition of 0~20 kΩ, the positioning accuracy, high-resistance fault detection rate, and average handling time of this method have significantly improved compared with the traditional method. This method resolves the problem of information isolation in distribution networks through resource integration and logic transformation, thereby providing a standardized technical pathway for high-reliability power supply.

To mitigate the impact of disorderly EV charging on the power grid and enhance user charging experiences, an orderly charging strategy based on the hippopotamus optimization algorithm is proposed. Firstly, the disorderly charging load of electric vehicles is simulated using the Monte Carlo method. Subsequently, an orderly charging model is established with the dual objectives of minimizing the grid load peak-to-valley difference rate and user charging costs. Finally, the hippopotamus optimization algorithm is employed to solve this model. Simulation results demonstrate that compared with disorderly charging, the grid load peak-to-valley difference rate is reduced by 23%, while user charging costs is decreased by 38%. Consequently, the proposed strategy exhibits significant effectiveness in both reducing the grid load peak-to-valley difference rate and user charging costs.

Targeting the inefficiencies in storage battery capacity verification and large-scale impact of faults in conventional substations, an online capacity verification system based on a grouped parallel architecture and bidirectional LLC (Inductor-Inductor-Capacitor) resonant conversion is proposed. The system reconfigures a 108-cell (2 V each) series battery bank into nine independent 12 V modules, which are then connected in parallel to the DC bus via bidirectional LLC resonant converters to achieve charging and discharging control and fault isolation. A dynamic capacity model and parallel capacity verification strategy are developed in conjunction with edge computing terminals. Validation at a 110 kV substation demonstrates 90% reduction in capacity verification time with dual-group parallel operation, 100% success rate in fault isolation, 50% decrease in annual average operation and maintenance costs, and 100% accuracy in automated capacity verification report generation. By synergizing the grouped architecture with edge computing, this system achieves the transition of storage battery capacity verification from manual operations to fully automated and intelligent online operation, providing critical technical support for smart substation construction.

In response to the problem that the existing control strategies for combined frequency regulation systems of thermal power units and energy storage (referred to as "thermal power-energy storage" hereinafter) are difficult to adapt to the randomness and multi-time-scale characteristics of the automatic generation control(AGC) instructions of the power grid, a coordinated optimization method for combined thermal power-energy storage frequency regulation based on adaptive fuzzy control theory is designed. This method collects real-time operational parameters, including unit AGC commands, real-time load, main steam pressure, and energy storage's state of charge (SOC). A fuzzy controller is employed to calculate the boundary conditions for the unit’s variable load rate. Based on the energy storage SOC, the real-time output power and output priority sequence of the energy storage unit are determined, forming a dual-mode coordinated response architecture for thermal power and energy storage. This approach ensures the accuracy of single frequency regulation power tracking and maintains the stability of the energy storage SOC, guaranteeing the system’s long-term tracking capability for AGC commands. Simulations based on actual data from a 1 000 MW unit demonstrate that the proposed strategy improves the comprehensive performance indicator k-value by 2.32 times, increases frequency regulation revenue by 1 621.56 CNY, and maintains the energy storage SOC within a safe threshold range over a 12-hour control cycle, effectively addressing the challenge of long-term-scale coordinated control in combined thermal power-energy storage frequency regulation systems.

With the large-scale integration of renewable energy sources such as wind power and photovoltaic power into the power grid, the uncertainty and volatility of the power system operation have significantly increased, making it difficult to extract load sequence features and improve the accuracy of short-term power load forecasting. To address this issue, a short-term power load forecasting model based on the combination of wavelet transform convolution(WTC) and Informer model is proposed. The improved variational mode decomposition(VMD) is used to decompose and denoise the input data before feeding it into the wavelet convolution module for multi-level wavelet convolution, achieving multi-scale feature extraction of complex time series and reducing sequence complexity, thereby improving the forecasting accuracy. To verify the validity of the model, multiple sets of experiments are conducted. The experimental results show that the Mean Absolute Percentage Error(MAPE) of the model proposed in this paper is 1.893 1%. Compared with the methods of using the Informer model alone or only using GSWOA-VMD-informer, it is reduced by 1.059 4% and 0.504 8%, respectively, verifying the effectiveness of this model.

Power system inertia estimation is crucial to ensure the stability of the power grid and to cope with unexpected events. A Bird-mating Convolutional XGBoost Network(BCXNet) algorithm is proposed to accurately estimate the inertia value of the power system. Firstly, a convolutional neural network(CNN) is used to extract features of system inertia and construct high-level time series data features containing multidimensional features. Secondly, the eXtreme gradient boosting(XGBoost) is used for inertia regression analysis, and the XGBoost hyperparameters are optimized by combining the bird-mating optimizer(BMO), thereby further improving the accuracy of inertia estimation. Finally, the accuracy of inertia estimation by the proposed method is verified through experiments, and the coefficient of determination(R²) reaches 0.987 3, the root mean square error(RMSE) is 0.079 7, and the mean absolute percentage error(MPAE) is 2.6%, which indicates that the estimated model output is in high agreement with the real value.

During the design process of overhead transmission lines, manual calculation of tower leg coordinates and the generation of footprint Shapefile(SHP) files are inefficient and prone to errors. To address this issue, an automated mapping method based on open-source geographic information system (GIS) library is proposed, which not only establishes a geometric model for calculating tower leg coordinates but also utilizes the GDAL/OGR library to achieve batch generation and visualization from tower parameters to footprint SHP files. The application in a 500 kV transmission line project in Tibet demonstrates that the mean square error of the tower leg coordinates calculated by this method is 0.2 mm, which realizes the full process and automated tower position mapping, greatly improving work efficiency and data accuracy. Additionally, the software is easy to install and holds good promotion and application value.

To address the challenges posed by the complex, non-periodic, and multi-modal characteristics of climbing and working on energized poles for exoskeleton assistance control, a recognition method based on plantar pressure and deep reversible consistency learning(DRCL) is proposed. A plantar pressure data collection system is constructed, and the center of pressure(COP) and various derived features are extracted as the main inputs. Combined with the DRCL method, multi-modal feature fusion and robust recognition are achieved through selective prior learning, semantic consistency constraints, and modality-invariant representation reshaping. Experimental results show that the COP-derived features achieve F₁ values above 0.99 under both level ground walking and pre-pole-climbing preparation modes. After model training, the recognition accuracy for three subjects reaches a maximum of 98.1%, with an average accuracy of 94.4%. This method significantly improves the accuracy and robustness of mode recognition for climbing and working on energized poles, providing a reliable basis for exoskeleton robots to achieve real-time human-robot collaborative assistive control.

For the problem that there is scarce image annotation data and difficult feature extraction in the cracks of key parts such as chimneys, walls and metal equipment in thermal power plants under complex lighting conditions, a crack detection model named PowerCrackNet is proposed, which is based on a dual-branch prototype metric network with few-shot learning. The network integrates the reflection rate information irrelevant to lighting in the image using the Retinex theory, guiding the network to learn the invariance of lighting. During the feature extraction stage, a prior mask containing crack location and structural information is generated by measuring the similarity between query and support features in a high-dimensional space. Then, in the feature interaction stage, the convolution attention and atrous spatial convolution pooling pyramid mechanism are employed to integrate the feature information and prior information from different scales, thereby improving the spatial inconsistencies among samples in few-shot learning. Finally, in the prototype metric stage, a complementary attention mechanism is used to enhance the expressiveness of support prototype vectors for crack categories. Quantitative comparisons and visualization analysis on the public low-illumination crack dataset LCSD demonstrate that the proposed method achieves stronger crack segmentation performance under low-light conditions with limited samples(1-shot), outperforming the current other method. Further validation in real-world scenarios confirms that the proposed approach can provide efficient and reliable technical support for the early warning of cracks in key parts of thermal power plants, highlighting its significant practical value.

Aiming at the current problems such as low prediction accuracy of photovoltaic power and the lack of interpretability of the model, a photovoltaic power interval prediction model based on ISGA-XGBoost and SHAP is proposed. Firstly, the statistics of the sliding time window of photovoltaic power are introduced into the input features to capture the dynamic change trends and patterns of the time series, and a prediction model based on XGBoost is established. Through the regularization strategy and parallel computing optimization, high-dimensional features are processed and overfitting is suppressed. Secondly, ISGA is adopted to integrate the head goose rotation mechanism, the call guidance mechanism and the outlier boundary strategy to improve the hyperparameter optimization ability of the XGBoost model. Then, considering the uncertainty of photovoltaic power, the bootstrap method is adopted to quantify the prediction intervals at different confidence levels. Finally, the SHAP interpretable model is introduced to quantify the contributions of each feature variable and improve the interpretability of the prediction results. The results of the calculation examples show that the model proposed in this paper has higher prediction accuracy compared with other models, and has better generalization ability and interpretability.

To adress the issues of insufficient consideration of opportunity costs and limited resource allocation efficiency when energy storage participates in multi-market bidding, a joint bidding strategy for energy storage considering opportunity costs is proposed. Firstly, based on the spot market mechanism of Shandong Province's power system, the transaction mode for energy storage participating in the electricity energy and frequency regulation auxiliary service markets and quantification method of opportunity costs are analyzed. On this basis, a leader-follower game bilevel optimization model with independent energy storage as the upper-level leader and the trading center as the lower-level follower is constructed. The upper model aims to maximize the comprehensive revenue of energy storage, while the lower model aims to minimize the clearing cost of the power market. By combining the KKT conditions and the big M method, the mixed-integer nonlinear bilevel model is transformed into a solvable mixed-integer linear programming model, and a professional solver is used for solution. The case study results show that the proposed strategy can effectively quantify the impact of opportunity costs on bidding decisions, optimize the resource allocation of energy storage in multiple markets, and significantly improve the economic benefits of energy storage. This strategy provides theoretical and technical support for the market-oriented operation of energy storage in power systems with high proportions of renewable energy.

In order to enhance the interaction between enthusiasm electricity users and the power grid, and meet the electricity needs of different users, a pricing package design method based on dual game theory and user choice behavior is proposed. Firstly, a dual game theory framework for designing electricity pricing packages is established, which covers two levels of games: one is non cooperative games between power sales companies, and the other is master-slave games between power sales companies and electricity users. Then, a pricing package decision model for participants in the power market is constructed. Finally, the designed electricity price package model is solved through iterative loop nesting. The results of the calculation examples show that the electricity price package design method proposed in this paper can take into account the interests of both power sales companies and power users, which is conducive to reducing the purchase and sale prices of power for power sales companies, improving the satisfaction of power users, and achieving a win-win situation for both the sellers and users.

To achieve the smooth connection between the medium and long-term electricity market and the day-ahead market, a joint clearing model of energy and ancillary services in the day-ahead market considering the regulation performance of cascade hydropower stations is constructed. The model considers the relationship between the energy clearing volume and the ancillary service clearing volume among cascade hydropower stations, and aims to minimize the total procurement cost of the day-ahead market to obtain the clearing results of energy and ancillary services in the day-ahead market. The duality theorem is used to analyze the relationship between the ancillary service offer of cascade hydropower stations and the energy clearing price. The results show that the lower the ancillary service offer, the lower the energy clearing price. The settlement results using the difference contract method show that the established model can reduce the risk of recovering excessive profits for power generation entities and increase electricity revenue.