The rapid proliferation of renewable energy integration and escalating grid operational complexity have intensified demands for resilient self-healing mechanisms in modern power systems.Conventional approaches relying...The rapid proliferation of renewable energy integration and escalating grid operational complexity have intensified demands for resilient self-healing mechanisms in modern power systems.Conventional approaches relying on static models and heuristic rules exhibit limitations in addressing dynamic fault propagation and multimodal data fusion.This study proposes a Transformer-enhanced intelligent microgrid self-healing framework that synergizes large languagemodels(LLMs)with adaptive optimization,achieving three key innovations:(1)Ahierarchical attention mechanism incorporating grid impedance characteristics for spatiotemporal feature extraction,(2)Dynamic covariance estimation Kalman filtering with wavelet packet energy entropy thresholds(Daubechies-4 basis,6-level decomposition),and(3)A grouping-stratified ant colony optimization algorithm featuring penalty-based pheromone updating.Validated on IEEE 33/100-node systems,our framework demonstrates 96.7%fault localization accuracy(23%improvement over STGCN)and 0.82-s protection delay,outperforming MILP-basedmethods by 37%in reconfiguration speed.The system maintains 98.4%self-healing success rate under cascading faults,resolving 89.3%of phase-toground faults within 500 ms through adaptive impedance matching.Field tests on 220 kV substations with 45%renewable penetration show 99.1%voltage stability(±5%deviation threshold)and 40%communication efficiency gains via compressed GOOSE message parsing.Comparative analysis reveals 12.6×faster convergence than conventional ACO in 1000-node networks,with 95.2%robustness against±25%load fluctuations.These advancements provide a scalable solution for real-time fault recovery in renewable-dense grids,reducing outage duration by 63%inmulti-agent simulations compared to centralized architectures.展开更多
The rapid development and increased installed capacity of new energy sources such as wind and solar power pose new challenges for power grid fault diagnosis.This paper presents an innovative framework,the Intelligent ...The rapid development and increased installed capacity of new energy sources such as wind and solar power pose new challenges for power grid fault diagnosis.This paper presents an innovative framework,the Intelligent Power Stability and Scheduling(IPSS)System,which is designed to enhance the safety,stability,and economic efficiency of power systems,particularly those integrated with green energy sources.The IPSS System is distinguished by its integration of a CNN-Transformer predictive model,which leverages the strengths of Convolutional Neural Networks(CNN)for local feature extraction and Transformer architecture for global dependency modeling,offering significant potential in power safety diagnostics.TheIPSS System optimizes the economic and stability objectives of the power grid through an improved Zebra Algorithm,which aims tominimize operational costs and grid instability.Theperformance of the predictive model is comprehensively evaluated using key metrics such as Root Mean Square Error(RMSE),Mean Absolute Percentage Error(MAPE),and Coefficient of Determination(R2).Experimental results demonstrate the superiority of the CNN-Transformer model,with the lowest RMSE and MAE values of 0.0063 and 0.00421,respectively,on the training set,and an R2 value approaching 1,at 0.99635,indicating minimal prediction error and strong data interpretability.On the test set,the model maintains its excellence with the lowest RMSE and MAE values of 0.009 and 0.00673,respectively,and an R2 value of 0.97233.The IPSS System outperforms other models in terms of prediction accuracy and explanatory power and validates its effectiveness in economic and stability analysis through comparative studies with other optimization algorithms.The system’s efficacy is further supported by experimental results,highlighting the proposed scheme’s capability to reduce operational costs and enhance system stability,making it a valuable contribution to the field of green energy systems.展开更多
基金the project“Research on Power SafetyDecision Support SystemBased on Large Language Models”(Science and Technology Project of Huaian Hongneng Group Co.,Ltd.)under Contract No.SGTYHT/23-JS-001.
文摘The rapid proliferation of renewable energy integration and escalating grid operational complexity have intensified demands for resilient self-healing mechanisms in modern power systems.Conventional approaches relying on static models and heuristic rules exhibit limitations in addressing dynamic fault propagation and multimodal data fusion.This study proposes a Transformer-enhanced intelligent microgrid self-healing framework that synergizes large languagemodels(LLMs)with adaptive optimization,achieving three key innovations:(1)Ahierarchical attention mechanism incorporating grid impedance characteristics for spatiotemporal feature extraction,(2)Dynamic covariance estimation Kalman filtering with wavelet packet energy entropy thresholds(Daubechies-4 basis,6-level decomposition),and(3)A grouping-stratified ant colony optimization algorithm featuring penalty-based pheromone updating.Validated on IEEE 33/100-node systems,our framework demonstrates 96.7%fault localization accuracy(23%improvement over STGCN)and 0.82-s protection delay,outperforming MILP-basedmethods by 37%in reconfiguration speed.The system maintains 98.4%self-healing success rate under cascading faults,resolving 89.3%of phase-toground faults within 500 ms through adaptive impedance matching.Field tests on 220 kV substations with 45%renewable penetration show 99.1%voltage stability(±5%deviation threshold)and 40%communication efficiency gains via compressed GOOSE message parsing.Comparative analysis reveals 12.6×faster convergence than conventional ACO in 1000-node networks,with 95.2%robustness against±25%load fluctuations.These advancements provide a scalable solution for real-time fault recovery in renewable-dense grids,reducing outage duration by 63%inmulti-agent simulations compared to centralized architectures.
基金The research project,“Research on Power Safety Assisted Decision System Based on Large Language Models”(Project Number:JSDL24051414020001)acknowledges with gratitude the financial and logistical support it has received.
文摘The rapid development and increased installed capacity of new energy sources such as wind and solar power pose new challenges for power grid fault diagnosis.This paper presents an innovative framework,the Intelligent Power Stability and Scheduling(IPSS)System,which is designed to enhance the safety,stability,and economic efficiency of power systems,particularly those integrated with green energy sources.The IPSS System is distinguished by its integration of a CNN-Transformer predictive model,which leverages the strengths of Convolutional Neural Networks(CNN)for local feature extraction and Transformer architecture for global dependency modeling,offering significant potential in power safety diagnostics.TheIPSS System optimizes the economic and stability objectives of the power grid through an improved Zebra Algorithm,which aims tominimize operational costs and grid instability.Theperformance of the predictive model is comprehensively evaluated using key metrics such as Root Mean Square Error(RMSE),Mean Absolute Percentage Error(MAPE),and Coefficient of Determination(R2).Experimental results demonstrate the superiority of the CNN-Transformer model,with the lowest RMSE and MAE values of 0.0063 and 0.00421,respectively,on the training set,and an R2 value approaching 1,at 0.99635,indicating minimal prediction error and strong data interpretability.On the test set,the model maintains its excellence with the lowest RMSE and MAE values of 0.009 and 0.00673,respectively,and an R2 value of 0.97233.The IPSS System outperforms other models in terms of prediction accuracy and explanatory power and validates its effectiveness in economic and stability analysis through comparative studies with other optimization algorithms.The system’s efficacy is further supported by experimental results,highlighting the proposed scheme’s capability to reduce operational costs and enhance system stability,making it a valuable contribution to the field of green energy systems.
