Smart grid cybersecurity:anomaly detection in solar power systems using deep learningOA
This study explores the application of deep learning techniques,specifically long short-term memory(LSTM)networks,to detect anomalies in solar power production within a smart grid framework.Given the increasing integration of renewable energy sources such as solar power into electrical grids,ensuring cybersecurity and operational stability has become a critical challenge.The LSTM model was developed to capture the temporal dependencies in energy production data,demonstrating its ability to accurately predict production values and identify significant deviations.The model achieved a mean squared error(MSE)of 0.846 with a precision of 79%,although it showed a relatively lower recall of 33%,indicating the need for further optimization to enhance sensitivity.In a comparative analysis,the LSTM model was evaluated against a feedforward neural network(FFNN)and traditional isolation forest model.The FFNN outperformed the LSTM in key metrics,including accuracy,precision,recall,F1-score,and area under the curve(AUC)-receiver operating characteristic(ROC),highlighting its robustness in anomaly detection tasks.Conversely,the isolation forest model demonstrated a limited effectiveness.Despite the potential of the LSTM model,these results suggest that FFNNs currently provide a more reliable solution for detecting anomalies in solar power systems.The study also emphasized the importance of real-time anomaly detection and its implications for grid stability and reliability.By integrating advanced deep learning models into monitoring systems,operators can proactively manage and mitigate the impact of anomalies,thereby ensuring a stable power supply.
Jagbir Singh;Owais Ahmad Shah;Sujata Arora
Department of Electrical and Electronics Engineering,Noida International University,Gautam Budh Nagar,203201,IndiaDepartment of Electronics&Communication Engineering,Dayananda Sagar University,Bengaluru,562112,IndiaDepartment of Electrical and Electronics Engineering,Noida International University,Gautam Budh Nagar,203201,India
信息技术与安全科学
Smart grid cybersecurityAnomaly detectionDeep learningLong short-term memory(LSTM)networksSolar power systems
《Energy Storage and Saving》 2026 (1)
P.16-26,11
评论