Dynamic Line Rating Forecasting Using Bi-LSTM

Transmission systems worldwide are being upgraded to accommodate massive amounts of renewable energy. The cost of installing new transmission lines is too high, and using static line ratings (SLR) in power system operations only partially utilizes transmission systems. Dynamic line rating (DLR) may improve the power transfer capacity of existing transmission lines, resulting in greater renewable energy utilization and fewer curtailments. DLR depends highly on volatile weather conditions, so it must be forecasted for day-ahead and real-time power system operations. This paper proposes a novel bi-directional long-short-term memory (Bi-LSTM) model-based DLR forecasting methodology considering uncertain weather parameters. The productiveness of the stated procedure is displayed through experimental validation on real-world DLR datasets. A comparison with existing forecasting techniques reflects the superiority of the proposed methodology in terms of predictive accuracy and robustness. The improved DLR forecasts using the proposed model can increase power transmission network resilience and stability.