Electric load forecasting under False Data Injection Attacks using deep learning

Precise electric load forecasting at different time horizons is an essential aspect for electricity producers and consumers who participate in energy markets order to maximize their economic efficiency. Moreover, accurate prediction of the contributes toward robust resilient power grids due error minimization generators scheduling schemes. The accuracy existing methods relies on data quality noisy real-world environments, integrity malicious cyber-attacks. This paper proposes a cyber-secure deep learning framework that accurately predicts horizon spanning from hour week. proposed systematically integrates Autoencoder (AE), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM) models (AE-CLSTM). feasibility solution validated by using realistic grid acquired distribution network Tabriz, Iran. Compared other methods, method shows highest both normal case with noise stealthy False Data Injection Attack (FDIA). practical suitable mitigating attacks.