Accuracy improvement of electrical load forecasting against new cyber-attack architectures

•Developing contributions to consider data integrity aspects in load forecasting problem.•Proposing robust schemes to perform accurate load forecasting under new cyber-attacks.•A novel cyber-attack named Civil Attack (CA) is employed to strength the forecasting.•Using accurate non-linear techniques in zones with high load covariance.•Gaussian Process Regression and Random Forest Regression is used to increase accuracy.