Enhancing E-Governance In The Ministry Of Electricity In Iraq Using Artificial Intelligence

Forecasting the electricity demand is essential to op-timize future production. This forecasting can be boosted by uti-lizing machine learning as an enhancement to the e-Governance at the Ministry of Electricity in Iraq. This research aims to investigate the effects of machine learning on e-Governance using electricity consumption and exogenous variables (weather data) from 2020 to 2021, in the city of Mosul in northern Iraq. This study utilizes several neural network and statistical models, including (Stacked and Bi-) LSTM and statistical models (ARIMA and SARIMAX), to forecast future consumption and to increase the effectiveness of Governance in detecting anomalies. The results show that deep learning models have a better result in comparison with statistical models in terms of the root mean square error rate. The best model of LSTM achieves a result of 11.2 RMSE while the statistical models comprised of ARIMA and SARIMAx obtain an RMSE value of 15 and 12.2 respectively.