Forecasting Load at Residential, Industrial, and Commercial Stations for Dubai Electricity and Water Authority: A Machine Learning Approach to Managing Generation During Peak and Off-Peak Times Over the Coming Years

In Dubai, with the rapid growth of energy demand, efficient demand side management is necessary to optimize the distribution of electricity, reduce peak loads, and improve grid stability. Traditional DSM strategies are based on historical data and reactive control mechanisms that cannot adapt to evolving consumption patterns. This research uses Machine Learning techniques to enhance DSM in residential, commercial, and industrial sectors by developing predictive models that forecast energy demand based on seasonal variations. The dataset used was from Dubai Electricity and Water Authority (DEWA), covering the consumption patterns across Summer, Winter, Transition from Winter to Summer, and Transition from Summer to Winter. PCA, Principal Component Analysis, was carried out to reduce the number of input variables into four key principal components: FAC1_1, FAC2_1, FAC3_1, FAC4_1, enhancing the model's efficiency with no loss of information that is essential. Twelve ML models are developed for each sector, trained using GLM, Regression, LSVM, Linear-AS, and ANN. To this end, both a 70-30 train-test split and k-fold cross-validation were realized to ensure that the models have been well gauged for their performance. Performance was analyzed in terms of RMSE, MAE, and R² scores. Results showed sharp variations in the performance of these models across sector and seasonal divides. While some models performed well for capturing long-term dependencies, others modeled the short-term fluctuations quite well. However, the best models varied depending on seasons and sectors, and no single model worked for all. Indeed, this proves that load forecasting needs a tailored approach. This opens up the possibility of ML driven DSM toward optimization of electricity distribution for grid efficiency and helps DEWA make data-driven energy management decisions for peak and off-peak hours. This work helps towards the further advance of AI for energy management in showing how such data-driven strategies for DSM might support the achievement of sustainability in Dubai while warranting the required grid resilience.