Abstract

The United Arab Emirates (UAE) has aims to triple its renewable energy capacity to 14 GW to attain the objective of 30% contribution of clean energy by 2031. Energy consumption, on the other hand, will increase from 5.2 to 5.8 exajoules by 2028. The grid infrastructure for which was developed to facilitate the centralized generation of power, is not conducive to dealing with the variability of the decentralized renewable resources that lead to inefficiencies in operation and reliability problems. The Dubai power grid is analyzed using historical data of 105,120 residential, commercial and industrial supplies. In this analysis machine learning models using Random Forest and LSTM were found to be superior in respect of risk assessment characteristics and load forecasting in so far as the capability to predict 89% of the capacity variation and reduce forecast error to MSE 0.0639 in respect of risk levels. An artificial intelligence enabled digital twin was set up to optimize current time grid optimization including predictive maintenance and renewable portfolio planning. This digital twin was effective in respect of risk level prediction and next step predictions were in the Cautious range (2.59–2.72) in 80% of cases and more effective in predicting surplus energy and fault detection. The seasonal analysis indicated that peak energy always requires in summer as well as from the industrial area where peak demand capacity is (peaking to 4.91 MW) and surplus whilst being of significance in respect of intervention point criticalities. The findings indicate increased efficiency, reliability and integration of renewables by utilization of AI translational digital twins applied to the grid reduce risk of operation whilst assisting in the UAE energy transition. Demonstration samples consist of standard calibrated recalibration advice for sensor readings and seasonal load management techniques as well as worldwide predictive analytical advisement implementation leading to realization of a smart adaptive grid.

Publication Date

12-2025

Document Type

Thesis

Student Type

Graduate

Degree Name

Professional Studies (MS)

Department, Program, or Center

Graduate Programs & Research

Advisor

Sanjay Modak

Advisor/Committee Member

Hammou Messatfa

Campus

RIT Dubai

Download

Share

COinS