Research Projects
E-Cast: Electricity Demand Forecasting Platform
In order to supply the electricity demand, the utility company requires accurate predictions of the demand signal. In this project, we develop E-cast, a forecasting tool that is capable of performing short-term and long-term predictions of the electricity demand of KSA. Different machine learning algorithms, e.g. deep neural networks such as Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU) are used to produce the prediction which is then deployed in a web-based tool that is used by the Saudi Electricity Company (SEC).
Reinforcement Learning for Operational Decisions of Power Systems
In multiple disciplines across science and engineering, practitioners often face the challenge of making sequential decisions or actions in the face of uncertainty of the outcomes at each stage. In this project, we try to leverage recent advancements in RL agortihms such as Deep Q-Netwriks, Actor-Critics, and Policy Gradient methods, in order to learn optimal strategies in a power system operational setting. The goal is to design algorithms that learn the optimal strategy in an efficient and robust manner, while also being applicable to a variety of real-world scenarios.
Impact of Variable Generation on the Flexibility Requirement of the Saudi PowerSystem
The main aim of the research is to develop methodological analysis and quantitative tools that allow exploring and enlightening an efficient evolution of the electric power system in the Kingdom. The project is concerned with studying the flexibility of the Saudi power system and examining how it would change with increased variable generation penetration utilizing a unit commitment and economic dispatch optimization model and a dataset for the generation mix in KSA.
Cost Reflective Electricity Tariff Design
In this project, we explore different possible designs of the Saudi electricity tariff and examine the effect of each component of the tariff on the behavior of the consumer and on the system. We have simulated the response of the end users to the change in electricity prices, and developed an optimization model to find the optimal tariff that maximizes the additional utility to the consumers and minimizes the additional cost to the electricity company. the project aims to give insight to regulators on possible alternatives for the electricity tariff design that tries to send the right signals to the consumers, promoting an efficient use of the power network.
COVID‑19 Decision Support Platform
In this project, a digital twin of the kingdom is created that account for many special factors related to the Saudi context and is able to capture a variety of its complex dynamics using millions of individual and point-of-interest data points. Then, an agent based model is created to simulate the spread of disease in the kingdom. The model is able to simulate different intervention strategies and its effect on the spread of a disease. The model is then encapsulated in a decision support web platform providing an interactive analytic front end for decision makers.
Construction of an Electrostatic 3D Nano‑Printer
Nanotechnology's potential is restricted by its difficult and expensive fabrication methods. In this project, we leverage advances in the nanotechnology literature to assemble an electrostatic nano-printer able to create useful gold nanostructures utilizing less expensive techniques and without significant performance compromise. This project is a step towards printing nano structures using organic material instead of gold, which can have a great impact on fields such as medicine and biophysics.