Enhancing Modeling and Simulation Tools Using Uncertainty Quantification and Machine Learning for Nuclear and Conventional Thermal-hydraulics

Principal Investigator
Imran Afgan
Department
Mechanical Engineering
Focus Area
Robotics, AI, & Data Science
Enhancing Modeling and Simulation Tools Using Uncertainty Quantification and Machine Learning for Nuclear and Conventional Thermal-hydraulics

Nuclear and hydro power generation techniques have an important role to play in the UAE’s 2030 energy vision of low carbon sources to meet future energy demand. The UAE is currently investing in the installation of a number of conventional hydro power plants and its first nuclear power plant, Barakah, costing around 25 billion USD. For these plants to operate at high efficiencies and to satisfy stringent safety requirements, a step-change is consequently needed in modeling and simulation capabilities to accurately describe the multi-physics (coupled thermal hydraulics and solid mechanics) processes.

The current proposed project not only targets the wider thermal-hydraulics modeling/design planned activities of the UAE’s 2030 energy vision but also directly supports the ‘UAE Strategy for Artificial Intelligence (AI),’ which emphasizes strongly on the automation of renewable energy and space sectors.

The main objectives of the current project are to:

    1. Enhance the modeling and simulation tools for Fluid Structure Interactions (FSI) and Conjugate-Heat-Transfer (CHT), which govern plant steady and transient operations.
    2. Develop reliable and versatile physical models for modeling and simulation algorithms.

The project is divided into multiple work packages (WP) with an aim to enhance the applicability and reliability of current modeling capabilities targeting the multi-physics problems encountered in nuclear and renewable energy sectors.

Enhancing Modeling and Simulation Tools Using Uncertainty Quantification and Machine Learning for Nuclear and Conventional Thermal-hydraulics