Industrial metaverse for freen data center (II): digital twin via omniverse

• Main Author: Pasaol, Cogal Jacques Tracy
• Other Authors: Wen Yonggang
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science
• Online Access: https://hdl.handle.net/10356/175258
• Institution: Nanyang Technological University

In the modern technological landscape, digital twins play a pivotal role in bridging the gap between the physical world and the data that drives it. They serve as tangible simulations of systems for improvement in processes such as product design, infrastructure planning and/or renewable energy management. This project focuses on improving set up processes of data center digital twins for the DcWiz platform; focusing primarily on chiller plants. Data centers produce copious amount of heat and consume even more energy. A data center digital twin can aid a user in pinpointing problem areas, monitoring operating conditions and simulating scenarios for future improvement. AI models can be trained on the data fed to the digital twin. These models can manage system configurations and reduce power consumption by making accurate decisions based on real-time data. The output of the models could be improved data center configurations and so quick turnover of digital twin configurations must be supported to enable this. The DcWiz platform is built upon WebGL, Unity and C#. A major pain point for the team when starting new projects or changing the layout is having to hard code it on Unity and C#. This is a time-consuming process which requires the architect to have in-depth technical knowledge of the Unity and C# assets available in the project. My focus area is on making the chiller plant 3D configuration more portable and configurable. This could then extend to other configuration types supported by DcTwin. As it is, chiller plant redesign requires technical knowledge of working with C# and Unity assets; skills stakeholders may not possess. By simplifying the design to a JSON and providing a more approachable Graphical User Interface (GUI), the turnover rate for redesign and testing is increased. This will help quicken development and optimisation for the chiller plant.