一、Opening defense question：DATA CENTER ENERGY COST MINIMIZATION TECHNIQUES THROUGH EFFECTIVE UTILIZATION OF RENEWABLE ENERGY SOURCES USING DEEP LEARNING AND REINFORCEMENT LEARNING
二、Opening defense：Abubakar Rafique
三、Opening and defense time: 11:30 am on June 15, 2024
四、Opening and defense location: Conference Room 430, School of Automation
五、Introduction to the content of the opening defense：
The surge in internet services has escalated reliance on cloud data centers, leading to increased energy consumption and costs. Our research introduces renewable energy-powered data centers utilizing distributed sources to meet energy demands. Traditional optimization methods often neglect renewables due to their intermittent nature. We address this with a Deep Learning-based renewable energy supply prediction algorithm, complemented by a data center energy demand prediction and job scheduling algorithm. The project includes three work packages: (a) Develop a zonal-based tier-III data center architecture to optimize workload distribution, ensuring maximum uptime, minimal job queue time, optimal heat rate, and reduced carbon emissions. (b) Create an advanced renewable energy forecasting algorithm to manage renewable source intermittency. (c) Implement a distributed renewable energy sources mapping strategy to minimize reliance on conventional energy. This integration aims to enhance data center sustainability and efficiency, addressing environmental and economic challenges.
六、Introduction of the Proposing Defender
Abubakar Rafique, PhD candidate in Electrical Engineering, School of Automation, Northwestern Polytechnical University. Research focuses on machine learning, deep learning, optimization and energy management.
