NewHydrogen Features Insights on AI’s Role in Power Grids and Data Centers

NewHydrogen, Inc. (OTCMKTS: NEWH), the company behind ThermoLoop™, has unveiled a podcast featuring CEO Steve Hill and Dr. Vladimir Dvorkin, an Assistant Professor of Electrical Engineering and Computer Science at the University of Michigan. The discussion delves into the evolving relationship between artificial intelligence (AI), data centers, and power grids, along with challenges in data security and critical infrastructure integration.

Dr. Dvorkin noted that data centers are becoming significant components of modern infrastructure, capable of impacting power grids dramatically. He explained that electrical loads can fluctuate rapidly due to the demands of AI processing, and emission regulations may determine which data centers should handle heavier workloads at specific times. He highlighted the adaptability of AI training algorithms, which can be modified to reduce electricity consumption at data centers, effectively smoothing load profiles with minimal performance impact. He emphasized that electricity availability is less location-dependent compared to natural gas.

Addressing data security in power systems, Dr. Dvorkin acknowledged the sensitivity of information tied to the electric grid, a critical national infrastructure. He explained that while grid models exist, the need to maintain confidentiality can complicate management. Modern technologies such as differential privacy and optimization, he said, offer solutions by enabling the synthesis of power grid models without over-disclosing sensitive data. He described it as a delicate trade-off between revealing enough information to refine trading strategies and ensuring system resilience.

On the integration of AI with critical infrastructure, Dr. Dvorkin discussed the need for data center operators, such as Google, Microsoft, and Amazon, to anticipate long-term power demands. He cited the recent agreement between Microsoft and the Three Mile Power Plant, which secures energy supply for 20 years, as an example of proactive planning. He also highlighted short-term efforts, such as Google’s coordination with local power companies to smooth out data center demand during peak hours, thereby minimizing grid stress. He pointed out the necessity for flexibility to accommodate scenarios like sudden power plant outages.

Dr. Vladimir Dvorkin, who earned his Ph.D. in Electrical Engineering from the Technical University of Denmark, now serves as an Assistant Professor at the University of Michigan. Prior to his current role, he worked as a postdoctoral fellow with the Laboratory for Information and Decision Systems (LIDS) and the Energy Initiative at the Massachusetts Institute of Technology. Leading the ⏻ptiML group, Dr. Dvorkin focuses on harnessing optimization and machine learning to advance digital and carbon-free power systems. His scholarly work is available through Google Scholar under his profile, Vladimir Dvorkin.

Have you read?
Countries: Women in the workforce.
Countries: Personal space.
World’s Most (And Least) Religious Countries.
Best Countries to Invest In Travel, Tourism, and Hospitality.
Most Forested Countries In The World.