SZSong Zhang

Song Zhang

*This is an algorithmically generated author record based on publicly available information.

Song Zhang has a total of 15 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Grid Modernization in the Age of Artificial Intelligence and Cloud: Synergies Between Grid Modernization, Artificial Intelligence, and Cloud

    by Song Zhang, Amritanshu Pandey, Deepa Pahuja, Tom Nudell, Junhui Zhao, Ranjan Banerji, Xiaochuan Luo, Shaohui Liu

    The electric power grid faces unprecedented challenges as it evolves to meet rising energy demands, integrate renewable sources, and adapt to new technologies. This paper explores how cloud computing and artificial intelligence (AI) are emerging as transformative solutions for grid modernization. Cloud platforms provide the scalable infrastructure and computing power needed to process vast amounts of grid data, while AI enables advanced analytics and decision-making capabilities. Together, these technologies are revolutionizing grid operations across multiple domains. Key applications include intelligent solutions for data center integration, accelerating generator interconnection processes, facilitating distributed energy resource management at scale, enabling data-driven predictive maintenance, enhancing grid resilience and self-healing, optimizing energy generation and distribution, and implementing advanced cybersecurity measures. The article discusses several real-world implementations demonstrating significant operational improvements and cost savings for utilities. It also examines emerging trends like generative AI and multiagent systems that promise to further enhance grid capabilities. While highlighting the immense potential of cloud and AI technologies, the authors emphasize that successful adoption requires a holistic approach encompassing organizational change, workforce development, and regulatory evolution. Looking ahead, the convergence of cloud and AI is poised to create a more intelligent, efficient, and reliable power grid to meet future energy needs. However, realizing this vision will require ongoing collaboration among utilities, technology providers, regulators, and researchers.


  • Practical Adoption of Cloud Computing in Power Systems—Drivers, Challenges, Guidance, and Real-World Use Cases

    by Song Zhang, Amritanshu Pandey, Xiaochuan Luo, Maggy Powell, Ranjan Banerji, Lei Fan, Abhineet Parchure, Edgardo Luzcando

    Motivated by the Federal Energy Regulatory Commission’s (FERC) recent direction and ever-growing interest in cloud adoption by power utilities, a Task Force was established to assist power system practitioners with secure, reliable and cost-effective adoption of cloud technology to meet various business needs. This paper summarizes the business drivers, challenges, guidance, and best practices for cloud adoption in power systems from the Task Force’s perspective, after extensive review and deliberation by its members, including grid operators, utility companies, software vendors, and cloud providers. The paper begins by enumerating various business drivers for cloud adoption in the power industry. It follows with the discussion of the challenges and risks of migrating power grid utility workloads to the cloud. Next, for each corresponding challenge or risk, the paper provides appropriate guidance. Notably, the guidance is directed toward power industry professionals who are considering cloud solutions and are yet hesitant about the practical execution. Finally, to tie all the sections together, the paper documents various real-world use cases of cloud technology in the power system domain, which both the power industry practitioners and software vendors can look toward to design and select their own future cloud solutions. We hope that the information in this paper will serve as helpful guidance for the development of NERC guidelines and standards relevant to cloud adoption in the industry.


  • Synchrophasor-Based Emergency Generation Control for Area Balancing

    by Eugene Litvinov, Song Zhang, Xiaochuan Luo, Tongxin Zheng

    In North America, Balancing Authority (BA) maintains the demand-supply balance and frequency stability within its responsible area. In case of an emergency such as loss of Supervisory Control and Data Acquisition (SCADA), Automatic Generation Control (AGC), State Estimation and Energy Management System (EMS), system operators lose the automatic monitoring and control capability, as a result of which they have to manually dispatch units to maintain area balancing. This could be inefficient and at the risk of failure to comply with NERC’s balancing control performance standards. We propose an emergency control strategy to help BA balance the system cost-effectively based on the independent synchrophasor infrastructure in the event of SCADA/EMS failure. Depending on the availability of Electronic Dispatch (ED) network, the proposed emergency control decides if a synchrophasor-based AGC-like function should be performed every four seconds to regulate Area Control Error (ACE). It also executes economic dispatch on a 5-minute basis by solving the Desired Dispatch Point (DDP) of units with PMU installed through a linear programming problem to meet forecasted load change in five minutes. Closed-loop simulations demonstrate that the proposed scheme is able to successfully balance the system at a lower cost to fulfill the BA’s obligations under emergency operation conditions.


  • Practical Design and Implementation of Cloud Computing for Power System Planning Studies

    by Xiaochuan Luo, Song Zhang, Eugene Litvinov

    The increased penetration of renewable resources and retirement of traditional generation units are posing new challenges to power system operations and planning to meet the ever-growing demand for more computing resources to run large-scale, computationally intensive simulations. As the traditional on-premises infrastructure loses pace to the constantly increasing computing needs, cloud computing has become an attractive addition for increasing average computing capacity, and alternative for meeting peak demands. ISO New England (ISO-NE) pioneered the use of cloud computing technology for large-scale power system simulations, believing it to be the only viable solution that is capable of keeping pace with the growing challenge in the foreseeable future. This paper shares the practical design and implementation of an elastic, on-demand, application-independent cloud platform that ISO-NE deployed in production at the end of 2017. As a secure, cost-effective and scalable platform for addressing the intensive computing demand in planning studies, this platform made using Amazon Web Service (AWS) as easy as using on-premises computers by completely automating the access to cloud based resources. This frees ISO engineers to focus on power system analysis, and thus significantly increased their productivity. To accommodate simulation jobs with varying urgency requirements, this platform allows the users to make the tradeoff between cost and completion time. The platform also further strengthened its support for cyber security and data privacy.