Public Data Repositories for Power Systems Operation and Machine Learning Applications

Machine learning and artificial intelligence (ML/AI) applications to power systems engineering problems have gained a lot of attention in recent years, however, applications have been limited by several factors. First, there is a lack of publicly available datasets in power engineering. Secondly, use cases and methodologies for ML/AI applications using real-world datasets that are already being collected by utilities are unclear, despite a significant amount of effort in academia and private industry. This panel will discuss some of the efforts underway to address these limitations, including the curation and publication of an open data repository for grid disturbance signatures, and an ML/AI application using a large amount of data for scalable demand response. Specifically, we will discuss data collection and anonymization efforts towards public repositories, and their potential applications for power systems operation such as demand response using ML/AI technologies on these datasets. Presentations in this panel session: - An Overview of Public Datasets and Tools for Artificial Intelligence (23PESGM4211)