Aim and Scope
Machine learning (ML) has seen tremendous recent successes in many areas of artificial intelligence such as vision, natural language processing (NLP), and robotics. It has since sparked great interests in its potential use in power systems. This is enabled by both a) the continuing proliferation of data collected in the power systems from, e.g., Phasor Measurement Units (PMUs) and smart sensors, and b) the significant growth of computational resources that are essential for processing the massive amount of data. It is of great interest to effectively derive actionable information from power system data and to thereby significantly enhance system efficiency, reliability, security and resilience. Beyond processing the big data, machine learning has also played a major role in leveraging information from the physical models of power systems. Notable examples include learning-based power system state estimation given low observability, initializing optimal power flow, complex load modeling, asset monitoring and bad data detection. Data analytics can also help power system operators to calibrate and verify their physical system models for more effective operation and planning. The marriage of data-driven and model-based approaches under a unified ML framework has further enabled us to attack many more challenging problems, e.g., addressing rare and extreme events in power systems where data collection is fundamentally limited. Last but not least, ML has seen tremendous potential in designing more effective control policies in a shorter time, in particular via reinforcement learning.
As such, it is evident that ML is on the brink of having a tremendous impact on enhancing power system operation and planning. This workshop aims to bring together researchers and practitioners in the field of machine learning and big data analytics for power transmission systems. Topics of interest in which ML plays a major role include but are not limited to:
|
|
