Data-Driven State and Parameter Estimation in Power Distribution Systems

With continually increasing renewable energy penetration, higher requirements are being imposed on the modeling and monitoring of distribution networks, which calls for transformative solutions to a number of issues including missing information on topology, phase connection and network parameters, lack of sensing and communication infrastructure to cover observability, diverse sampling rates and accuracy classes of measurement data, and frequent unmonitored events. Recent advancement of data analytics and machine learning are providing new opportunities to address these long-standing challenges. This panel will discuss the rise of data-driven approaches in state and parameter estimation of distribution networks. Panelists will present theoretical frameworks and practical results to showcase the possibility of handling the aforementioned challenges by mining a variety of data assets in distribution network operation.