Physics-guided Deep Learning for Power System State Estimation

In the past decade,dramatic progress has been made in the field of machine learning.This paper explores the possibility of applying deep learning in power system state estimation.Traditionally,physics-based models are used including weighted least square(WLS)or weighted least absolute value(WLAV).These models typically consider a single snapshot of the system without capturing temporal correlations of system states.In this paper,a physics-guided deep learning(PGDL)method is proposed.Specifically,inspired by autoencoders,deep neural networks(DNNs)are used to learn the temporal correlations.The estimated system states from DNNs are then checked against physics laws by running through a set of power flow equations.Hence,the proposed PGDL is both data-driven and physics-guided.The accuracy and robustness of the proposed PGDL method are compared with traditional methods in standard IEEE cases.Simulations show promising results and the applicability is further discussed.

An HPC Based Real-time Path Rating Calculation Tool for Congestion Management with High Penetration of Renewable Energy

Every year, transmission congestion costs billions ofdollars for electricity customers. This clearly identifies the criticalneed for more transmission capacity and also poses big challengesfor power grid reliability in stressed conditions due to heavyloading and in uncertain situations due to variable renewableresources and responsive smart loads. However, it becomesincreasingly difficult to build new transmission lines, whichtypically involve both economic and environmental constraints.In this paper, advanced computing techniques are developedto enable a non-wire solution that realizes unused transfercapabilities of existing transmission facilities. An integratedsoftware prototype powered by high-performance computing(HPC) is developed to calculate ratings of key transmission pathsin real time for relieving transmission congestion and facilitatingrenewable integration, while complying with the North AmericanElectric Reliability Corporation (NERC) standards on assessingtotal transfer capabilities. The innovative algorithms include: (1)massive contingency analysis enabled by dynamic load balancing,(2) parallel transient simulation to speed up single dynamicsimulation, (3) a non-iterative method for calculating voltagesecurity boundary and (4) an integrated package consideringall NERC required limits. This tool has been tested on realisticpower system models in the Western Interconnection of NorthAmerica and demonstrates satisfactory computational speedusing parallel computers. Various benefits of real-time path ratingare investigated at Bonneville Power Administration using realtime EMS snapshots, demonstrating a significant increase in pathlimits. These technologies would change the traditional goals ofpath rating studies, fundamentally transforming how the grid isoperated, and maximizing the utilization of national transmissionassets, as well as facilitating integration of renewable energy andsmart loads.