TY - GEN
T1 - Invertible Neural Network for Consistent State Estimation in Distribution Grid with Unobservability
AU - Yuan, Jingyi
AU - Weng, Yang
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - State estimation (SE) serves as the basis for monitoring and control, but the performance is challenged when both prior knowledge and observability are limited due to increasing distribution system extension and renewable penetration, typically on system edges. To solve the problem, machine learning approaches have been recently applied to approximate the mapping from measurements to system states for SE. However, such direct approximation of the inverse system process lacks physical consistency with the forward underlying model (i.e., power flow equations), leading to inaccurate or even physically infeasible SE solutions. Thus, we propose a two-way learning method by designing tractably invertible paths in structural neural networks, which build a perfectly matched forward-inverse system model to estimate states. For the physical consistency of learning, we make the most of prior power system knowledge to compensate for information loss against unobservability and contract feasible SE solutions in the inverse learning process. Specifically, the designs are threefold to regularize SE performance: embedding interpretable power flow basis functional, regularizing dimensional homogeneity, and complementing latent variables. We test the SE performance of invertible learning via extensive simulations on IEEE test systems and a utility distribution grid. Numerical results show high accuracy, degradability of the data-driven model, and robustness to data quality issues.
AB - State estimation (SE) serves as the basis for monitoring and control, but the performance is challenged when both prior knowledge and observability are limited due to increasing distribution system extension and renewable penetration, typically on system edges. To solve the problem, machine learning approaches have been recently applied to approximate the mapping from measurements to system states for SE. However, such direct approximation of the inverse system process lacks physical consistency with the forward underlying model (i.e., power flow equations), leading to inaccurate or even physically infeasible SE solutions. Thus, we propose a two-way learning method by designing tractably invertible paths in structural neural networks, which build a perfectly matched forward-inverse system model to estimate states. For the physical consistency of learning, we make the most of prior power system knowledge to compensate for information loss against unobservability and contract feasible SE solutions in the inverse learning process. Specifically, the designs are threefold to regularize SE performance: embedding interpretable power flow basis functional, regularizing dimensional homogeneity, and complementing latent variables. We test the SE performance of invertible learning via extensive simulations on IEEE test systems and a utility distribution grid. Numerical results show high accuracy, degradability of the data-driven model, and robustness to data quality issues.
KW - Distribution system edges
KW - invertible NN
KW - physical consistency
KW - state estimation
KW - two-way learning
KW - unobservability
UR - http://www.scopus.com/inward/record.url?scp=85174690565&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85174690565&partnerID=8YFLogxK
U2 - 10.1109/PESGM52003.2023.10253055
DO - 10.1109/PESGM52003.2023.10253055
M3 - Conference contribution
AN - SCOPUS:85174690565
T3 - IEEE Power and Energy Society General Meeting
BT - 2023 IEEE Power and Energy Society General Meeting, PESGM 2023
PB - IEEE Computer Society
T2 - 2023 IEEE Power and Energy Society General Meeting, PESGM 2023
Y2 - 16 July 2023 through 20 July 2023
ER -