Probabilistic Power Flow Based on a Gaussian Process Emulator

Journal Article · Wed Apr 01 00:00:00 EDT 2020 · IEEE Transactions on Power Systems

DOI:https://doi.org/10.1109/tpwrs.2020.2983603· OSTI ID:1668501

Xu, Yijun ^[1]; Hu, Zhixiong ^[2]; Mili, Lamine ^[1]; Korkali, Mert ^[3]; Chen, Xiao ^[3]

---

1. Virginia Polytechnic Inst. and State Univ. (Virginia Tech), Falls Church, VA (United States)
2. Univ. of California, Santa Cruz, CA (United States)
3. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)

In this letter, a novel Gaussian process emulator is proposed, for the first time, to conduct the probabilistic power-flow calculation. Based on Bayesian inference, a Gaussian process emulator is trained and served as a nonparametric, reduced-order model of the nonlinear power-flow model. This emulator has allowed us to evaluate the time-consuming power-flow solver at the sampled values with a negligible computational cost. The simulations reveal the excellent performance of this method.

• Research Organization: Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
• Sponsoring Organization: National Science Foundation (NSF); USDOE National Nuclear Security Administration (NNSA); USDOE Office of Electricity (OE)
• Grant/Contract Number: AC52-07NA27344
• OSTI ID: 1668501
• Report Number(s): LLNL-JRNL--790158; 988765
• Journal Information: IEEE Transactions on Power Systems, Journal Name: IEEE Transactions on Power Systems Journal Issue: 4 Vol. 35; ISSN 0885-8950
• Publisher: IEEE
• Country of Publication: United States
• Language: English

Related Subjects

42 ENGINEERING
Gaussian process emulator
Latin hypercube sampling
copula
probabilistic power flow