Papers

Conference paper

M. S. Samhan, J. Guerrero Orbe, N. Haghdadi, D. Sim, A. Bruce and B. Yildiz, "Reverse power flow sign correction for meter-limited substation data," accepted for the Australasian Universities Power Engineering Conference (AUPEC), November 2026.

Journal article (in preparation)

A journal article on the current method, M9, is in preparation. This page will cite it once it is published.

Citing the software

If you use PyNRPF, please cite it using the metadata in CITATION.cff, or the archived release:

Samhan, M. S., Bruce, A., & Yildiz, B. (2026). PyNRPF: Detecting and rectifying wrong reverse power flow readings [Software]. Zenodo. https://doi.org/10.5281/zenodo.18764943

Show details

About the conference paper

The paper identifies the wrong reverse power flow (RPF) sign that magnitude-only substation meters produce in Australian distribution networks. It presents two earlier methods, a deterministic rule (M7) and a machine-learning classifier (M8), which detect the error at both the day level and the reading level. It also reports a case study in the Ausgrid network on how correcting the sign changes minimum demand. The code and data that produced it are archived, unchanged, under publication/1_conference_paper/, and the software version it used is release 0.3.1 on Zenodo.

About the journal article

The journal article covers how M9 works, its validation on reviewed real-world errors, and the effect of sign correction on net-load forecasting. Its code and data are under publication/2_journal_article/; Data, Evaluation and Reproduce the paper describe them.

Version DOIs

The software DOI above always resolves to the latest archived release; each release also has its own version DOI on Zenodo.