Comparison of scenario reduction approaches for reservoir inflow timeseries generated by a Bayesian Neural Network
Ja-Ho Koo, Edo Abraham, Andreja Jonoski, Dimitri P. Solomatine, Namal Rathnayake, Babak Mohammadi, Anurag Barthwal, Ziqiang Zeng, Nezir Aydin, Nezir Aydin · 2026
Read Comparison of scenario reduction approaches for reservoir inflow timeseries generated by a Bayesian Neural Network by Ja-Ho Koo, Edo Abraham, Andreja Jonoski, Dimitri P. Solomatine, Namal Rathnayake, Babak Mohammadi, Anurag Barthwal, Ziqiang Zeng, Nezir Aydin, Nezir Aydin (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13215500.txt.
Description
Ja-Ho Koo and colleagues present a 2026 methods study on shrinking large sets of possible reservoir inflow futures down to a manageable few. Flood control decisions need these scenarios, yet too many of them make optimization impractical.
The team generated inflow time series with a Bayesian neural network, which captured short-term patterns well but missed peak flows. They then compared four distance measures for picking a smaller representative set. Energy distance best preserved the overall statistics, while Manhattan and Euclidean distances kept the extreme high-water scenarios that matter most for flood safety. The authors recommend those two measures with about thirty scenarios as a practical balance.
Catalog record
| CREATOR | Ja-Ho Koo, Edo Abraham, Andreja Jonoski, Dimitri P. Solomatine, Namal Rathnayake, Babak Mohammadi, Anurag Barthwal, Ziqiang Zeng, Nezir Aydin, Nezir Aydin |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 112 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |