Developing a data-driven model for dynamic reservoir operation using a combined hidden Markov-decision tree and classification tree algorithms
Chen, Yanan
Content Files
CHEN-THESIS-2021.pdf
Loading…
Download Files
Loading…
Download Counts (All Files)
Loading…
Edit File
Loading…
Permalink
https://hdl.handle.net/2142/113159
Description
Title
Developing a data-driven model for dynamic reservoir operation using a combined hidden Markov-decision tree and classification tree algorithms
Author(s)
Chen, Yanan
Issue Date
2021-07-13
Director of Research (if dissertation) or Advisor (if thesis)
Cai, Ximing
Department of Study
Civil & Environmental Eng
Discipline
Civil Engineering
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
M.S.
Degree Level
Thesis
Date of Ingest
2022-01-12T22:35:06Z
Keyword(s)
reservoir operation simulation
hidden Markov-decision tree model
classification tree
Abstract
Reservoir operations are faced with greater challenges than before due to growing water demands and climate change, and thus understanding and improvement of reservoir operations are critical. This study extends the hidden-Markov-decision tree (HM-DT) model developed by Zhao and Cai (2020) and proposes a data-driven reservoir operation model (DROM). The HM-DT model is first applied to individual reservoirs to derive sets of representative operation modules. Then a module classification model based on the Classification and Regression-tree algorithm is developed to determine which module to use for a day. DROM combines the derived operation modules and the module classification model to realize daily release prediction. DROM is tested with 25 reservoirs operated by USBR in north Great Plains regions, and it is shown that DROM can achieve acceptable accuracy in simulating historical releases (NSE > 0.4) and predicting future releases (NSE > 0.2) for 23 reservoirs. Compared with existing data-driven models, DROM shows several advantages including easily satisfied data requirements, transparent model structure, and broad applicability to various reservoirs. Especially, DROM can simulate the dynamic operation patterns through choosing the modules, while other previous models can only derive static operation rules. DROM can be used to better understand real-world reservoir operation behaviors and to explore the improvement of operation via combining with optimization models.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
CHEN-THESIS-2021.pdf