The analysis of reservoir operation rules in the southwestern United States using a hidden Markov-decision tree model

Realistic reservoir operation is an important component in basin hydrological models. However, the operation rules can be complex under various situations and are not well understood yet. The objective of the study is to find out patterns of actual reservoir operation rules in a specific region under different conditions. The results are helpful to understand realistic reservoir operation behaviors. Moreover, the work will contribute to building a reservoir component for hydrological models and provide references for improving reservoir operation decisions in the future. A hidden Markov-decision tree (HM-DT) model is applied in this study. Hidden Markov Model is used to describe the transition between operation states while each state corresponds to a decision tree model to determine daily reservoir releases. This study collected the historical records of 66 reservoirs in the southwestern United States and 8 reservoirs in the northern United States and identified their regional representative operation modules. A comparison between the representative operation modules was conducted to show the spatial variability of reservoir operations. Further, by analyzing the historical operation patterns of several reservoirs, the transitions between operation rules in different time periods were explored, including periodicity and special responses to floods. The operation rules adopted by reservoirs with different operation functions are also analyzed and compared. A robustness analysis was performed to confirm the robustness of the HM-DT model outputs. Finally, several attempts have been made to link the HM-DT model with Soil and Water Assessment Tool (SWAT) and future work is proposed.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Civil Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2021

Author dc:creator

• Lyu, Lingqi

Contributors dc:contributor

• Cai, Ximing

• Reservoir operation
• hidden Markov model

Statement dc:rights

• Copyright 2021 Lingqi Lyu

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/110740

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/110740