Title:
A new optimization method for reservoir systems
A new optimization method for reservoir systems
Author(s)
Iftikhar, Bilal
Advisor(s)
Georgakakos, Aris P.
Editor(s)
Collections
Supplementary to
Permanent Link
Abstract
Reservoir system management has been an active research area for several decades and has generated many notable optimization methods. However, the transfer of optimization methods to general practice has proven to be challenging. This research introduces a new approach suitable for realistically large systems that (i) overcomes computational shortcomings of traditional management methods and (ii) can be implemented by practicing engineers with traditional educational backgrounds.
The new approach, called Orthogonal Function Dynamic Programming (OFDP), brings together and extends classical results from function approximation theory to circumvent the computational limitations of ordinary DP (also known as the “curse of dimensionality”). Chebyshev polynomials are used as basis functions to develop accurate and efficient approximations of key DP functionals, markedly reducing the DP computational requirements while maintaining solution optimality and implementation ease. OFDP applications for progressively larger river basins in the southeast U.S. (including the Apalachicola-Chattahoochee-Flint (ACF), Alabama-Coosa-Tallapoosa (ACT), and Savannah river basins) demonstrate that the new approach is reliable and efficient. The second Thesis contribution leverages the new method to optimize the reservoir operation zones commonly employed (but empirically delineated) by reservoir management agencies, including the U.S. Army Corps of Engineers and the Bureau of Reclamation in the U.S. and water and power utilities and ministries worldwide. The relative performance of optimized versus currently used operation zones is assessed with respect to various performance metrics (including system hydropower output, reliability of meeting water supply targets and environmental flows, flood risk, and reservoir depletion and recovery during and after droughts, among others) in detailed, head-to-head, simulations with HEC-ResSim, an independent modeling system developed by the U.S. Army Corps of Engineers. These assessments demonstrate that the optimized zones (a) improve reservoir management and (b) provide an effective and easily implementable means to incorporate optimization and forecasting scientific advances in reservoir system management.
Sponsor
Date Issued
2022-05-10
Extent
Resource Type
Text
Resource Subtype
Dissertation