Many-objective Shuffled Complex-Self Adaptive Hybrid EvoLution (MSC-SAHEL): A flexible optimization framework for water-energy systems
Abstract
Both water and energy systems operation require multi-criterion decision making to mitigate conflicting goals in practice. With the recently developed Shuffled Complex Evolution-Self Adaptive Hybrid EvoLution (SC-SAHEL) algorithm, and its ancestor - the Shuffled Complex Evolution developed at the University of Arizona (SCE-UA), we develop a many-version of the algorithm called Many-SC-SAHEL (MSC-SAHEL) to address the importance of Pareto optimality and trade-offs in water resources management. The core concept of the MSC-SAHEL is to utilize multiple Evolutionary Algorithms (EAs) in a parallel scheme to boost the efficiency of identifying global non-dominated solutions. The new algorithm can tackle a wide range of optimization problems by selecting the most suitable EAs for the problem in hand. In this process, MSC-SAHEL reveals the pros and cons of EAs for class of optimization problems in an "award" and "punishment" scheme, which allocates more resources to the most efficient and effective EAs for the problem. We benchmark the new MSC-SAHEL algorithm on several conceptual test functions and a flexible, many-reservoir model to the performance of the algorithm. Our study reveals the potential of the newly developed MSC-SAHEL algorithm for solving a broad class of complex Many-optimization problems. The new algorithm is also flexible for various EAs, boundary-handing, and initial sampling methods. Hence, the algorithm provides an arsenal of tools for solving a wide range of water-energy optimization problems in an effective and efficient way.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFM.H52C..04R
- Keywords:
-
- 3305 Climate change and variability;
- ATMOSPHERIC PROCESSES;
- 1655 Water cycles;
- GLOBAL CHANGE;
- 1878 Water/energy interactions;
- HYDROLOGY;
- 1880 Water management;
- HYDROLOGY