A Computationally Efficient Technique for Source Identification Problems in Three-Dimensional Aquifer Systems Using Neural Networks and Simulated Annealing
| Autor: | S. V. N. Rao |
|---|---|
| Rok vydání: | 2006 |
| Předmět: |
Simulation optimization
geography Mathematical optimization Engineering geography.geographical_feature_category Artificial neural network business.industry Aquifer Management Monitoring Policy and Law Identification (information) Decision variables Flow (mathematics) Simulated annealing business Waste Management and Disposal |
| Zdroj: | Environmental Forensics. 7:233-240 |
| ISSN: | 1527-5930 1527-5922 |
| Popis: | A non-gradient–based simulation/optimization technique is presented for source identification problems involving discrete and continuous decision variables. In this approach an artificial neural network is used as surrogate simulator of an existing density-dependent flow and transport model and simulated annealing algorithm as the optimizer. The source location and its strength are determined explicitly in a three-dimensional aquifer system. The methodology is efficient and robust and reduces the computational time burden to be negligible. Although a single source is demonstrated in this study, the methodology can be extended to multiple sources and problems involving unknown aquifer parameters. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|