HR: 0800h
AN: H21A-1310    [Abstracts]
TI: Parameter estimation using Multiobjective Particle Swarm Optimization (MOPSO)
AU: * Gill, M
EM: kashif@cc.usu.edu
AF: Utah Water Research Laboratory, 8200 Old Main Hill, Logan, UT 84322-8200 United States
AB: In the current application, a multiobjective optimization approach is presented for estimation of parameters of hydrologic models. The complexity of hydrologic processes demands efficient and effective tools to fully determine system characteristics. A relatively new optimization algorithm, known as particle swarm optimization (PSO) has been employed here for parameter estimation. The PSO algorithm comes from the family of evolutionary computation techniques and has been applied in various other fields. The approach was initially devised for a single objective function, but in the current application we introduce a multiobjective algorithm, called multiobjective particle swarm optimization (MOPSO), and test it on two different kinds of modeling efforts in hydrology, namely a support vector machine (SVM) model for predicting soil moisture, and a well known conceptual rainfall-runoff (CRR) model, the Sacramento Soil Moisture Accounting (SAC-SMA) model, for estimating streamflow. The algorithm is modified to address multiobjective problems by introducing the Pareto rank concept. The performance of the algorithm is also tested for two test functions.
DE: 0555 Neural networks, fuzzy logic, machine learning
DE: 0560 Numerical solutions (4255)
DE: 1860 Streamflow
DE: 1866 Soil moisture
DE: 1894 Instruments and techniques: modeling
SC: Hydrology [H]
MN: Fall Meeting 2005