Calibration of the modified Bartlett-Lewis model using global optimization techniques and alternative objective functions

摘要

The calibration of stochastic point process rainfall models, such as of theBartlett-Lewis type, suffers from the presence of multiple local minima whichlocal search algorithms usually fail to avoid. To meet this shortcoming, fourrelatively new global optimization methods are presented and tested for theirability to calibrate the Modified Bartlett-Lewis Model. The list of testedmethods consists of: the Downhill Simplex Method, Simplex-SimulatedAnnealing, Particle Swarm Optimization and Shuffled Complex Evolution. Theparameters of these algorithms are first optimized to ensure optimalperformance, after which they are used for calibration of the Modified Bartlett-Lewis model.Furthermore, this paper addresses the choice of weights in the objectivefunction. Three alternative weighing methods are compared to determinewhether or not simulation results (obtained after calibration with the bestoptimization method) are influenced by the choice of weights.