A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Determining rainfall-intensity-duration-frequency relationship using Particle Swarm Optimization
Abstract This study proposes a Particle Swarm Optimization (PSO) algorithm to model the Rainfall-Intensity-Duration-Frequency (RIDF) relationship. The study is carried out under two scenarios. In scenario I, a data set with a length of 50 years is used. In Scenario II, the data set is extended to 68 years by adding the values of the recent 18 years. Scenario I is used for testing the robustness of the proposed PSO-RIDF model. The PSO-RIDF algorithm gives the same objective function value for different runs and this shows that the proposed algorithm is robust. Scenario II is used to investigate the influence of data length on model performance. It has been observed that the proposed PSO-RIDF model gives the same performance results as that of the Genetic Algorithm (GA) according to various error evaluation criteria. The PSO-RIDF model shows better performance than GA formulas when the number of parameters increases. It has also been observed that the length of the data set and the chosen formulation are influential on model performance. The weighting parameters of the RIDF model may be determined with PSO algorithm in one-stage instead of any statistical computations and/or trial-error procedure.
Determining rainfall-intensity-duration-frequency relationship using Particle Swarm Optimization
Abstract This study proposes a Particle Swarm Optimization (PSO) algorithm to model the Rainfall-Intensity-Duration-Frequency (RIDF) relationship. The study is carried out under two scenarios. In scenario I, a data set with a length of 50 years is used. In Scenario II, the data set is extended to 68 years by adding the values of the recent 18 years. Scenario I is used for testing the robustness of the proposed PSO-RIDF model. The PSO-RIDF algorithm gives the same objective function value for different runs and this shows that the proposed algorithm is robust. Scenario II is used to investigate the influence of data length on model performance. It has been observed that the proposed PSO-RIDF model gives the same performance results as that of the Genetic Algorithm (GA) according to various error evaluation criteria. The PSO-RIDF model shows better performance than GA formulas when the number of parameters increases. It has also been observed that the length of the data set and the chosen formulation are influential on model performance. The weighting parameters of the RIDF model may be determined with PSO algorithm in one-stage instead of any statistical computations and/or trial-error procedure.
Determining rainfall-intensity-duration-frequency relationship using Particle Swarm Optimization
Karahan, Halil (author)
KSCE Journal of Civil Engineering ; 16 ; 667-675
2012-04-29
9 pages
Article (Journal)
Electronic Resource
English
Determining rainfall-intensity-duration-frequency relationship using Particle Swarm Optimization
| Online Contents | 2012
Extreme Rainfall Nonstationarity Investigation and Intensity-Frequency-Duration Relationship
| British Library Online Contents | 2014
Rainfall Intensity-Duration-Frequency Formula for India.
| Online Contents | 1993