A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Investigation of the Optimization Models Performance in Inverse Transient Leak Detection Method
https://orcid.org/http://orcid.org/0000-0002-4916-8520
Since one of the most important steps in leak detection method based on inverse transient analysis is optimization, in the current paper the effect of the accurate selection of the optimization model was investigated on inverse transient leak detection method results. Three different optimization models, the logarithmic barrier function method, the mesh adaptive direct search method and the genetic algorithm from three general structures including gradient-based, gradient-free and evolutionary algorithms, were developed. The leak detection model was developed especially for viscoelastic pipes, and during optimization processes by using the three suggested models, the leaks were determined simultaneously with the wave speed, steady flow friction factor and also the coefficients of creep function of the viscoelastic model. In this regard, experimental data with different discharges through the leak and different leak locations were considered. The mesh adaptive direct search model which is based on gradient-free algorithm showed the best operation in determining the location and size of the leaks. It was also shown that the pressure head affected by leaks is not only influenced by mechanical losses but also by waves reflections caused by valve closure.
Investigation of the Optimization Models Performance in Inverse Transient Leak Detection Method
https://orcid.org/http://orcid.org/0000-0002-4916-8520
Since one of the most important steps in leak detection method based on inverse transient analysis is optimization, in the current paper the effect of the accurate selection of the optimization model was investigated on inverse transient leak detection method results. Three different optimization models, the logarithmic barrier function method, the mesh adaptive direct search method and the genetic algorithm from three general structures including gradient-based, gradient-free and evolutionary algorithms, were developed. The leak detection model was developed especially for viscoelastic pipes, and during optimization processes by using the three suggested models, the leaks were determined simultaneously with the wave speed, steady flow friction factor and also the coefficients of creep function of the viscoelastic model. In this regard, experimental data with different discharges through the leak and different leak locations were considered. The mesh adaptive direct search model which is based on gradient-free algorithm showed the best operation in determining the location and size of the leaks. It was also shown that the pressure head affected by leaks is not only influenced by mechanical losses but also by waves reflections caused by valve closure.
Investigation of the Optimization Models Performance in Inverse Transient Leak Detection Method
https://orcid.org/http://orcid.org/0000-0002-4916-8520
Since one of the most important steps in leak detection method based on inverse transient analysis is optimization, in the current paper the effect of the accurate selection of the optimization model was investigated on inverse transient leak detection method results. Three different optimization models, the logarithmic barrier function method, the mesh adaptive direct search method and the genetic algorithm from three general structures including gradient-based, gradient-free and evolutionary algorithms, were developed. The leak detection model was developed especially for viscoelastic pipes, and during optimization processes by using the three suggested models, the leaks were determined simultaneously with the wave speed, steady flow friction factor and also the coefficients of creep function of the viscoelastic model. In this regard, experimental data with different discharges through the leak and different leak locations were considered. The mesh adaptive direct search model which is based on gradient-free algorithm showed the best operation in determining the location and size of the leaks. It was also shown that the pressure head affected by leaks is not only influenced by mechanical losses but also by waves reflections caused by valve closure.
Investigation of the Optimization Models Performance in Inverse Transient Leak Detection Method
Iran J Sci Technol Trans Civ Eng
Mousavifard, Maryam (author)
2021-09-01
11 pages
Article (Journal)
Electronic Resource
English
Leak detection in pipelines by inverse backward transient analysis
| Online Contents | 2009
Leak detection in pipelines by inverse backward transient analysis
| British Library Online Contents | 2009
Leak detection in pipelines by inverse backward transient analysis
| Online Contents | 2009
Inverse Transient Analysis for leak detection in a PVC pipe network
| British Library Conference Proceedings | 2007
Investigation of transient leak detection using a 2D transient flow model
| British Library Conference Proceedings | 2005