Genetic programming for hydrological applications: to model or to forecast that is the question
Genetic programming (GP) is a widely used machine learning (ML) algorithm that has been applied in water resources science and engineering since its conception in the early 1990s. However, similar to other ML applications, the GP algorithm is often used as a data fitting tool rather than as a model...
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
IWA Publishing
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/d711d7302fe54ce1a470f710110acec0 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:d711d7302fe54ce1a470f710110acec0 |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:d711d7302fe54ce1a470f710110acec02021-11-05T17:48:51ZGenetic programming for hydrological applications: to model or to forecast that is the question1464-71411465-173410.2166/hydro.2021.179https://doaj.org/article/d711d7302fe54ce1a470f710110acec02021-07-01T00:00:00Zhttp://jh.iwaponline.com/content/23/4/740https://doaj.org/toc/1464-7141https://doaj.org/toc/1465-1734Genetic programming (GP) is a widely used machine learning (ML) algorithm that has been applied in water resources science and engineering since its conception in the early 1990s. However, similar to other ML applications, the GP algorithm is often used as a data fitting tool rather than as a model building instrument. We find this a gross underutilization of the GP capabilities. The most unique and distinct feature of GP that makes it distinctly different from the rest of ML techniques is its capability to produce explicit mathematical relationships between input and output variables. In the context of theory-guided data science (TGDS) which recently emerged as a new paradigm in ML with the main goal of blending the existing body of knowledge with ML techniques to induce physically sound models. Hence, TGDS has evolved into a popular data science paradigm, especially in scientific disciplines including water resources. Following these ideas, in our prior work, we developed two hydrologically informed rainfall-runoff model induction toolkits for lumped modelling and distributed modelling based on GP. In the current work, the two toolkits are applied using a different hydrological model building library. Here, the model building blocks are derived from the Sugawara TANK model template which represents the elements of hydrological knowledge. Results are compared against the traditional GP approach and suggest that GP as a rainfall-runoff model induction toolkit preserves the prediction power of the traditional GP short-term forecasting approach while benefiting to better understand the catchment runoff dynamics through the readily interpretable induced models. HIGHLIGHTS GP is used as a rainfall-runoff model induction toolkit rather than as a short-term forecasting mechanism.; Hydrological knowledge has been integrated with the machine learning algorithm.; This guides algorithm to induce physically sound and consistent model configurations.; The model building blocks are inspired by the Sugawara TANK model template.; They represent the elements of incorporated hydrological knowledge.;Herath Mudiyanselage Viraj Vidura HerathJayashree ChadalawadaVladan BabovicIWA Publishingarticlegenetic programmingmachine learningrainfall-runoff model inductionsugawara tank modeltheory-guided data scienceInformation technologyT58.5-58.64Environmental technology. Sanitary engineeringTD1-1066ENJournal of Hydroinformatics, Vol 23, Iss 4, Pp 740-763 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
genetic programming machine learning rainfall-runoff model induction sugawara tank model theory-guided data science Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066 |
| spellingShingle |
genetic programming machine learning rainfall-runoff model induction sugawara tank model theory-guided data science Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066 Herath Mudiyanselage Viraj Vidura Herath Jayashree Chadalawada Vladan Babovic Genetic programming for hydrological applications: to model or to forecast that is the question |
| description |
Genetic programming (GP) is a widely used machine learning (ML) algorithm that has been applied in water resources science and engineering since its conception in the early 1990s. However, similar to other ML applications, the GP algorithm is often used as a data fitting tool rather than as a model building instrument. We find this a gross underutilization of the GP capabilities. The most unique and distinct feature of GP that makes it distinctly different from the rest of ML techniques is its capability to produce explicit mathematical relationships between input and output variables. In the context of theory-guided data science (TGDS) which recently emerged as a new paradigm in ML with the main goal of blending the existing body of knowledge with ML techniques to induce physically sound models. Hence, TGDS has evolved into a popular data science paradigm, especially in scientific disciplines including water resources. Following these ideas, in our prior work, we developed two hydrologically informed rainfall-runoff model induction toolkits for lumped modelling and distributed modelling based on GP. In the current work, the two toolkits are applied using a different hydrological model building library. Here, the model building blocks are derived from the Sugawara TANK model template which represents the elements of hydrological knowledge. Results are compared against the traditional GP approach and suggest that GP as a rainfall-runoff model induction toolkit preserves the prediction power of the traditional GP short-term forecasting approach while benefiting to better understand the catchment runoff dynamics through the readily interpretable induced models. HIGHLIGHTS
GP is used as a rainfall-runoff model induction toolkit rather than as a short-term forecasting mechanism.;
Hydrological knowledge has been integrated with the machine learning algorithm.;
This guides algorithm to induce physically sound and consistent model configurations.;
The model building blocks are inspired by the Sugawara TANK model template.;
They represent the elements of incorporated hydrological knowledge.; |
| format |
article |
| author |
Herath Mudiyanselage Viraj Vidura Herath Jayashree Chadalawada Vladan Babovic |
| author_facet |
Herath Mudiyanselage Viraj Vidura Herath Jayashree Chadalawada Vladan Babovic |
| author_sort |
Herath Mudiyanselage Viraj Vidura Herath |
| title |
Genetic programming for hydrological applications: to model or to forecast that is the question |
| title_short |
Genetic programming for hydrological applications: to model or to forecast that is the question |
| title_full |
Genetic programming for hydrological applications: to model or to forecast that is the question |
| title_fullStr |
Genetic programming for hydrological applications: to model or to forecast that is the question |
| title_full_unstemmed |
Genetic programming for hydrological applications: to model or to forecast that is the question |
| title_sort |
genetic programming for hydrological applications: to model or to forecast that is the question |
| publisher |
IWA Publishing |
| publishDate |
2021 |
| url |
https://doaj.org/article/d711d7302fe54ce1a470f710110acec0 |
| work_keys_str_mv |
AT herathmudiyanselagevirajviduraherath geneticprogrammingforhydrologicalapplicationstomodelortoforecastthatisthequestion AT jayashreechadalawada geneticprogrammingforhydrologicalapplicationstomodelortoforecastthatisthequestion AT vladanbabovic geneticprogrammingforhydrologicalapplicationstomodelortoforecastthatisthequestion |
| _version_ |
1718444130726576128 |