Prediction of rain garden runoff control effects based on multiple machine learning techniques DOI
Xingli Jia, Qi Yang,

Hui Liang

и другие.

Environmental Technology, Год журнала: 2025, Номер unknown, С. 1 - 12

Опубликована: Янв. 29, 2025

Due to the rapid development of urbanisation, cities frequently experience waterlogging during rainfall. Rain gardens are widely used in new urban construction because they effectively control surface runoff from rainwater, thereby reducing waterlogging. The effectiveness rain is influenced by multiple factors. This paper predicts effects using models. By constructing five experimental structures, 240 sets rates for garden structures were collected build a database. Feature correlation analysis identified four input parameters: rainfall recurrence interval, storage layer depth, catchment area, and infiltration rate. Using BP, SVM, Random Forest, initial predictive models established. To enhance accuracy models, Zebra Optimization Algorithm was employed optimisation, model performance characterised coefficient determination, mean squared error, absolute error. results show that ZOA-BP has best prediction on test set, (R2) 0.979, RMSE 2.331, which verifies validity model. research outcome can provide references application expected reduce design operational costs related projects.

Язык: Английский

Prediction of rain garden runoff control effects based on multiple machine learning techniques DOI
Xingli Jia, Qi Yang,

Hui Liang

и другие.

Environmental Technology, Год журнала: 2025, Номер unknown, С. 1 - 12

Опубликована: Янв. 29, 2025

Due to the rapid development of urbanisation, cities frequently experience waterlogging during rainfall. Rain gardens are widely used in new urban construction because they effectively control surface runoff from rainwater, thereby reducing waterlogging. The effectiveness rain is influenced by multiple factors. This paper predicts effects using models. By constructing five experimental structures, 240 sets rates for garden structures were collected build a database. Feature correlation analysis identified four input parameters: rainfall recurrence interval, storage layer depth, catchment area, and infiltration rate. Using BP, SVM, Random Forest, initial predictive models established. To enhance accuracy models, Zebra Optimization Algorithm was employed optimisation, model performance characterised coefficient determination, mean squared error, absolute error. results show that ZOA-BP has best prediction on test set, (R2) 0.979, RMSE 2.331, which verifies validity model. research outcome can provide references application expected reduce design operational costs related projects.

Язык: Английский

Процитировано

1