Comparative Analysis of ANN, GEP, and Water Advance Power Function for Predicting Infiltrated Water Volume in Furrow of Permeable Surface

Water, Год журнала: 2025, Номер 17(9), С. 1304 - 1304

Опубликована: Апрель 27, 2025

The present investigation utilized artificial neural networks (ANN) and gene expression programming (GEP) in comparison with the two-point method (TPM) to develop a generalized solution for predicting infiltrated water volume (∀Z) across various soil types under furrow conditions. This work assesses infiltration behavior respect experimental data from several temporal contexts. Data distribution model performance are evaluated via descriptive statistics correlation tests. Artificial intelligence (AI) models (ANN GEP) trained utilizing input variables—inflow rate (Qin); length (L); waterfront advance time at end of (TL); opportunity (To); cross-sectional area inflow (Ao) compared TPM performance. More precisely consistently than power function, AI-based algorithms hope be invading volume. Statistical analysis shows that ANN GEP have lower error metrics, increased generalizability, better representation complex dynamics. determination coefficient (R2) produced 98.1% testing 97.8% validation, while showed accuracy reductions 2.5% 4.6%, respectively. On other side, R2 95.7% 96.1% 0.7% 3%, During computation, TPMs root mean square (RMSE) 0.0135 m3/m exceeded all values. Errors within 10% relative deviation were displayed using ∀Z. Particularly, GEP, study revealed AI techniques predict irrigation penetration function. These adaptation, extrapolation, accuracy. Results show AI-driven modeling may maximize hydrological assessments control.

Язык: Английский

Research on Olympic medal prediction based on GA-BP and logistic regression model

F1000Research, Год журнала: 2025, Номер 14, С. 245 - 245

Опубликована: Март 28, 2025

Background Predicting the number and distribution of Olympic medals in future has become a hot topic, but predicting is not easy requires comprehensive consideration multiple factors such as historical data, athlete performance, host country effects. Method This article uses GA-BP algorithm model, combined with genetic (GA) backpropagation neural network (BPNN), to optimize weights bias parameters BP using global search capability algorithm, thereby improving training efficiency prediction performance. By estimating gold total medals, verifying accuracy medal table for 2028 Los Angeles Olympics. Meanwhile, based on synthetic control Estonia China were selected research subjects construct virtual group two experimental groups analysis. Result The results showed that won more head coach than without one. In 1992, 1 2 bronze under guidance excellent coaches, indicating significant role coaches athletes’ Conclusion study provides valuable insights decision-making Committee, revealing key distribution, optimizing allocation national strategic resources, performance countries at Games.

Язык: Английский