Performance improvement for infiltration rate prediction using hybridized Adaptive Neuro-Fuzzy Inferences System (ANFIS) with optimization algorithms

The infiltration process during irrigation is an essential variable for better water management and hence there a need to develop accurate model estimate the amount irrigation. However, fact that highly non-linear procedure required special modeling approach accurately mimic procedure. Therefore, ability of Adaptive Neuro-Fuzzy Interface System (ANFIS) models in estimating infiltrated furrow sustainable proposed. main innovation current research first attempt employ ANFIS predicating rates, addition, integrate with three new optimization algorithms. Three optimizing algorithms, viz. Sine Cosine Algorithm (SCA), Particle Swarm Optimization (PSO), Firefly (FFA) were used tune ANFIS-parameters. Experimental data from six different studies countries have been this study validate proposed model. inflow rate, length, opportunity time, cross-sectional area, waterfront advance time utilized as input parameters. results indicated ANFIS-SCA could provide estimation rate compared ANFIS-PSO. Mean Absolute Error (MAE) Percent Bias (PBIAS) errors computed ANIFS-SCA (0.007 m3/m 0.12) was significantly than those achieved ANFIS-FFA ANFIS-PSO In addition that, outperformed high level accuracy. Hybrid showed outstanding performance over other optimizer algorithms be applied systems management.