ABSTRACT
Determining uniformity coefficients of sprinkle irrigation systems, in general, depends on field trials, which require time and financial resources. One alternative to reduce time and expense is the use of simulations. The objective of this study was to develop an artificial neural network (ANN) to simulate sprinkler precipitation, using the values of operating pressure, wind speed, wind direction and sprinkler nozzle diameter as the input parameters. Field trials were performed with one sprinkler operating in a grid of 16 x 16, collectors with spacing of 1.5 m and different combinations of nozzles, pressures, and wind conditions. The ANN model showed good results in the simulation of precipitation, with Spearman's correlation coefficient (rs) ranging from 0.92 to 0.97 and Willmott agreement index (d) from 0.950 to 0.991, between the observed and simulated values for ten analysed trials. The ANN model shows promise in the simulation of precipitation in sprinkle irrigation systems.
Key words:
sprinkler; water distribution uniformity; artificial intelligence; computational model
