Evaluation of artificial intelligence models in calculating daily and monthly reference evapotranspiration (case study: Khorramabad station)

The FAO-penman-monteith method requires a lot of input data, which is difficult in many cases to access, it is necessary to replace the simpler models with low initial inputs and appropriate accuracy. Based on the effect of input parameters on reference evapotranspiration, six input patterns for modelling were determined. This result indicates the appropriate selection of input parameters and their effectiveness in modelling and increasing the number of effective variables in the input causes the expansion of the model memory to estimate the output values, which increases the number of data for network training and the network is well generalised.