Abstract
The work was carried out according to an innovative methodology that combines the approach of regression of Gaussian processes with the Broyden-Fletcher-Goldfarb optimization algorithm with limited memory in order to predict the turbidity parameter of water at the outlet of sand-filled filters used in micro-irrigation systems. The order of significance of the variables involved in predicting the “turbidity” parameter at the outlet of filters with sand filler has been established. In particular, the input variable “Turbidity” can be considered the most important parameter in making a forecast. The methodology applied in the work can be successfully applied to other filtration processes with the same or different types of filter media, but the characteristics of each filter and experiment must always be taken into account.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
