Comparison of Neuro-Fuzzy and Regression Models for Prediction of Outflow of on-farm Reservoir

摘要

Neuro-ftizzy and regression models predicting of outflow of an on farm reservoir of 300 m~3 capacity, located at Indian Agricultural Research Institute (IARI), New Delhi site of India, were developed and validated using crop water requirement, evaporation losses and farm pond inflow data. The outflow predicted by these two models was compared with each other. It is raveled that Neuro-Fuzzy model predicted the outflow with high coefficient of determination (R~2) of 0.98, model efficiency (E) of 0.97 and absolute average deviation (AAD) of 0.002. The regression model was found to be inferior with R~2 = 0.940.78, E = 0.72 and AAD = 0.031. MATLAB software was used for development of Neuro-Fuzzy and regression model.