Optimal reservoir operation for irrigation of multiple crops using fuzzy logic

Abstract

In this study, a Fuzzy based model using a non- linear programming to obtain optimal reservoir operation for irrigation of multiple crops is proposed. The reservoir level Fuzzy logic model can extract important features of the system from the input- output data set by non- linear programming and represents features as general operating rules. The developed model can serve not only as efficient decision making tool in easy and understandable Fuzzy inference systems but also can provide operators with a limited number of the most meaningful rules using clustering- based approach. The model is set properly in a yearly base and monthly steps. Results show that the changing trend of water releases in both models is the same with R2=0.97. Over the 12 months period, both trends had risen from October to May but since then they had fallen gradually. In general the amount of annual released water in Fuzzy model is almost less than NLP, especially in competitive months, May and June. The percentage of water deficit to the percentage of annual mean water deficit was respectively 0.57 and 0.81 in training and 0.93 and 1.145 in the test stage. In addition, the water deficit compared with the amount of cultivated crops acreage has more impact on net benefit. Also, allocating less water to wheat compared with barley and sorghum had significant effect on the yield production. The findings suggest that in the year with water deficit the amount of water release in competitive months to increase the net benefit should be more considered.