Limited water resources and poor performance of irrigation networks has attracted the attention of experts to improve water delivery management by application of improved control systems. Improved control systems in irrigation canals, because of complex hydraulic behavior, should be developed and tested using mathematical and hydrodynamic models. Fuzzy theory has been widely and successfully applied in several engineering control problems. In this research fuzzy theory has been applied for control of irrigation canals. For the development of fuzzy control system, the ICSS hydrodynamic model which could simulate unsteady hydraulic behavior in response of control systems in irrigation canals is used. The performance of developed mathematical model of fuzzy control system is tested in canal standard no. 2 introduced by ASCE, considering large flow variations. The performance of fuzzy control system is evaluated using several indicators such as, maximum absolute error, cumulative absolute error, and response time. The figures of flow, depth, and gate opening variations in time are analyzed as well. The results show that by applying large flow variations, the maximum depth deviation was 1.9%. The maximum response time for water depth to be stabilized within 1% range of target depth was 1.8 minutes. The results indicate that the developed fuzzy control system was successful in downstream control of irrigation canal and could be introduced for practical applications.