A genetic algorithm for optimizing off-farm irrigation scheduling

Abstract

This paper examines the use of genetic algorithm (GA) optimization to identify water delivery schedules for an open-channel irrigation system. Significant objectives and important constraints are identified for this system, and suitable representations of these within the GA framework are developed. Objectives include maximizing the number of orders that are scheduled to be delivered at the requested time and minimizing variations in the channel flow rate. If, however, an order is to be shifted, the irrigator preference for this to be by ±24 h rather than ±12 h is accounted for. Constraints include avoiding exceedance of channel capacity. The GA approach is demonstrated for an idealized system of five irrigators on a channel spur. In this case study, the GA technique efficiently identified the optimal schedule that was independently verified using full enumeration of the entire search space of possible order schedules. Results have shown great promise in the ability of GA techniques to identify good irrigation order schedules.