Improved Ant Colony Optimization for Optimal Crop and Irrigation Water Allocation by Incorporating Domain Knowledge

摘要

An improved ant colony optimization (ACO) formulation for the allocation of crops and water to different irrigation areas is developed. The formulation enables dynamic decision variable option (DDVO) adjustment and makes use of domain knowledge through visibility factors (VFs) to bias the search towards selecting crops that maximize net returns and water allocations that result in the largest net return for the selected crop, given a fixed total volume of water. The performance of this formulation is compared with that of other ACO algorithm variants (without and with domain knowledge) for two case studies, including one from the literature and one introduced in this paper for different water-availability scenarios within an irrigation district located in Loxton, South Australia near the River Murray. The results for both case studies indicate that the use of VFs (1)increases the ability to identify better solutions at all stages of the search; and (2)reduces the computational time to identify near-optimal solutions. Furthermore, the savings in computational time obtained by using VFs and DDVO adjustment should be considerable for ACO application to problems such as detailed irrigation scheduling that rely on more-complex crop models than those used in the case studies presented in the paper.