A novel greedy adaptive ant colony algorithm for shortest path of irrigation groups

Abstract

With the full-scale implementation of facility agriculture, the laying of a water distribution network (WDN) on farmland plays an important role in irrigating crops. Especially in large areas of farmland, with the parameters of moisture sensors, the staff can divide the WDN into several irrigation groups according to the soil moisture conditions in each area and irrigate them in turn, so that irrigation can be carried out quickly and efficiently while meeting the demand for irrigation. However, the efficiency of irrigation is directly related to the pipe length of each irrigation group of the WDN. Obtaining the shortest total length of irrigation groups is a path optimization problem. In this paper, a grouped irrigation path model is designed, and a new greedy adaptive ant colony algorithm (GAACO) is proposed to shorten the total length of irrigation groups. To verify the effectiveness of GAACO, we compare it with simple modified particle swarm optimization (SMPSO), chaos-directed genetic algorithms (CDGA) and self-adaptive ant colony optimization (SACO), which are currently applied to the path problem. The simulation results show that GAACO can effectively shorten the total path of the irrigation group for all cases from 30 to 100 water-demanding nodes and has the fastest convergence speed compared to SMPSO, CDGA and SACO. As a result, GAACO can be applied to the shortest pipeline path problem for irrigation of farmland groups.