Abstract: Drip irrigation has been one of the most effective water-conserving irrigation technologies in the arid and semiarid areas. Among them, the rotational irrigation mode has been the mainstream application in Xinjiang, Western China. The Rotation Irrigation Groups (RIGs) have been widely designed to deal with the water shortage in the farmlands irrigation. However, the RIGs arrangement can often be taken several days to calculate in large-scale projects, according to the experience or EXCEL software. Furthermore, these traditional approaches cannot acquire the reasonable solution. Therefore, a single-objective model was previously proposed to verify the feasibility of the intelligent algorithm on the manual drip irrigation. In this study, a mathematical model was proposed to determine the minimum flow standard deviation using a gravitational search algorithm (GSRV) and Genetic Algorithm (GA) with the hybrid variable neighborhood. A systematic analysis was made to clarify the differences between the automated and manual drip irrigation, particularly for the high calculation efficiency under the intelligent algorithm. The chromosome coding and population initialization were also established to acquire the problem characteristics of automated drip irrigation, according to the dividing and dispersion index. An offspring repair algorithm was also designed to reduce the constraint and illegal solutions from the crossover and variation using the penalty function method. The selection of operator was also combined with a simulated annealing mechanism to avoid the optimization process falling into the local optimum. Since the classical GA presented the global solid search but the weak local search, two neighborhood structures were constructed using the flow and location adjustment, while a search algorithm was designed using Variable Neighborhoods Search (VNS), further to enhance the GA searchability and accuracy. As such, the different neighborhood operators were selected for the VNS to jump out of the local minima. The shaking procedure and the Local Search (LS) procedure were iteratively alternated until a predefined stop condition, thereby to escape from local optimal solutions. The specific search procedure was as follows. 1) The parameters were first initialized to calculate the population fitness, then to select the coding bit with the probability for the dynamic performance of variable neighborhood search. The next step was check the constraints to calculate the fitness value and update the population for the subsequent search. Finally, the population encoding was output until the population traversal was completed or the algorithm iterations reached. The experimental results show that the improved model was effective and universality in this case. Specifically, the minimum standard deviation of the solution was 10.03 m3/h. The solution was also converged in about 350 generations, where the branch pipes were dispersed in the different sub-main pipes. The GSRV-GA demonstrated an excellent convergence and search accuracy, which verified the effectiveness of the proposed algorithm. Three groups of cases with different sizes were selected to carry out the verification of model universality, also indicating an excellent performance. The minimum standard deviations were 15.18, 13.93, and 7.52 m3/h, respectively. The average standard deviations of shaking experiment simulating the blockage of 1-5 branch pipes were 10.50, 11.09, 11.47, 12.15, and 12.81 m3/h, respectively. The jitter simulation experiment showed the influence of branch blockage on the flow equalization of the irrigation group, indicating the robustness of the model. At the same time, the approximate standard deviations of the flow rate were no more than 1%, and the dispersion values were 1 and 0.73 for the improved model and the previous one in the actual case, respectively. It infers that the different grouping requirements of rotation irrigation group were realized in the design. Consequently, the feasibility was achieved for the RIG division of automatic drip irrigation, thereby to fully meet the hydraulic calculation and engineering requirements in modern agriculture. The finding can provide the research foundation for the subsequent mining of water-saving technology in automatic drip irrigation.