Large-scale canals are worldwidely constructed to divert water from rich to arid areas to mitigate water shortages. Efficient control of canals is essential to improve water-diversion performance. Numerous model-based approaches have been proposed and made great progress for canal control. However, when the predictive model is unpromising for long time-step predictions, model-free approaches could be considered as a possible way. Since most existing model-free approaches are focused on control of small canals or reservoirs, this paper proposes a new control approach named PARL for large-scale canals. We leverage the idea of divide-and-conquer to decompose the control task of large-scale canals into policy-learning and action-learning sub-tasks, and develop PARL by means of hierarchical reinforcement learning. Extensive experiments are conducted via numerical simulation on the case study of Chinese South to North Water Transfer Project, and experimental results show that PARL can achieve desirable performance improvements over other model-free learning approaches.