Soil moisture movement reveals the hydrological environment and irrigation characteristics of plants, which is very important and basic hydrological problem. A soil moisture movement algorithm based on cellular automata suitable for greenhouse was proposed to simulate the lateral flow and vertical penetration of soil water in different soil layers. Artemisia annua was used as an experimental plant, and two numerical simulations were set up to determine the minimum water replenishment and the degree of upward soil modification. To verify the performance of the model, corn was used as an experimental plant for application simulation. The numerical simulation results show that the effect of only water for plant cells is superior to soil cells, and the disturbance behavior of the underlying soil layer should be minimized during the cultivation process. The average MAE/RMSE of all soil layers is 0.86/1.06 and in deep layer (50 cm and 60 cm) is 0.53/0.64. It shows the model has a certain prediction and simulation ability, especially in deep soil layers. The proposed algorithm can simply calculate soil flow, set the minimum water replenishment, and evaluate the water replenishment efficiency, which can provide a theoretical reference for the water replenishment and soil replacement scheme.

土壤水分运动揭示了植物的水文环境和灌溉特性，这是非常重要和基本的水文问题。提出了一种适用于温室的基于元胞自动机的土壤水分运动算法，模拟土壤水分在不同土层中的侧向流动和垂直渗透。以青蒿为试验植物，建立两次数值模拟，确定最小补水量和土壤改良度上升。为了验证模型的性能，使用玉米作为实验植物进行应用模拟。数值模拟结果表明，仅水对植物细胞的影响优于土壤细胞，在栽培过程中应尽量减少下层土层的干扰行为。所有土层的平均 MAE/RMSE 为 0.86/1.06，深层（50 厘米和 60 厘米）为 0.53/0.64。表明该模型具有一定的预测和模拟能力，尤其是在深层土层中。所提算法可以简单地计算土壤流量，设定最小补水量，并评价补水效率，可为补水和土壤置换方案提供理论参考。