DOI：10.12343/Agrixiv.202607.00006  CSTR：31252.36.agrixiv.202607.00006

Research on Intelligent Irrigation Decision-Making System for Cotton in Arid Areas Based on Soil Moisture Prediction

• 摘要：新疆作为我国核心棉花产区，区域气候干旱、水土资源供需矛盾突出，现有棉田膜下滴灌模式多依靠农户经验开展灌溉作业，存在水肥利用效率偏低、人工管理成本高、灌水定额动态调控不足等现实难题。本文构建一套融合多层土壤墒情动态监测与作物短期需水量预测的棉花智慧灌溉闭环决策系统。系统依托时域反射法（TDR）土壤水分传感器与田间一体化微型气象站采集多源环境数据，采用径向基函数（RBF）神经网络构建棉花 3～5 日需水量预测模型，结合棉花不同生育期临界墒情阈值自动输出标准化灌溉调控方案。以新疆阿克苏规模化棉田为试验区域开展完整生长季定点对比试验，结果表明：相较于传统经验灌溉模式，本系统可实现田间灌溉用水量削减 14%～26%；籽棉产量提升 22.6%，水分利用效率提升 76.3%；灌溉环节人工投入降低 72%，水肥综合生产成本下降 12%～27%，每亩可新增综合收益约 196 元，且棉花纤维长度、断裂比强度、马克隆值等核心品质指标无显著劣变。该系统兼顾节水增产、降本省工多重效益，为西北干旱区棉花规模化精准灌溉提供可落地的智能化技术方案。

  关键词： 干旱区; 棉花; 土壤墒情; 智慧灌溉; 灌溉决策系统

   

  Abstract: Xinjiang is China’s core cotton-producing region, characterized by arid climate and prominent contradiction between supply and demand of soil and water resources. The existing film-mulched drip irrigation mode in cotton fields mostly relies on farmers’ experience for irrigation, which leads to low water and fertilizer utilization efficiency, high manual management costs and insufficient dynamic regulation of irrigation quota. This paper constructs a closed-loop intelligent cotton irrigation decision-making system integrating dynamic monitoring of multi-layer soil moisture and short-term crop water demand prediction. The system collects multi-source environmental data through Time Domain Reflectometry (TDR) soil moisture sensors and integrated field micro weather stations. A Radial Basis Function (RBF) neural network is adopted to establish a 3–5 day cotton water demand prediction model, and standardized irrigation regulation schemes are automatically output combined with critical soil moisture thresholds of cotton at different growth stages. Fixed-point comparative experiments were carried out in large-scale cotton fields in Aksu, Xinjiang throughout the whole growing season. The results show that compared with the traditional empirical irrigation mode, the system can reduce field irrigation water consumption by 14%–26%, increase seed cotton yield by 22.6%, and improve water use efficiency by 76.3%. Manual input in irrigation is reduced by 72%, the comprehensive production cost of water and fertilizer drops by 12%–27%, and the additional comprehensive income per mu reaches about 196 yuan. Moreover, there is no significant deterioration in core quality indicators of cotton such as fiber length, breaking tenacity and micronaire value. Taking water saving, yield increase, cost reduction and labor saving into account, the system provides a practical intelligent technical scheme for large-scale precision irrigation of cotton in arid areas of Northwest China.

  Key words: arid area; cotton; soil moisture; intelligent irrigation; irrigation decision-making system

  提交时间：2026-07-13

  版权声明：作者本人独立拥有该论文的版权，预印本系统仅拥有论文的永久保存权利。任何人未经允许不得重复使用。
• html

• 图表

• 朱俊卓, 陈铭, 张若雨菲, 穆再排尔・吐尔孙, 玉米提江・阿布杜热依木. 干旱区智慧农业智能水肥一体化系统设计与应用研究. 2026. doi: 10.12343/Agrixiv.202607.00002

  Abdul Rahim Junejo, Shakeel Ahmed Soomro, Aftab Ahmed Memon, Wenquan Niu, Yasmin Junejo, Jahangeer Dahri, Jamshed Ali Channa, Zahid Ali Channa, N. Indianraj, Hyder Bux Khoso, Khadija Urooj, Aamir Hussain Bhutt. Effects of sub-surface wick irrigation system on crop yield, irrigation intervals, base period and crop water productivity of sponge gourd and bitter gourd.. 2023. doi: 10.31220/agriRxiv.2023.00178

  Abdul Rahim Junejo, Shakeel Ahmed Soomro, Komal Jamil Gujjar, N. IndianRaj, Jamshed Ali Channa, Jahangeer Dahri, Yasmin Junejo, Saba Qureshi, Muhammad Ibrahim Junejo, Khadija Uroo. Water saving and crop yield under sub-surface wick irrigation system for sponge gourd ( Luffa aegyptiaca ) and bitter gourd ( Momordica charantia ) crops.. 2023. doi: 10.31220/agriRxiv.2023.00177

  Azam Lashkari, Jun-guo Li. Impact of climatic factors and poverty on wheat yields via machine learning in a semi-arid region across West Asia.. 2024. doi: 10.31220/agriRxiv.2024.00239

  Bibiana Betancur‐Corredor, Birgit Lang, David J. Russel. Reducing tillage intensity benefits the soil micro- and mesofauna in a global meta-analysis.. 2022. doi: 10.31220/agriRxiv.2022.00146

• 序号	提交日期	编号	操作
  1	2026-07-10	10.12343/Agrixiv.202607.00006V1	下载

• 公开评论  匿名评论  仅发给作者 提交评论