| 摘要: |
| 综合考虑农业系统的不确定性以及灌区地表水来水量的动态性，结合区间多阶段随机规划模型、双侧随机机会约束规划模型和联合概率规划模型，建立区间多阶段双侧随机联合概率机会约束规划模型，将该模型应用于民勤县红崖山灌区，结果表明:1)模型可以解决区间不确定性和约束两侧参数随机不确定性，反映系统的动态变化过程；2)预先设置反映约束条件违规风险的独立违规概率和反映独立违规概率之和的联合违规概率，根据3种联合违规概率水平和3种独立违规概率变化趋势，模型得到在9种情景下的灌溉方案、灌溉面积和灌溉目标；3)随着联合违规概率的增大，系统的灌溉效益逐渐增大，联合违规概率为0.1时的灌溉效益比联合违规概率为0.01时的灌溉效益上下限值分别增大5.31亿元和6.55亿元。同时在不同的情景下，灌区的瓜果类经济作物种植面积最大，粮食作物种植面积次之，蔬菜的种植面积最小。本研究能够为决策者提供不同的灌溉方案，并为当地水资源管理部门制定主要作物灌溉面积分配方案提供参考。 |
| 关键词: 红崖山灌区 区间多阶段随机规划 机会约束规划 双侧随机不确定性 动态性 |
| DOI：10.11841/j.issn.1007-4333.2022.11.20 |
| 投稿时间：2022-03-02 |
| 基金项目:国家自然科学基金项目(41871199) |
|
| Interval multi-stage double-sided stochastic joint-probabilistic chance-constrained programming: A case study of the optimal allocation of water resources in Hongyashan irrigation area in Minqin County |
|
BIAN Lekang,ZHANG Chenglong,GUO Ping,PAN Qi
|
| (College of Water Resources & Civil Engineering, China Agricultural University, Beijing 100083, China) |
| Abstract: |
| Considering the uncertainty of agricultural system and the dynamics of surface water inflow in irrigation area, this study formulates an interval multi-stage double-sided stochastic joint-probabilistic chance-constrained programming model by combining the techniques of interval multi-stage stochastic programming, double-sided stochastic chance-constrained programming and joint-probabilistic programming. This model is applied to Hongyashan irrigation area in Minqin County. The results showed that: 1)The model not only solved the interval uncertainty of the model and random uncertainties in the double-sided constraints, but also reflected the dynamic variations of the system. 2)This study set the individual violation probability that reflected the violation risk of constraint conditions and the joint violation probability that reflects the sum of individual violation probability. Through the given three levels of joint constraint violations and three kinds of individual probability scenarios before solving the developed model, optimal solutions of irrigation schemes, irrigation area and irrigation targets under nine scenarios were obtained. 3)With the increased constraint-violation probability levels, the irrigation benefit of the system gradually increases. When the joint violation probability is raised from 0. 01 to 0. 1, the upper bound value increased by 531 million Yuan and the lower bound value increased by 655 million Yuan. Under different scenarios, the planting area of economic crops such as fruits in the irrigation area was the largest, followed by grain crops, and the planting area of vegetables was the smallest. In conclusion, this research can provide different irrigation schemes for decision makers and a reference for water resources management department to formulate the allocation scheme of main crop irrigation area. |
| Key words: Hongyashan irrigation area interval multi-stage stochastic programming chance-constrained programming double-sided randomness dynamic |
