论文部分内容阅读
莺落峡是黑河干流出山口径流量的重要控制站,莺落峡径流量的多少直接影响着该流域经济、社会的发展和生态环境保护,水资源分配和调度的管理和决策。论文基于人工神经网络,对莺落峡径流进行了模拟预测。将月径流分为汛期和非汛期,分别建立BP人工神经网络,通过对径流分类前后的模型进行比较,发现分类后的月径流BP模型的性能显然优于未分类的模型,故此设计了4种不同气候情景,采用分类后的模型对莺落峡2030年的径流量进行了预测。即,在降水量不变、气温增加0.5℃,2030年莺落峡年径流量将增加8.92%;气温增加1℃、降水量不变,年径流量将减少5.414%;气温不变、降水量增加10%,年径流量将增加9.905%;气温增加0.5℃、降水量增加10%,年径流量将增加8.98%。
Yingluo Gorge is an important control station for the runoff volume of the Heihe main stream. The amount of runoff in Yingluoxia directly affects the economic and social development, ecological environment protection, water resources allocation and scheduling management and decision-making in this valley. Based on the artificial neural network, this paper simulated the runoff of the Yingluoxia Gorge. The monthly runoff was divided into flood season and non-flood season, BP artificial neural network was established. By comparing the models before and after the runoff classification, it was found that the performance of the classified monthly runoff BP model was better than the uncategorized model. Therefore, According to different climatic scenarios, the runoff of 2030 in Yingluoxia Gorge was predicted by using the classified model. That is, when the precipitation is constant and the temperature is increased by 0.5 ℃, the annual runoff of the Niaoxiaxia Gorge will increase by 8.92% in 2030. If the temperature increases by 1 ℃ and the precipitation will remain unchanged, the annual runoff will decrease by 5.414%; the temperature will not change and precipitation The increase of 10% will increase the annual runoff by 9.905%. When the temperature increases by 0.5 ℃, the precipitation will increase by 10% and the annual runoff will increase by 8.98%.