摘要
流域水系统平衡、自然生态保护、经济社会发展等均可对径流量变化产生影响,为更加合理优化配置水资源有必要开展径流量模拟分析。文章依据BP网络法和英那河长序列逐日径流资料,对小尺度流域的月径流量状况进行拟合分析,在此基础上验证了该模型的可行行。结果显示:对于降水径流模拟BP神经网络具有明显的优势和良好的应用前景,其Nash效率系数为0.92,大于新安江和径流系数法计算结果,而与HSPF(0.95)相当;BP神经网络能够较为准确的预测变化趋势且模型运行简便,但模拟结果精确度仍存在一定的提升空间。
Watershed water system balance of natural ecological protection and economic and social development can affect the change of runoff,in order to optimize the allocation of water resources,it is necessary to carry out runoff simulation analysis.In this paper,based on the BP network method and the daily runoff data of the Yingna river length sequence,the monthly runoff of the small-scale watershed is simulated and analyzed,and the feasibility of the model is verified.The results show that the BP neural network has obvious advantages and good application prospect,the Nash efficiency coefficient is 0.92,which is higher than the calculation result of Xin'an river and runoff coefficient method,and is equivalent to HSPF(0.95);BP neural network can predict the changing trend accurately and the model is simple to run,but there is still some room for improvement in the accuracy of simulation results.
作者
高清震
Gao Qing-zhen(Zhuanghe Urban Water Affairs Service Center, Zhuanghe 116400, China)
出处
《黑龙江水利科技》
2020年第3期16-20,共5页
Heilongjiang Hydraulic Science and Technology
关键词
BP网络
径流模拟
确定性系数
英那河
BP network
runoff simulation
certainty coefficient
Yingna river