| 姜宝良,李林晓,李腾超.基于BP神经网络的新乡百泉逐月泉水流量动态分析[J].矿产勘查,2018,9(3):516-521 |
| 基于BP神经网络的新乡百泉逐月泉水流量动态分析 |
| Dynamic analysis of monthly spring flow of Baiquan at Xinxiang based on BP neural network |
| 投稿时间：2017-09-26 |
| DOI： |
| 中文关键词: BP神经网络 泉水流量 逐步回归模型 百泉 新乡市 |
| 英文关键词: BP neural network,dynamic analysis,spring flow,stepwise regression model,Baiquan,Xinxiang city |
| 基金项目:国家自然科学基金项目(编号:41372260)资助。 |
|
| 摘要点击次数: 1962 |
| 全文下载次数: 980 |
| 中文摘要: |
| 根据1964.01—1978.12新乡百泉泉水流量动态分析,泉水流量动态受大气降水控制,反映了天然状态下的泉水流量动态特征,并建立了百泉逐月泉水流量的逐步回归预测模型。用BP神经网络方法,建立了不同时段的逐月泉水流量动态预测模型,并与逐步回归模型进行比较,BP神经网络模型拟合效果优于逐步回归模型。根据逐步回归模型和BP神经网络模型预测的结果,极枯降水年份的泉水流量2.727m3·s-1作为百泉泉域裂隙岩溶地下水开采资源量,可在平水年和丰水年的丰水期有泉水自然涌出,重现自然美景,保护风景泉水旅游资源,做到先观后用,达到保泉和供水的目的。 |
| 英文摘要: |
| According to dynamic analysis spring flow of Baiquan at Xinxiang from 1964.01 to 1978.12, dynamic of spring flow are controlled by atmospheric precipitation, reflecting the dynamic of spring flow characteristics of natural condition, and the stepwise regression prediction model of monthly spring flow was established .The dynamic prediction model of monthly spring flow in different period is established applying BP neural network method, and compared with stepwise regression model.The fitting effect of BP neural network model is better than the stepwise regression model.According to the predicted results of the stepwise regression model and the results of BP neural network model, regarding 2.727 m3·s-1 of spring flow in a very low precipitation years as the resource of groundwater exploitation of fissure karst at Baiquan spring territory, the spring can naturally flow in the wet year and the normal flow year during the wet season and recreate natural beauty and protect spring scenery tourism resources.The purpose of the water supply and protecting water can be reached after watching it. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|