基于IL-HMMs预测模型的地下水埋深预测研究

发布时间：2018-05-03 22:33

本文选题：地下水埋深 + 磴口县　；参考：《农业机械学报》2017年12期

【摘要】：以西北干旱典型县域磴口县为研究区,基于增量学习的改进隐马尔可夫预测模型(IL-HMMs),对区域地下水埋深进行了预测研究。为检验IL-HMMs模型预测效果,将模型预测结果与2013年长观井的实测数据进行了比较;同时为检验模型的优劣性,与未经增量学习的隐马尔可夫模型(HMMs)、加权马尔可夫链(WMCP)和BP神经网络(BP neural network,BPNN)预测模型的预测结果进行了比较。结果表明:与其他几种预测模型相比,IL-HMMs模型预测精度显著提高,误差更小,有较好的鲁棒性。并使用IL-HMMs模型对2018年地下水埋深进行了预测,预测结果表明,2018年地下水年平均埋深略有增加、局部区域地下水埋深增量加剧。基于IL-HMMs模型的地下水埋深预测具有很好稳定性的同时对新数据加入又有很好的鲁棒性,可为地下水埋深动态预测提供思路与方法补充,为区域地下水资源开发利用和保护提供重要依据。
[Abstract]:Taking Dengkou County, a typical arid county in Northwest China, as the research area, an improved hidden Markov prediction model based on incremental learning was used to predict the groundwater depth in the region. In order to test the prediction effect of IL-HMMs model, the prediction results of the model are compared with the measured data in 2013, and the advantages and disadvantages of the model are also tested. The prediction results are compared with those of the hidden Markov model (HMMsN), the weighted Markov chain (neural) and the BP neural network (BP neural network). The results show that the prediction accuracy of IL-HMMs model is significantly higher than that of other prediction models, the error is smaller, and the prediction model has better robustness. The IL-HMMs model is used to predict the groundwater depth in 2018. The results show that the average groundwater depth increases slightly in 2018 and the increment of groundwater depth in local area is increased. The prediction of groundwater depth based on IL-HMMs model has good stability and robustness to the new data, which can be used to provide ideas and methods for dynamic prediction of groundwater depth. It provides an important basis for the development, utilization and protection of regional groundwater resources.
【作者单位】：北京林业大学精准林业北京市重点实验室;佛罗里达大学地理系;
【基金】：国家自然科学基金项目(41371189) “十二五”国家科技支撑计划项目(2012BAD16B00)
【分类号】：P641

【相似文献】