不确定遗传神经网络在滑坡危险性预测中的研究与应用

发布时间：2018-05-09 15:23

本文选题：不确定数据 + 滑坡　；参考：《计算机工程》2017年02期

【摘要】：针对滑坡危险性预测中降雨等不确定因素难以获取,以及有效处理和标准反向传播算法存在局部极小值和训练速度慢等问题,为提高滑坡危险性的预测精度,提出一种不确定遗传神经网络滑坡预测方法。基于改进遗传算法和反向传播神经网络分类算法,结合滑坡灾害预测相关理论,考虑到与滑坡灾害密切相关的降雨等不确定因素,给出不确定数据分离度的概念,阐述不确定属性数据的处理方法,构建不确定遗传神经网络,建立滑坡灾害预测模型,以延安宝塔区为例进行验证。实验结果显示,该方法的有效精度和总体精度分别为92.1%和86.7%,验证了不确定遗传神经网络算法在滑坡灾害预测中的可行性。
[Abstract]:In order to improve the prediction accuracy of landslide risk, it is difficult to obtain uncertain factors such as rainfall in landslide risk prediction, and there are some problems in effective treatment and standard back-propagation algorithm, such as local minimum value and slow training speed. An uncertain genetic neural network landslide prediction method is proposed. Based on improved genetic algorithm and back-propagation neural network classification algorithm, combined with the related theory of landslide disaster prediction, considering the uncertain factors such as rainfall closely related to landslide disaster, the concept of uncertain data separation degree is given. The processing method of uncertain attribute data is expounded, the uncertain genetic neural network is constructed, and the landslide disaster prediction model is established, which is verified by taking Baota area, Yan'an as an example. The experimental results show that the effective accuracy and overall accuracy of the method are 92.1% and 86.7% respectively. The feasibility of the uncertain genetic neural network algorithm in landslide disaster prediction is verified.
【作者单位】：江西理工大学信息工程学院;江西理工大学资源与环境工程学院;江西理工大学应用科学学院;
【基金】：国家自然科学基金“基于不确定数据挖掘的滑坡区域地质灾害危险性评价方法”(41362015)
【分类号】：P642.22;TP18

【参考文献】