CAJ | 학술논문

鉴于各种传统的大坝变形监测方法必然造成信息量的丢失，有必要寻找一种可以最大化利用各种拟合方法所包含的信息，并最终达到提高高程拟合精度的方法。提出了基于BP神经网络方法与逐步回归方法的组合模型，研究了将该组合模型用于大坝变形监测的精度和可行性，通过具体工程实例验证，详细介绍了该组合模型在大坝变形监测中的应用。
감우각충전통적대패변형감측방법필연조성신식량적주실，유필요심조일충가이최대화이용각충의합방법소포함적신식，병최종체도제고고정의합정도적방법。제출료기우BP신경망락방법여축보회귀방법적조합모형，연구료장해조합모형용우대패변형감측적정도화가행성，통과구체공정실례험증，상세개소료해조합모형재대패변형감측중적응용。
ABSTRACT:In the light of that various traditional methods of monitoring dam deformation will inevitably cause the loss of information, it is necessary to find a method that can maximize the use of the information contained in all kinds of fitting method and eventually to improve elevation fitting precision. This paper proposes a combination model based on BP neural network method and stepwise regression method, studies the feasibility and precision of this combination model applied in dam’s deformation monitoring, and through an engineering example’s verification, introduces in detail the application of this combination model in dam’s deformation monitoring.