基于马尔科夫链的曲线拟合法在尾矿坝沉降预测中的应用  

Application of Curve Fitting Method Based on Markov Chain in Tailings Dam Settlement Prediction

作　　者：汪宏宇 龚循强 鲁铁定[1] 陈志平[1] WANG Hong-yu;GONG Xun-qiang;LU Tie-ding;CHEN Zhi-ping(School of Geomatics,East China University of Technology,Nanchang 330013,China;Key Laboratory of Radioactive Geology and Exploration Technology,Fundamental Science for National Defense,East China University of Technology,Nanchang 330013,China)

机构地区：[1]东华理工大学测绘工程学院,江西南昌330013 [2]东华理工大学放射性地质与勘探技术国防重点学科实验室,江西南昌330013

出　　处：《东华理工大学学报（自然科学版）》2021年第4期364-369,共6页Journal of East China University of Technology(Natural Science)

基　　金：国家自然科学基金项目(41904031,42101457);江西省自然科学基金项目(20202BABL213033);东华理工大学放射性地质与勘探技术国防重点学科实验室开放基金项目(RGET1905)。

摘　　要：为了对尾矿坝进行高效准确的沉降预测,笔者提出一种基于马尔科夫链的曲线拟合法。将曲线拟合法中的指数曲线法和Logistic曲线法分别与马尔科夫链构建组合预测模型,在保留曲线拟合法建模简单、易于操作的优势的基础上,提高模型的预测精度。结合尾矿坝沉降监测数据,并与传统的曲线拟合法、后向传播神经网络、基于遗传算法的后向传播神经网络进行比较。实验结果表明,提出的方法预测误差小,预测精度明显优于传统预测方法。In order to predict the settlement of tailings dams efficiently and accurately,a curve fitting method based on Markov chain was proposed.On the basis of retaining the advantages of simple and easy operation of curve fitting method,the exponential curve and Logistic curve of curve fitting method and Markov chain were used separately to build combined prediction model for improving prediction accuracy.Experimental evaluation with settlement data of the tailings dam has been carried out.The results were compared with the traditional curve fitting method,back propagation(BP)neural network,and genetic algorithm-based back propagation(GA-BP)neural network.The experimental results show that the smaller prediction error can be obtained using the proposed method and the prediction accuracy is significantly better than the traditional prediction methods.

关 键 词：马尔科夫链 尾矿坝 沉降预测 指数曲线法 Logistic曲线法 

分 类 号：P258[天文地球—测绘科学与技术]