基于支持向量机的矿区开采沉降的预测

Subsidence prediction of diggings based on support vector machine

针对矿区开采沉降预测方法问题,在分析了矿区开采沉降因素的基础上,利用统计学习的新方法——支持向量机,结合最小二乘算法,提出了矿区沉降的预测模型,预测结果与神经元网络,多项式拟合结果进行比较,结果表明支持向量机在沉降预测方面准确性高,泛化能力强。

Directed at the subsidence prediction of diggings, this paper, based on the analysis of the factors of subsidence, presents an attempt to develop the forecast model of the subsidence, depending on the new statistics-support vector machine, combined with least square. The comparison of the prediction results obtained in this way with those done with the neural networks and polynomial fit shows that the support vector machine performs with higher veracity and better extensiveness in case of subsidence prediction.