论文部分内容阅读
矿柱是地下矿山支撑顶板围岩、维持采场稳定的关键结构要素。为迅速准确地判别矿柱稳定性,选取矿柱宽度、矿柱高度、矿柱宽高比、矿岩单轴抗压强度和矿柱承受载荷作为影响指标,利用高斯过程机器学习算法建立矿柱状态与其主要影响因素之间的非线性映射关系,进而提出一种基于高斯过程二元分类(GPC)的矿柱状态识别模型。结合工程实例,以40组样本数据进行训练,以10组样本数据对该模型进行检验,并与ANN和SVM进行对比。结果表明,矿柱状态识别的高斯过程模型是科学可行的,该模型具有参数自适应化获取、分类精度高、计算复杂度低等优点,还可对矿柱状态判别结果的不确定性或可信度进行定量化评价。
Pillar is the mine underground mine roof support, to maintain the stability of the key structural elements. In order to quickly and accurately determine the stability of the pillars, the pillar width, pillar height, pillar aspect ratio, uniaxial compressive strength of rock and pillar bearing load were taken as the influencing indexes. Gauss process machine learning algorithm was used to establish pillar State and its main influencing factors, and then proposed a pillars state identification model based on Gaussian Processes Binary Classification (GPC). Combined with engineering examples, 40 sets of sample data were used for training. The model was tested with 10 sets of sample data and compared with ANN and SVM. The results show that the Gaussian process model of ore pillar status identification is scientifically feasible. The model has the advantages of adaptive parameter acquisition, high classification accuracy and low computational complexity. Reliability of quantitative evaluation.