基于投影寻踪插值模型的围岩稳定性分类研究

由于受到多种因素的综合作用，因此对围岩稳定性进行准确分类是一个难题。选取影响围岩分类的5个主要因素，根据分类标准，采用在每级标准中随机内插的方法，得到50个标准样本，用于构建分类模型。针对上述复杂优化模型，采用遗传算法进行优化，以获得的最佳投影方向为基础，得到围岩稳定性分类的投影寻踪插值模型。实例分析表明：该模型的分类结果是准确和可靠的，与采用其他几种分类方法所得结论一致。此外，该模型可以得出每个样本的具体得分值，即使对属于同一级的样本也可以排出它们的优劣顺序，因而具有较高的分类精度，是一种更具有应用前景的新方法。

Classification of surrounding rock stabilities based on projection pursuit interpolation model

Because of multifactor synthetic effect，accurate classification of surrounding rock mass stability has always been a difficult problem. Five key factors are chosen to construct the model. In addition, because of few standard samples, fifty standard samples are produced by the way of stochastic interpolation on basis of classification standard. Aiming at the above complex optimization, genetic algorithm is used to get the optimal projection direction. On the basis of the direction, projection pursuit interpolation model is given. Subsequently, five unknown samples are classified based on the model. The classification results are consistent with that from other methods. Therefore, it is effective and reliable. Moreover, particular score of each sample could be computed by the model; so it has more precise classification and is a more promising method.