论文部分内容阅读
利用核主成分分析法对滑坡位移影响因子进行特征提取,以获得的主成分作为支持向量机的特征向量建立支持向量机模型,其中模型参数通过粒子群算法进行选择优化,构建出核主成分分析和粒子群优化支持向量机协同模型,对滑坡相对位移进行预测。预测结果的平均绝对误差和相对误差分别为0.760和7.563%,与其他预测模型相比,其拟合和泛化能力最优,表明核主成分分析和粒子群优化支持向量机协同模型的预测结果与实际监测值具有很好的一致性。
Principal component analysis (PCA) is used to extract the feature of landslide displacement influence factor, and the principal component is used as the feature vector of support vector machine to build support vector machine model. The model parameters are selected and optimized by particle swarm optimization to construct the principal component analysis And PSO SVM collaborative model to predict the relative displacement of landslide. The average absolute error and relative error of the prediction results are 0.760 and 7.563%, respectively. Compared with other prediction models, the fitting and generalization abilities are the best, which shows that the principal component analysis and PSO SVM collaborative model predicts It has good consistency with the actual monitoring value.