论文部分内容阅读
影响滑坡稳定性的因素较多,利用滑坡稳定性影响因素快速预测滑坡稳定状态是当前滑坡研究的重要内容。利用相关系数、支持向量机、交叉验证法、遗传算法、粒子群优化算法等理论建立支持向量机模型对滑坡稳定性进行了研究。以湖北竹溪县197个滑坡为例,研究结果表明:遗传算法优化的支持向量机滑坡稳定性预测模型预测效果最好,与实际情况吻合得最好。最佳参数c为3.001 6、g为0.041 008,训练集滑坡稳定性预测的正确率为84%,测试集滑坡稳定性预测的正确率为79.32%。因此所提遗传算法优化的支持向量机滑坡稳定性预测模型对于滑坡稳定性分析具有一定参考价值。
There are many factors that affect the stability of landslide. It is an important part of current landslide research to use the factors that affect the stability of landslides to predict the stability of landslides quickly. The stability of landslide is studied by using the support vector machine (SVM) model such as correlation coefficient, support vector machine, cross validation, genetic algorithm and particle swarm optimization algorithm. Taking 197 landslides in Zhuxi County, Hubei Province as an example, the results show that the genetic algorithm optimized support vector machine landslide stability prediction model has the best prediction effect, which is in good agreement with the actual situation. The optimal parameter c is 3.001 6, g is 0.041 008, the correct rate of training set landslide stability prediction is 84%, and the correct rate of test set landslide stability prediction is 79.32%. Therefore, the genetic algorithm optimization support vector machine landslide stability prediction model for the stability of landslide has some reference value.