摘要
为准确预测泥石流危险度,提出了基于主成分分析法(PCA)和遗传算法(GA)优化的支持向量机(SVM)模型。首先利用主成分分析法对7个泥石流危险度影响因子进行数据降维,将提取出的主成分作为支持向量机模型的输入向量,以泥石流危险度作为输出向量,并运用遗传算法寻优获得最佳支持向量机模型参数,最终建立了基于PCA-GA-SVM的泥石流危险度预测模型,并对9条泥石流沟的危险度进行预测,结果表明:PCA-GA-SVM模型的预测准确率达88.9%,满足工程要求。
In order to predict debris flow risk accurately, support vector machine （SVM） model optimized by genetic algorithm （GA） based on principle component analysis （PCA） was proposed. First, using PCA to make data dimension reduction for 7 influencing factors of debris flow risk, then the extracted principle components were used as model input vectors, and debris risk degree as model output vectors, and the best SVM parameters were optimized by GA, finally the prediction model for debris flow risk based on PCA-GA-SVM was established, and it was used to predict the risk of 9 debris flow gullies. The result shows that the prediction accuracy of PCA-GA-SVM model is 88.9%, which meets general requirements of engineering.
出处
《河北地质大学学报》
2017年第2期20-24,共5页
Journal of Hebei Geo University
基金
河北地质大学第十三届学生科技基金重点科研项目(KAG201607)
关键词
泥石流危险度
主成分分析法
遗传算法
支持向量机
debris flow risk
principal component analysis
genetic algorithm
support vector machine
