基于双树复小波-LSSVM的大坝沉降预测

Dam Settlement Prediction Based on Dual-tree Complex Wavelet and LSSVM

大坝沉降往往受多种因素影响,对此提出一种基于双树复小波-最小二乘支持向量机(LSSVM)的大坝沉降预测方法。利用双树复小波有效分离出大坝原始序列中隐含的不同频率分量,分析影响大坝沉降的水位、温度与各频率分量的关系,建立相应的LSSVM预测模型。实验表明,该方法有较高的预测精度。

Dam settlement is often affected by many factors,and a dam settlement prediction method based on dual-tree complex wavelet and LSSVM was proposed.The dual-tree complex wavelet was applied to effectively separate the different frequency components in the original sequence of the dam.The relationship between the water level,temperature and the frequency components affecting the dam settlement was analyzed,and the corresponding LSSVM prediction model as established.The experimental results showed that the method has high prediction accuracy.