基于PSR-WSVM模型的边坡位移预测

为建立高精度的边坡位移预测模型，采用相空间重构（PSR）将边坡位移时间序列数据转换为多维数据，同时构造小波核函数改进的支持向量机模型，建立PSR-WSVM模型并应用于边坡位移预测。将PSR-WSVM模型预测结果与传统支持向量机（SVM）模型、小波支持向量机（WSVM）模型和基于相空间重构的支持向量机（PSR-SVM）模型预测结果进行对比，通过平均绝对误差（MAE）、平均绝对误差百分比（MAPE）和均方根误差（RMSE）3个精度评价指标验证PSR-WSVM模型的可行性。工程实例结果表明，PSR-WSVM模型预测结果的3个精度评价指标都优于另外3种模型，边坡位移预测的精度明显提升。     

Variability of Soft X—ray Spectral Shape in Blazars Observed by ROSAT

1 INTRODUCTION Blazars, including BL Lac objects, highly polarized and optically violently variable quasars,and flat-spectrum radio quasars (FSRQs), are characterized by highly variable non-thermalemission which dominates their characteristics from radio to y-ray bands. The mechanismbelieved to be responsible for their broadband emission is synchrotron radiation followed by in-verse Compton (IC) scattering at higher energies (e.g. Blandford & Konigl 1979). Relativisticbeaming of a jet…