An improved nonlinear smooth twin support vector regression based‐behavioral model for joint compensation of frequency‐dependent transmitter nonlinearities

2021 ◽  
Author(s):  
Tianfu Cai ◽  
Mingyu Li ◽  
Yao Yao ◽  
Changzhi Xu ◽  
Yi Jin ◽  
...  
Keyword(s):  
Support Vector Regression ◽  
Behavioral Model ◽  
Support Vector ◽  
Twin Support Vector Regression ◽  
Frequency Dependent ◽  
Nonlinear Smooth

An adaptive twin support vector regression machine based on rough and fuzzy set theories

2018 ◽  
Vol 32 (9) ◽  
pp. 4709-4732
Author(s):  
Zhenxia Xue ◽  
Roxin Zhang ◽  
Chuandong Qin ◽  
Xiaoqing Zeng
Keyword(s):  
Support Vector Regression ◽  
Fuzzy Set ◽  
Support Vector ◽  
Twin Support Vector Regression

The Research of Fuzzy Weighted Twin Support Vector Regression in the MBR Simulation Prediction

2014 ◽  
Vol 556-562 ◽  
pp. 3648-3653 ◽  
Author(s):  
Chan Juan Ji ◽  
Chun Qing Li ◽  
Tao Wang
Keyword(s):  
Support Vector Regression ◽  
Membrane Fouling ◽  
Principal Component ◽  
Support Vector ◽  
Support Vector Data Description ◽  
Vector Data ◽  
Negative Effects ◽  
Twin Support Vector Regression ◽  
Output Layer ◽  
Input Layer

This paper using the way of Support Vector Data Description (SVDD) and considering the tightness between the Membrane Bio-Reactor (MBR) samples, applies the Fuzzy Weighted Twin Support Vector Regression (FTSVR) to the MBR simulation prediction research. Firstly,adopt the principal component analysis (PCA) on membrane fouling factors to achieve dimension reduction and de-correlation, then put the PCA output layer as the input layer of FTSVR, flux as the output layer, eventually, the MBR Membrane Fouling Prediction Model is built. This method considers the different effects on the regression hyperplane of different MBR samples,and effectively eliminates the negative effects due to error even outliers in the process of MBR data measurement.

Sign in / Sign up

Export Citation Format

Share Document