Research Article

Adaptive Control Using Fully Online Sequential-Extreme Learning Machine and a Case Study on Engine Air-Fuel Ratio Regulation

Table 1

Comparison among OS-ELM, ReOS-ELM, and FOS-ELM.

OS-ELMReOS-ELMFOS-ELM

Objective function and and
ParametersHidden node number Hidden node number
Regularization parameter
Hidden node number
Regularization parameter
Training methodInitial offline batch training and followed by online sequential trainingInitial offline batch training and followed by online sequential trainingOnline sequential training
Initial sample number
Overfitting riskYesNoNo
Learning performanceSame as ELMSame as ReELMSame as ReELM