Data driven soft sensor development for complex chemical processes using extreme learning machine

A novel double parallel extreme learning machine with Pearson correlation coefficient based independent subnets (PCCIS-DPELM) is proposed for accurately modeling complex chemical processes.