Using the kriging method based on cross-validation stepwise variable selection to optimize anaerobic wastewater treatment process

Abstract

In this study, process parameters for wastewater treatment are optimised in order to improve the efficiency of energy recovery from wastewater treatment. In order to identify process parameters that have a significant impact on carbon recovery, we propose a stepwise forward variable selection method based on K-fold cross kriging to validate the loss ER.The results show that for complex wastewater treatment systems, the cross-validated kriging-based stepwise variable selection method can predict carbon recovery quickly and accurately. Finally, this paper combines the kriging prediction model with the PSO optimisation strategy to determine the optimal set points of process parameters with the objective of maximising carbon recovery.