Nonlinear state and parameter estimation in batch biological wastewater treatment

Abstract

Wastewater treatment is an active research area from both the technological and the control points of view. Various types of supervisory and low-level control approaches can be applied in this field. However, the inaccessibility of sensors which should provide the data needed for control is one of the major obstacles in current development. Estimation is therefore a prerequisite for performing those control tasks that rely on important process variables which cannot be measured directly. In this paper an estimation algorithm for simultaneous state and parameter estimation of a nonlinear process is derived for a batch biological wastewater treatment process. The aim is to estimate two important process variables related to dissolved oxygen dynamics during the aerobic phase of a batch cycle: the oxygen transfer function and biomass respiration rate. The main contributions of the work presented are providing on-line information about these variables for improved process control and diagnosis, and the application of the chosen nonlinear estimation algorithms for simultaneous state and parameter estimation in process control.