State estimation-based linear model-predictive control (MPC) with augmented disturbance model was applied to a continuous wastewater neutralization process for rejecting disturbances in feed pH and feed flow rate of the inlet process stream. The wastewater neutralization process is characterized by its inherent time-varying nonlinearity, complexity with varying time delays. These processes were efficiently controlled through the use of step-ahead predictions to make proper control decisions in order to eliminate the effects caused by entering disturbances. Accuracy of the process model used in the predictive control algorithms greatly impacted the performance of the controller. Implementation of MPC with linear process model approximations for non-linear processes commonly leads to plant-model mismatch due to rapidly changing dynamics and external disturbances, which is known to pose a constraints on usage of traditional linear MPC for pH neutralization process. To address this mismatch between plant and model predictions, due to the nonlinearity of the process, the states were corrected at each sampling instant using Kalman filtering. To account for external disturbances, additional disturbance states were augmented to the process model and were estimated along with the process states and propagated in step-ahead predictions. This augmentation introduced an integral action in the control implementation and ensured the offset-free setpoint tracking.