Composite Control of Post-Chlorine Dosage During Drinking Water Treatment

Chlorination is an effective disinfection method of drinking water treatment. Optimizing this process poses certain challenges, as the objectives of reaching the required disinfection effect and limiting the formation of disinfection by-products must be met simultaneously. These need adjusting the chlorine dosage in real time to cope with fluctuations in water quality and maintaining a free chlorine residual at the clear-water reservoir outlet that does not exceed regulatory limits. This paper proposes a composite control scheme that combines disturbance observer (DOB) and model predictive control (MPC) to meet the aforementioned objectives. MPC is employed for the feedback control of post-chlorine dosage, and DOB is adapted to monitor fluctuations in influent water quality. The estimated value from the DOB is applied in the feed-forward control of post-chlorine dosage. Both simulation and experimental results demonstrate that the proposed DOB-MPC composite control scheme provides a better disturbance rejection effect against fluctuations in influent water quality than MPC-only and PID-only control schemes, and it can effectively stabilize the quality of treated water.