A Linear Model Framework to Predict and Emulate Residual Chlorine in Water Treatment Plants Based on Laplace Transform

Chlorination disinfection is a critical process for ensuring water quality safety. However, intricate flow dynamics in water treatment plants significantly influence disinfection efficacy, often leading to fluctuating residual chlorine levels in treated water. Both elevated and insufficient residual chlorine concentrations present potential health hazards. In practical water treatment plants, the disinfectant dosage is mainly controlled manually based on human experience. This situation not only wastes disinfectants, but also has the potential to compromise water quality and trigger the production of a large number of by-products. In this work, we propose a data-driven Dual Linear Model framework based on Laplace Transform (DLMLT) to accurately predict and emulate the residual chlorine levels in water treatment plants. The model comprises two Linear Time-Invariant (LTI) system models tailored for residual chlorine dynamics prior to and post-clear water reservoir storage. Through rigorous data acquisition, model development, and validation, the proposed framework enables precise prediction of residual chlorine concentrations and effective simulation of disinfection dosing strategies. The proposed DLMLT can help optimize chlorine utilization and prevent overdosing, thereby ensuring safe drinking water standards across diverse treatment facilities.