Accurate residual-chlorine prediction in pressurized water systems can be affected by the way one-dimensional models aggregate radial transport and pipe-wall processes. This study develops an empirical Reynolds-number-dependent correction for chlorine transport using controlled loop experiments. A 20 m PVC loop with a nominal diameter of 2 in was operated at eleven Reynolds numbers from 500 to 30,000. Each condition was tested in triplicate; chlorine was restored to 1.5 mg/L before each run and residual concentration was recorded hourly for 72 h. The analysis used a dimensionless wall-related term Z and a worksheet-defined percentage error E. For each Reynolds-number group, the normalized response was calculated as ? = E?/Z? and fitted with a power law in logarithmic space. Parameter uncertainty and predictive performance were assessed using 95% confidence intervals, MAE, RMSE, MAPE, and leave-one-Reynolds-group-out cross-validation. ? decreased from 0.300650 at Re = 500 to 0.003466 at Re = 30,000. The fitted relationship, ? = 0.1306(Re/1000)?1.063, produced R² = 0.9967, RMSE = 0.0094, and MAPE = 5.69%; cross-validation yielded RMSE = 0.0140 and MAPE = 6.95%. The formulation is therefore proposed as a compact diagnostic correction within the tested PVC-loop range and should not be used as a substitute for system-specific hydraulic and water-quality calibration.
Copyrights © 2026