In an effort to help policy makers and manufacturers understand the impact of parameter uncertainties on UV reactor performance, a numerical bioassay model was developed by integrating a UV reactor model based on computational fluid dynamics with a Monte Carlo model developed to account for parameter uncertainty. For the model implemented, it was determined that reactor performance uncertainty was less than 6%. The integrated model was used to evaluate several checkpoint bioassay criteria including one currently used by the California Department of Public Health. The model showed that these criteria failed to take into account the fact that in an ideal case, a full scale reactor will pass a single checkpoint test 50% of the time. In reality, differences in equipment measurement errors between the system validation and checkpoint bioassay, and limitations of the power law form of the dose monitoring equation in accurately representing system validation data will result in poorer than expected performance. It was suggested that such checkpoint criteria be modified by crediting the inherent over-sizing of full scale reactors.

This content is only available as a PDF.