PERFORMANCE VALIDATION OF UV SYSTEMS:IMPLICATIONS FOR SYSTEM DESIGN AND OPERATION

摘要

In December 2000, the National Water Research Institute and the American Water Works Association Research Foundation published the Ultraviolet Disinfection Guidelines for Drinking Water and Water Reuse (NWRI/AWWARF, 2000). One of the primary purposes of the 2000 Guidelines is to provide a standard approach for evaluation and implementation of ultraviolet (UV) disinfection technologies. In the two and one half years since publication of the 2000 Guidelines, a handful of performance validations of full-scale UV systems for water reuse have been completed (WBA, 2001a, 2001b, 2002, 2003a, 2003b). The purpose of these validations has been to evaluate system performance for either permitting of site specific UV systems or obtaining regulatory agency acceptance of a specific UV technology. The performance validations involved seeding filtered secondary effluent with a test organism, MS-2 bacteriophage, and measuring the degree of inactivation achieved by the UV system under various flow conditions and number of UV lamps. The data were analyzed by comparing the inactivation achieved by the full-scale UV system with static bioassay results obtained in the laboratory using a collimated beam apparatus, which allows accurate measurement of UV dose. This comparison allows the assignment of a delivered dose to each field bioassay result. As specified in the 2000 Guidelines, the delivered dose data must be analyzed as a function of flow per lamp. A typical performance validation plot is shown in Figure 1. Unfortunately, a number of issues arise in interpreting the data presented in these figures before these data can be applied to the design and operation of UV disinfection systems. Three important issues are evaluated and discussed in this paper: (1) how to account for the improved performance of multiple UV banks in series at the same flow per lamp, (2) how to account for varying UV transmittance (UVT) in determining system performance curves, and (3) application of the testing results to day-to-day system operation.