DEVELOPMENT OF A CREDIBLE RISK MODEL FOR A RECLAIMED WATER REUSE ANALYSIS FOR THE F. WAYNE HILL WATER RESOURCES CENTER

As part of a multi-phased wastewater reuse evaluation for the F. Wayne Hill Water Resources Center (the Hill Plant), a risk model was developed, similar to that used in the nuclear power industry. This model is capable of determining the likelihood of a wastewater treatment plant providing insufficient treatment of the influent sewage. The development and implementation of the model were deemed to be critical steps in evaluating the risks associated with use of reclaimed water as source water for potable water production.The risk model was based on well-established reliability and statistical techniques used in numerous industries worldwide. The risk model consists of two components, a process component that mapped the efficiency of each individual process in the plant and a reliability component used to derive the status of all individual parts of the process during a given period of time.The process model evaluates the efficiency of individual processes in removing the pollutants modeled. A commercially available Monte Carlo simulation tool, Crystal Ball®, was used to provide a large number of random numbers to simulate the variations in the influent qualities and the process effectiveness of the Hill Plant. The process model was used to determine the impact of the plant on the levels of seven parameters subject to effluent limits contained in the existing National Pollutant Discharge Elimination System (NPDES) permit for the treatment plant. The seven parameters were chemical oxygen demand (COD), dissolved oxygen (DO), pH, turbidity, ammonia, total phosphorus and fecal coliform bacteria.The reliability model component was used to determine the status of all parts of the treatment plant for a given timeline. The reliability model was constructed with a number of fault conditions to determine the impact of various degraded states of the plant.To determine the levels of parameters in the effluent, the model simulation was allowed to run for a period of five years. Comparing the risk model outputs to actual levels of the parameters in the effluent of the Hill Plant showed that the model produced reasonable estimates of the average levels of parameters and reasonable estimates of the expected variations.The frequency of producing effluent not meeting permit requirements was determined to be driven by the frequency of producing COD in the effluent not meeting the 30-day average limit. The model was further used for sensitivity studies to identify an appropriate set point for COD monitoring equipment to trigger diversion of treated effluent to off-spec storage thereby reducing the likelihood of the effluent exceeding the permit levels. The plant staff adjusted the operating set point to the level recommended by the model. No excursions have been experienced in the plant to date. With the exception of COD, no single point credible failure conditions were identified.Risk modeling provides design engineers, facility owners and operators with a tool to help evaluate and eliminate potential points of failure in wastewater treatment processes. Future developments in this arena could include linking one of the sophisticated, scientifically-based process models currently on the market with the type of reliability model used here.With critical water resources already being stressed, the importance of protecting and efficiently using our water resources will only continue to grow. Water reuse will likely play a greater and greater role in water resource management. Designing for and demonstrating the reliability of treatment facilities will be key to gaining public acceptance, as well as to protecting public health and the aquatic environment.