Stochastic Analysis of Trickling Filter

In most conventional methods of analysis and design for wastewater processes the stochastic nature of the process inputs and parameters is ignored. Stochastic treatment of a process is more difficult than a deterministic treatment but can provide more significant information concerning the design parameters. A stochastic approach to trickling filter design is described based on a two-phase two-component model of the filter. Random variables were introduced to represent random BOD inputs and flow rates as well as uncertainty in the model coefficients. This study indicates under what circumstances the sample estimate of the expected value of the output of a stochastic model of the trickling filter differs from the output of the related deterministic model based on replacing the random variables by their expected values. It also shows how to use the measurements of sample dispersion of the output BOD to compute safety factors for design.