Keywords

design, monte carlo, uncertainty analysis, wastewater treatment plant

Start Date

1-7-2012 12:00 AM

Abstract

This paper presents a prototype tool for design of Wastewater Treatment Plants (WWTPs). It is a model-based approach that explicitly accounts for both temporal variability and uncertainty. It evaluates a set of WWTP designs for a predefined configuration and estimates the probability of compliance (POC) given some sources of uncertainty. The proposed tool uses two nested loops: the outer loop tests different scenarios looking for the optimal combination of the design variables; the inner loop uses a Monte Carlo simulation propagating the uncertain model inputs to obtain the POC. The Benchmark Simulation Model no.1 was used as basic plant configuration. The design variables were: total volume of the plant; aerobic fraction; waste flow rate; recycle flow rate; and internal recirculation flow rate. The sources of uncertainty included parameters related to the biochemical model and the secondary clarifier settling. A set of 100 designs was evaluated by comparing the distributions of the average and maximum concentrations of NH4, TN and TSS with respect to typical effluent requirements. It was found that the effect of the sources of uncertainty is quite important to evaluate the performance of the plant. For example, while the BSM1 plant fulfills some effluent requirements using the default parameter values, it was shown that it only can guarantee them with a POC below 0.85 when considering the uncertainty. This is an example of the potential that this tool has for better informing the engineering firms about their design proposals.

COinS
 
Jul 1st, 12:00 AM

A tool for optimum design of WWTPs under uncertainty: Estimating the Probability of Compliance

This paper presents a prototype tool for design of Wastewater Treatment Plants (WWTPs). It is a model-based approach that explicitly accounts for both temporal variability and uncertainty. It evaluates a set of WWTP designs for a predefined configuration and estimates the probability of compliance (POC) given some sources of uncertainty. The proposed tool uses two nested loops: the outer loop tests different scenarios looking for the optimal combination of the design variables; the inner loop uses a Monte Carlo simulation propagating the uncertain model inputs to obtain the POC. The Benchmark Simulation Model no.1 was used as basic plant configuration. The design variables were: total volume of the plant; aerobic fraction; waste flow rate; recycle flow rate; and internal recirculation flow rate. The sources of uncertainty included parameters related to the biochemical model and the secondary clarifier settling. A set of 100 designs was evaluated by comparing the distributions of the average and maximum concentrations of NH4, TN and TSS with respect to typical effluent requirements. It was found that the effect of the sources of uncertainty is quite important to evaluate the performance of the plant. For example, while the BSM1 plant fulfills some effluent requirements using the default parameter values, it was shown that it only can guarantee them with a POC below 0.85 when considering the uncertainty. This is an example of the potential that this tool has for better informing the engineering firms about their design proposals.