executive summary wastewater treatment
The observed variation in sludge rheology weakly correlated to the change of seasons.
of pressure drop for the selected pipeline. The Herschel-Bulkley model was used to fit the sludge flow curve (i.e., relationship between the shear stress and shear rate). Among the rheological parameters of the model, variation in consistency impacts more than others on the variation of pressure drop in the pipeline. Consistency shows 34% relative variation in digested sludge, 54% in primary sludge and 61% in thickened waste activated sludge. While yield stress shows 20% relative variation in digested sludge, 40% in primary sludge and 40% in thickened waste activated sludge. We showed that the residual error of sludge flow curve fitting where HerschelBulkley and Bingham models are used is not normally distributed and increases with shear rate. This means both models have errors in high range data which impact the fittings. To make fitting better for this system, one solution is to use robust algorithms (which systematically weight the data at the lower range). Herschel-Bulkley model best fit sludge flow curve data between 10-300 s-1 with a robust fitting method while modified Herschel-Bulkley best fit sludge flow curve data between 0.1-1000 s-1 with either robust or conventional fitting method. Bingham Plastic model only shows acceptable performance for primary sludge. Because of significant viscoplasticity (the state of showing yield stress which is minimum stress for material to flow) in sludge, Re3 friction loss model used for this calculation which was originally
developed for mineral slurries with similar rheological behaviour. In the original version of Re3 friction loss model, there are parameters appropriate for mineral slurries, one of them is pipe roughness correlated to the size of mineral particles. To estimate this parameter, we set up optimisation to change the roughness values of the model to achieve the least average of errors for pressure drop predictions in a 900m sludge pipeline. The result suggests a roughness of 256 μm provides the least error between the calculations and observations. But the accuracy of this optimisation needs to be verified in future investigations. The observed variation in sludge rheology weakly correlated to the change of seasons. A black-box model can be used to correlate rheology with the parameters of sludge composition. The Re3 friction loss model estimated the actual pressure drop of dilute digested sludge flow through the ETP sludge pipeline with 10% error. The pressure drop calculations for higher solid concentration suggested high variations of pressure drop, which requires further study. Dr Ehsan Farno is a Research Fellow in Chemical and Environmental Engineering at RMIT University. Dr Nicky Eshtiaghi is an Associate Professor in Chemical Engineering at RMIT University. To read the full article, visit the Water e-Journal at bit.ly/water_ejournal
www.awa.asn.au
95