Optimal sampling design for a sludge blanket interface settling model

• Published in: Water research (Oxford) Vol. 31; no. 5; pp. 1148 - 1154
• Main Authors: Renko, Esa K., Sirviö, Hannu
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.05.1997; Elsevier Science
• Subjects: Applied sciences; Biological and medical sciences; Biological treatment of waters; Biotechnology; D-optimality; design of experiments; Environment and pollution; Exact sciences and technology; Fundamental and applied biological sciences. Psychology; General purification processes; Industrial applications and implications. Economical aspects; modelling; Pollution; sludge settling; Wastewaters; Water treatment and pollution; Wynn-Fedorov algorithm; Wynn-Fedorov algorithm; D-optimality; sludge settling; modelling; design of experiments; Activated sludge; Optimal sampling; Experimental design; Waste water purification; Modeling; Decantation; Interface; Biological purification
• ISSN: 0043-1354; 1879-2448
• DOI: 10.1016/S0043-1354(96)00359-4

An experimental design for a nonlinear sludge blanket interface settling model was examined. The model describes sludge settling in a batch reactor as a function of time having two parameters. The criterion applied in detecting the minimal sampling schedule was the D-optimality. The design was constructed on the basis of the Wynn-Fedorov algorithm. The algorithm proved its validity leading to a logical result. The optimal sampling design consists of two measurement points which is equal to the number of parameters to be estimated. This is a generally observed result in D-optimality and can be interpreted that in the D-optimality the information per implemented observation is maximized. In addition to the determination of optimal sampling schedule, the experimental design can be used in examining models. The two measurement points have the same weights indicating that the parameters are equally important in the examined model. The results also show that a small change in the nominal values of the parameters have a minor effect on the output of the studied model. The model output is more sensitive to the changes in the parameter α than to the changes in the parameter C. In the field of environmental engineering, model oriented observational design is rarely utilized, although it might be helpful in achieving better understanding of models and more effective utilization of resources.