SOM-Based Modelling for an Activated Sludge Treatment Process

The control of an activated sludge treatment plant depends on on-line process data and measurements made in the laboratory. The most important quality indicators of activated sludge are settling properties. Even though the behaviour of activated sludge is basically known, long lags and the complex characteristic of biological activity set a challenge for controlling the process. In this paper, the process lags of an activated sludge treatment plant are determined by a cross-correlation method. We also demonstrate the use of Self-Organizing Maps (SOM) for analysing an activated sludge treatment process. A data set of the process was used for training a SOM, which could be used for estimating the diluted sludge volume index (DSVI). The results of a validation data set showed a good agreement between the actual and estimated values of DSVI. The results indicate that the SOM method can also be efficiently applied to the analysis of the activated sludge treatment process and so towards better process control.