Transformers for interfacing anaerobic digestion models to pre- and post-treatment processes in a plant-wide modelling context

https://doi.org/10.1016/j.envsoft.2005.11.002Get rights and content

Abstract

In view of the growing importance of integrated and plant-wide modelling of wastewater treatment plants, this work reviews, applies and compares two transforming/interfacing methods by connecting anaerobic digestion and activated sludge models. The two methods are systematic approaches to transform state variables of one model to another and vice versa. The theory of the first method was presented before (Vanrolleghem et al., 2005. Wat. Sci. Technol., 52(1–2), 493–500.) as a general approach for interfacing any two models presented by Petersen matrices. The present work is the first application and therefore validation of this general approach. The theory of the second method was specifically developed for connecting ASM1 and ADM1, both standard IWA models. As an illustration, in this work a specific simulation example is presented in which the COST/IWA activated sludge benchmark plant is extended by sludge treatment and digestion facilities.

Introduction

Integrated modelling of wastewater systems comprising the collection network (sewer system), the treatment plant and the receiving water has been growing and advancing since the late 1990s (Butler and Schütze, 2005, Meirlaen et al., 2001). The treatment plant processes too should be dealt with in an integrated manner. Plant-wide modelling including anaerobic sludge digestion which is closely integrated to the activated sludge system has been proposed (Jeppsson et al., 2004, Zaher et al., 2002).

Anaerobic digestion receives a growing attention in the field of wastewater treatment, both in industrial and municipal sectors. For the latter, anaerobic digestion is employed for sludge treatment and stabilization. It considerably reduces the amount of sludge produced. However, the cost of waste sludge treatment constitutes approximately 35% of the capital cost and 55% of the annual operation and maintenance costs of a wastewater treatment plant (Cinar et al., 2004, Knezevic et al., 1995). As a compensation, the lower amounts of sludge reduce the cost of final disposal. The reduction of sludge volumes is turned mainly to biogas that can be used as clean energy source, e.g. generating electrical power for plant operation. For industries, anaerobic treatment could be a good option for wastewater treatment as an alternative to connecting to the sewer and paying higher tariffs. Other loads that can be connected to the digesters are truck loads from decentralised systems or sludge from other treatment plants. To study such scenarios and evaluate plant-wide control systems and operating strategies, integrated modelling of anaerobic digestion and other plant processes is very valuable.

The question tackled in this paper is how to connect models so that control systems, sensors and subsequent treatment processes can be evaluated at the plant-wide level. Also, the paper aims at enabling both the domestic and industrial sectors to assess the aforementioned possible solutions. For this purpose, a standard benchmark system, the BSM1 developed for activated sludge systems (Spanjers et al., 1998, Copp, 2002), is extended with the sludge treatment line including the anaerobic digester, which is also fed, for instance, with externally supplied truck loads of organic waste (Fig. 1).

The considered models are ASM1 (Henze et al., 2000), the standard model for BSM1, as description of the activated sludge process and ADM1 (Batstone et al., 2002) for the digester. Two transformers are built for each of the two interfacing methodologies that were compared and implemented in this study. The activated sludge plant is considered as a pre-treatment that concentrates the pollutants in the form of thickened secondary sludge to the digester; for this an ASM1–ADM1 transformer is needed. The activated sludge plant is also considered as post-treatment of the return liquors originating from sludge digestion and drying; for this, an ADM1–ASM1 transformer is needed.

In this paper, two methodologies to interface these two different processes (process models) are implemented. The extended benchmark example is used as a case study to compare both methodologies in terms of the overall plant output and simulated dynamics of the components in the digester.

Section snippets

Interfacing methodologies

Two methodologies are implemented in this work to interface the anaerobic model ADM1 to the activated sludge model ASM1 and vice versa. The first method is the general Continuity-Based Interfacing Method (CBIM) for models of wastewater systems described by Petersen matrices (Vanrolleghem et al., 2005). Since it is the first time the CBIM is applied and validated, the CBIM itself and its results will be illustrated in detail. The second method was specifically developed for ASM1–ADM1–ASM1

Results and discussions

The simulation results are discussed in two parts. In the first part, using the practical plant-wide example, the CBIM transformers are designed and the standard MCN transformers are applied. The first part aims to study the practical issues related to the use of the CBIM transformers and presents the ad hoc assumptions implemented in the first CBIM transformer to produce additional fluxes of Sfa, Ssu and Xsu. The second part of the results discusses the main differences between the two

Conclusions

Two methods recently proposed to interface ASM1 and ADM1 models were compared in a case study of a plant-wide model of a treatment works. Both lead to similar results in terms of the plant-wide output. Both interfacing methods lead to almost the same output of the activated sludge plant, biogas from the digester and sludge production. However, with the additional flexibility of the CBIM approach, the transformation of “origin” components can be distributed over a larger number of components on

Acknowledgement

The authors would like to thank the financial support of the EU projects: TELEMAC IST-2000-28156 and CD4WC – EVK1-CT-2002-00118.

References (18)

There are more references available in the full text version of this article.

Cited by (0)

View full text