Mariano Nicolas Cruz Bournazou, Harvey Arellano-Garcia, Günter
Wozny, Gerasimos Lyberatos, Costas Kravaris
ASM3 extended for two-step
nitrification–denitrification: a model reduction for
sequencing batch reactors
Open Access via institutional repository of Technische Universität Berlin
Document type
Journal article | Accepted version
(i. e. final author-created version that incorporates referee comments and is the version accepted for
publication; also known as: Author’s Accepted Manuscript (AAM), Final Draft, Postprint)
This version is available at
https://doi.org/10.14279/depositonce-15645
Citation details
This is the peer reviewed version of the following article:
Cruz Bournazou, M. N., Arellano-Garcia, H., Wozny, G., Lyberatos, G., Kravaris, C. (2012). ASM3 extended for
two-step nitrification-denitrification: a model reduction for sequencing batch reactors. In Journal of Chemical
Technology amp; Biotechnology (Vol. 87, Issue 7, pp. 887–896). Wiley. https://doi.org/10.1002/jctb.3694,
which has been published in final form at https://doi.org/10.1002/jctb.3694. This article may be used for
non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions.
Terms of use
This work is protected by copyright and/or related rights. You are free to use this work in any way permitted by
the copyright and related rights legislation that applies to your usage. For other uses, you must obtain
permission from the rights-holder(s).
For Peer Review
1
ASM3 extended for two-step nitrification-denitrification: 1
a model reduction for Sequencing Batch Reactors 2
The ASM3 extended for two step nitrification-denitrification is reduced in order to match the 3
specific conditions of Sequencing Batch Reactor systems with Shortcut Biological nitrogen 4
Removal 5
M. N. Cruz Bournazou*, H. Arellano-Garcia*, G. Wozny*, G. Lyberatos**, C. Kravaris**
6
* Chair of Process Dynamics and Operation, Berlin Institute of Technology, Sekr. KWT-9
7
Str. des 17. Juni 135, D-10623 Berlin, Germany
8
** Department of Chemical Engineering, University of Patras, 26504 Patras, Greece
9
m.nicolas.cruz@mailbox.tu-berlin.de
10
Tel.:+49 3031421634
11
Fax: +49 3031426915
12
KWT 9,Raum KWT-A 110
13
Straße des 17. Juni 135
14
D-10623 Berlin
15
16
Abstract 17
BACKGROUND: The ASM3 extended for two-step nitrification-denitrification represents the 18
most accurate model for the description of the Activated Sludge Process with Nitrate Bypass 19
Nitrification-Denitrification. This model includes 20 reaction rates, 15 state variables, and 20
more than 35 parameters, which make its calculation costly and difficult to estimate. The lack 21
of a fast and accurate model, able to predict both, concentration of nitrite and nitrate over 22
time, is the principal obstacle for efficient model-based optimization and model-based control. 23
Page 1 of 47
http://mc.manuscriptcentral.com/jctb-wiley
Journal of Chemical Technology & Biotechnology
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For Peer Review
2
RESULTS: In this work, a fast and accurate model for the activated sludge process in a 1
Sequencing Batch Reactor is proposed. For this purpose, the ASM3 extended for two step 2
nitrification-denitrification is reduced in order to match the specific conditions of Sequencing 3
Batch Reactor systems with Shortcut Biological nitrogen Removal. The proposed model 4
considers a two-step nitrification-denitrification process and can thus describe the bypass of 5
nitrate. Different approaches for model reduction together with an exhaustive analysis of the 6
extended ASM3 model as well as the process are discussed. 7
CONCLUSIONS: The resulting model with only five differential equations reduces the 8
calculation time up to one order of magnitude while maintaining a high description accuracy, 9
which states the advantages of model reduction. 10
Keywords:
model reduction, activated sludge, ASM3, two-step nitrification-
11
denitrification, nitrate bypass, nitrification-denitrification, invariant reaction, time scale
12
analysis
13
1. Introduction 14
Modeling of the Activated Sludge Process (ASP) for wastewater treatment represents a mayor 15
bioengineering challenge. A wide consortium of microorganisms consumes carbonaceous 16
matter producing energy and new products in a complex symbiotic process. Biological 17
nitrogen removal through nitrification/denitrification is currently widely practiced in order to 18
produce a clarified effluent to be discharged to sensitive receiving water bodies, preventing 19
eutrophication. Furthermore, the description of the nitrification-denitrification process as a 20
reaction with at least two steps is gaining importance for the simulation of optimal ASPs. It 21
has been suggested that the Nitrate Bypass Nitrification-Denitrification (NBND), also known 22
as shortcut Biological Nitrogen Removal (SBNR), process shows important advantages over 23
conventional nitrification-denitrification
1
. It is evident that nitrate bypassing process can only 24
Page 2 of 47
http://mc.manuscriptcentral.com/jctb-wiley
Journal of Chemical Technology & Biotechnology
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For Peer Review
3
be represented by a model which considers nitrate production and consumption during 1
nitrification-denitrification. 2
Although an exact description of the multiple biochemical reactions taking place in the 3
process is impossible, some models achieve to describe the general behavior of the system 4
and have been widely implemented in practice. The ASM3 extended for two-step 5
nitrification-denitrification (referred in this work as extended ASM3
2
) represents the most 6
accurate model for the description of the ASP with NBND, other similar works can be found 7
in literature
3
. This model includes 20 reaction rates, 15 state variables, and more than 35 8
parameters, which make its calculation costly and the model highly unidentifiable. The lack of 9
a fast and accurate model, able to predict both concentration of nitrite and nitrate, is the 10
principal obstacle for an efficient model based optimization and model-based control strategy 11
considering NBND. 12
Many efforts have been made to create simple models for Waste Water Treatment (WWT) 13
processes, in particular for the Activated Sludge Process (ASP)
4
. Some examples represent 14
the boundary relocation
5
, neural networks
6
, and model reduction for steady state model 15
based control
7
. Nevertheless, a systematic model reduction of an Activated Sludge Model 16
(ASM) able to describe the bypass nitrate nitrification-denitrification while maintaining the 17
important dynamics of the mechanistic model has to our best knowledge never been reported. 18
Page 3 of 47
http://mc.manuscriptcentral.com/jctb-wiley
Journal of Chemical Technology & Biotechnology
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For Peer Review
4
Aim of this work is to create a model for an efficient simulation of the SBR process under 1
NBND. To achieve this goal, the extended ASM3 is reduced to match the specific conditions 2
of SBR systems under NBND conditions. 3
2. Background 4
2.1. Activated Sludge 5
The most applied method for biological treatment of waste water is the ASP
8
. The family of 6
Activated Sludge Models (ASM) represents the state-of-the-art model framework for ASP 7
simulation
6
. ASM1 is the most widely used in practice
9
, ASM2
10
is applied to simulate 8
processes that include biological phosphorus removal
11
and the latest version, ASM3
12
, 9
includes the quantification of energy storage in order to describe ready biodegradable 10
substrate and oxygen uptake with higher accuracy. Finally, a newer version of ASM3, 11
referred to in this contribution as extended ASM3, where nitrification and denitrification are 12
considered as two-step processes, taking into account nitrite as an intermediate, has recently 13
been presented
2
. 14
In order to extend the ASM3, 7 process equations were included, resulting in a stoichiometric 15
matrix with dimension 15x20. In addition to a low parameter identifiability, the extended 16
ASM3 demands significant computation time. These drawbacks represent the main obstacle 17
for efficient optimization and model-based control. On the other hand, the extended ASM3 18
describes many states, which can be neglected for the specific case of SBR process
13, 14
. 19
Page 4 of 47
http://mc.manuscriptcentral.com/jctb-wiley
Journal of Chemical Technology & Biotechnology
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Loading more pages...