Dynamic modelling of the activated sludge process: Improving prediction using neural networks

Côté, Mélissa; Grandjean, Bernard P. A.; Lessard, Paul; Thibault, Jules

doi:10.1016/0043-1354(95)93250-w

Cited by 149 publications

(35 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Operation, control and supervision of WWTPs have been approached from many different points of view, including classical control methods [4][5][6][7], mechanistic models [8,9], knowledge-based systems [10][11][12], case-based reasoning [13], neural nets [3,14] and hybrid approaches [15,16]. However, a direct cause-effect relationship for WWTP performance has been established only in a few cases and, even in those, experimental results could lead to contradictory conclusions, avoiding the formulation of deterministic cause-effect relationships that could be used as prediction models.…”

Section: Previous Work and Objectivesmentioning

confidence: 99%

Prediction of the bulking phenomenon in wastewater treatment plants

Belanche

Valdés

Comas

et al. 2000

Artificial Intelligence in Engineering

View full text Add to dashboard Cite

The control and prediction of wastewater treatment plants poses an important goal: to avoid breaking the environmental balance by always keeping the system in stable operating conditions. It is known that qualitative information -coming from microscopic examinations and subjective remarks -has a deep influence on the activated sludge process. In particular, on the total amount of effluent suspended solids, one of the measures of overall plant performance. The search for an input-output model of this variable and the prediction of sudden increases (bulking episodes) is thus a central concern to ensure the fulfillment of current discharge limitations. Unfortunately, the strong interrelation between variables, their heterogeneity and the very high amount of missing information makes the use of traditional techniques difficult, or even impossible. Through the combined use of several methods -rough set theory and artificial neural networks, mainly -reasonable prediction models are found, which also serve to show the different importance of variables and provide insight into the process dynamics. ᭧

show abstract

Section: Previous Work and Objectivesmentioning

confidence: 99%

Prediction of the bulking phenomenon in wastewater treatment plants

Belanche

Valdés

Comas

et al. 2000

Artificial Intelligence in Engineering

View full text Add to dashboard Cite

show abstract

“…Comparable observations were made by Cote et al (1995). Cote et al (1995) compared different types of model by which the effluent from a wastewater treatment plant was predicted.…”

Section: Resultsmentioning

confidence: 99%

“…(Hagan and others 1996;Yen and Langari 1999). A large quantity of data is required for further calculation by artificial neural network (ANN) model (Cote et al 1995;Häck and Köhne 1996;Zhang and Stanley 1999;Gontarski et al 2000;Choi and Park 2001).…”

Section: Introductionmentioning

confidence: 99%

Prediction of the effluent from a domestic wastewater treatment plant of CASP using gray model and neural network

Chen

2009

Environ Monit Assess

View full text Add to dashboard Cite

When a domestic wastewater treatment plant (DWWTP) is put into operation, variations of the wastewater quantity and quality must be predicted using mathematical models to assist in operating the wastewater treatment plant such that the treated effluent will be controlled and meet discharge standards. In this study, three types of gray model (GM) including GM (1, N), GM (1, 1), and rolling GM (1, 1) were used to predict the effluent biochemical oxygen demand (BOD), chemical oxygen demand (COD), and suspended solids (SS) from the DWWTP of conventional activated sludge process. The predicted results were compared with those obtained using backpropagation neural network (BPNN). The simulation results indicated that the minimum mean absolute percentage errors of 43.79%, 16.21%, and 30.11% for BOD, COD, and SS could be achieved. The fitness was higher when using BPNN for prediction of BOD (34.77%), but it required a large quantity of data for constructing model. Contrarily, GM only required a small amount of data (at least four data) and the prediction results were analogous to those of BPNN, even lower than that of BPNN when predicting COD (16.21%) and SS (30.11%). According to the prediction, results suggested that GM could predict the domestic effluent variation when its effluent data were insufficient.

show abstract

“…This is due to the inherent feature of black box modeling methodology; simplicity, fault and noise tolerance, plasticity property and versatility to process changes (Shahaf and Marom, 2001). Recently many literatures have been published using Artificial Neural Networks (ANN) for modeling biological wastewater treatment processes (Cote et al, 1995;Lee and Park, 1999;Gontarski et al, 2000;Cinar et al, 2006). ANNs have also been applied successfully for a wide variety of SBR based applications; process monitoring and control (Zhao and Kummel, 1995), soft sensors (Lee and Park, 1999) and online control of process variables (Cho et al, 2001).…”

Section: Introductionmentioning

confidence: 99%

Experimental Results and Neural Prediction of Sequencing Batch Reactor Performance under Different Operational Conditions

Rene

Joo²,

Park

2008

J ENV INFORM

View full text Add to dashboard Cite

ABSTRACT. Three lab scale sequencing batch reactors (SBR) were simultaneously operated at different process conditions to understand the dynamics of organic and nitrogen removal from a synthetic wastewater source. The SBRs were operated continuously for 255 days at different C/N ratio (3 -6), aeration time (4 -10 hr) and salt concentrations (0.5 -2%). The COD removal efficiencies under steady state operation were consistently greater than 80%, while nitrogen removal efficiencies (10 -98%) were inhibited by high salt concentrations. Back propagation neural network was applied to model this experimental data using influent COD, influent nitrogen, salt concentration, aeration time, MLSS concentration and C/N ratio as the input parameters to predict the performance parameters, viz., COD removal efficiency (COD-RE), total nitrogen removal efficiency (T-RE), NH 4 + -N, NO 3 --N and NO 2 --N formed. The data points were randomized and divided into training (190 × 3) and testing set (65 × 3). The internal network parameters were selected using the 2 k full factorial design of experiments. The appropriate network topology for this system (6-12-5) was selected by estimating the best correlation coefficient (R) value (0.8482) achieved during prediction of the testing set. The result from this study showed that a neural network based model can be used as an efficient data driven model to predict the performance of a SBR unit.

show abstract

Dynamic modelling of the activated sludge process: Improving prediction using neural networks

Cited by 149 publications

References 15 publications

Prediction of the bulking phenomenon in wastewater treatment plants

Prediction of the bulking phenomenon in wastewater treatment plants

Prediction of the effluent from a domestic wastewater treatment plant of CASP using gray model and neural network

Experimental Results and Neural Prediction of Sequencing Batch Reactor Performance under Different Operational Conditions

Contact Info

Product

Resources

About