Neural networks: an efficient approach to predict on-line the optimal coagulant dose

2004 ◽  
Vol 4 (5-6) ◽  
pp. 87-94 ◽  
Author(s):  
S. Deveughèle ◽  
Z. Do-Quang
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Water Treatment ◽  
Treatment Plant ◽  
Water Treatment Plant ◽  
Raw Water ◽  
Modeling Tools ◽  
Water Parameters ◽  
Physico Chemical ◽  
On Line

The problem under study was the on-line prediction of the optimal coagulant dose from raw water parameters; it has been tackled by using powerful modeling tools: Artificial Neural Networks (ANNs). Such tools do not rely on physico-chemical relationships; the model is built by using an historical dataset available on the plant (raw water parameters and Jar-tests data). A prototype has been implemented on a full-scale water treatment plant in France. The approach is explained, some relevant results are shown and the industrial benefits are discussed. The expected OPEX reduction (coagulant) is about 10%.

scholarly journals Modeling of Chlorine and Coagulant Dose in a Water Treatment Plant by Artificial Neural Networks

2019 ◽  
Vol 9 (3) ◽  
pp. 4176-4181
Author(s):  
A. S. Kote ◽  
D. V. Wadkar
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Water Treatment ◽  
Treatment Plant ◽  
Water Treatment Plant ◽  
Feed Forward Neural Network ◽  
Feed Forward ◽  
Dose Model ◽  
Artificial Neural

Coagulation and chlorination are complex processes of a water treatment plant (WTP). Determination of coagulant and chlorine dose is time-consuming. Many times WTP operators in India determine the coagulant and chlorine dose approximately using their experience, which may lead to the use of excess or insufficient dose. Hence, there is a need to develop prediction models to determine optimum chlorine and coagulant doses. In this paper, artificial neural networks (ANN) are used for prediction due to their ability to learn and model non-linear and complex relationships. Separate ANN models for chlorine and coagulant doses are explored with radial basis neural network (RBFNN), feed-forward neural network (FFNN), cascade feed forward neural network (CFNN) and generalized regression neural network (GRNN). For modeling, daily water quality data of the last four years are collected from the plant laboratory of WTP in Maharashtra (India). In order to improve performance, these models are established by varying input variables, hidden nodes, training functions, spread factor, and epochs. The best models are selected based on the comparison of performance measures. It is observed that the best performing chlorine dose model using defined statistics is found to be RBFNN with R=0.999. Similarly, the CFNN coagulant dose model with Bayesian regularization (BR) training function provided excellent estimates with network architecture (2-40-1) and R=0.947. Based on the above models, two graphical user interfaces (GUIs) were developed for real-time prediction of chlorine and coagulant dose, which will be useful for plant operators and decision makers.

Artificial neural networks for cost optimization of coagulation, sedimentation and filtration in drinking water treatment

2008 ◽  
Vol 8 (4) ◽  
pp. 383-388
Author(s):  
H.-J. Mälzer ◽  
S. Strugholtz
Keyword(s):  
Neural Networks ◽  
Drinking Water ◽  
Artificial Neural Networks ◽  
Water Treatment ◽  
Cost Optimization ◽  
Treatment Plant ◽  
Drinking Water Treatment ◽  
Water Treatment Plant ◽  
Artificial Neural ◽  
Safety And Stability

The applicability of Artificial Neural Networks (ANN) for process and costs optimization in drinking water treatment by coagulation, sedimentation and rapid filtration was investigated. The results showed that besides a considerable cost reduction, an improvement of process safety and stability can be expected. For further testing, the ANN will be installed at a water treatment plant for online coagulation control and process optimization.

scholarly journals Artificial neural networks to control chlorine dosing in a water treatment plant

2018 ◽  
Vol 40 (1) ◽  
pp. 37275
Author(s):  
André Felipe Librantz ◽  
Fábio Cosme Rodrigues dos Santos ◽  
Cleber Gustavo Dias
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Water Treatment ◽  
Treatment Plant ◽  
Water Treatment Plant ◽  
Artificial Neural

Fuzzy control of coagulation reaction through streaming current monitoring

1997 ◽  
Vol 36 (4) ◽  
pp. 127-134 ◽  
Author(s):  
J. C. Liu ◽  
M. D. Wu
Keyword(s):  
Fuzzy Logic Controller ◽  
Treatment Plant ◽  
Control Process ◽  
Water Treatment Plant ◽  
Critical Parameters ◽  
Flow Mode ◽  
Raw Water ◽  
Streaming Current ◽  
On Line ◽  
Current Monitoring

A fuzzy logic controller (FLC) incorporating the streaming current detector (SDC) was utilized in the automatic control of the coagulation reaction. Kaolinite was used to prepare synthetic raw water, and ferric chloride was used as the coagulant. The control set point was decided at a streaming current (SC) of −0.05 and pH of 8.0 from jar tests, zeta potential and streaming current measurements. A bench-scale water treatment plant with rapid mix, flocculation, and sedimentation units, operated in a continuous-flow mode, was utilized to simulate the reaction. Two critical parameters affecting the coagulation reaction, i.e., pH and streaming current, were chosen as process outputs; while coagulant dose and base dose were chosen as control process inputs. They were on-line monitored and transduced through a FLC. With raw water of initial turbidity of 110 NTU, residual turbidity of lower than 10 NTU before filtration was obtained. Results show that this combination functions satisfactorily for coagulation control.

Sign in / Sign up

Export Citation Format

Share Document