The Use of Artificial Neural Networks (ANNs) in Modelling Sewerage Systems for Management in Real Time: Volume 1 - Main Report
HR Wallingford led a consortium of 5 contractors on behalf of UKWIR to carry out an investigation into the use of ANN models (Artificial Neural Networks) for producing 'instant' predictions of the performance of sewerage systems in real time.
An ANN model is trained against a series of recorded events for a point of interest (eg CSO spills and rainfall), and then used to predict its performance during current conditions.
The outcome of the project indicated that these models could achieve high levels of accuracy.
This tool is seen as being of value to the operations sections of water companies in providing rapid, continuous predictions during a rainfall event. This would allow instant intervention to be taken if a network incident is indicated.
The next stage for testing these models would be to carry out a field trial.
|UKWIR Reference :- 12/SW/01/2||Published Date :- 19/11/2012|
|Retail Price (£) :- 23||ISBN :- 1 84057 656 1|
Supporting Material and Reports
| The Use of Artificial Neural Networks (ANNs) in Modelling Sewerage Systems for Management in Real Time: Volume 1 - Main Report CD|