Manage Sewer In-Line Storage Control Using Hydraulic Model and Recurrent Neural Network
- 284 Downloads
This paper described manage sewer in-line storage control for the city of Drammen, Norway. The purpose of the control is to use the free space of the pipes to reduce overflow at the wastewater treatment plant (WWTP). This study combined the powerful sides of the hydraulic model and neural networks. A detailed hydraulic model was developed to identify which part of the sewer system have more free space. Subsequently, the effectiveness of the proposed control solution was tested. Simulation results showed that intentionally control sewer with free space could significantly reduce overflow at the WWTP. At last, in order to enhance better decision making and give enough response time for the proposed control solution, Recurrent Neural Network (RNN) was employed to forecast flow. Three RNN architectures, namely Elman, NARX (nonlinear autoregressive network with exogenous inputs) and a novel architecture of neural networks, LSTM (Long Short-Term Memory), were compared. The LSTM exhibits the superior capability for time series prediction.
KeywordsSewer system Hydraulic model Recurrent neural network LSTM
- Brownlee J (2017) Chapter 25: Time series prediction with LSTM recurrent neural networks. In Deep Learning with Python [e-book]. https://machinelearningmastery.com/deep-learning-with-python2/. Accessed 24 Mar 2017
- Cho, K., Van Merriënboer, B., Bahdanau, D., and Bengio, Y. (2014). On the properties of neural machine translation: Encoder-decoder approaches. arXiv preprint arXiv:1409.1259.Google Scholar
- DHI group (2014) https://www.mikepoweredbydhi.com/products/mike-urban Accessed 24 Mar 2017
- Gers F (2001) Long short-term memory in recurrent neural networks. Dissertation, Universität HannoverGoogle Scholar
- Grum M, Thornberg D, Christensen ML, Shididi SA, Thirsing C (2011) Full-scale real time control demonstration project in Copenhagen’s largest urban drainage catchments. In: Proceedings of the 12th international conference on urban drainage, Porto Alegre.Google Scholar
- Keras Documentation (2015) https://keras.io/getting-started/faq/. Accessed 24 Mar 2017.
- Lipton, Z. C., Berkowitz, J., and Elkan, C. (2015). A critical review of recurrent neural networks for sequence learning. arXiv preprint arXiv:1506.00019.Google Scholar
- Martinez N (2016) Analyse av. fordrøyningstiltak på eksisterende avløpsnett i Solumstrand rensedistrikt. Dissertation, Norwegian University of Life Sciences. In Norwegian.Google Scholar
- Ødegård J, Persson M, Mathiesen TB (2013) Investeringsbehov i vann og avløpssektoren. Norsk Vann. In Norwegian.Google Scholar
- Remesan R, Mathew J (2014) Hydrological data driven modelling: a case study approach. Springer, Switzerland, pp 71–110Google Scholar
- Rossman, L. A. (2010). Storm water management model user’s manual, version 5.0. Cincinnati: National Risk Management Research Laboratory, Office of Research and Development, US Environmental Protection Agency.Google Scholar
- Schmidt V (2016) Keras recurrent tutorial. https://github.com/Vict0rSch/deep_learning/tree/master/keras/recurrent. Accessed 24 Mar 2017.
- Seggelke K, Rosenwinkel KH, Vanrolleghem PA, Krebs P (2005) Integrated operation of sewer system and WWTP by simulation-based control of the WWTP inflow. Water Sci Technol 52(5):195–203Google Scholar
- Xingjian SHI, Chen Z, Wang H, Yeung DY, Wong WK, Woo WC (2015) Convolutional LSTM network: a machine learning approach for precipitation nowcasting. In: Advances in neural information processing systems, pp. 802–810.Google Scholar
- Zaytar MA, Amrani CE (2016) Sequence to sequence weather forecasting with long short-term memory recurrent neural networks. Int J Comput Appl 143:7–11Google Scholar