Short-term forecasting of turbidity in trunk main networks.

Water discolouration is an increasingly important and expensive issue due to rising customer expectations, tighter regulatory demands and ageing Water Distribution Systems (WDSs) in the UK and abroad. This paper presents a new turbidity forecasting methodology capable of aiding operational staff and enabling proactive management strategies. The turbidity forecasting methodology developed here is completely data-driven and does not require hydraulic or water quality network model that is expensive to build and maintain. The methodology is tested and verified on a real trunk main network with observed turbidity measurement data. Results obtained show that the methodology can detect if discolouration material is mobilised, estimate if sufficient turbidity will be generated to exceed a preselected threshold and approximate how long the material will take to reach the downstream meter. Classification based forecasts of turbidity can be reliably made up to 5 h ahead although at the expense of increased false alarm rates. The methodology presented here could be used as an early warning system that can enable a multitude of cost beneficial proactive management strategies to be implemented as an alternative to expensive trunk mains cleaning programs.

[1]  Neil Davey,et al.  Time Series Prediction and Neural Networks , 2001, J. Intell. Robotic Syst..

[2]  Joby Boxall,et al.  Discoloration Material Accumulation in Water Distribution Systems , 2011 .

[3]  Adrian J. Saul,et al.  Empirical travel time estimation in a distribution network , 2002 .

[4]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[5]  Manfred Morari,et al.  The false nearest neighbors algorithm: An overview , 1997 .

[6]  Joby Boxall,et al.  The role of trunk mains in discolouration , 2010 .

[7]  L. Tsimring,et al.  The analysis of observed chaotic data in physical systems , 1993 .

[8]  J. Vreeburg,et al.  Impact of particles on sediment accumulation in a drinking water distribution system. , 2008, Water research.

[9]  Joby Boxall,et al.  Modelling both the continual erosion and regeneration of discolouration material in drinking water distribution systems , 2014 .

[10]  Fei Li,et al.  Burst Detection Using an Artificial Immune Network in Water-Distribution Systems , 2014 .

[11]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[12]  S Husband,et al.  Understanding and managing discolouration risk in trunk mains. , 2016, Water research.

[13]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[14]  A. Rango,et al.  MERITS OF STATISTICAL CRITERIA FOR THE PERFORMANCE OF HYDROLOGICAL MODELS1 , 1989 .

[15]  Joby Boxall,et al.  Online data processing for proactive UK water distribution network operation , 2013 .

[16]  J. C. van Dijk,et al.  Particles in the drinking water system: from source to discolouration , 2004 .

[17]  Ivan Stoianov,et al.  Investigating the Impact of Sectorized Networks on Discoloration , 2015 .

[18]  Joby Boxall,et al.  Analysis of multi-parameter water quality data using event detection software on laboratory simulated events , 2012 .

[19]  Edward Keedwell,et al.  Short-term Forecasting of Turbidity in a UK Water Distribution System☆ , 2016 .

[20]  Joby Boxall,et al.  Operational management of trunk main discolouration risk , 2016 .

[21]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..

[22]  Pierre Baldi,et al.  Assessing the accuracy of prediction algorithms for classification: an overview , 2000, Bioinform..

[23]  E. Blokker.  Stochastic water demand modelling for a better understanding of hydraulics in water distribution networks , 2010 .

[24]  Yumiko Abe,et al.  Cohesiveness and hydrodynamic properties of young drinking water biofilms. , 2012, Water research.

[25]  Andreas S. Weigend,et al.  The Future of Time Series: Learning and Understanding , 1993 .

[26]  A J Saul,et al.  Laboratory studies investigating the processes leading to discolouration in water distribution networks. , 2008, Water research.

[27]  Joby Boxall,et al.  Regeneration of Discolouration in Distribution Systems , 2003 .

[28]  Jan Vreeburg.  Discolouration in drinking water systems: a particular approach , 2007 .

[29]  Avi Ostfeld,et al.  A dynamic thresholds scheme for contaminant event detection in water distribution systems. , 2013, Water research.

[30]  Joby Boxall,et al.  Improved representation of water age in distribution networks to inform water quality. , 2009 .

[31]  P. G. Schaap,et al.  Particle Accumulation Rate of Drinking Water Distribution Systems Determined by Incoming Turbidity , 2015 .

[32]  J H G Vreeburg,et al.  Discolouration in potable water distribution systems: a review. , 2007, Water research.

[33]  J. C. van Dijk,et al.  Simulating Nonresidential Water Demand with a Stochastic End-Use Model , 2013 .

[34]  Rob J Hyndman,et al.  Another look at measures of forecast accuracy , 2006 .

[35]  Joby Boxall,et al.  Online modelling of water distribution systems: a UK case study , 2009 .

[36]  D. Stephenson,et al.  Criteria for comparison of single event models , 1986 .

[37]  Thomas G. Dietterich.  Machine Learning for Sequential Data: A Review , 2002, SSPR/SPR.

[38]  Ian Goulter,et al.  Relationship between Velocity Profiles and Turbidity Problems in Distribution Systems , 2001 .

[39]  J. Vreeburg,et al.  Radial transport processes as a precursor to particle deposition in drinking water distribution systems. , 2011, Water research.

[40]  Pierre Geurts,et al.  Extremely randomized trees , 2006, Machine Learning.

[41]  Drainage Division,et al.  Criteria for Evaluation of Watershed Models , 1993 .

[42]  Joby Boxall,et al.  DISCOLOURATION RISK MANAGEMENT FOR TRUNK MAINS , 2011 .

[43]  J. Armstrong,et al.  Evaluating Forecasting Methods , 2001 .

[44]  J B Boxall,et al.  Asset deterioration and discolouration in water distribution systems. , 2011, Water research.