Over the last ten years, Multi-Objective Evolutionary Algorithms (MOEAs) have undergone tremendous development and have been successfully applied in various fields to solve complex problems that were formulated typically with two or three objectives. However, an increasingly expanding basis of our knowledge of natural processes and engineering requirements, and the need of taking into account sustainability indicators in order to compensate for the socio-environmental damage caused by the human activity, has increasingly led to formulation of the real world optimisation problems requiring a large number of management criteria.

Water industries within the United Kingdom and worldwide are becoming increasingly interested in the development and utilisation of Real Time Monitoring/Control (RT-M/C) of distribution and waste water systems, and flooding issues. This thesis describes the investigations undertaken by the author as part of the Pennine Water Group at the University of Sheffield, to develop a suite of RT-M/C algorithms, utilising quasi-steady state methodology for use within specific (problematic) wastewater system scenarios. The algorithms were developed by utilising, a novel laboratory scale (1:10) prototype research apparatus, in the form of a simple manhole junction.

Urban drainage networks contain a large number of structures; the most common of which is a manhole. Previous studies investigating solute transport through manholes identified a threshold level of surcharge at which the hydraulic conditions in the manholes sharply altered. At high levels of surcharge, a dead zone formed above the main jet, potentially retaining solute for later discharge. This phenomenon could have significant impacts on the results obtained from 1D urban drainage network models which only consider pure advection. Previous research (including other CFD based studies) concentrated on steady flow conditions through these structures which are not thought to be representative of the dynamic flow conditions experienced in practice. During large storm events, the capacity of combined sewer networks may be exceeded. This leads to the discharge of untreated sewage to receiving watercourses. Discharges such as these are governed by legislation and, therefore, an understanding of pollutant transfer through urban drainage networks is required. Using existing laboratory and Computational Fluid Dynamics (CFD) data, this study produced a primary and secondary validated methodology for meshing and modelling manholes using CFD. This was then extended to free-surface multiphase modelling allowing for more realistic unsteady conditions to be considered. This included an in-depth comparison of suitable turbulence models and solute-modelling options. It was found that, whilst it was possible to model multiphase free-surface flow within such a structure, it was unnecessarily computationally expensive as temporal changes in the flow field are rapid when compared with a practical time-step for use with 1D models. The validated modelling methodology was then used to recreate the original study that identified the hydraulic threshold. This included a selection of manholes with a range of inlet to manhole diameter ratios. However, this was extended further to include manholes with a ratio less than five (commonly found in UK sewer systems). It was shown that the hydraulic threshold broke down below this limit, removing the need for this to be modelled in 1D systems.

Suspended sediment, transported in urban stormwater sewers, is examined in order to determine its source, size, mineralogy, form and surface texture characteristics. The transport history is studied in relation to the hydrological parameters of rainfall and discharge in one catchment. The catchment is situated in North West London where field sampling was carried out over the period from March 1980 to December 1981. A Coulter Counter is used for particle size determinations; methods of sampling and the choice of dispersant and electrolyte are discussed. Particle surface texture analysis employs Scanning Electron Microscopy and preparation methods are discussed. Elemental composition is examined by energy dispersive x-ray analysis. Particle textures are described and quantified using a detailed surface area method and the Fuzzy Technique is employed in the analysis of a large number of particles. Sediment sources in the catchment include roads, buildings, open spaces and airborne material. Sediment is washed off land surfaces during rainfall and transported along the storm sewer to the outfall. Suspended sediment sampled at the outfall is commonly in the size range 1 to 40 mm and predominantly consists of quartz particles from roads tone erosion which have undergone considerable alteration by abrasion, silica precipitation and solution during drain transport. Storms and their sediment load fall into four groups : I. Intense rainfall of short duration generates moderately high total rainfall and discharge. Sediment comprises fresh-faced, angular, particles rapidly entrained from the land surface and of unimodal size distribution. II. Long periods of rainfall of moderate intensity create high rainfall totals and moderately high discharge. Drain deposited aggregates and surface particles are transported first; silica precipitates develop later, leading to aggregation as the discharge falls: size distributions are bimodal. III. Moderate rainfall and discharge transport sediment of similar characteristics to Group II but of moderated form. IV. Low rainfall and discharge for short period transports severly altered drain sediment of bimodal size distributions. Progressive sediment alteration along the storm sewer was simulated in a flume.

No description available.

Les travaux décrits dans le présent rapport concernent l’étude d’un ouvrage de traitement en ligne des eaux pluviales issues d’un bassin versant fortement urbanisé. L’ouvrage Charles Keller, d’une capacité de 7000 m3, est conçu pour traiter les eaux pluviales issues du bassin versant de Boudonville, situé dans la Communauté Urbaine du Grand Nancy. L’ouvrage est installé à l’exutoire de bassin versant en aval d’autres ouvrages de stockage temporaires. Les eaux traitées par l’ouvrage Charles Keller sont ensuite renvoyées vers la Meurthe et les boues produites sont traitées au niveau de la station d’épuration du Grand Nancy à Maxéville. La stratégie adoptée dans ce travail a permis de d’aborder l’ouvrage sous plusieurs angles : 1) d’abord l’étude de l’ouvrage seul en se focalisant sur l’ensemble de ses composantes unitaires (dessableurs, chambre d’injection des réactifs, réacteurs et décanteurs) ; 2) la partie physico-chimique (non opérationnelle jusqu’à présent) a fait l’objet de plusieurs compagnes d’analyse à travers les essais en Jar test effectués au laboratoire sur des eaux issues du bassin versant de Boudonville dont les caractéristiques couvrent toute la gamme que l’ouvrage Charles Keller peut traiter ; 3) ensuite l’ouvrage a été resitué dans l’ensemble du système de gestion des eaux pluviales (bassin versant - ouvrage Charles Keller - station d’épuration) et son fonctionnement simulé. Plusieurs configurations ont été testées pour reproduire les conditions réelles de fonctionnement de l’ouvrage. Les résultats obtenus ont montré d’abord la complexité de faire fonctionner dans la pratique un tel ouvrage, notamment avec sa partie coagulation/floculation mais a fait également ressortir des défauts de conception. / An online urban storm-water treatment system has been studied in this work. This system is able to treat the runoff from a highly impervious watershed (Boudonville) in Greater Nancy (North-East of France). It has a capacity of 7000 cubic meters. Some storage tanks are existing in the watershed. The treated water is discharged into the Meurthe River, while the sludge is treated in the Greater Nancy wastewater treatment plant in Maxéville.The strategy adopted in this work allowed for the consideration of the treat ment system from several angles:• first of all, the system is analyzed through its elementary components (sand removal unit, reagents injection and reactors for flocculation-coagulation and clarifiers): this has been done by observing the variations in water quality at the different treatment levels with online instrumentation. This has been completed with an offline characterization.• the flocculation-coagulation section, which was not yet operational during our work, was studied in the laboratory by jar tests. The water to be treated was sampled from the Boudonville watershed to be close to the conditions that should be observed in the Charles Keller treatment system.• finally the stormwater treatment was simulated as part of a full storm-water management system (watershed-Charles Keller treatment system-wastewater treatment plant).• The results that were obtained showed in the one hand the complexity of using such a treatment system in real-life conditions, especially with a coagulation / flocculation part to it, which has some conceptual issues.